using a GMC-300 for a project detector?


Advanced search

Message boards : Number crunching : using a GMC-300 for a project detector?

Author Message
apotheca
Send message
Joined: 7 Apr 12
Posts: 5
Credit: 243,321
RAC: 86

Message 1394 - Posted: 25 Oct 2012 | 2:05:22 UTC

I have a GMC-300 geiger counter from http://www.gqelectronicsllc.com/
is there a way to use this as my detector? it comes with USB to serial drivers that covert it to a com port on the PC. The software that comes with the detector reads the real time data coming from the detector much like the bonic project does. Is there a way for radioactive@home to connect to the detector?

Dagorath
Avatar
Send message
Joined: 4 Jul 11
Posts: 151
Credit: 42,738
RAC: 0

Message 1396 - Posted: 25 Oct 2012 | 5:54:00 UTC - in response to Message 1394.

It would be a miracle if your device presents the data the radioactive@home software expects when it reads the device so I am sure the project admins will say the answer is no. The software could probably be modified to read your device but I doubt anybody wants to take that job on for just 1 device. For 100 devices it would probably be worth it but there are potential hurdles that might make the job impossible. It depends what info the device's engineers/designers have decided to make available to the public and how readable the document is.

A better way to attempt to use the device with radioactive@home might be to massage the data it saves to your disk into the format this project expects and get BOINC to upload that data to R@H servers. It's a hack but there is a good chance it could work reliably. Again the trick is to find someone who is willing to do it for just 1 device. You might be able to convince the people who make/sell the device to do the work as they might be able to justify the project as a way to add value to their product and increase sales. Or they may laugh at the idea.

I've played with the R@H app a little and it's not terribly complicated. Also there appears to be a software simulator of the device available for free download. If the simulator creates a data file like the actual device does I'll have something to work with; if it does not then I might ask you to send me a copy of one of the data files from your disk. Perhaps I can find a way to make it work. I must add that if the project admins don't approve the results I won't release my work.

____________

Dagorath
Avatar
Send message
Joined: 4 Jul 11
Posts: 151
Credit: 42,738
RAC: 0

Message 1398 - Posted: 26 Oct 2012 | 4:55:15 UTC

I checked out the simulator software they provide and it's cool but it doesn't write a demo data file to disk like I thought it might. That doesn't put the hack I mentioned out of reach but after looking around their website and support forum it appears their customers are writing apps for their detectors. If so then there is probably enough info available to the public to write an app that will allow the GQ brand detectors to work with this project. I posted an invitation on their forum, let's see what becomes of it. It might be a way to bring many detectors into the project with minimal effort/expense. Any thoughts from R@H admins?

____________

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1404 - Posted: 26 Oct 2012 | 11:53:14 UTC - in response to Message 1398.

The easiest way to use this sensor would be to write a simple app/script just for this device and use it as anonymous platform, like optimized apps in other projects.
All you need is the data.bin present at exit and formatted trickle messages which are then sent back to the server (right now they're required, but if needed I could make them optional). I bet that 75% of the code could be just copied & pasted from our app sources.

apotheca
Send message
Joined: 7 Apr 12
Posts: 5
Credit: 243,321
RAC: 86

Message 1406 - Posted: 26 Oct 2012 | 19:50:49 UTC

I reading though my manual that I downloaded for the unit. It referred to the protocol on the download page so one can use it for their own projects. It is 8 items down on the download page on the web site. I tried to hyper terminal to my unit using the settings it states. but it did not respond to my command. I will try again this weekend when I have more time to play with it. That darn work thing is always getting in the way.

this is a little bit from that protocol document that might be used.

The serial port communication is based on a fixed baud rate.
Baud: 57600
Data bit: 8
Parity: None
Stop bit: 1
Control: None

3. Turn on the GQ GMC heartbeat

Note: This command enable the GQ GMC unit to send count per second data to host every second automatically.

Command: <HEARTBEAT1>>

Return: A 16 bit unsigned integer is returned every second automatically. Each data package consist of 2 bytes data from GQ GMC unit. The first byte is MSB byte data and second byte is LSB byte data.


e.g.: 10 1C the returned 1 second count is 28. Only lowest 14 bits are used for the valid data bit. The highest bit 15 and bit 14 are reserved data bits.

Dagorath
Avatar
Send message
Joined: 4 Jul 11
Posts: 151
Credit: 42,738
RAC: 0

Message 1411 - Posted: 27 Oct 2012 | 16:38:25 UTC

@TJM,

If source for GQ devices is available then yes the anonymous platform idea is the easiest.


@apotheca,

Thanks for the info. I'm not quite sure what to do with it yet or what it means but I read a little about the GQ devices every day, I'll get it eventually. I'm sure in time someone from the GQ forums will respond with more info or else I'll stumble across what I need. I am not about to buy a GMC-300 but I've been looking at the less expensive GMC-080 Geiger Counter Kit . I could justify that, maybe. What do you know about it? Is it just a stripped down version of the GMC-300? If it's just a GMC-300 without the LCD and it uses the same basic circuit then I could justify purchasing one and making it work with R@H but only if the code I develop will work with GMC-300 too.

____________

apotheca
Send message
Joined: 7 Apr 12
Posts: 5
Credit: 243,321
RAC: 86

Message 1418 - Posted: 29 Oct 2012 | 3:53:22 UTC

Tonight I had some time to play with the data output of the GMC-300. I was able to use hyperterminal to connect to the unit and start a data output with the command <HEARTBEAT1>>. It then gave an output every second as described in the communication protocol document. This is where my knowlege of what to do with ends. making an app/scrips for anonymous platform is not something I know how to do.

I didn't realize when I ordered the unit on e-bay that GQ electronics is in Seattle,WA. I live an hour south of them. I sometimes have to do service calls in Seattle. I might drive by their location to see if they have a store front when i'm up there.

@Dagorath
As far as the GMC-080 goes I think it is a sripped down version of the GMC-200. I noticed the main difference between the GMC-200 and GMC-300 besides the LCD screen is that the GMC-300 has a USB output only. The GMC-200 and GMC-080 is an audio jack output that plugs in to the mic port on your ipod. I did see a audio to USB cable listed as part of the GMC-200 package. Since both GMC-200 and GMC-300 use the same software, I would expect a scrip written for one will work on all models.

Dagorath
Avatar
Send message
Joined: 4 Jul 11
Posts: 151
Credit: 42,738
RAC: 0

Message 1419 - Posted: 29 Oct 2012 | 5:05:12 UTC - in response to Message 1418.

Yes, the GMC-080 and GMC-300 are quite different, I see it now after reading more of their docs. I'm convinced writing an app that would enable you to use it with R@H would be very easy if I had a GMC-300 to test with and verify the code but I can't justify the expense. I don't want to attempt it without having a device myself so I think we should talk someone else into doing it. For all we know that someone else could already be working on it. Let's give it a few more days and see what response we get from the message I posted in the GC forum. If nobody responds I'll contact two GC customers who have written apps for the device and see if I can talk one of them into it.

The other alternative is that me or perhaps TJM writes the app and gets you to test the code. That kind of arrangement can be troublesome, it's best if the author does the initial testing but if that turns out to be the only way then that's what we'll do if you're willing.

____________

Dagorath
Avatar
Send message
Joined: 4 Jul 11
Posts: 151
Credit: 42,738
RAC: 0

Message 1446 - Posted: 6 Nov 2012 | 6:24:19 UTC

apotheca, you have a PM
____________

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1449 - Posted: 6 Nov 2012 | 11:47:08 UTC - in response to Message 1406.


The serial port communication is based on a fixed baud rate.
Baud: 57600
Data bit: 8
Parity: None
Stop bit: 1
Control: None

3. Turn on the GQ GMC heartbeat



Can you run the hyperterminal (or anything else that can talk via COM port), send the heartbeat command and see what the hardware returns ?
If it only sends the data every second without anything else done from software side, then it will be easy to get the data with modified rad@home app.

Dagorath
Avatar
Send message
Joined: 4 Jul 11
Posts: 151
Credit: 42,738
RAC: 0

Message 1454 - Posted: 6 Nov 2012 | 22:21:54 UTC - in response to Message 1449.


The serial port communication is based on a fixed baud rate.
Baud: 57600
Data bit: 8
Parity: None
Stop bit: 1
Control: None

3. Turn on the GQ GMC heartbeat



Can you run the hyperterminal (or anything else that can talk via COM port), send the heartbeat command and see what the hardware returns ?
If it only sends the data every second without anything else done from software side, then it will be easy to get the data with modified rad@home app.


It appears there is an app for download at the GQ site that works nicely with the GMC-300 and can deliver data in various formats. I'm thinking the R@H app could be modified to detect which detector is installed and if it finds a GMC-300 then spawn the GQ app with whatever parameters are required to get output that mimics the standard output.

It appears the GMC-300 has a fair bit of RAM and can store a good chunk of data onboard. If so then the R@H app might lay dormant for X hours then read 2 hours worth of data from the device then terminate the task and report it.
____________

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1455 - Posted: 6 Nov 2012 | 23:28:48 UTC - in response to Message 1454.

The only way to support a device that requires drivers and a whole different communication approach is the app_info + dedicated app. The BOINC app has to work for everyone on all systems, not just fort those who have the drivers installed and that was one of the reason to pick up HID as the interface.
At some point the stable version of the app could be added as a second one with a switch in user prefs to choose which one to run.

Spawning another executable might look easy at first, but I highly doubt it's easily doable via BOINC considering all the data that API and server requires. IMO writing an app which would talk to the sensor via COM interface is the quickest way to get the data, I might give it a try but I'd like to see raw output from the COM port first.



Dagorath
Avatar
Send message
Joined: 4 Jul 11
Posts: 151
Credit: 42,738
RAC: 0

Message 1466 - Posted: 8 Nov 2012 | 16:56:14 UTC - in response to Message 1455.

The only way to support a device that requires drivers and a whole different communication approach is the app_info + dedicated app. The BOINC app has to work for everyone on all systems, not just fort those who have the drivers installed and that was one of the reason to pick up HID as the interface.
At some point the stable version of the app could be added as a second one with a switch in user prefs to choose which one to run.

Spawning another executable might look easy at first, but I highly doubt it's easily doable via BOINC considering all the data that API and server requires. IMO writing an app which would talk to the sensor via COM interface is the quickest way to get the data, I might give it a try but I'd like to see raw output from the COM port first.




I agree 100%... the best way is an app that uses the COM interface though I have some reservations due to some of the things I've heard about that interface. Sounds to me like they're doing something funny with that interface but I don't have any details to confirm that hunch. I requested details
at the GQ site and was advised to use their app instead. Maybe the details are not public domain, maybe he didn't understand my request, maybe the details are in plain sight but I missed them, I don't know.

Spawning their app would be my second or third choice because I too am aware of potential problems with that approach. In the absence of details about the COM interface I am willing to try spawning their app. I think it would require very little time to try it and if it works then it works. If not then I won't have wasted much time on it.
____________

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1467 - Posted: 8 Nov 2012 | 19:56:01 UTC - in response to Message 1466.

The protocol is nicely described on their website. Receiving data from COM port is relatively easy.

Spawning another process is much harder IMO, it might work in standalone mode but keep in mind that in order to use it with BOINC, you have to put all the files where they should be - executables, input and output files. Under normal conditions the BOINC client uses info from input/output templates so it knows what is placed where. If you change the app, the template is no longer correct and I don't know if it can be simulated with app_info.
Also, generally running another process under BOINC control requires the wrapper (or you won't be able to controll the process in any way).

Dagorath
Avatar
Send message
Joined: 4 Jul 11
Posts: 151
Credit: 42,738
RAC: 0

Message 1470 - Posted: 10 Nov 2012 | 12:29:25 UTC - in response to Message 1467.

The protocol is nicely described on their website. Receiving data from COM port is relatively easy.

Spawning another process is much harder IMO, it might work in standalone mode but keep in mind that in order to use it with BOINC, you have to put all the files where they should be - executables, input and output files. Under normal conditions the BOINC client uses info from input/output templates so it knows what is placed where. If you change the app, the template is no longer correct and I don't know if it can be simulated with app_info.
Also, generally running another process under BOINC control requires the wrapper (or you won't be able to controll the process in any way).


Doh! I don't know how I missed it, I must half been half asleep ;-) I always knew you were the right man for this job.

I don't know exactly how this will play out but there was mention of using the anonymous platform mechanism. That will work and might be the easiest way for you to make it work but GMC-300 users probably will have their hands full just installing BOINC and attaching to R@H. Remember they are totally new to the concept of DC. Asking them to setup an anonymous platform probably won't work unless you put all the commands into a script or provide perfectly clear and simple steps.

Wish I had a GMC-300 to test with, apotheca has disappeared? Maybe cajole one of the users at the GQ site into testing?

____________

miki_e
Send message
Joined: 28 Sep 12
Posts: 20
Credit: 225,510
RAC: 74

Message 1525 - Posted: 18 Dec 2012 | 13:22:48 UTC

Hi,

Any progress concerning the GMC-300 support?
I have just received one and I am ready to test it with boinc...

Some users would perhaps prefer bying this more sophisticated detector (standalone, battery operated, logging, etc.) and are still ready to leave it on connected to the boinc network when not using the device for other purposes....

This would increase the density of the detectors with minimal work on you side...

Anyway: you are doing an excellent job !!!

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1539 - Posted: 21 Dec 2012 | 9:23:33 UTC - in response to Message 1525.

If you're willing to spare some time and test the app, I might be able to add support for GMC-300 and possibly other sensors from the same family.
What I got now is a simple framework which continuously reads data from the COM port, without any processing. It still need some work (for example adding a configuration file), I think I'll build a test version next week.

miki_e
Send message
Joined: 28 Sep 12
Posts: 20
Credit: 225,510
RAC: 74

Message 1540 - Posted: 21 Dec 2012 | 13:47:20 UTC - in response to Message 1539.

Great!

Good news. Just drop me instructions what needs to be tested....

Nice Christmas and a Happy New year to the whole team!

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1546 - Posted: 29 Dec 2012 | 12:29:18 UTC - in response to Message 1540.

There will be a delay because the serial port communication works completely different way on Windows and Linux and I'd like the app to be portable if possible.
Once I get the basic read/write working the rest shouldn't take too long.


miki_e
Send message
Joined: 28 Sep 12
Posts: 20
Credit: 225,510
RAC: 74

Message 1638 - Posted: 18 Feb 2013 | 8:55:44 UTC

Hello,

50 days from the last message.... Any progress??

chillomatic
Send message
Joined: 8 Feb 13
Posts: 14
Credit: 0
RAC: 0
Message 1639 - Posted: 21 Feb 2013 | 12:50:02 UTC

I've ordernd a GMC-300 aswell. I Hope someday ist is possible to use ist with BOINC and R@H.
Best regards

chillomatic
Send message
Joined: 8 Feb 13
Posts: 14
Credit: 0
RAC: 0
Message 1652 - Posted: 2 Mar 2013 | 0:15:45 UTC

Hi!
I received my GMC-300 today and it seems to work correct.
I asked about the progress on implementing this device with the Radioactive@home project in BOINC at the ge electronics forum:
( http://www.gqelectronicsllc.com/forum/topic.asp?TOPIC_ID=3665 )
and got an answer and a download link.
"Yes. The new version data viewer Re.1.30 and Data Logger PRO Re5.25 have added Radioactive@Home file format."

so i got the tool and it saves the readings in realtime to a .R@H file.

My question is, how do i import it into BOINC and do my part in the project.

Have a nice weekend!

chillomatic
Send message
Joined: 8 Feb 13
Posts: 14
Credit: 0
RAC: 0
Message 1655 - Posted: 3 Mar 2013 | 20:19:59 UTC

Hi guys,
I exported 7h20m of realtime measurements to all available formats
(.R@H, .csv, .rbn, .htm, .html) for you to have a look at if you want.

Here is a download link.
http://dl.dropbox.com/u/2985968/7h%2012m%20GMC-300%20realtime%20data%20export.rar

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1684 - Posted: 7 Apr 2013 | 19:29:28 UTC - in response to Message 1655.

Sorry for the delay, I was quite busy during last weeks.

The 'rad@h' format output looks good.
The problem is that the project does not really use the file for anything, as all the real data is sent via trickles (almost in realtime), formatted by XML.
I think it would be possible to write a wrapper for the app, pick up the data file and report it to the server, however to do this, the app would have to support at least minimum control via command line (launch, stop after certain time).

I'll release my own "logger" soon, it should work with GMC-300.

Profile T-Armstrong
Send message
Joined: 23 Dec 11
Posts: 1
Credit: 0
RAC: 0
Message 1700 - Posted: 18 Apr 2013 | 20:23:27 UTC - in response to Message 1684.
Last modified: 18 Apr 2013 | 20:26:30 UTC

Hello

I want to buy a device that I can also participate in Radioactiv @ home. What is the device ? , and where to buy it in Germany?

Please give me the right Name for this device.

Thomas G
Founder of Astronomen.net

Profile Saenger
Send message
Joined: 16 Jun 11
Posts: 134
Credit: 376,083
RAC: 148

Message 1701 - Posted: 19 Apr 2013 | 4:58:17 UTC - in response to Message 1700.

Hallo Thomas!
Das ist ein selbstgebauter Geigerzähler der nur hier erhältlich ist. Und die Produktion läuft sehr schleppend.

Hi Thomas!
It's a self-made detector build by the project and only here available. The production is rather slow going.
____________
Gruesse vom Saenger

miki_e
Send message
Joined: 28 Sep 12
Posts: 20
Credit: 225,510
RAC: 74

Message 1702 - Posted: 19 Apr 2013 | 9:22:53 UTC

Hello,

Just to clear up the situation: the subject of this thread is the possible use of

GMC-300 ready made, standalone detector that can be purchased directly from the company manufacturing it:

http://www.gqelectronicsllc.com/comersus/store/comersus_viewItem.asp?idProduct=4542

Actually it cannot be used with the project, but TJM is working on the SW module to interface it and hopefully we will be able to test it soon....

If your are willing to buy rather the project's own special detector, please see the current status of the production and ordering in another thread...

Best regards,

miki_e

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1744 - Posted: 28 May 2013 | 10:40:00 UTC - in response to Message 1702.
Last modified: 28 May 2013 | 10:41:16 UTC

Could anyone who owns GMC-300 connect to it using HyperTerminal, send the commands http://www.gqelectronicsllc.com/download/GQ-RFC1201.txt one by one and capture the output ?
I'd like to see the real data sent by the sensor.
I do not have one, so I made a simple device simulator which in theory should be compatible with GMC-300. The problem is that it doesn't work with the software and I have to find out why.

I'd also like to see what's returned when you send a couple of <GETCPM>> commands at intervals of 1-2 minutes, even better if you could also write down what value was shown on the display. My first attempt to read the data uses heartbeat mode, but it's complicated. I think using GETCPM might be better way. If the sensor sends the current value, I can ask it every <n> minutes, then use the data to get the number of pulses (for compatibility with the server).

miki_e
Send message
Joined: 28 Sep 12
Posts: 20
Credit: 225,510
RAC: 74

Message 1772 - Posted: 6 Jun 2013 | 14:48:16 UTC - in response to Message 1744.

Hello,

...after such a long waiting I was almost not checking the messages any more...

I will try to log the coms in next days....

Best regards.

miki_e
Send message
Joined: 28 Sep 12
Posts: 20
Credit: 225,510
RAC: 74

Message 1776 - Posted: 12 Jun 2013 | 15:19:37 UTC - in response to Message 1744.

Hello,

Sorry, I was quite busy recently....

Here we have the coms:

Just to check communication: 2 times getversion:

TX IN ASCII: <GETVER>><GETVER>>
EQUALS TX IN HEXA: 3C 47 45 54 56 45 52 3E 3E 3C 47 45 54 56 45 52 3E 3E

device will reply:
RX ASCII: GMC-300Re 2.30GMC-300Re 2.30
=RX HEXA: 47 4D 43 2D 33 30 30 52 65 20 32 2E 33 30 47 4D 43 2D 33 30 30 52 65 20 32 2E 33 30


Command to get counts per minute repeated 4 times in interval of some 15 seconds

TX ASCII: <GETCPM>><GETCPM>><GETCPM>><GETCPM>>
TX HEXA: 3C 47 45 54 43 50 4D 3E 3E 3C 47 45 54 43 50 4D 3E 3E 3C 47 45 54 43 50 4D 3E 3E 3C 47 45 54 43 50 4D 3E 3E

device will reply with 2 (and only 2) bytes immediately after each command:
RX HEXA: 00 0E 00 0F 00 04 00 05

Looks to me that <getcpm>> is VERY easy to use...

If you need more logs, please let me know....

Best regards,

miki_e

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1783 - Posted: 16 Jun 2013 | 20:45:21 UTC - in response to Message 1776.

Thanks. I'm slowly making progress.
Few days ago I was able to talk to my 'simulator' via loopback interface, yet trying to communicate using another PC and a serial cable didn't work at all.

Today I've achieved another success, I was able to communicate with another PC and it responded to GMC-300 commands (I had the 'simulator' running on the other end).

I'll add a simple prefs parser to read COM port number from XML config file and then I'll release a simple test app, I'd like to see if it will be able to get response from a real GMC-300.


miki_e
Send message
Joined: 28 Sep 12
Posts: 20
Credit: 225,510
RAC: 74

Message 1789 - Posted: 21 Jun 2013 | 7:59:34 UTC - in response to Message 1783.

Hello TJM,

please check your private messages and comment...

Best regards,

miki_e

miki_e
Send message
Joined: 28 Sep 12
Posts: 20
Credit: 225,510
RAC: 74

Message 1797 - Posted: 25 Jun 2013 | 15:12:38 UTC

Hello TJM,

System is ready and running. Check you PM.

Best regards,

miki_e

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1798 - Posted: 25 Jun 2013 | 16:04:59 UTC - in response to Message 1797.
Last modified: 25 Jun 2013 | 17:23:09 UTC

The good news: app is able to talk to GMC-300 via COM port.

The bad news: I have to figure out why <GETCPM>> does not work for me.
I've just noticed that if the GMC was in heartbeat mode, it's still sending packets every second even if the app that requested heartbeat is closed.
This is something I did not expect and I think that's why my app does not get any values, in fact it gets stuck after doing the first read. This should be easy to fix just by sending <HEARTBEAT0>> as soon as the remote device is identified as sensor. Hopefully I'll get the <GETCPM>> working today.


17:55:24 (2740): Can't open init data file - running in standalone mode
Debug: Using port number: 3
COM3 opened for read/write
Debug: Port timeouts:
ReadTotalTimeoutConstant: 1000
ReadIntervalTimeout: 0
ReadTotalTimeoutMultiplier: 100
WriteTotalTimeoutMultiplier: 100
WriteTotalTimeoutConstant: 1000
<GETVER>> returned GMC-300Re 2.30€


EDIT: Now it seems to be working, I'm adding the 'BOINC stuff' to the app. Most of the code will be just copy&paste from the standard app with just minor tweaks.


<GETVER>> returned GMC-300Re 2.30
<GETCPM>> returned 0 19
<GETCPM>> returned 0 20
<GETCPM>> returned 0 22
<GETCPM>> returned 0 23
<GETCPM>> returned 0 21
<GETCPM>> returned 0 23
<GETCPM>> returned 0 24
<GETCPM>> returned 0 23


Chances are pretty high that I'll be able to return first chunk of data to the server today.

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1800 - Posted: 25 Jun 2013 | 19:51:03 UTC - in response to Message 1798.
Last modified: 25 Jun 2013 | 22:05:43 UTC

EDIT: So finally here is the first beta app which works with GMC-300:

http://www.enigmaathome.net/static/3468x3wmec/radioactiveathome.zip

How to use:

-attach to project

-wait until it downloads default app, set the project to no new work, abort task, update project and close BOINC manager

-go to the BOINC data folder, find radioactive@home folder - inside you'll find radac_1.77something.exe and sensors_1.77.xml. Delete both files and place the files from archive there (app_info.xml, gmc.xml, gmc300.exe).

-edit the port number inside gmc.xml. Other settings are ignored.

-restart the BOINC manager, the 'event log' should say that it found app_info.xml. Allow new work from the project & do update, it should get a task.


Known bugs and issues:

- The current version reads CPM every 4 minutes and recalculates the data to radioactive@home format (raw number of counts + timeline). In theory it works, but I think I'll change it to read 4 values at sample_time/4 and then I'll sum them together to get better accuracy/resolution.

- Stderr is chaotic :P Sorry about that but for now it's forced to print lots of debug info.

- Sensor ID in both trickle data and data.bin is missing.

- The values shown on graphs, map & tables are probably too low. That's because the server uses fixed divider to convert cpm->nSv/h (it assumes that the sensor uses SBM-20). This is going to be changed, the server will allow manual tube selection and "autodetection" based on sensor type. I can't say when it's going to be done, as it is a lot of work.
____________

miki_e
Send message
Joined: 28 Sep 12
Posts: 20
Credit: 225,510
RAC: 74

Message 1802 - Posted: 26 Jun 2013 | 13:39:41 UTC

Great news...

Would be nice if some other owners of GMC-300 (and similar models) try to install and run the application and post comments and observations...

Please note that you need to install USB driver from gqelectronicsllc.com site and find out the COM number assigned to GMC-300.... For the rest follow TJM's instructions....

....................

TJM, I will keep the system running as it is at least until next week... so if you need to do some more tests, please go ahead....

Perhaps you can also make an announcement of the new detector supported in "news" and/or on the homepage ?

miki_e


Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1803 - Posted: 26 Jun 2013 | 15:48:51 UTC - in response to Message 1802.

I think it's too early for news, as the app has lots of bugs :p I'm sure Murphy's laws would get me quickly at this stage.
However it would be nice to find more testers, especially someone who could check the app on Win Vista/7/8. I wrote the code from scratch, it uses Win API and as far as I know should support XP or newer (perhaps even 2k) but that's not verified yet.

I've just added a bit of code to extract the GQ GMC-300 revision and return it to the server as 2 bytes bcd data (for compatibility with server backend), so eventually it will be shown on graphs. I also tweaked the code that converts CPM to nSv/h, now I'm going to let it run for 2-3 hours to grab some data.

miki_e
Send message
Joined: 28 Sep 12
Posts: 20
Credit: 225,510
RAC: 74

Message 1804 - Posted: 26 Jun 2013 | 15:58:11 UTC - in response to Message 1803.

Hmm, you are right. We may need some more tests, specially with other OS... I can set up a test system with Win7 and eventually with w2k if needed. I do not have (and do not like) Vista and so far I have no experience with Win8...

JP_Austin
Send message
Joined: 27 Jun 13
Posts: 35
Credit: 56,630
RAC: 0

Message 1805 - Posted: 27 Jun 2013 | 3:44:20 UTC - in response to Message 1804.

Hello. I have a GQ CPM-300 and I was searching on hoiw to network the unit into a central mapping application just like this. I was going down the h**p://www.radiationnetwork.com/ path until I realized that the CPM-300 can not be configured for pulse output. It sends packets as you have been dealing with it here!

Anyway I am open to testing the new application software. I just downloaded the Bionic software and now setup to connect to the GQ device. I'll let you know how it goes...

JP_Austin
Send message
Joined: 27 Jun 13
Posts: 35
Credit: 56,630
RAC: 0

Message 1806 - Posted: 27 Jun 2013 | 4:13:09 UTC - in response to Message 1805.
Last modified: 27 Jun 2013 | 4:24:34 UTC

I have the Bionic running and I started the attached application but not sure it's doing anything. The command looking screen is just blank with no text messages shown. Is this how the application is supposed to work?

BTW running in Win7 here.

23:22:13 (3864): Can't open init data file - running in standalone mode
Debug: Using port number: 6
Radioactive@Home app rev $Rev: 585 $ for GMC-300 starting...
COM6 opened for read/write
Debug: Port timeouts:
ReadTotalTimeoutConstant: 1000
ReadIntervalTimeout: 0
ReadTotalTimeoutMultiplier: 100
WriteTotalTimeoutMultiplier: 100
WriteTotalTimeoutConstant: 1000
<GETVER>> returned GMC-300Re 2.36
<GETCPM>> returned 0 17
CPM value: 17
1000,1,2013-6-27 4:22:17,0,r,0


Unit was reporting about 17 CPM at the time! Not sure what else it needs but it may be working. I next need to check online and see if it's communicating with the server.

JP_Austin
Send message
Joined: 27 Jun 13
Posts: 35
Credit: 56,630
RAC: 0

Message 1807 - Posted: 27 Jun 2013 | 5:39:46 UTC - in response to Message 1806.
Last modified: 27 Jun 2013 | 5:41:32 UTC

Perhaps I need to backup but i could not find any radioactive@home folder in the Boinc folder. I manually placed the download above in there but did not see anything close to a file like stated. Do I need to uninstall the Boinc and start over?

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1809 - Posted: 27 Jun 2013 | 12:08:26 UTC - in response to Message 1807.
Last modified: 27 Jun 2013 | 12:41:59 UTC

You have to find where the BOINC stores it's data.
The easiest way to do it - restart the BOINC client, go to advanced view, search for 'event log' in menus and somewhere at the beginning of the log look for a line like this:

"2013-06-27 11:51:33 | | Data directory: I:\BOINC\data"

Inside you'll find a sub-folder 'projects' which stores all the apps and other files. Look for 'radioactiveathome.org_boinc', delete the files inside and replace them with the experimental app. This is what's inside the folder you have to find, and a second screen showing replaced files:

http://www.enigmaathome.net/static/3482xbvihm/files1.png
http://www.enigmaathome.net/static/3483x2c6ay/files2.png

EDIT: Everything has to be done with the BOINC manager/client stopped.
If there is no 'radioactiveathome.org_boinc' inside your 'projects' subfolder, you have to attach to the project first.

I know it's complicated process, but it's the only way to run experimental apps under BOINC control. At some point in future I'll add an application select field to the project preferences which will make the process much easier.

JP_Austin
Send message
Joined: 27 Jun 13
Posts: 35
Credit: 56,630
RAC: 0

Message 1812 - Posted: 27 Jun 2013 | 15:44:22 UTC
Last modified: 27 Jun 2013 | 15:49:25 UTC

Yes I found the data directroy was in another location within the PC's space under c:/Program Data. I did find the original files and removed them and replaced with the three from the download. I made sure the BOINC was shutdown at the time. I restarted the process and it shows the BOINC starting along with the BOINC manager and more important the GMC300 application was launched this time. I was later asked to update my location on the map. Perhaps it will start to work the next time it checks in? I'm getting closer.

Now I need to return the other software! Saved my $65.00!

EDIT - Just checked the map. I'm live and showing up here in Central Texas!

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1813 - Posted: 27 Jun 2013 | 17:27:16 UTC - in response to Message 1812.

Very nice, so now we know that the app works on Win 7.

I'm going to add some more code so the server will be able to identify sensor type, later it will be used to select different CPM->uSv dividers. The sensor type will be also included in our data exports.

JP_Austin
Send message
Joined: 27 Jun 13
Posts: 35
Credit: 56,630
RAC: 0

Message 1816 - Posted: 28 Jun 2013 | 3:54:41 UTC
Last modified: 28 Jun 2013 | 3:59:24 UTC

Yes it's working well. I do have some questions related to access. If my network goes down for periods of time how does the system react? Does it just store the results for later transmission when the network is available again? My system may power off from time to time and wanted to know how that was handled.

Yes it will be nice if you can report the sensor type. What else would it take to make the whole process easier to configure? I sort of hacked and slashed my way to the finish line but it should be easier once this is out of alpha software stage!

BTW - thank you for writing the code and the effort so far. It's appreciated. This counter was just sitting here only recording background levels now it's actually useful in a much larger way.

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1820 - Posted: 28 Jun 2013 | 9:44:20 UTC - in response to Message 1816.
Last modified: 28 Jun 2013 | 9:51:38 UTC

Due to the limitations of BOINC client / server software, when the network goes down the client continues until the end of current workunit, the data is stored and sent to the server as soon as the network goes back up. The client can't get new WU if the network is offline. You can tweak the WU lenght in project preferences, just do not set it too high because sometimes it can cause problems.

When the host goes off, no data is returned to the server. I think that the GMC-300 can store a large number of samples, but our sensor and app were not designed for offline operation and the server will not accept any 'out of sequence' data, especially without timestamps.

Both apps (for our USB sensor and the GMC-300) do identify sensors and they report the sensor type to the server. However, currently the server does not know what to do with it. I might be able to add some functionality over the weekend.

In theory the server is able to send user-selected app, and then it is 'installed' automatically. Just like the default app.
This requires major changes in the server code - customizing the scheduler and probably feeder, adding second data return path (even if it will be just copy&paste) and who knows what else.
Right now I can't say when it will be done, I will probably have to study lots of docs before trying and my free time is limited.

Btw, the 'anonymous platform' apps, like the experimental one for GMC-300, aren't automatically updated, so it is a good idea to check every now and then if there is a new version available. Updating is easy, XMLs are already in correct location so only the executable has to be overwritten.

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1821 - Posted: 28 Jun 2013 | 12:38:25 UTC - in response to Message 1820.
Last modified: 28 Jun 2013 | 12:40:37 UTC

I patched the server to read the divider from database, based on the sensor ID. Of course this won't handle any 'custom' sensors with swapped tubes, but at least it supports one divider per sensor type. Now I need to know what's the correct divider for GMC-300.



EDIT: it requires the 'rev 5' app, I'll upload it later.
____________

JP_Austin
Send message
Joined: 27 Jun 13
Posts: 35
Credit: 56,630
RAC: 0

Message 1825 - Posted: 29 Jun 2013 | 5:54:21 UTC - in response to Message 1821.

Thank you for the effort. I really is helpful.

JP_Austin
Send message
Joined: 27 Jun 13
Posts: 35
Credit: 56,630
RAC: 0

Message 1829 - Posted: 29 Jun 2013 | 15:47:38 UTC - in response to Message 1825.

Not sure what happened but my dot on the map went away? Did the change of the server now require the new application as well? Was working well for the last couple of days but this morning looks like I dropped off. I will check my side but appears to be working with updated counts and all. I will try the familiar reboot in Windows but the BOINC seems to be happy. No red boxes.

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1830 - Posted: 29 Jun 2013 | 16:24:43 UTC - in response to Message 1829.

The dot is there, I have no idea why it was gone.

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1832 - Posted: 30 Jun 2013 | 15:33:46 UTC - in response to Message 1830.

Here is the Rev 6 app which add/fixes a few things:

- For stderr, it uses divider 111 instead of 171 (I read on the GQ forums that 111 is the standard value for 0-1000CPM while using default M4011 tube).
- Fixes sensor ID, returns sensor type and it's software revision to the server
- Shorter (3min) sample time

JP_Austin
Send message
Joined: 27 Jun 13
Posts: 35
Credit: 56,630
RAC: 0

Message 1833 - Posted: 30 Jun 2013 | 19:24:55 UTC

Downloaded and applied. Will see if it reports the sensor type now. Thanks.

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1834 - Posted: 30 Jun 2013 | 21:23:11 UTC - in response to Message 1833.

Not sure if you noticed, but currently the app can't see your sensor and as far as I can see, the problem started before upgrade:
http://radioactiveathome.org/boinc/result.php?resultid=2012989

It also helped me to find a bug, the app 'wait time' is far too low, it is supposed to wait for like ~6 hours if the sensor is not connected.

JP_Austin
Send message
Joined: 27 Jun 13
Posts: 35
Credit: 56,630
RAC: 0

Message 1835 - Posted: 1 Jul 2013 | 6:18:56 UTC
Last modified: 1 Jul 2013 | 6:20:40 UTC

Yes something went badly and the system did not respond for half a day or more. I restarted and seems to be running now. Reports the proper model number and version of software. I like it!

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1836 - Posted: 1 Jul 2013 | 14:10:13 UTC

Yep, also the graphs shows that the server reads conversion value from the database, because the average reported value jumped up.

JP_Austin
Send message
Joined: 27 Jun 13
Posts: 35
Credit: 56,630
RAC: 0

Message 1837 - Posted: 2 Jul 2013 | 2:09:12 UTC
Last modified: 2 Jul 2013 | 2:48:21 UTC

Think it's slightly high as I have granite counters in the office and the sensor is about a foot away from the counter top. I have noted the count goes down just a few CPM if I move it out into the room. Will have to be where it is as the wife has proclaimed it will not be out in the middle of the room! She who must be obeyed wins that one!

BTW - My cable was disconnected to the sensor for a few hours without my knowledge. I did get an automated email asking to check the connection. HOw do i see this status on the BOINC site or can I?

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1838 - Posted: 2 Jul 2013 | 14:19:20 UTC - in response to Message 1837.
Last modified: 2 Jul 2013 | 16:43:08 UTC

There is no 'status' because the connection between client(+sensor) and server is not 100% realtime.
The server checks the status of the returned workunit and compares it with the previous one.
If the sensor was seen earlier and after the next workunit it's gone, the automated mail is sent.
I think there is also a limit of 1 mail per day.

JP_Austin
Send message
Joined: 27 Jun 13
Posts: 35
Credit: 56,630
RAC: 0

Message 1839 - Posted: 4 Jul 2013 | 3:10:32 UTC

Something is funny. I let the thing operate all day long and it shows that I did a lot of tasks but the dot is not showing up. Has not been on the map for a couple days now. I uninstalled BOINC and reinstalled with the admin only having the ability to run the BOINC and manager. I have a couple other accounts on this machine and BOINC was attempting to start a new session for each account as it was logged into. After the install the other accounts do not try to auto start the manager now.

Could it be something with the install and having to reset something or permissions? What is odd is the thing appears like its working but just not on the map.

miki_e
Send message
Joined: 28 Sep 12
Posts: 20
Credit: 225,510
RAC: 74

Message 1840 - Posted: 4 Jul 2013 | 8:14:10 UTC

Hi TJM,

I am about to stop the GMC-300 system and move the detector to the new location... Is it OK for you or do you need to make some more tests? Eventually I could leave the system running until Monday....

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1841 - Posted: 4 Jul 2013 | 16:25:48 UTC - in response to Message 1840.

It's ok, right now I don't have much spare time anyway and hopefully I'll be able to tweak the app a bit without access to the sensor.
Perhaps I'll even buy one in the nearest future, as I'm interested in porting the app to Linux (standard COM + maybe Raspberry specific version).


____________

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1842 - Posted: 4 Jul 2013 | 16:28:35 UTC - in response to Message 1839.
Last modified: 4 Jul 2013 | 16:28:56 UTC

Something is funny. I let the thing operate all day long and it shows that I did a lot of tasks but the dot is not showing up. Has not been on the map for a couple days now. I uninstalled BOINC and reinstalled with the admin only having the ability to run the BOINC and manager. I have a couple other accounts on this machine and BOINC was attempting to start a new session for each account as it was logged into. After the install the other accounts do not try to auto start the manager now.

Could it be something with the install and having to reset something or permissions? What is odd is the thing appears like its working but just not on the map.


The BOINC server provides stderr output from all the tasks that were completed. Here are yours: http://radioactiveathome.org/boinc/results.php?hostid=7898

It looks like the app has serious problems accessing the sensor for some reason.

JP_Austin
Send message
Joined: 27 Jun 13
Posts: 35
Credit: 56,630
RAC: 0

Message 1843 - Posted: 5 Jul 2013 | 1:43:55 UTC - in response to Message 1842.

Yeah I ended up nearly starting from scratch. I had to remove the old account and started a new one. Only then it started to read the sensor properly again. I think there was some odd stuff going on when I moved the BOINC on my machine from "every account can start it" to "only the admin can start it". I also started the BOINC with admin privileges. Not sure if that is all of it but it's working now!

JP_Austin
Send message
Joined: 27 Jun 13
Posts: 35
Credit: 56,630
RAC: 0

Message 1862 - Posted: 7 Jul 2013 | 18:05:52 UTC - in response to Message 1843.

Says it's completed a number of tasks properly yet never shows up on the map? I let it run 24 hours and yet no dot. Very frustrating. Don't believe it's on my end.

The new ID after starting over is http://radioactiveathome.org/boinc/results.php?hostid=8011

I do see this this type of entry at the start of Boinc in the stdoutae log file:
06-Jul-2013 20:11:02 [---] No config file found - using defaults
06-Jul-2013 20:11:02 [---] Starting BOINC client version 7.0.64 for windows_x86_64
06-Jul-2013 20:11:02 [---] log flags: file_xfer, sched_ops, task
06-Jul-2013 20:11:02 [---] Libraries: libcurl/7.25.0 OpenSSL/1.0.1 zlib/1.2.6
06-Jul-2013 20:11:02 [---] Data directory: C:\ProgramData\BOINC
06-Jul-2013 20:11:02 [---] Running under account xxxxxxxxxxxxxxx
06-Jul-2013 20:11:02 [---] Processor: 4 AuthenticAMD AMD Phenom(tm) 9500 Quad-Core Processor [Family 16 Model 2 Stepping 2]
06-Jul-2013 20:11:02 [---] Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 htt pni cx16 popcnt syscall nx lm svm sse4a osvw ibs page1gb rdtscp 3dnowext 3dnow
06-Jul-2013 20:11:02 [---] OS: Microsoft Windows 7: Ultimate x64 Edition, Service Pack 1, (06.01.7601.00)
06-Jul-2013 20:11:02 [---] Memory: 7.75 GB physical, 15.50 GB virtual
06-Jul-2013 20:11:02 [---] Disk: 688.87 GB total, 221.09 GB free
06-Jul-2013 20:11:02 [---] Local time is UTC -5 hours
06-Jul-2013 20:11:02 [---] CAL: ATI GPU 0: ATI Radeon HD 2300/2400/3200/4200 (RV610) (CAL version 1.4.1385, 256MB, 224MB available, 80 GFLOPS peak)
06-Jul-2013 20:11:02 [Radioactive@Home] Found app_info.xml; using anonymous platform
06-Jul-2013 20:11:02 [Radioactive@Home] URL http://radioactiveathome.org/boinc/; Computer ID 8011; resource share 100
06-Jul-2013 20:11:02 [---] No general preferences found - using defaults
06-Jul-2013 20:11:02 [---] Reading preferences override file
06-Jul-2013 20:11:02 [---] Preferences:
06-Jul-2013 20:11:02 [---] max memory usage when active: 3967.59MB
06-Jul-2013 20:11:02 [---] max memory usage when idle: 7141.66MB
06-Jul-2013 20:11:02 [---] max disk usage: 220.99GB
06-Jul-2013 20:11:02 [---] don't use GPU while active
06-Jul-2013 20:11:02 [---] suspend work if non-BOINC CPU load exceeds 25 %
06-Jul-2013 20:11:02 [---] (to change preferences, visit a project web site or select Preferences in the Manager)
06-Jul-2013 20:11:02 [---] Not using a proxy
Initialization completed
06-Jul-2013 20:11:02 [---] Suspending GPU computation - computer is in use
06-Jul-2013 20:11:02 [Radioactive@Home] Restarting task sample_2026615_0 using radac version 170 in slot 0
06-Jul-2013 20:16:02 [---] Resuming GPU computation
06-Jul-2013 20:18:10 [---] Suspending GPU computation - computer is in use
06-Jul-2013 20:21:11 [---] Resuming GPU computation
06-Jul-2013 20:41:34 [Radioactive@Home] Computation for task sample_2026615_0 finished
06-Jul-2013 20:41:36 [Radioactive@Home] Started upload of sample_2026615_0_0
06-Jul-2013 20:41:38 [Radioactive@Home] Finished upload of sample_2026615_0_0
06-Jul-2013 20:41:38 [Radioactive@Home] Sending scheduler request: To fetch work.
06-Jul-2013 20:41:38 [Radioactive@Home] Reporting 1 completed tasks
06-Jul-2013 20:41:38 [Radioactive@Home] Requesting new tasks for CPU
06-Jul-2013 20:41:40 [Radioactive@Home] Scheduler request completed: got 1 new tasks
06-Jul-2013 20:41:42 [Radioactive@Home] Starting task sample_2026822_0 using radac version 170 in slot 0
06-Jul-2013 21:12:13 [Radioactive@Home] Computation for task sample_2026822_0 finished
06-Jul-2013 21:12:15 [Radioactive@Home] Started upload of sample_2026822_0_0
06-Jul-2013 21:12:16 [Radioactive@Home] Finished upload of sample_2026822_0_0
06-Jul-2013 21:12:20 [Radioactive@Home] Sending scheduler request: To fetch work.
06-Jul-2013 21:12:20 [Radioactive@Home] Reporting 1 completed tasks


Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1863 - Posted: 7 Jul 2013 | 18:39:54 UTC - in response to Message 1862.

Host is shown on map if it returned at least a couple of samples during the last 24 hours.

The tasks are completed (validated) even if there is no sensor attached*, so please do not use 'completion' to verify if things are working. Just click on the result number to see stderr out. Also, 0 credits mean something went wrong.

This is your host's result page, navigate to second or third page and you'll see when the app stopped talking to the sensor.

http://radioactiveathome.org/boinc/results.php?hostid=8011

Perhaps it's a bug in the app, but I can't find any obvious reason why would it suddenly stop communicating.

JP_Austin
Send message
Joined: 27 Jun 13
Posts: 35
Credit: 56,630
RAC: 0

Message 1865 - Posted: 8 Jul 2013 | 5:22:26 UTC

Yes this is odd. When it is not reporting the percentage is always stuck at 0% completed even when the tasks say they are completing. I closed BOINC down and started the GMC-300 application to make sure the unit was still reporting data. That worked just fine. I then started up BOINC again and asked it to reset the project. Started up but I noticed an odd few readings

Debug: Using port number: 6
Radioactive@Home app rev $Rev: 585 $ for GMC-300 starting...
COM6 opened for read/write
Debug: Port timeouts:
ReadTotalTimeoutConstant: 1000
ReadIntervalTimeout: 0
ReadTotalTimeoutMultiplier: 100
WriteTotalTimeoutMultiplier: 100
WriteTotalTimeoutConstant: 1000
<GETVER>> returned GMC-300Re 2.36
<GETCPM>> returned -64 0
CPM value: -16384
1000,4294966204,2013-7-8 4:57:11,0,f,0
<GETCPM>> returned 0 13
CPM value: 13
242000,4294966256,2013-7-8 5:1:12,0,n,0
<GETCPM>> returned 0 17
CPM value: 17
483000,4294966324,2013-7-8 5:5:13,0,n,0
<GETCPM>> returned 0 17
CPM value: 17
724000,4294966392,2013-7-8 5:9:14,0,n,0
Debug: trickle sent
Trickle sent


Is it possible if you read negative values that the program ignores them or tosses them out as invalid? I see that at times when the GMC-300 is first starting up communication. I hooked a Hyperterminal to the port and watched the data stream out. Seems like the unit will sometimes report garbage for the first query of <GETCPM>. Is it possible this is what is messing up the BOINC?

miki_e
Send message
Joined: 28 Sep 12
Posts: 20
Credit: 225,510
RAC: 74

Message 1867 - Posted: 9 Jul 2013 | 8:21:00 UTC
Last modified: 9 Jul 2013 | 8:21:20 UTC

Hi TJM,

I have moved the GMC sensor to the new location and I had some problems to establish communication between the sensor and your application. Successful communication does not show in boinc log and I had to wait until the task was finished to see the result and the debug report.......>>>....and it was not working...

This morning I aborted the task (which generated a nice crash report in the debug log). I verified that the GMC was turned on, communicating correctly with GQ's SW and requested new work in boinc... It is possible that the sensor has been left in the heartbeat mode when I ended the GQ's SW... (????!!!!)

Now I am checking the report and everything seems to work.

Please see http://radioactiveathome.org/boinc/results.php?hostid=8070

Suggestions:
- maybe you should verify that com is OK (reasonable reply from GMC) and reset com and retry (5x) if not...
- successful com with GMC should be logged in boinc log as well as event of communication lost... (if possible)
- the result should not be marked validated if there was NO sensor connected...

miki_e

P.S

-- if you need I can reconnect the sensor to the test system again for your trials....
-- I can also put you online raspbery PI with GMC sensor (still the same one) if you are interested....

JP_Austin
Send message
Joined: 27 Jun 13
Posts: 35
Credit: 56,630
RAC: 0

Message 1875 - Posted: 10 Jul 2013 | 16:08:11 UTC
Last modified: 10 Jul 2013 | 16:30:28 UTC

Agree things are still a bit unstable. I have noticed that if my machine is reset the BOINC manager needs to startup with the proper permissions or it will not start getting new jobs. Not sure that is your code but something to mention. I killed it and restarted under the admin privileged and it properly starts the manager and the tasks (your code) is properly launched.

Once it starts this is where we need a bit of improvement. The GMC-300 is likely in an unknown state. Do you put it into the <HEARTBEAT0> and then get the version?
I had an instance where the version was not returned properly. See below:

Debug: Using port number: 6
Radioactive@Home app rev $Rev: 585 $ for GMC-300 starting...
COM6 opened for read/write
Debug: Port timeouts:
ReadTotalTimeoutConstant: 1000
ReadIntervalTimeout: 0
ReadTotalTimeoutMultiplier: 100
WriteTotalTimeoutMultiplier: 100
WriteTotalTimeoutConstant: 1000
<GETVER>> returned À
<GETVER>> returned string which is too long
<GETCPM>> returned 51 54
CPM value: 13110
1000,437,2013-7-10 15:45:18,0,f,0

The count was then all messed up as well. Perhaps a delay to allow the GMC-300 to process the command to go into the heartbeat0 mode? There may be a serial fifo buffer that needs to be emptied. I believe the GMC-300 uses a third party RS-232 to USB chip so it may have some overhead processing. I'd try a delay between commands.

The next one is related. See the version properly returned this time but I think it was already put into HEARTBEAT0 from the first attempt. But there is still a bit of a timing issue or something with the first GETCPM command. See this follow up request:

Debug: Using port number: 6
Radioactive@Home app rev $Rev: 585 $ for GMC-300 starting...
COM6 opened for read/write
Debug: Port timeouts:
ReadTotalTimeoutConstant: 1000
ReadIntervalTimeout: 0
ReadTotalTimeoutMultiplier: 100
WriteTotalTimeoutMultiplier: 100
WriteTotalTimeoutConstant: 1000
<GETVER>> returned GMC-300Re 2.36
<GETCPM>> returned -64 1
CPM value: -16383

1000,4294966477,2013-7-10 15:46:30,0,r,0
<GETCPM>> returned 0 23
CPM value: 23
242000,4294966569,2013-7-10 15:50:31,0,n,0


The follow on request returned valid data. Just that first one was a bit odd. I feel your really close with the code but may just need a few tweeks for the unique timing of this sensor.

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1877 - Posted: 10 Jul 2013 | 16:46:40 UTC - in response to Message 1875.
Last modified: 10 Jul 2013 | 16:55:35 UTC

I'm patching the app right now and hopefully I'll release new version today. It fixes a couple of known bugs.

The app sends <HEARTBEAT0>> after reading version from the hardware.
Does the GMC-300 boot up with hearbeat enabled ?

Any 'heartbeat' packet sent while the app is looking for version may cause problems. I did it this way (read version then talk to hardware) to avoid sending anything to unknown device, but since the user has to set the port manually anyway, I'll change it.

There is a 1 second delay between write -> read commands.


It is also possible that in nearest future we'll have an installer which will swap, update and perhaps even configure the app automatically.

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1878 - Posted: 10 Jul 2013 | 21:39:52 UTC - in response to Message 1877.

Please try this:

http://www.enigmaathome.net/static/3500xfd5qk/gmc300.exe

just run it as standalone first to see if it is able to ID sensor and get the data. I added more delays, moved the hearbeat0 and (hopefully) fixed a bug in get_cpm.

JP_Austin
Send message
Joined: 27 Jun 13
Posts: 35
Credit: 56,630
RAC: 0

Message 1880 - Posted: 11 Jul 2013 | 2:49:52 UTC - in response to Message 1878.
Last modified: 11 Jul 2013 | 3:03:44 UTC

Radioactive@Home app rev $Rev: 585 $ for GMC-300 starting...
COM6 opened for read/write
Connected to GMC-300 Re 2.36 (revision int: 566)
2000,1,2013-7-11 2:45:18,r,0.0 minutes,18.0 cpm,0.27 uSv/h
184000,79,2013-7-11 2:48:20,n,3.0 minutes,26.0 cpm,0.15 uSv/h
366000,130,2013-7-11 2:51:22,n,3.0 minutes,17.0 cpm,0.10 uSv/h
548000,175,2013-7-11 2:54:24,n,3.0 minutes,15.0 cpm,0.09 uSv/h
730000,235,2013-7-11 2:57:26,n,3.0 minutes,20.0 cpm,0.12 uSv/h
912000,313,2013-7-11 3:0:28,n,3.0 minutes,26.0 cpm,0.15 uSv/h



Looks much better so far! I need to play with it a bit more by re-booting machine and cycling power on the GMC-300. It essentially ran right off the start no problem. Is the first reading correct on the calculation of the uSv/h?

BTW - The GMC-300 defaults with <HEARTBEAT1> or constantly spitting out data all the time. If you did a version check before the HEARTBEAT0 you likely got lucky reading the proper data back. I would first issue the heaartbeat0 then delay a bit and read the version and then launch into the read CPM code.

Overall a big improvement!

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1881 - Posted: 11 Jul 2013 | 4:58:59 UTC - in response to Message 1880.

The first reading has very short sample time, so the value is not always correct. It's marked with 'f' and ignored by all non-averaging functions on the server.

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 1886 - Posted: 13 Jul 2013 | 3:44:07 UTC

Very interesting thread.

Appears there is currently no Linux support?

I'd be willing to purchase one of these and help test. Could boinc interface with a perl script in place of the .exe being used above?

I'm 100% newb with boinc programming and any api's or project interfaces for r@h but know my way around linux.

PM me info and if it sounds possible or something you're interested in working together on I'll get the GMC-300 ordered and on the way!

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1887 - Posted: 13 Jul 2013 | 9:08:57 UTC - in response to Message 1886.

Yep, currently the app is for Win only, but the Linux version is planned. To port the app to Linux only the COM I/O has to be rewritten (as the Win version uses functionality provided by Win API and Visual C++), the rest is already compatible with gcc.

I have no idea if BOINC could run perl script, but I guess it would be rather hard to do because the BOINC API is C++ and the app uses it a lot.

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 1888 - Posted: 13 Jul 2013 | 14:50:04 UTC - in response to Message 1887.

I see they do have a linux version of the software that provides the com access to the gmc-300. Will you have time to work on the gcc portion? I can test whatever you have and recompile tweeks to get it working.

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 1889 - Posted: 13 Jul 2013 | 14:53:32 UTC - in response to Message 1887.

You also mentioned a raspberry version. I don't think this will work on rpi because the device is powered from the USB from what I see and the rpi doesn't have powered usb ports. I do have a rpi though if you'd like to test on that also I can setup and get you shell access to whatever os you like to run on it.

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1890 - Posted: 13 Jul 2013 | 15:28:20 UTC - in response to Message 1889.

Our 'native' USB sensor is powered from USB, requires around 50-60mA of current at 5V and has no problem working with Raspberry.

I think the GMC-300 uses external power supply, but I'm not sure.

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 1891 - Posted: 13 Jul 2013 | 15:34:05 UTC - in response to Message 1890.

Just placed the order should have one on hand next week.

JP_Austin
Send message
Joined: 27 Jun 13
Posts: 35
Credit: 56,630
RAC: 0

Message 1892 - Posted: 13 Jul 2013 | 15:58:52 UTC

The GMC-300 uses USB to power the device but actually that just charges the internal Lipo battery. It can run stand-alone for about 12 hours or more to act like a logger. Handy if you want to slip it into your luggage for a long flight. you can log the higher CPM readings while at high elevation.

The BeagleBone Black and I assume a Rpi can both use a powered hub if you need more USB ports. My son recently got a BB Black and added a powered hub to it. Not a problem as he was running a keyboard, mouse, USB Audio, and think he still had an extra port. They can plug directly into your router and would make a terrific monitoring node.

miki_e
Send message
Joined: 28 Sep 12
Posts: 20
Credit: 225,510
RAC: 74

Message 1896 - Posted: 15 Jul 2013 | 8:57:17 UTC

There still seems to be some problem to establish communication with GMC-300 after system restart.

1) connection was lost (for unknown reason...PC problem and reboot ???)

<GETCPM>> returned 0 20
CPM value: 20
4098000,1153,2013-7-13 22:55:19,0,n,0
<GETCPM>> returned 0 8
CPM value: 8
4339000,1185,2013-7-13 22:59:20,0,n,0
Debug: trickle sent
Trickle sent
<GETCPM>> returned 0 15
CPM value: 15
4580000,1245,2013-7-13 23:3:21,0,n,0
Debug: Using port number: 5
Radioactive@Home app rev $Rev: 585 $ for GMC-300 starting...
Found previously created output file, workunit almost completed - exiting
23:10:03 (596): called boinc_finish

</stderr_txt>
]]>

..... after this the comm with GMC-300 was not working:

<core_client_version>7.0.28</core_client_version>
<![CDATA[
<stderr_txt>
Debug: Using port number: 5
Radioactive@Home app rev $Rev: 585 $ for GMC-300 starting...
COM5 opened for read/write
Debug: Port timeouts:
ReadTotalTimeoutConstant: 1000
ReadIntervalTimeout: 0
ReadTotalTimeoutMultiplier: 100
WriteTotalTimeoutMultiplier: 100
WriteTotalTimeoutConstant: 1000
gmc_GetVersion(): read failed
Unable to open port COM523:40:24 (180): called boinc_finish

</stderr_txt>
]]>

Please see log of computer 8070.

2) I installed the latest version of the GMC300.exe (as posted), restarted the system hoping that everything would be OK.

3) After some time (needed to check the workunit result on the server) I noticed that communication was not working. I checked the communication with GQ's communication SW and everything was OK.

4) Quit the GQ's SW, restart boinc and everything is OK. Looks like that R@H have still some problem to initialize the communication, while GQ's SW is successful.

This evening I will try to spy and log all communication between the GMC-300 and GQ's SW. It will perhaps help us to understand how to initialize communication in the bullet-proof way....

miki_e
Send message
Joined: 28 Sep 12
Posts: 20
Credit: 225,510
RAC: 74

Message 1897 - Posted: 15 Jul 2013 | 9:05:36 UTC

Just a precision concerning the GMC-300 power:

- there is no separate P/S plug
- USB connection is used in quite usual way to:

--- communication AND
--- power supply the unit permanently when connected AND
--- recharge the integrated LiPo battery

It is using a standard USB port, thus the power taken should not exceed 500mA...

I have not tried, but normally RPi should not have any problem to P/S this unit (even directly from the native RPi port... ) ...

miki_e
Send message
Joined: 28 Sep 12
Posts: 20
Credit: 225,510
RAC: 74

Message 1902 - Posted: 17 Jul 2013 | 14:56:36 UTC

TJM, here are communication logs between GQ's SW (real time monitoring) and GMC-300 as well as log between Boinc and GMC-300. Data is logged in hexa.

If you need some more data, just let me know.

https://www.dropbox.com/s/tl54ttjlyngx6ss/boinc-capture.LOG
https://www.dropbox.com/s/1pevdaa1maqu3y5/gmc-sw-capture.LOG

miki_e

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1907 - Posted: 18 Jul 2013 | 20:01:50 UTC - in response to Message 1902.

That's interesting, the original app sends the data byte after byte, while my app sends the whole command at once. I think it does not matter, as the sensor handles that without problems, I think it is transparently handled by the COM port at hardware or driver level.

I also see that the original app sets the COM speed and parameters after opening the port. My app does that on open (at least according to the MS docs, because I haven't seen any evidence). It sets 57600 8/n/1 with no flow control.

I'll have to study the stderr outputs to guess where the app fails, so far I've seen the communication failure directly after or during GETCPM, and then for some reason the app was unable to talk to the sensor again.

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 1908 - Posted: 19 Jul 2013 | 3:00:50 UTC

GMC-300 in hand.

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1911 - Posted: 20 Jul 2013 | 17:20:38 UTC - in response to Message 1908.
Last modified: 20 Jul 2013 | 17:22:37 UTC

I'm studying possibilities for 'easy' Linux port.
I think I'll try using termios.h - it provides all the necessary functions to read/write from serial port.
Perhaps I'll be able to build first Linux version tomorrow.

JP_Austin
Send message
Joined: 27 Jun 13
Posts: 35
Credit: 56,630
RAC: 0

Message 1912 - Posted: 21 Jul 2013 | 4:15:06 UTC

Yes I have initialization issues at times. If I used the GMC-300 data viewer first then shut that application down and launch the BOINC then things work properly. Once up and running it's solid. Been running for several days without problems. Only the initial sequence that is a bit unstable.

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1913 - Posted: 21 Jul 2013 | 10:22:05 UTC - in response to Message 1912.
Last modified: 21 Jul 2013 | 10:24:28 UTC

Are you using the latest app ?

I think one of the bugs that caused the communication failures is fixed already, another one will be fixed soon.
I did various communication tests with my device simulator, using laptop connected to COM port to simulate GMC-300.
When for some reason the communication fails, chances are there is something left in the COM port buffer and it might be read next time the app polls the device.

Any excess data read by the app at startup will cause communication failure. I'll try to find a way to check if there is something in the buffer before sending commands. A quick fix would be reading one char at a time in a loop until read fails just to be sure that there is nothing left in the buffer. Perhaps the Windows API provides better solution.

One thing I don't understand is why the app was unable to find the sensor. The first work unit after read failure may not work due to the bug described above. However on the next retry it should work again, unless there is another bug/glitch.

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 1914 - Posted: 21 Jul 2013 | 23:51:45 UTC

Attempting to connect to the unit. I have revision 2.38 and the linux version of gqgmc says:

"The command to read the version number of the firmware failed."

Also while reading through the forums for more input I came across this

http://www.gqelectronicsllc.com/forum/topic.asp?TOPIC_ID=3762

Where it is suggested taking the battery out if constantly plugged in which I assume most here would be looking to do on the project.

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1918 - Posted: 22 Jul 2013 | 19:54:09 UTC - in response to Message 1914.
Last modified: 22 Jul 2013 | 20:05:44 UTC

I didn't have enough time to finish the Linux app during the weekend, mid week I don't have much free time so the progress is slow.

I also fixed a couple more bugs in the Windows app, I think it did not set all the COM parameters correctly.

Here is the latest snapshot, but please keep the older version somewhere just in case (this version is not fully tested yet). It throws away any unexpected data at startup, at the cost of slight delay.

http://www.enigmaathome.net/static/3523x05wwu/gmc300.exe

JP_Austin
Send message
Joined: 27 Jun 13
Posts: 35
Credit: 56,630
RAC: 0

Message 1920 - Posted: 23 Jul 2013 | 4:34:53 UTC - in response to Message 1918.

Seems to be a bit better. I rebooted the PC and besides the normal BOINC issues needing to start with admin rights it ran right off. In the past it would read the serial number a bit strange first time up. This time just worked without having to prime the pump with the GMC-300 software first!

Radioactive@Home app rev $Rev: 179 $ for GMC-300 starting...
COM6 opened for read/write
Connected to GMC-300 Re 2.36 (revision int: 566)
2000,1,2013-7-23 4:28:33,f,0.0 minutes,12.0 cpm,0.27 uSv/h

JP_Austin
Send message
Joined: 27 Jun 13
Posts: 35
Credit: 56,630
RAC: 0

Message 1924 - Posted: 27 Jul 2013 | 1:34:06 UTC

Only concern is that the GMC-300 is reading slightly higher than many of the other units out in the field. My CPM numbers are not all that different but the BOINC graphs show a elevated level compared to similar CPM counts. Not sure if there is some algorithm used to calculate the uSv/h numbers but it may need to be updated.

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1925 - Posted: 28 Jul 2013 | 17:57:48 UTC - in response to Message 1924.

The server uses fixed divider 111, I think it's the correct value for MD-4011. If it's not, it can be very easily tweaked. Just a single field in database.


I'm still working on the Linux app. Basically I have the same issue I had previously on Windows. When using software emulated com pair, the app seems to be working. It writes/reads with no problems at all.
The problem starts when talking to real hardware. The remote end does not receive a single byte on write() and the app gets stuck on read(). I think it's a problem with COM settings, most probably flow control.


JP_Austin
Send message
Joined: 27 Jun 13
Posts: 35
Credit: 56,630
RAC: 0

Message 1926 - Posted: 28 Jul 2013 | 19:31:45 UTC
Last modified: 28 Jul 2013 | 19:32:48 UTC

As I understand it the GMC-300 does not use any flow control. Only standard RS-232 type communications with 8 bits, 1 stop bit, no parity or flow control. Flow control typically is established with a special cable that correctly routes the CTS/RTS pins in a typical serial interface. With the prolific RS-232 to USB chip doing the interface were a bit fixed in how they wired up the unit.

Yes i just looked at the GMC app and it shows flow control turned off for interfacing to the unit.

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1927 - Posted: 28 Jul 2013 | 19:46:39 UTC - in response to Message 1926.
Last modified: 28 Jul 2013 | 20:05:14 UTC

Yep, I know, but still the app did not like talking to the hardware at all.
However after some tweaking I finally managed to talk to my device emulator and it seems to be working:


Radioactive@Home app rev $Rev: 191 $ for GMC-300 starting...
COM10 opened for read/write
<GETVER>> returned GMC-300Re 2.36
Connected to GMC-300 Re 2.36 (revision int: 566)
<GETCPM>> returned 10 57
4000,436,2013-7-28 19:42:57,n,0.1 minutes,2617.0 cpm,58.86 uSv/h


Does anyone feel brave enough to test the Linux version ? I'm sure it has lots of bugs :P

http://www.enigmaathome.net/static/3535xwuuib/gmc300.tgz

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 1928 - Posted: 29 Jul 2013 | 2:47:27 UTC

Would the process be similar to running the windows version?

Connect to project, delete files, extract files then run?

I'll test linux version first thing tomorrow!

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 1929 - Posted: 29 Jul 2013 | 3:05:05 UTC - in response to Message 1927.

What port number would be used?

/dev/ttyUSB0?

22:56:34 (9790): Can't open init data file - running in standalone mode
Radioactive@Home app rev $Rev: 178 $ for GMC-300 starting...
COM10 opened for read/write
gmc_GetVersion(): failed to send <HEARTBEAT0>>, write error or no device on the other end
gmc_GetVersion(): failed to send <GETVER>>, write error or no device on the other end
gmc_GetVersion(): read failed
(revision int: 0)
22:58:52 (10053): Can't open init data file - running in standalone mode
gmc_ReadPortNumber(): Wrong port number
22:58:52 (10053): called boinc_finish

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1930 - Posted: 29 Jul 2013 | 4:45:54 UTC - in response to Message 1929.
Last modified: 29 Jul 2013 | 5:00:32 UTC

Just a number, 0-99.

EDIT: Well I found a bug in the app, it won't open the port because it's looking for /dev/ttyS<x>. I'll fix that later today.

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 1931 - Posted: 29 Jul 2013 | 16:16:43 UTC - in response to Message 1930.

ln -s /dev/ttyUSB0 /dev/ttyS99

Set <portnumber>99</portnumber>

10:29:44 (31111): Can't open init data file - running in standalone mode
Radioactive@Home app rev $Rev: 178 $ for GMC-300 starting...
COM99 opened for read/write
<GETVER>> returned GMC-300Re 2.38
Connected to GMC-300 Re 2.38 (revision int: 568)
<GETCPM>> returned 0 17
3000,2,2013-7-29 14:29:53,r,0.1 minutes,17.0 cpm,0.36 uSv/h
<GETCPM>> returned 0 15
187000,48,2013-7-29 14:32:57,n,3.1 minutes,15.0 cpm,0.09 uSv/h
<GETCPM>> returned 0 21
371000,112,2013-7-29 14:36:1,n,3.1 minutes,21.0 cpm,0.12 uSv/h
<GETCPM>> returned 0 16
555000,161,2013-7-29 14:39:5,n,3.1 minutes,16.0 cpm,0.09 uSv/h
<GETCPM>> returned 0 17
739000,213,2013-7-29 14:42:9,n,3.1 minutes,17.0 cpm,0.10 uSv/h

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1932 - Posted: 29 Jul 2013 | 16:21:26 UTC - in response to Message 1931.
Last modified: 29 Jul 2013 | 16:21:41 UTC

I've updated the app, it now connects to correct port:


http://www.enigmaathome.net/static/3537xqafzk/gmc300.tgz

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 1935 - Posted: 29 Jul 2013 | 17:50:08 UTC - in response to Message 1934.

I set portnumber=0

13:49:34 (15997): Can't open init data file - running in standalone mode
gmc_ReadPortNumber(): Wrong port number
13:49:34 (15997): called boinc_finish

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 1933 - Posted: 29 Jul 2013 | 17:51:25 UTC - in response to Message 1932.
Last modified: 29 Jul 2013 | 18:19:29 UTC

The new version will not connect to my symbolic link /dev/ttyS99 and doesn't respond well to 0 it says COM0 in the output.

Using the first version with /dev/ttyS99 symbolic link when I run it manually i see output. Running from BOINC no output.

I've got everything setup on BOINC similar to windows by changing the app_info.xml to the correct binary names.

29-Jul-2013 12:56:03 [Radioactive@Home] Found app_info.xml; using anonymous platform
29-Jul-2013 12:51:48 [Radioactive@Home] work fetch resumed by user
29-Jul-2013 12:51:50 [Radioactive@Home] Sending scheduler request: To fetch work.
29-Jul-2013 12:51:50 [Radioactive@Home] Requesting new tasks for CPU
29-Jul-2013 12:51:53 [Radioactive@Home] Scheduler request completed: got 1 new tasks
29-Jul-2013 12:51:55 [Radioactive@Home] Starting task sample_2121707_0 using radac version 170

I see the binary running...

boinc 11835 0.0 0.0 1304 604 ? SNl 13:04 0:00 ../../projects/radioactiveathome.org_boinc/gmc300 read

But don't see any new output in stderr or data.bin

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 1936 - Posted: 29 Jul 2013 | 17:52:43 UTC - in response to Message 1934.

Set to portnumber=0 and received

13:49:34 (15997): Can't open init data file - running in standalone mode
gmc_ReadPortNumber(): Wrong port number
13:49:34 (15997): called boinc_finish

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 1937 - Posted: 29 Jul 2013 | 17:54:37 UTC

Set to portnumber=0 and received

13:49:34 (15997): Can't open init data file - running in standalone mode
gmc_ReadPortNumber(): Wrong port number
13:49:34 (15997): called boinc_finish

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1934 - Posted: 29 Jul 2013 | 18:30:03 UTC - in response to Message 1933.

The app said it can't open COM<x> because the device name was hardcoded, but on Linux it actually used /dev/ttyUSB<x>.
I defined separate device name for Linux, but it's just a cosmetic change.

http://www.enigmaathome.net/static/3539xpvgh9/gmc300.tgz

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 1938 - Posted: 29 Jul 2013 | 19:14:23 UTC

Unsure why our times are out of order here but I used the 3rd version provided and see the results:

Set to portnumber=0 and received

13:49:34 (15997): Can't open init data file - running in standalone mode
gmc_ReadPortNumber(): Wrong port number
13:49:34 (15997): called boinc_finish

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1939 - Posted: 29 Jul 2013 | 19:26:00 UTC - in response to Message 1938.

Can I see your gmc.xml ?

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 1940 - Posted: 29 Jul 2013 | 19:36:13 UTC

<gmc>
<comsettings>
<comment>Edit port number below</comment>
<portnumber>0</portnumber>
<comment>On Windows, the app tries to open COM<x> where <x> is the specified port number</comment>
<comment>On Linux, the app tries to open /dev/ttyUSB<x> where <x> is the specified port number</comment>
<baud>57600</baud>
<bits>8</bits>
<parity>n</parity>
<stopbits>1</stopbits>
</comsettings>
<options>
</options>
</gmc>

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 1941 - Posted: 29 Jul 2013 | 19:38:44 UTC
Last modified: 29 Jul 2013 | 19:40:34 UTC

There has been 1 work unit returned to the project that says validated.

http://radioactiveathome.org/boinc/workunit.php?wuid=2121707
http://radioactiveathome.org/boinc/result.php?resultid=2140993

But don't see any output in the stderr.txt or data.bin files that are in the r@h project folder.

This is using the first version with the symbolic link to ttyS99.

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1942 - Posted: 29 Jul 2013 | 20:19:58 UTC - in response to Message 1941.

stderr.txt and data.bin are auto deleted after upload to the server is done (the 'slot\<x>' folder is cleaned on exit).

The completed workunit shows that the app is at least partially working, 12 samples were successfully validated.

JP_Austin
Send message
Joined: 27 Jun 13
Posts: 35
Credit: 56,630
RAC: 0

Message 1943 - Posted: 30 Jul 2013 | 1:51:30 UTC

Good job guys. This will enable a slough of embedded processors to be used. Thinking of using the BeagleBone Black as a standalone processor to free up my machine now running Boinc.

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 1944 - Posted: 31 Jul 2013 | 1:02:33 UTC
Last modified: 31 Jul 2013 | 1:25:35 UTC

It seems that not every time when the binary runs a connection is made to the device. I found the output in slots/4

/var/lib/boinc-client/slots/4# tail -f stderr.txt
Debug: num_samples: 10
Debug: Using port number: 99
Radioactive@Home app rev $Rev: 178 $ for GMC-300 starting...
Debug: num_samples: 10
Debug: Using port number: 99
Radioactive@Home app rev $Rev: 178 $ for GMC-300 starting...
COM99 opened for read/write
Debug: num_samples: 10
Debug: Using port number: 99
Radioactive@Home app rev $Rev: 178 $ for GMC-300 starting...
COM99 opened for read/write
<GETVER>> returned GMC-300Re 2.38
Connected to GMC-300 Re 2.38 (revision int: 568)
<GETCPM>> returned 0 18
3000,2,2013-7-31 1:22:30,568,f,GMC-300


I "kill <ps>" and it will just say starting. I then open minicom and connect to the port. Then kill <ps> again and it connected to the port. It's as if it may need some sort of clearing/reset before attempting connection?

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 1945 - Posted: 31 Jul 2013 | 2:00:18 UTC

Example: This task started after some intervention as mentioned above:

http://radioactiveathome.org/boinc/result.php?resultid=2146027

Once it completed the next task started but gets stuck:

Debug: num_samples: 10
Debug: Using port number: 99
Radioactive@Home app rev $Rev: 178 $ for GMC-300 starting...


I then run minicom which is configured to connect to the port and this prints

COM99 opened for read/write


Still sticks and nothing happens

I perform the following:

#ps aux | grep radio
boinc 9561 0.2 0.0 1304 640 ? SNl 21:54 0:00 ../../projectsradioactiveathome.org_boinc/gmc300 read
root 9951 0.0 0.0 23032 924 pts/4 R+ 21:58 0:00 grep --color=auto radio
# kill 9561


As soon as I 'kill 9561' the output of stderr.txt shows:

Debug: num_samples: 10
Debug: Using port number: 99
Radioactive@Home app rev $Rev: 178 $ for GMC-300 starting...
COM99 opened for read/write
<GETVER>> returned GMC-300Re 2.38
Connected to GMC-300 Re 2.38 (revision int: 568)
<GETCPM>> returned 0 21
4000,3,2013-7-31 1:58:39,568,f,GMC-300


And continues until the task finishes.

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 1946 - Posted: 31 Jul 2013 | 18:02:09 UTC

From the gqgmc.cc code example he uses a clearUSB function and maybe that is what is needed here and is why the device needs to be reinitialized through minicom before it works again.

// clearUSB is public method to clear the read (input) buffer of
// the serial port. This method exists due to vagaries of the
// turn_on_cps_cmd. Since there is no protocol for the returned
// data (ie, no start, no stop, no ack, no nak), the synchronization
// of the getAutoCPS() method with the GQ GMC is flakey. Consequently,
// when the turn_off_cps_cmd happens, there may be left over
// data in the USB input buffer. So this method simply reads the
// USB input buffer until it is empty.
void
GQGMC::clearUSB()
{
uint32_t rcvd(0);
char inp;

// Assume that there isn't that much left over from any previous
// read. In other words, we couldn't ever get that far off.
// So read only 10 bytes.
const
uint16_t kMaxtries(10);

// Read returned data until input buffer is empty as indicated by
// rcvd = 0 or not. If rcvd = 0 then the buffer empty and that is
// what we want.
for(uint16_t i=0; i<kMaxtries; i++)
{
rcvd = read(usb_serial, &inp, 1);
if (rcvd == 0) break;
} // end for

// Not good, there are still more characters in input buffer,
// so declare error.
if (rcvd > 0)
error_code = eClear_USB;

return;
}// end clearUSB()

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 1947 - Posted: 31 Jul 2013 | 18:09:04 UTC

Also when I open minicom it instantly gives me the version number without typing <GETVER>>. So it's as if that is still in the buffer but not being passed to the gmc300 binary?

Any possibility of viewing the source and build instructions on the gmc300 so I can hack on it through the week?

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 1951 - Posted: 5 Aug 2013 | 16:55:01 UTC

Having heard nothing I was eager to pass back results. I put together a script that when run in cron will detect a new task, init the port, restart the apps and keep results running all day using the first version of the linux code provided here and by creating a symbolic link to ttyS99 as referenced above.



#!/bin/bash
BOINC_PATH="/var/lib/boinc-client/slots"

SLOT=`grep -rH radio $BOINC_PATH/* | egrep -o 's/[0-9]{1,2}/i'`
arrSLOT=(${SLOT//// })
SLOT=${arrSLOT[1]}

RAH_OUT="$BOINC_PATH/$SLOT/stderr.txt"
tail -n 1 $RAH_OUT | grep -q 'starting...'
if [ $? == 0 ]
then
screen -dmS boincclearport /dev/ttyS99 57600
sleep 1
kill $(ps aux | grep '[b]oincclearport' | awk '{print $2}')
kill $(ps aux | grep '[g]mc300 read' | awk '{print $2}')
fi


Which has keep the jobs running without pause.

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1952 - Posted: 6 Aug 2013 | 20:25:03 UTC - in response to Message 1951.

Unfortunately last week my time was very limited.

I don't understand why the app can't connect to the sensor each time it is started. Perhaps that's because it does not restore the COM setting to previous values on exit, but then the issue should be easily repeatable, not random.

Another thing I don't understand is why the second/third version does not work with ttyUSB0 directly. On my test system it works with any device name.

I'll try to build a version with lots of debug info tomorrow, perhaps it will help to find what's wrong.
I'll share the sources somewhere after a quick clean up.


foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 1953 - Posted: 6 Aug 2013 | 20:41:46 UTC - in response to Message 1952.
Last modified: 6 Aug 2013 | 20:42:07 UTC

Sounds good. I think as stated in that code above about the clearusb. The device has an odd way of sending the data and it's not clearing out properly so I think if before you attempt the connection you perform something similar to that gqgmc function that would probably get this working as expected.

When I open minicom or screen when the gmc300 binary is stuck it immediately displays the version number without <GETVER>>.

JP_Austin
Send message
Joined: 27 Jun 13
Posts: 35
Credit: 56,630
RAC: 0

Message 1956 - Posted: 7 Aug 2013 | 3:25:28 UTC
Last modified: 7 Aug 2013 | 3:29:07 UTC

Ditto have had initialization issues on the PC where the first read from the unit is garbled. If I kill it and restart it a second time it normally starts just fine. For some reason my PC got rebooted today and the app did not start up properly. Been offline for most of the day. Just now got it communicating again.

Would a search of the expected string in a limited loop be OK? Perhaps poll the port a time or two looking for the correct response. Once it's OK then move on. If not good after four or five attempts give up. Either that or properly clear out the receive buffer as indicated earlier.

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1957 - Posted: 7 Aug 2013 | 19:58:07 UTC - in response to Message 1956.
Last modified: 7 Aug 2013 | 19:58:57 UTC

The latest Win app snapshot issues <HEARTBEAT0> on startup, waits 2s and then reads the COM buffer and throws the data away without checking what was there. After doing that it resumes normal operation.

I *think* the Linux app does the same with tcflush().

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1958 - Posted: 7 Aug 2013 | 20:42:22 UTC - in response to Message 1957.

foovax can you try this one ? http://www.enigmaathome.net/static/3545xbxq0x/gmc.tgz I added lots of debug info around COM Open.

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 1959 - Posted: 7 Aug 2013 | 22:07:39 UTC - in response to Message 1958.

18:06:40 (30187): Can't open init data file - running in standalone mode
gmc_ReadPortNumber(): Wrong port number
18:06:40 (30187): called boinc_finish


<gmc>
<comsettings>
<comment>Edit port number below</comment>
<portnumber>0</portnumber>
<comment>On Windows, the app tries to open COM<x> where <x> is the specified port number</comment>
<comment>On Linux, the app tries to open /dev/ttyUSB<x> where <x> is the specified port number</comment>
<baud>57600</baud>
<bits>8</bits>
<parity>n</parity>
<stopbits>1</stopbits>
</comsettings>
<options>
</options>
</gmc>

geox
Send message
Joined: 31 Jul 13
Posts: 5
Credit: 1,124
RAC: 0
Message 1961 - Posted: 8 Aug 2013 | 13:14:47 UTC

I'm using a GMC-300 since a week but I can't understand the graphics: other sensor graphics all over the world seems to be not far from the daily average value but the graph of GMC-300 stays always above 0.22 while the daily average is never more than 0.16uSv/h. Why happen this? can someone explain what are the the blue triangles? peaks or the average of each sample?

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 1962 - Posted: 8 Aug 2013 | 13:39:04 UTC - in response to Message 1961.

I'm running the same in Linux and have the same results on the chart with ~.1 higher than average.

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 1963 - Posted: 8 Aug 2013 | 13:41:03 UTC - in response to Message 1925.

The server uses fixed divider 111, I think it's the correct value for MD-4011. If it's not, it can be very easily tweaked. Just a single field in database.


I think once the app's are finished and move from beta then calibration can fix the charts based on this comment above in this thread.

geox
Send message
Joined: 31 Jul 13
Posts: 5
Credit: 1,124
RAC: 0
Message 1964 - Posted: 8 Aug 2013 | 18:12:57 UTC

The server uses fixed divider 111, I think it's the correct value for MD-4011. If it's not, it can be very easily tweaked. Just a single field in database.




I think once the app's are finished and move from beta then calibration can fix the charts based on this comment above in this thread.


Maybe but I think is a little less than 0.1uSv/h higher... maybe is around 0.07 higher.

the last sample and 24 hours average values are correct in my opinion as they match the values of the GMC-300 stored data.

in this image is possible to see that the 24 hours average is 0.16uSv/h and the chart giving about 0.07uSv/h more

http://imageshack.us/photo/my-images/196/54lh.png/

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1965 - Posted: 8 Aug 2013 | 18:19:37 UTC - in response to Message 1964.

I'll take a look at this, if the sensor is detected as GMC-300, the server should know which divider to use.

And here is the 'rev 188' Linux app, I think it fixes the 'wrong port number' bug. It's also the first revision which shares code with Win app (both app now use the same sources).
http://www.enigmaathome.net/static/3546xp7j4g/gmc300.tgz

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 1967 - Posted: 8 Aug 2013 | 22:47:28 UTC - in response to Message 1965.

Radioactive@Home app rev $Rev: 178 $ for GMC-300 starting...
Trying to open: /dev/ttyUSB0
Open retval: 7
tcgetattr() returned 0
tcsetattr() returned 0
tcflush() returned 0
/dev/ttyUSB0 opened for read/write
<GETVER>> returned GMC-300Re 2.38
Connected to GMC-300 Re 2.38 (revision int: 568)
<GETCPM>> returned 0 22
4000,3,2013-8-8 22:45:55,f,0.1 minutes,22.0 cpm,0.41 uSv/h


Port issue appears resolved. Running in boinc now with interval set to .5 and my keep alive script disabled. Will see if it can end 1 work unit and proceed with the next work unit.

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 1968 - Posted: 8 Aug 2013 | 23:22:46 UTC - in response to Message 1967.

Radioactive@Home app rev $Rev: 178 $ for GMC-300 starting...
Trying to open: /dev/ttyUSB0
Open retval: 7
tcgetattr() returned 0
tcsetattr() returned 0
tcflush() returned 0
/dev/ttyUSB0 opened for read/write
<GETVER>> returned GMC-300Re 2.38
Connected to GMC-300 Re 2.38 (revision int: 568)
<GETCPM>> returned 0 22
4000,3,2013-8-8 22:45:55,f,0.1 minutes,22.0 cpm,0.41 uSv/h
<GETCPM>> returned 0 16
188000,52,2013-8-8 22:48:59,n,3.1 minutes,16.0 cpm,0.09 uSv/h
<GETCPM>> returned 0 15
372000,98,2013-8-8 22:52:3,n,3.1 minutes,15.0 cpm,0.09 uSv/h
<GETCPM>> returned 0 20
556000,159,2013-8-8 22:55:7,n,3.1 minutes,20.0 cpm,0.12 uSv/h
<GETCPM>> returned 0 16
740000,208,2013-8-8 22:58:11,n,3.1 minutes,16.0 cpm,0.09 uSv/h
<GETCPM>> returned 0 21
924000,272,2013-8-8 23:1:15,n,3.1 minutes,21.0 cpm,0.12 uSv/h
<GETCPM>> returned 0 13
1108000,311,2013-8-8 23:4:19,n,3.1 minutes,13.0 cpm,0.07 uSv/h
<GETCPM>> returned 0 20
1292000,372,2013-8-8 23:7:23,n,3.1 minutes,20.0 cpm,0.12 uSv/h
Trickle sent
<GETCPM>> returned 0 11
1476000,405,2013-8-8 23:10:27,n,3.1 minutes,11.0 cpm,0.06 uSv/h
<GETCPM>> returned 0 25
1660000,481,2013-8-8 23:13:31,n,3.1 minutes,25.0 cpm,0.14 uSv/h
<GETCPM>> returned 0 17
1844000,533,2013-8-8 23:16:35,n,3.1 minutes,17.0 cpm,0.10 uSv/h
Trickle sent
Done - calling boinc_finish()
19:16:35 (14240): called boinc_finish


First job ran to completion as expected. Second didn't start:

Radioactive@Home app rev $Rev: 178 $ for GMC-300 starting...
Trying to open: /dev/ttyUSB0

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 1976 - Posted: 20 Aug 2013 | 18:59:24 UTC - in response to Message 1968.

This is a problem with COM port settings and I'm trying to figure out what's wrong.
Port works fine until it is closed, close() returns success, everything looks fine from the operating system point of view but then the port can't be opened again and any app that tries to access it gets stuck.
It's not a problem with USB->COM, because it also affects standard serial ports, so it must be something I do to init the port. Without any kind of init it works every time, but then the app won't talk to the sensor (COM settings mismatch).

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 1977 - Posted: 20 Aug 2013 | 23:19:25 UTC - in response to Message 1976.

I posted the above a few weeks ago and it appears to reference this issue?

// clearUSB is public method to clear the read (input) buffer of
// the serial port. This method exists due to vagaries of the
// turn_on_cps_cmd. Since there is no protocol for the returned
// data (ie, no start, no stop, no ack, no nak), the synchronization
// of the getAutoCPS() method with the GQ GMC is flakey. Consequently,
// when the turn_off_cps_cmd happens, there may be left over
// data in the USB input buffer. So this method simply reads the
// USB input buffer until it is empty.
void
GQGMC::clearUSB()
{
uint32_t rcvd(0);
char inp;

// Assume that there isn't that much left over from any previous
// read. In other words, we couldn't ever get that far off.
// So read only 10 bytes.
const
uint16_t kMaxtries(10);

// Read returned data until input buffer is empty as indicated by
// rcvd = 0 or not. If rcvd = 0 then the buffer empty and that is
// what we want.
for(uint16_t i=0; i<kMaxtries; i++)
{
rcvd = read(usb_serial, &inp, 1);
if (rcvd == 0) break;
} // end for

// Not good, there are still more characters in input buffer,
// so declare error.
if (rcvd > 0)
error_code = eClear_USB;

return;
}// end clearUSB()



The above code comes from the gqgmc linux driver code that is linked to from the official website for the gmc-300. I'd try implementing this function in and calling it before opening the ports.

JP_Austin
Send message
Joined: 27 Jun 13
Posts: 35
Credit: 56,630
RAC: 0

Message 1983 - Posted: 29 Aug 2013 | 4:18:12 UTC

Any time to update this? I can get the preset code to work but will make rolling it out to others a but more problematic as it does have start-up issues at times.

Has anyone got the Beagle-Bone Black working with this unit?

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 1985 - Posted: 29 Aug 2013 | 17:01:14 UTC - in response to Message 1983.

I have no access to source code. I've presented the code above that comes from the gmc-300 linux driver code to assist with the startup problems but no implementation as of yet.

I run the bash script which is also posted above and that keeps the work going on my machine.

Profile wd9dui
Send message
Joined: 3 Feb 13
Posts: 8
Credit: 0
RAC: 0
Message 1995 - Posted: 2 Sep 2013 | 18:16:01 UTC

I have ordered a GMC-300, the web site says that they are on back order, maybe 10 September? This will replace my old C6979 Geiger kit that sits and blinks all day…
Gary
____________

JP_Austin
Send message
Joined: 27 Jun 13
Posts: 35
Credit: 56,630
RAC: 0

Message 1997 - Posted: 2 Sep 2013 | 23:20:01 UTC - in response to Message 1995.

They are decent detectors. For this application where it's being plugged in all the time I unplugged the battery. It's a 9 volt rechargeable so it has a clip like a standard 9 volt. This way you won't overcharge the battery and possibly damage it with extended overcharge. This was reported as a potential problem in the GMC forums.


The only problems I have had lately is when I re-boot the PC. Sometimes the initialization of the serial port does not occur properly. I have got good at finding the processes and killing them and restarting! Worst case I have to start the GMC application and verify I can communicate with the unit prior to starting BOINC. TO be honest I have not had to do that in a while since it starts up OK about half the time. I do run the app under admin rights since I setup the app to only allow the admin to run it. I have a couple other people on this machine and I really didn't want them messing with the app or trying to start it.

Let us know how it works out once you get one.

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 2005 - Posted: 5 Sep 2013 | 21:21:07 UTC - in response to Message 1997.

I think I have fixed the Linux 'port hang' issue.
At this moment I'm unable to test the app with any kind of COM port hardware, because my Linux/dev netbook spends most of the time at my workplace and I only have remote console access available.
The app does not hang the port anymore - I was able to verify that because it happened even with nothing connected. I hope I didn't introduce any new bugs.

http://www.enigmaathome.net/static/3602xhx3am/gmc.tgz

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 2006 - Posted: 5 Sep 2013 | 23:22:31 UTC - in response to Message 2005.

Still stuck after the end of one unit moving to the next

Radioactive@Home app rev $Rev: 188 $ for GMC-300 starting...
Trying to open: /dev/ttyUSB0


I also ran a full work unit with Rev:188 and received same message above with no progress. Still have to manually start.

Profile wd9dui
Send message
Joined: 3 Feb 13
Posts: 8
Credit: 0
RAC: 0
Message 2014 - Posted: 13 Sep 2013 | 18:04:39 UTC - in response to Message 1995.

I have just received my GCC-300E yesterday, took some time to get the drivers to work with the unit and Windows 7 etc. I manage to get the comm. port 5 to work with the unit and the GQ GMC Data Viewer. –Kind of neat, better than just watching a blinking LED!

I guess I will down load the SW and see if I can get the unit to work with Boinc soon.
Gary

____________

Profile wd9dui
Send message
Joined: 3 Feb 13
Posts: 8
Credit: 0
RAC: 0
Message 2021 - Posted: 14 Sep 2013 | 21:28:30 UTC - in response to Message 2014.

I am not having much luck with this, for now I think I will hold off on using my GMC-300E until a working model is available, I do not want to try and reinvent the wheel…... I will use http://ec2.netc.com/ as it is plug and play… thanks….
5Debug: Using port number: 5
Radioactive@Home app rev $Rev: 585 $ for GMC-300 starting...
COM5 opened for read/write
Debug: Port timeouts:
ReadTotalTimeoutConstant: 1000
ReadIntervalTimeout: 0
ReadTotalTimeoutMultiplier: 100
WriteTotalTimeoutMultiplier: 100
WriteTotalTimeoutConstant: 1000
gmc_GetVersion(): read failed
Unable to open port COM5Debug: Using port number: 5
Radioactive@Home app rev $Rev: 585 $ for GMC-300 starting...
Unable to open port COM508:34:42 (5292): called boinc_finish


____________

JP_Austin
Send message
Joined: 27 Jun 13
Posts: 35
Credit: 56,630
RAC: 0

Message 2036 - Posted: 23 Sep 2013 | 4:31:34 UTC
Last modified: 23 Sep 2013 | 4:33:00 UTC

Did you have the GMC application opened at the same time you started the Boinc client? If you did then the port will not be free and the Boinc will fail. You can check the unit is working with the COM port using the GMC application but then exit that application prior to starting boinc. Also make sure all of the boinc files are not running in the background. Ctrl-alt-del and verify that no boinc code or the GMC300 code is running. Then start it fresh and see how it does.

Go to the C:\ProgramData\BOINC\slots\0 directory and report what the errors look like. I think your very close. This version of code actually works fairly well with the GMC unit.

Profile wd9dui
Send message
Joined: 3 Feb 13
Posts: 8
Credit: 0
RAC: 0
Message 2042 - Posted: 30 Sep 2013 | 2:04:32 UTC - in response to Message 2036.

It seems to be working now? Still need to see if I can put it on the map?
18:38:52 (1036): Can't open init data file - running in standalone mode
Debug: Using port number: 5
Radioactive@Home app rev $Rev: 585 $ for GMC-300 starting...
COM5 opened for read/write
Debug: Port timeouts:
ReadTotalTimeoutConstant: 1000
ReadIntervalTimeout: 0
ReadTotalTimeoutMultiplier: 100
WriteTotalTimeoutMultiplier: 100
WriteTotalTimeoutConstant: 1000
<GETVER>> returned GMC-300ERe3.10
<GETCPM>> returned -128 0
CPM value: -32768
1000,4294965658,2013-9-30 1:38:55,0,f,0
<GETCPM>> returned 0 22
CPM value: 22
242000,4294965746,2013-9-30 1:42:56,0,n,0
<GETCPM>> returned 0 25
CPM value: 25
483000,4294965846,2013-9-30 1:46:57,0,n,0
<GETCPM>> returned 0 16
CPM value: 16
724000,4294965910,2013-9-30 1:50:58,0,n,0
<GETCPM>> returned 0 15
CPM value: 15
965000,4294965970,2013-9-30 1:54:59,0,n,0
<GETCPM>> returned 0 15
CPM value: 15
1206000,4294966030,2013-9-30 1:59:0,0,n,0
____________

apotheca
Send message
Joined: 7 Apr 12
Posts: 5
Credit: 243,321
RAC: 86

Message 2044 - Posted: 6 Oct 2013 | 19:23:31 UTC

WOW! I am impressed! So a year ago when I got my GMC-300 I posed the question if the unit would work with Radioactive@home. The little bit that I got was it maybe be done but would be a lot of work. Then a project at work took over my life for a month and I forgot about this. Yesterday I came across the Radioactive@home web page that referred to the GMC-300. I was shocked to say the least. I read all the post in the thread and followed the instructions. Now I'm online using my GMC-300. You guys rock! I wish I could have been more involved in the testing but sometimes life just gets in the way.

I do have to add that I had some of the same problems that others had in connecting to the unit the first time after a reboot. It takes some shutting down of bonic and checking with the GMC-300 software. Then shutting down the GMC-300 software and restarting the bonic. This seems to work best. Also I disconnected the 9 volt battery in the unit so it won't be over charged. It's running just fine on USB power only.

Again, Thank you TJM, Dagorath, miki_e, chillomatic, JP_Austin, foovax, and wd9dui for all the work you guys put in to this.

JP_Austin
Send message
Joined: 27 Jun 13
Posts: 35
Credit: 56,630
RAC: 0

Message 2057 - Posted: 16 Oct 2013 | 11:44:06 UTC - in response to Message 2044.

Glad we could help! The real heavy lifting was done by TJM with testing by the rest of us. I wouldn't call it 100% solid but seems to restart easy enough for me. My problem is my machine will power off for a day or so and I'm looking to pick up a Beaglebone Black board to run the software and do all of the BOINC interface. That will likely be the next project to try.

[AF>Amis des Lapins] Papoun37
Send message
Joined: 19 Oct 13
Posts: 8
Credit: 2,186
RAC: 0
Message 2061 - Posted: 19 Oct 2013 | 18:30:39 UTC

English Version

I ordered a GMC-320 today. Although this is not the GCM-300 I will try to make it work for Radioactive @ home. I should receive it on November 22. The location will be in France at 5 km from the sensor 7905. See you soon.


French version:

J'ai commandé un GMC-320 aujourd'hui. Même si ce n'est pas le GCM-300 j'essaierai de le faire fonctionner pour Radioactive@home. Je doit le recevoir le 22 novembre. La localisation sera en france à 5 kilometres du sensor 7905. A bientot.

[AF>Amis des Lapins] Papoun37
Send message
Joined: 19 Oct 13
Posts: 8
Credit: 2,186
RAC: 0
Message 2065 - Posted: 29 Oct 2013 | 11:33:00 UTC - in response to Message 2061.

Yippee

My GMC-320 works now under number 9476 sensor. It's a bit difficult to understand that message # 1800 is the final message for the download of the application and its installation procedure, but with a little patience and google translation is done by understanding
____________

[AF>Amis des Lapins] Papoun37
Send message
Joined: 19 Oct 13
Posts: 8
Credit: 2,186
RAC: 0
Message 2067 - Posted: 30 Oct 2013 | 11:58:24 UTC - in response to Message 1800.
Last modified: 30 Oct 2013 | 12:19:45 UTC

Do Someone can help me.
Whith GMC exe for a GMC-320 and Boinc 7.0.64,when i turn on my PC after a turn off in a middle of a work unit, Boinc don't restart a new work of R@H automatically.
Example today: my PC is turn on at 10h08 but now 12h58 R@H is always stopped:



But if I restart Boinc, R@H restart and only at this time i can see canonical result on my profile:

<core_client_version>7.0.64</core_client_version>
<![CDATA[
<stderr_txt>
Debug: Using port number: 5
Radioactive@Home app rev $Rev: 585 $ for GMC-300 starting...
COM5 opened for read/write
Debug: Port timeouts:
ReadTotalTimeoutConstant: 1000
ReadIntervalTimeout: 0
ReadTotalTimeoutMultiplier: 100
WriteTotalTimeoutMultiplier: 100
WriteTotalTimeoutConstant: 1000
<GETVER>> returned GMC-320Re 3.12
<GETCPM>> returned 0 27
CPM value: 27
1000,0,2013-10-29 22:21:25,0,f,0
<GETCPM>> returned 0 17
CPM value: 17
242000,68,2013-10-29 22:25:26,0,n,0
....
....
<GETCPM>> returned 0 22
CPM value: 22
6274000,2118,2013-10-30 0:5:58,0,n,0
<GETCPM>> returned 0 21
CPM value: 21
6515000,2202,2013-10-30 0:9:59,0,n,0
Debug: trickle sent
Trickle sent
Debug: Using port number: 5
Radioactive@Home app rev $Rev: 585 $ for GMC-300 starting...
Found previously created output file, workunit almost completed - exiting
10:08:03 (3216): called boinc_finish

</stderr_txt>
]]>



Is it a bad setting of Boinc?
____________

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 2068 - Posted: 5 Nov 2013 | 21:46:38 UTC - in response to Message 2067.

Sorry guys, I've been gone for a while because I had way too many things to do at work.

I've just ordered GMC-300e, so maybe I'll be able to fix all the remaining bugs. I wonder how long I'll have to wait for the shipment.



[AF>Amis des Lapins] Papoun37
Send message
Joined: 19 Oct 13
Posts: 8
Credit: 2,186
RAC: 0
Message 2071 - Posted: 7 Nov 2013 | 18:11:40 UTC - in response to Message 2068.

Two weeks is the daily time to receive it.
If you want, see my sensor graph :sensor number 9476. Boinc say task is invalid but it keep it on graph

See that, I think we should add tests in the application to stop and restart if the com port forwarding errors of this kind and perhaps Boinc start to fast the application and a daily time to run gmc300.exe open com port is necessary...

(sorry for my English..)

Stderr output
<core_client_version>7.0.64</core_client_version>
<![CDATA[
<stderr_txt>
Debug: Using port number: 5
Radioactive@Home app rev $Rev: 585 $ for GMC-300 starting...
COM5 opened for read/write
Debug: Port timeouts:
ReadTotalTimeoutConstant: 1000
ReadIntervalTimeout: 0
ReadTotalTimeoutMultiplier: 100
WriteTotalTimeoutMultiplier: 100
WriteTotalTimeoutConstant: 1000
<GETVER>> returned GMC-320Re 3.12
<GETCPM>> returned -128 0
CPM value: -32768
1000,4294966204,2013-11-7 4:29:24,0,f,0
<GETCPM>> returned 0 22
CPM value: 22
242000,4294966292,2013-11-7 4:33:25,0,n,0
<GETCPM>> returned -128 0
CPM value: -32768
483000,4294834673,2013-11-7 4:37:26,0,n,0
<GETCPM>> returned -128 0
CPM value: -32768
724000,4294703054,2013-11-7 4:41:27,0,n,0
Debug: trickle sent
Trickle sent
<GETCPM>> returned -128 0
CPM value: -32768
965000,4294571435,2013-11-7 4:45:28,0,n,0
<GETCPM>> returned -128 0
CPM value: -32768
1206000,4294439816,2013-11-7 4:49:29,0,n,0
<GETCPM>> returned -128 1
CPM value: -32767
1447000,4294308201,2013-11-7 4:53:30,0,n,0
Debug: trickle sent
Trickle sent
<GETCPM>> returned -128 0
CPM value: -32768
1688000,4294176582,2013-11-7 4:57:31,0,n,0
<GETCPM>> returned -128 1
CPM value: -32767
1929000,4294044967,2013-11-7 5:1:32,0,n,0
<GETCPM>> returned -128 0
CPM value: -32768
2170000,4293913348,2013-11-7 5:5:33,0,n,0
Debug: trickle sent
Trickle sent
<GETCPM>> returned -128 0
CPM value: -32768
2411000,4293781729,2013-11-7 5:9:34,0,n,0
<GETCPM>> returned -128 1
CPM value: -32767
2652000,4293650114,2013-11-7 5:13:35,0,n,0
<GETCPM>> returned -128 0
CPM value: -32768
2893000,4293518495,2013-11-7 5:17:36,0,n,0
Debug: trickle sent
Trickle sent
<GETCPM>> returned -128 2
CPM value: -32766
3134000,4293386884,2013-11-7 5:21:37,0,n,0
<GETCPM>> returned -128 0
CPM value: -32768
3375000,4293255265,2013-11-7 5:25:38,0,n,0
<GETCPM>> returned -128 0
CPM value: -32768
3616000,4293123646,2013-11-7 5:29:39,0,n,0
Debug: trickle sent
Trickle sent
<GETCPM>> returned -128 0
CPM value: -32768
3857000,4292992027,2013-11-7 5:33:40,0,n,0
<GETCPM>> returned -128 1
CPM value: -32767
4098000,4292860412,2013-11-7 5:37:41,0,n,0
<GETCPM>> returned -128 1
CPM value: -32767
4339000,4292728797,2013-11-7 5:41:42,0,n,0
Debug: trickle sent
Trickle sent
<GETCPM>> returned -128 0
CPM value: -32768
4580000,4292597178,2013-11-7 5:45:43,0,n,0
<GETCPM>> returned -128 0
CPM value: -32768
4821000,4292465559,2013-11-7 5:49:44,0,n,0
<GETCPM>> returned -128 0
CPM value: -32768
5062000,4292333940,2013-11-7 5:53:45,0,n,0
Debug: trickle sent
Trickle sent
<GETCPM>> returned -128 1
CPM value: -32767
5303000,4292202325,2013-11-7 5:57:46,0,n,0
<GETCPM>> returned -128 0
CPM value: -32768
5544000,4292070706,2013-11-7 6:1:47,0,n,0
<GETCPM>> returned -128 0
CPM value: -32768
5785000,4291939087,2013-11-7 6:5:48,0,n,0
Debug: trickle sent
Trickle sent
<GETCPM>> returned -128 0
CPM value: -32768
6026000,4291807468,2013-11-7 6:9:49,0,n,0
<GETCPM>> returned -128 0
CPM value: -32768
6267000,4291675849,2013-11-7 6:13:50,0,n,0
<GETCPM>> returned -128 0
CPM value: -32768
6508000,4291544230,2013-11-7 6:17:51,0,n,0
Debug: trickle sent
Trickle sent
<GETCPM>> returned -128 1
CPM value: -32767
6749000,4291412615,2013-11-7 6:21:52,0,n,0
<GETCPM>> returned -128 1
CPM value: -32767
6990000,4291281000,2013-11-7 6:25:53,0,n,0
<GETCPM>> returned -128 0
CPM value: -32768
7231000,4291149381,2013-11-7 6:29:54,0,n,0
Debug: trickle sent
Trickle sent
Done - calling boinc_finish()
07:29:54 (4900): called boinc_finish

</stderr_txt>
]]>




Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 2074 - Posted: 13 Nov 2013 | 18:49:23 UTC - in response to Message 2071.
Last modified: 13 Nov 2013 | 23:16:24 UTC

I think that the negative readings bug will be fixed in the next app release, hopefully this weekend.

I received my GMC-300E today, unfortunately I won't have time for debugging/programming session until weekend.

EDIT: I found a few minutes of spare time and fixed the ID bug, now the GMC-300E is correctly detected.

I can also confirm the bug which was reported long time ago - sometimes the app is unable to detect the sensor on startup, running the GQ Data Logger fixes it.
It has never happened with my software emulator, so I guess there must be something wrong with the COM settings on the app side.

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 2076 - Posted: 15 Nov 2013 | 6:48:22 UTC - in response to Message 2074.

Two more bugs were fixed. I think the app will now init the COM port correctly without running the Data Logger app first (I still have to verify that on another PC).
GMC-300, 320 and 300E are now correctly recognized.



For some reason the values on graphs are wrong, this will be fixed next.

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 2077 - Posted: 15 Nov 2013 | 13:26:58 UTC - in response to Message 2076.

I'm still watching the thread and ready to test any new linux versions that you have.

Thanks!

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 2082 - Posted: 15 Nov 2013 | 23:47:23 UTC - in response to Message 2077.
Last modified: 16 Nov 2013 | 13:30:43 UTC

Linux version will be fixed next, as soon as I'll finish tweaking the Windows app. Right now it looks pretty stable, at least with the 300E.

Today I've fixed large number of bugs related to map and graph display, most of them were affecting GMC-3xx sensors due to non-standard CPM to uSv/h divider.

EDIT:

Refreshed build (Windows) for GMC-300, 300E with app info included: http://www.enigmaathome.net/static/3745x0av13/radioactiveathome.orgboinc.zip

Papoun37 - can you test it with GMC-320 ? Would be nice to know if it uses the same data format as the 300/300E.

JP_Austin
Send message
Joined: 27 Jun 13
Posts: 35
Credit: 56,630
RAC: 0

Message 2083 - Posted: 17 Nov 2013 | 1:49:40 UTC

Loaded this along with the latest rev of the BOINC app and working right out of the chute. No messing with anything so seems to be better on the serial port setup. I just renamed the old gmc300 app to a different name and dropped the new one in and started it up. Working like a charm.

[AF>Amis des Lapins] Papoun37
Send message
Joined: 19 Oct 13
Posts: 8
Credit: 2,186
RAC: 0
Message 2084 - Posted: 18 Nov 2013 | 17:56:55 UTC - in response to Message 2082.

Error on open com, reading version is not correct.
(But after close gmc300.exe, the official "GQ GMC Data Viewer" can read heartbeat)

Radioactive@Home app rev $Rev: 178 $ for GMC-300 starting...
COM5 opened for read/write
gmc_GetVersion(): read failed
Connected to («! (revision int: 1)
GetCommState failed: error 6
Unable to open port COM5


OR

Radioactive@Home app rev $Rev: 178 $ for GMC-300 starting...
COM5 opened for read/write
gmc_GetVersion(): read failed
Connected to ,ª. (revision int: 1)

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 2085 - Posted: 18 Nov 2013 | 18:14:29 UTC - in response to Message 2084.

Can you post a screenshot with your COM port settings in the Data Viewer app ?

[AF>Amis des Lapins] Papoun37
Send message
Joined: 19 Oct 13
Posts: 8
Credit: 2,186
RAC: 0
Message 2086 - Posted: 18 Nov 2013 | 18:35:33 UTC - in response to Message 2085.

View1 with error gmc300.exe




view2 with GMC official hearbeat without disconnect GMC320:




____________

JP_Austin
Send message
Joined: 27 Jun 13
Posts: 35
Credit: 56,630
RAC: 0

Message 2087 - Posted: 19 Nov 2013 | 2:39:50 UTC
Last modified: 19 Nov 2013 | 2:43:57 UTC

Not sure if this helps but I run the app with admin privileges. I was having trouble early on and that helped. I do see you have the GMC-320 device so they may have changed the serial/usb?

Just checked and this is the differences:

http://www.gqelectronicsllc.com/support/GQ_Geiger_Counter_Selection_Guide.png

[AF>Amis des Lapins] Papoun37
Send message
Joined: 19 Oct 13
Posts: 8
Credit: 2,186
RAC: 0
Message 2088 - Posted: 19 Nov 2013 | 6:14:08 UTC - in response to Message 2087.
Last modified: 19 Nov 2013 | 6:26:57 UTC

The protocol is the same for all GQ 3xx and it works I tested it on visual basic


http://www.gqelectronicsllc.com/download/GQ-RFC1201.txt

may be that he will guide you: I found the old version of gmc300.exe started up every time I booted the offcial GQ heatbeat properly before and I took care of them out using function exit

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 2089 - Posted: 23 Nov 2013 | 18:05:37 UTC - in response to Message 2088.

The Linux app (the last one I posted in this thread) seems to be working just fine with GMC-300E, only the sensor ID is broken.

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 2090 - Posted: 23 Nov 2013 | 20:19:55 UTC - in response to Message 2089.
Last modified: 23 Nov 2013 | 20:21:34 UTC

Should I use the new version or wait till sensor id is fixed?

EDIT: Disregard the last comment was intended for a different version than what I have.

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 2092 - Posted: 24 Nov 2013 | 18:37:57 UTC - in response to Message 2090.
Last modified: 25 Nov 2013 | 21:57:49 UTC

I think I fixed the problem with blocking COM port on startup/exit (Linux). Not all machines were affected so it was a bit hard to find.

This is my test machine running with GMC-300E connected, running one task after another. I had to chmod +rw /dev/ttyUSB0 for 'others', other than that the app does not require any hacks or tweaks.

http://radioactiveathome.org/boinc/results.php?hostid=914

The ID bug is still there (it only affects GMC-300E), I'll try to fix it today.

Foovax, could you try this with GMC-300 ?
http://www.enigmaathome.net/static/3774xntf12/gmc300.txz

doubleAA
Send message
Joined: 4 Oct 13
Posts: 2
Credit: 11,203
RAC: 0

Message 2095 - Posted: 27 Nov 2013 | 22:45:49 UTC - in response to Message 2092.

Hi TJM,
I've just received my GMC-300E and was about to prepare the app_info, etc. for my Windows box. However, many of the links containing your code drops (http://www.enigmaathome.net/static/... ) appear to be broken.

Can you please forward me to the latest drop for Windows.

Thanks in advance!!

doubleAA
Send message
Joined: 4 Oct 13
Posts: 2
Credit: 11,203
RAC: 0

Message 2096 - Posted: 27 Nov 2013 | 23:32:59 UTC - in response to Message 2095.

Please disregard my last post - after scrubbing through, I think I found the correct link.

After following the thread and getting past initialization pitfalls - it appears my first task is now running (fingers crossed)! Thanks to everyone for their effort to support this.

Cheers!

JP_Austin
Send message
Joined: 27 Jun 13
Posts: 35
Credit: 56,630
RAC: 0

Message 2104 - Posted: 30 Nov 2013 | 2:42:02 UTC

Good going! Yes the app appears to be much more stable. I even had the cable come out of the unit and it powered down. Re-plugged and the app properly picked up the communication. Great job guys.

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 2107 - Posted: 2 Dec 2013 | 6:17:26 UTC - in response to Message 2088.

Papoun37, can you try this version: http://www.enigmaathome.net/static/3786x0d6xp/gmc300.zip ? It works with GMC-300 and 300E, for now I can't find any reason why it would not work with GMC-320.

Frank
Send message
Joined: 22 Jun 13
Posts: 16
Credit: 0
RAC: 0
Message 2147 - Posted: 1 Jan 2014 | 22:34:46 UTC

Wow, that was a lot of reading from the bottom up. I have a GMC-280 that I hope this will work with, but as mentioned below, when I clicked on them links, it doesn't come back with anything, just a white page with a line of text on it. How or where can I download this latest app? Thanks

Frank
Send message
Joined: 22 Jun 13
Posts: 16
Credit: 0
RAC: 0
Message 2148 - Posted: 2 Jan 2014 | 2:37:18 UTC - in response to Message 1802.

Hello, I know these posts are old, but trying to grasp the update in this and see if its been in place and working. I have a 280, similar to the 300 model and want to try this and see if it works since its taking forever to get a sensor from here. Where can I find the latest updated app to try... Thanks

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 2149 - Posted: 2 Jan 2014 | 3:25:56 UTC - in response to Message 2148.

If using windows try the last one posted by TJM.

http://www.enigmaathome.net/static/3786x0d6xp/gmc300.zip

But there are some differences on how this is used and they are at the top of the thread.

Frank
Send message
Joined: 22 Jun 13
Posts: 16
Credit: 0
RAC: 0
Message 2153 - Posted: 2 Jan 2014 | 23:48:00 UTC

I'm using windows, when I put that link in, all I get is a white page and this at the top "Nie ma/nie bylo tu takiego pliku, lub zostal juz usuniety" whatever that means?

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 2154 - Posted: 3 Jan 2014 | 7:36:36 UTC - in response to Message 2153.

The link has expired.
I'll upload working version later today, but I have no idea if it'll work with GMC-280 (not tested yet).

Frank
Send message
Joined: 22 Jun 13
Posts: 16
Credit: 0
RAC: 0
Message 2155 - Posted: 3 Jan 2014 | 13:18:17 UTC - in response to Message 2154.

Ya, not sure myself either, its similar to the 300 models, but think its just that it doesn't have memory in it if I'm not mistaken but the same hookup. Let me know on the app and link. I'll be off for a good chunk of today to dig out from a major snowstorm we just had over night :| Did I see another posting about perhaps sensors might be available after Feb? sure hope I am on the list since I requested for it over a year ago. Sure would love to get one and get online with it than trying to get another one to work. My section of the country doesn't have one and would be nice to add to the map and provide data. I'm on couple of other sites now but like this one for the map and chart plotting. Thank you .. Frank USA

Frank
Send message
Joined: 22 Jun 13
Posts: 16
Credit: 0
RAC: 0
Message 2166 - Posted: 5 Jan 2014 | 21:14:39 UTC - in response to Message 2154.

TJM, did you by chance upload the new link for this app? or is it posted somewhere else? I would like to give this a try on my 280 unit and if I have too, I will go out and purchase a 300 unit as well just so I can get online because the outlook for getting the real one for this site looks impossible as its been well over a year since I requested and signed up for one.. whatever is missing, tell us what it is and we'll try to finish it up on our end. How much of the sensor is available, what is holding these up? Can you release it minus the missing parts and let us go forward with it if that's possible? thank you Frank

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 2168 - Posted: 7 Jan 2014 | 22:35:52 UTC - in response to Message 2166.
Last modified: 9 Jan 2014 | 18:45:41 UTC

EDIT: Link to app build 188 deleted (it had critical bugs) - please scroll down for a working version.

Please let me know if it works with GMC-280. Don't forget to edit gmc.xml
____________

Profile Fabio Tomaselli (Sicilia)
Send message
Joined: 9 Dec 12
Posts: 21
Credit: 182,274
RAC: 0

Message 2174 - Posted: 9 Jan 2014 | 17:40:33 UTC
Last modified: 9 Jan 2014 | 17:48:04 UTC

TJM , I bought a GMC- 300E . Yesterday I downloaded and installed your new software following the instructions in your post :

http://radioactiveathome.org/boinc/forum_thread.php?id=160&nowrap=true # 1800

This is my gmc.xml:

<gmc>
<comsettings>
<comment>Edit port number below</comment>
<portnumber>6</portnumber>
<comment>The settings below aren't used for anything yet</comment>
<baud>57600</baud>
<bits>8</bits>
<parity>n</parity>
<stopbits>1</stopbits>
</comsettings>
<options>
</options>
</gmc>

This is my Boinc Log event:

09/01/2014 18.44.20 | | cc_config.xml not found - using defaults
09/01/2014 18.44.20 | | Starting BOINC client version 7.2.33 for windows_intelx86
09/01/2014 18.44.20 | | log flags: file_xfer, sched_ops, task
09/01/2014 18.44.20 | | Libraries: libcurl/7.25.0 OpenSSL/1.0.1 zlib/1.2.6
09/01/2014 18.44.20 | | Data directory: C:\ProgramData\BOINC
09/01/2014 18.44.20 | | Running under account Meteo Pedara
09/01/2014 18.44.20 | | No usable GPUs found
09/01/2014 18.44.20 | Radioactive@Home | Found app_info.xml; using anonymous platform
09/01/2014 18.44.20 | | Host name: MeteoPedara-PC
09/01/2014 18.44.20 | | Processor: 2 GenuineIntel Intel(R) Core(TM)2 Duo CPU E6550 @ 2.33GHz [Family 6 Model 15 Stepping 11]
09/01/2014 18.44.20 | | Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss htt tm pni ssse3 cx16 nx lm vmx smx tm2 pbe
09/01/2014 18.44.20 | | OS: Microsoft Windows Vista: Business x86 Edition, Service Pack 2, (06.00.6002.00)
09/01/2014 18.44.20 | | Memory: 2.99 GB physical, 6.20 GB virtual
09/01/2014 18.44.20 | | Disk: 74.53 GB total, 29.13 GB free
09/01/2014 18.44.20 | | Local time is UTC +1 hours
09/01/2014 18.44.20 | Radioactive@Home | URL http://radioactiveathome.org/boinc/; Computer ID 5340; resource share 100
09/01/2014 18.44.20 | | Preferences:
09/01/2014 18.44.20 | | max memory usage when active: 1529.85MB
09/01/2014 18.44.20 | | max memory usage when idle: 2753.73MB
09/01/2014 18.44.20 | | max disk usage: 29.14GB
09/01/2014 18.44.20 | | don't compute while active
09/01/2014 18.44.20 | | don't use GPU while active
09/01/2014 18.44.20 | | (to change preferences, visit a project web site or select Preferences in the Manager)
09/01/2014 18.44.20 | | Not using a proxy
09/01/2014 18.44.21 | | Suspending computation - computer is in use
09/01/2014 18.44.21 | | Suspending network activity - computer is in use
09/01/2014 18.44.40 | Radioactive@Home | update requested by user
09/01/2014 18.44.41 | Radioactive@Home | Sending scheduler request: Requested by user.
09/01/2014 18.44.41 | Radioactive@Home | Not requesting tasks
09/01/2014 18.44.44 | Radioactive@Home | Scheduler request completed

Then I checked the file stderr.txt who writes this :
Radioactive @ Home app rev $ Rev : $ 188 for GMC- 300 starting ...
SetCommTimeouts failed : error 0
Unable to open port COM6GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
Radioactive @ Home app rev $ Rev : $ 188 for GMC- 300 starting ...
SetCommTimeouts failed : error 0
Unable to open port COM6GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
.....
But if I use the original software GMC300E, the com port 6 opens and communicates with the GMC.
Where am I wrong? Help me please.

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 2175 - Posted: 9 Jan 2014 | 18:43:06 UTC - in response to Message 2174.

The last version has a minor bug in the COM handling, it won't find the sensor. I'm fixing it right now.

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 2176 - Posted: 9 Jan 2014 | 18:58:59 UTC - in response to Message 2175.

What is the status on the linux version?

And adjusting the charts/graphs to show a more accurate result ~(-0.7)

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 2177 - Posted: 9 Jan 2014 | 19:11:17 UTC - in response to Message 2176.

My Linux host was running for nearly a month without a single glitch (with GMC-300E).
The graphs problem is also fixed, but the sensor has to be recognised by the app - without ID the server uses default dividers.

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 2178 - Posted: 9 Jan 2014 | 19:22:04 UTC - in response to Message 2177.



This chart seems a bit high.

In the stderr.txt i see a version of:

<GETVER>> returned GMC-300Re 2.38

Valter
Send message
Joined: 20 Jan 12
Posts: 19
Credit: 151,433
RAC: 69

Message 2180 - Posted: 9 Jan 2014 | 19:44:23 UTC

Hi, everyone!

I have tried to download the gmc300.zip app file but it returns the message "Nie ma/nie bylo tu takiego pliku, lub zostal juz usuniety". Where can the app be found for download?

Thanks,
Valter.

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 2181 - Posted: 9 Jan 2014 | 19:48:13 UTC - in response to Message 2178.
Last modified: 9 Jan 2014 | 19:49:52 UTC

Updated windows app: rev 195. This one should also work with GMC-280.

I'll take a look at the graph issue probably tomorrow after work. I think it's just the graph code that does not read the correct divider, because here: http://radioactiveathome.org/boinc/test123a.php the value seems to be correct.


I have tried to download the gmc300.zip app file but it returns the message "Nie ma/nie bylo tu takiego pliku, lub zostal juz usuniety". Where can the app be found for download?


If you are looking for Win app, try the build 196 from this post.

Valter
Send message
Joined: 20 Jan 12
Posts: 19
Credit: 151,433
RAC: 69

Message 2182 - Posted: 9 Jan 2014 | 20:03:30 UTC - in response to Message 2181.

Yes, Windows (sorry). Have just downloaded 195 and 196. Many thanks!

Profile Fabio Tomaselli (Sicilia)
Send message
Joined: 9 Dec 12
Posts: 21
Credit: 182,274
RAC: 0

Message 2185 - Posted: 9 Jan 2014 | 20:19:38 UTC - in response to Message 2181.

Updated windows app: rev 195. This one should also work with GMC-280.

I'll take a look at the graph issue probably tomorrow after work. I think it's just the graph code that does not read the correct divider, because here: http://radioactiveathome.org/boinc/test123a.php the value seems to be correct.


I have tried to download the gmc300.zip app file but it returns the message "Nie ma/nie bylo tu takiego pliku, lub zostal juz usuniety". Where can the app be found for download?


If you are looking for Win app, try the build 196 from this post.


I Have just downloaded 195, but doesn't work
now my stderr.txt file:
Radioactive@Home app rev $Rev: 188 $ for GMC-300 starting...
SetCommTimeouts failed: error 0
Unable to open port COM6GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
GetCommState failed: error 6
Radioactive@Home app rev $Rev: 195 $ for GMC-300 starting...
GetCommState failed: error 6
Unable to open port COM2GetCommState failed: error 6
GetCommState failed: error 6

What can I do?

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 2186 - Posted: 9 Jan 2014 | 20:27:32 UTC - in response to Message 2185.
Last modified: 9 Jan 2014 | 20:27:59 UTC

Have you tried restarting the PC or at least reconnecting the sensor ?
Error 6 means the app is unable to open the port.

Profile Fabio Tomaselli (Sicilia)
Send message
Joined: 9 Dec 12
Posts: 21
Credit: 182,274
RAC: 0

Message 2187 - Posted: 9 Jan 2014 | 20:55:58 UTC - in response to Message 2186.

Have you tried restarting the PC or at least reconnecting the sensor ?
Error 6 means the app is unable to open the port.


Yes, i had restarted pc and reconnected sensor, but nothing to do.

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 2188 - Posted: 9 Jan 2014 | 21:20:11 UTC - in response to Message 2187.
Last modified: 9 Jan 2014 | 21:20:55 UTC

Did you change the COM port number in gmc.xml ?

Profile Fabio Tomaselli (Sicilia)
Send message
Joined: 9 Dec 12
Posts: 21
Credit: 182,274
RAC: 0

Message 2189 - Posted: 9 Jan 2014 | 21:24:27 UTC - in response to Message 2188.

Did you change the COM port number in gmc.xml ?


I did change the com port from 2 to 6 thati is the com port of my GMC-300E

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 2190 - Posted: 9 Jan 2014 | 21:46:42 UTC - in response to Message 2189.

But take a look at your log, the rev 195 app tries to open COM2.

Radioactive@Home app rev $Rev: 195 $ for GMC-300 starting...
GetCommState failed: error 6
Unable to open port COM2GetCommState failed: error 6
GetCommState failed: error 6


Profile Fabio Tomaselli (Sicilia)
Send message
Joined: 9 Dec 12
Posts: 21
Credit: 182,274
RAC: 0

Message 2191 - Posted: 9 Jan 2014 | 22:00:25 UTC - in response to Message 2190.

But take a look at your log, the rev 195 app tries to open COM2.

Radioactive@Home app rev $Rev: 195 $ for GMC-300 starting...
GetCommState failed: error 6
Unable to open port COM2GetCommState failed: error 6
GetCommState failed: error 6




it seems that now read/write
this is the file:

Radioactive@Home app rev $Rev: 195 $ for GMC-300 starting...
COM6 opened for read/write
Connected to GMC-300E 3.14 (revision int: 788)
12000,3,2014-1-9 21:0:36,f,0.2 minutes,14.0 cpm,0.14 uSv/h
194000,63,2014-1-9 21:3:38,n,3.0 minutes,20.0 cpm,0.12 uSv/h
376000,150,2014-1-9 21:6:40,n,3.0 minutes,29.0 cpm,0.17 uSv/h
558000,247,2014-1-9 21:9:42,n,3.0 minutes,32.0 cpm,0.19 uSv/h
740000,316,2014-1-9 21:12:44,n,3.0 minutes,23.0 cpm,0.13 uSv/h
922000,373,2014-1-9 21:15:46,n,3.0 minutes,19.0 cpm,0.11 uSv/h
1109000,457,2014-1-9 21:18:53,n,3.1 minutes,27.0 cpm,0.16 uSv/h
1292000,508,2014-1-9 21:21:56,n,3.0 minutes,17.0 cpm,0.10 uSv/h
Trickle sent
1474000,599,2014-1-9 21:24:58,n,3.0 minutes,30.0 cpm,0.18 uSv/h
1656000,671,2014-1-9 21:28:0,n,3.0 minutes,24.0 cpm,0.14 uSv/h
1838000,780,2014-1-9 21:31:2,n,3.0 minutes,36.0 cpm,0.21 uSv/h
2025000,851,2014-1-9 21:34:9,n,3.1 minutes,23.0 cpm,0.13 uSv/h
2207000,951,2014-1-9 21:37:11,n,3.0 minutes,33.0 cpm,0.19 uSv/h
2389000,1020,2014-1-9 21:40:13,n,3.0 minutes,23.0 cpm,0.13 uSv/h
2571000,1120,2014-1-9 21:43:15,n,3.0 minutes,33.0 cpm,0.19 uSv/h
Trickle sent
2753000,1214,2014-1-9 21:46:17,n,3.0 minutes,31.0 cpm,0.18 uSv/h
2936000,1293,2014-1-9 21:49:20,n,3.0 minutes,26.0 cpm,0.15 uSv/h
3118000,1384,2014-1-9 21:52:22,n,3.0 minutes,30.0 cpm,0.18 uSv/h
3300000,1481,2014-1-9 21:55:24,n,3.0 minutes,32.0 cpm,0.19 uSv/h

Valter
Send message
Joined: 20 Jan 12
Posts: 19
Credit: 151,433
RAC: 69

Message 2196 - Posted: 10 Jan 2014 | 22:59:22 UTC

Hi!

I have installed the USB port, connected the GMC-300E and it was defined as COM3 port.

I have downloaded your package, followed your instructions on this thread and changed the port to "3".

The following messages were received:

10/1/2014 20:36:43 | | File projects/radioactiveathome.org_boinc/radac_1.77_windows_intelx86.exe not found
10/1/2014 20:36:43 | | File projects/radioactiveathome.org_boinc/sensors_1.77.xml not found
10/1/2014 20:38:01 | Radioactive@Home | update requested by user
10/1/2014 20:39:03 | Radioactive@Home | Sending scheduler request: Requested by user.
10/1/2014 20:39:03 | Radioactive@Home | Requesting new tasks for CPU
10/1/2014 20:39:07 | Radioactive@Home | Scheduler request completed: got 1 new tasks
10/1/2014 20:39:09 | Radioactive@Home | Starting task sample_2791425_0 using radac version 170 in slot 10

And the WU is now running...

Is that it? Is it all correct, and the new sensor should be seen on the map, say by tomorrow?

Thanks and regards,
Valter.

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 2197 - Posted: 10 Jan 2014 | 23:30:56 UTC - in response to Message 2196.

Check if the progress indicator is going up.

____________

Valter
Send message
Joined: 20 Jan 12
Posts: 19
Credit: 151,433
RAC: 69

Message 2198 - Posted: 11 Jan 2014 | 0:07:02 UTC

It shows now 0.000% and does not move - elapsed time 01:27:25 and increasing.

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 2199 - Posted: 11 Jan 2014 | 0:17:59 UTC - in response to Message 2198.

Try restarting the computer or unplugging then plugging in the USB to the GMC-300.

Valter
Send message
Joined: 20 Jan 12
Posts: 19
Credit: 151,433
RAC: 69

Message 2203 - Posted: 11 Jan 2014 | 7:28:16 UTC
Last modified: 11 Jan 2014 | 7:37:27 UTC

I have restarted the computer and unplugged/plugged the sensor. Now the WU reads: progress, 11.111%, elapsed 03:37:34, remaining 01:47:30.

Frank
Send message
Joined: 22 Jun 13
Posts: 16
Credit: 0
RAC: 0
Message 2222 - Posted: 12 Jan 2014 | 3:06:48 UTC

I see Rev 195 app in the listings, but were is Rev 196? I can't seem to locate in this thread? This would be so much easier if we just had the detectors to go with this service... Thanks

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 2225 - Posted: 12 Jan 2014 | 10:32:03 UTC - in response to Message 2222.

Just try the 195, perhaps I made a mistake posting the revision numbers.

Frank
Send message
Joined: 22 Jun 13
Posts: 16
Credit: 0
RAC: 0
Message 2229 - Posted: 12 Jan 2014 | 15:42:30 UTC - in response to Message 2225.

OK, ya, I saw 195 and 196 posted but couldn't locate the 196, even in a search. I was put on the waiting list some time ago for 2 sensors, would even settle for one now, are these available again and what happened to the ones that added their names to the waiting list. I'm seeing all sorts of user posting ordering one. Are they available again? what is the cost to the USA? I was on the list well over a year ago.. Thanks

JP_Austin
Send message
Joined: 27 Jun 13
Posts: 35
Credit: 56,630
RAC: 0

Message 2255 - Posted: 14 Jan 2014 | 5:41:37 UTC - in response to Message 2191.

Looks like you have it running! Way to go.


But take a look at your log, the rev 195 app tries to open COM2.

Radioactive@Home app rev $Rev: 195 $ for GMC-300 starting...
GetCommState failed: error 6
Unable to open port COM2GetCommState failed: error 6
GetCommState failed: error 6




it seems that now read/write
this is the file:

Radioactive@Home app rev $Rev: 195 $ for GMC-300 starting...
COM6 opened for read/write
Connected to GMC-300E 3.14 (revision int: 788)
12000,3,2014-1-9 21:0:36,f,0.2 minutes,14.0 cpm,0.14 uSv/h
194000,63,2014-1-9 21:3:38,n,3.0 minutes,20.0 cpm,0.12 uSv/h
376000,150,2014-1-9 21:6:40,n,3.0 minutes,29.0 cpm,0.17 uSv/h
558000,247,2014-1-9 21:9:42,n,3.0 minutes,32.0 cpm,0.19 uSv/h
740000,316,2014-1-9 21:12:44,n,3.0 minutes,23.0 cpm,0.13 uSv/h
922000,373,2014-1-9 21:15:46,n,3.0 minutes,19.0 cpm,0.11 uSv/h
1109000,457,2014-1-9 21:18:53,n,3.1 minutes,27.0 cpm,0.16 uSv/h
1292000,508,2014-1-9 21:21:56,n,3.0 minutes,17.0 cpm,0.10 uSv/h
Trickle sent
1474000,599,2014-1-9 21:24:58,n,3.0 minutes,30.0 cpm,0.18 uSv/h
1656000,671,2014-1-9 21:28:0,n,3.0 minutes,24.0 cpm,0.14 uSv/h
1838000,780,2014-1-9 21:31:2,n,3.0 minutes,36.0 cpm,0.21 uSv/h
2025000,851,2014-1-9 21:34:9,n,3.1 minutes,23.0 cpm,0.13 uSv/h
2207000,951,2014-1-9 21:37:11,n,3.0 minutes,33.0 cpm,0.19 uSv/h
2389000,1020,2014-1-9 21:40:13,n,3.0 minutes,23.0 cpm,0.13 uSv/h
2571000,1120,2014-1-9 21:43:15,n,3.0 minutes,33.0 cpm,0.19 uSv/h
Trickle sent
2753000,1214,2014-1-9 21:46:17,n,3.0 minutes,31.0 cpm,0.18 uSv/h
2936000,1293,2014-1-9 21:49:20,n,3.0 minutes,26.0 cpm,0.15 uSv/h
3118000,1384,2014-1-9 21:52:22,n,3.0 minutes,30.0 cpm,0.18 uSv/h
3300000,1481,2014-1-9 21:55:24,n,3.0 minutes,32.0 cpm,0.19 uSv/h

Profile Fabio Tomaselli (Sicilia)
Send message
Joined: 9 Dec 12
Posts: 21
Credit: 182,274
RAC: 0

Message 2256 - Posted: 14 Jan 2014 | 8:26:38 UTC - in response to Message 2255.

Looks like you have it running! Way to go.


But take a look at your log, the rev 195 app tries to open COM2.

Radioactive@Home app rev $Rev: 195 $ for GMC-300 starting...
GetCommState failed: error 6
Unable to open port COM2GetCommState failed: error 6
GetCommState failed: error 6




it seems that now read/write
this is the file:

Radioactive@Home app rev $Rev: 195 $ for GMC-300 starting...
COM6 opened for read/write
Connected to GMC-300E 3.14 (revision int: 788)
12000,3,2014-1-9 21:0:36,f,0.2 minutes,14.0 cpm,0.14 uSv/h
194000,63,2014-1-9 21:3:38,n,3.0 minutes,20.0 cpm,0.12 uSv/h
376000,150,2014-1-9 21:6:40,n,3.0 minutes,29.0 cpm,0.17 uSv/h
558000,247,2014-1-9 21:9:42,n,3.0 minutes,32.0 cpm,0.19 uSv/h
740000,316,2014-1-9 21:12:44,n,3.0 minutes,23.0 cpm,0.13 uSv/h
922000,373,2014-1-9 21:15:46,n,3.0 minutes,19.0 cpm,0.11 uSv/h
1109000,457,2014-1-9 21:18:53,n,3.1 minutes,27.0 cpm,0.16 uSv/h
1292000,508,2014-1-9 21:21:56,n,3.0 minutes,17.0 cpm,0.10 uSv/h
Trickle sent
1474000,599,2014-1-9 21:24:58,n,3.0 minutes,30.0 cpm,0.18 uSv/h
1656000,671,2014-1-9 21:28:0,n,3.0 minutes,24.0 cpm,0.14 uSv/h
1838000,780,2014-1-9 21:31:2,n,3.0 minutes,36.0 cpm,0.21 uSv/h
2025000,851,2014-1-9 21:34:9,n,3.1 minutes,23.0 cpm,0.13 uSv/h
2207000,951,2014-1-9 21:37:11,n,3.0 minutes,33.0 cpm,0.19 uSv/h
2389000,1020,2014-1-9 21:40:13,n,3.0 minutes,23.0 cpm,0.13 uSv/h
2571000,1120,2014-1-9 21:43:15,n,3.0 minutes,33.0 cpm,0.19 uSv/h
Trickle sent
2753000,1214,2014-1-9 21:46:17,n,3.0 minutes,31.0 cpm,0.18 uSv/h
2936000,1293,2014-1-9 21:49:20,n,3.0 minutes,26.0 cpm,0.15 uSv/h
3118000,1384,2014-1-9 21:52:22,n,3.0 minutes,30.0 cpm,0.18 uSv/h
3300000,1481,2014-1-9 21:55:24,n,3.0 minutes,32.0 cpm,0.19 uSv/h



Yes, it now works fine

Stderr output

<core_client_version>7.2.33</core_client_version>
<![CDATA[
<stderr_txt>
Radioactive@Home app rev $Rev: 195 $ for GMC-300 starting...
COM5 opened for read/write
Connected to GMC-300E 3.14 (revision int: 788)
13000,12,2014-1-14 2:49:8,f,0.2 minutes,37.0 cpm,0.50 uSv/h
195000,133,2014-1-14 2:52:10,n,3.0 minutes,40.0 cpm,0.23 uSv/h
378000,221,2014-1-14 2:55:13,n,3.0 minutes,29.0 cpm,0.17 uSv/h
560000,318,2014-1-14 2:58:15,n,3.0 minutes,32.0 cpm,0.19 uSv/h
743000,397,2014-1-14 3:1:18,n,3.0 minutes,26.0 cpm,0.15 uSv/h
925000,463,2014-1-14 3:4:21,n,3.0 minutes,22.0 cpm,0.13 uSv/h
1108000,554,2014-1-14 3:7:23,n,3.0 minutes,30.0 cpm,0.17 uSv/h
1291000,630,2014-1-14 3:10:26,n,3.0 minutes,25.0 cpm,0.15 uSv/h
Trickle sent
1473000,724,2014-1-14 3:13:28,n,3.0 minutes,31.0 cpm,0.18 uSv/h
1655000,793,2014-1-14 3:16:30,n,3.0 minutes,23.0 cpm,0.13 uSv/h
1837000,844,2014-1-14 3:19:32,n,3.0 minutes,17.0 cpm,0.10 uSv/h
2019000,931,2014-1-14 3:22:34,n,3.0 minutes,29.0 cpm,0.17 uSv/h
2202000,1053,2014-1-14 3:25:37,n,3.0 minutes,40.0 cpm,0.23 uSv/h
2384000,1131,2014-1-14 3:28:39,n,3.0 minutes,26.0 cpm,0.15 uSv/h
2566000,1228,2014-1-14 3:31:41,n,3.0 minutes,32.0 cpm,0.19 uSv/h
Trickle sent
2748000,1319,2014-1-14 3:34:43,n,3.0 minutes,30.0 cpm,0.18 uSv/h
2930000,1416,2014-1-14 3:37:45,n,3.0 minutes,32.0 cpm,0.19 uSv/h
3112000,1507,2014-1-14 3:40:47,n,3.0 minutes,30.0 cpm,0.18 uSv/h
3294000,1585,2014-1-14 3:43:49,n,3.0 minutes,26.0 cpm,0.15 uSv/h
3476000,1679,2014-1-14 3:46:51,n,3.0 minutes,31.0 cpm,0.18 uSv/h
3658000,1782,2014-1-14 3:49:53,n,3.0 minutes,34.0 cpm,0.20 uSv/h
3840000,1857,2014-1-14 3:52:55,n,3.0 minutes,25.0 cpm,0.14 uSv/h
Trickle sent
4022000,1963,2014-1-14 3:55:57,n,3.0 minutes,35.0 cpm,0.20 uSv/h
4204000,2035,2014-1-14 3:58:59,n,3.0 minutes,24.0 cpm,0.14 uSv/h
4386000,2113,2014-1-14 4:2:1,n,3.0 minutes,26.0 cpm,0.15 uSv/h
4568000,2188,2014-1-14 4:5:3,n,3.0 minutes,25.0 cpm,0.14 uSv/h
4750000,2285,2014-1-14 4:8:5,n,3.0 minutes,32.0 cpm,0.19 uSv/h
4932000,2354,2014-1-14 4:11:7,n,3.0 minutes,23.0 cpm,0.13 uSv/h
5115000,2436,2014-1-14 4:14:10,n,3.0 minutes,27.0 cpm,0.16 uSv/h
Trickle sent
5297000,2520,2014-1-14 4:17:12,n,3.0 minutes,28.0 cpm,0.16 uSv/h
5479000,2601,2014-1-14 4:20:14,n,3.0 minutes,27.0 cpm,0.16 uSv/h
5661000,2679,2014-1-14 4:23:16,n,3.0 minutes,26.0 cpm,0.15 uSv/h
5843000,2779,2014-1-14 4:26:18,n,3.0 minutes,33.0 cpm,0.19 uSv/h
6025000,2860,2014-1-14 4:29:20,n,3.0 minutes,27.0 cpm,0.16 uSv/h
6207000,2929,2014-1-14 4:32:22,n,3.0 minutes,23.0 cpm,0.13 uSv/h
6389000,3007,2014-1-14 4:35:24,n,3.0 minutes,26.0 cpm,0.15 uSv/h
Trickle sent
6572000,3119,2014-1-14 4:38:27,n,3.0 minutes,37.0 cpm,0.21 uSv/h
6754000,3191,2014-1-14 4:41:29,n,3.0 minutes,24.0 cpm,0.14 uSv/h
6936000,3275,2014-1-14 4:44:31,n,3.0 minutes,28.0 cpm,0.16 uSv/h
7118000,3381,2014-1-14 4:47:33,n,3.0 minutes,35.0 cpm,0.20 uSv/h
7300000,3453,2014-1-14 4:50:35,n,3.0 minutes,24.0 cpm,0.14 uSv/h
7482000,3547,2014-1-14 4:53:37,n,3.0 minutes,31.0 cpm,0.18 uSv/h
7664000,3598,2014-1-14 4:56:39,n,3.0 minutes,17.0 cpm,0.10 uSv/h
Trickle sent
7846000,3664,2014-1-14 4:59:41,n,3.0 minutes,22.0 cpm,0.13 uSv/h
8028000,3748,2014-1-14 5:2:43,n,3.0 minutes,28.0 cpm,0.16 uSv/h
8210000,3811,2014-1-14 5:5:45,n,3.0 minutes,21.0 cpm,0.12 uSv/h
8392000,3926,2014-1-14 5:8:47,n,3.0 minutes,38.0 cpm,0.22 uSv/h
8574000,4023,2014-1-14 5:11:49,n,3.0 minutes,32.0 cpm,0.19 uSv/h
8756000,4074,2014-1-14 5:14:51,n,3.0 minutes,17.0 cpm,0.10 uSv/h
8938000,4158,2014-1-14 5:17:53,n,3.0 minutes,28.0 cpm,0.16 uSv/h
Trickle sent
9120000,4221,2014-1-14 5:20:55,n,3.0 minutes,21.0 cpm,0.12 uSv/h
9302000,4293,2014-1-14 5:23:57,n,3.0 minutes,24.0 cpm,0.14 uSv/h
9485000,4384,2014-1-14 5:27:0,n,3.0 minutes,30.0 cpm,0.17 uSv/h
9667000,4450,2014-1-14 5:30:2,n,3.0 minutes,22.0 cpm,0.13 uSv/h
Radioactive@Home app rev $Rev: 195 $ for GMC-300 starting...
COM5 opened for read/write
Radioactive@Home app rev $Rev: 195 $ for GMC-300 starting...
COM5 opened for read/write
Connected to GMC-300E 3.14 (revision int: 788)
12000,7,2014-1-14 5:42:7,r,0.2 minutes,24.0 cpm,0.32 uSv/h
195000,107,2014-1-14 5:45:10,n,3.0 minutes,33.0 cpm,0.19 uSv/h
377000,201,2014-1-14 5:48:12,n,3.0 minutes,31.0 cpm,0.18 uSv/h
559000,273,2014-1-14 5:51:14,n,3.0 minutes,24.0 cpm,0.14 uSv/h
741000,360,2014-1-14 5:54:16,n,3.0 minutes,29.0 cpm,0.17 uSv/h
923000,426,2014-1-14 5:57:18,n,3.0 minutes,22.0 cpm,0.13 uSv/h
Trickle sent
Done - calling boinc_finish()
06:57:18 (3444): called boinc_finish

Valter
Send message
Joined: 20 Jan 12
Posts: 19
Credit: 151,433
RAC: 69

Message 2266 - Posted: 15 Jan 2014 | 9:45:59 UTC

Good morning!

I believe I have done wrong in opening a new thread for the message below. I apologize for the wrong behavior.

Having settled down installation issues with your kind help in this thread, it was running smoothly until today, when various WUs started to show "Computing error". Sometimes, the following messages is shown:

14/1/2014 08:44:10 | Radioactive@Home | Output file sample_2805809_0_0 for task sample_2805809_0 absent

It seems that the trickles are being sent correctly, but the WUs now continuously show computing errors, and no credits are now being given.

Any ideas?

Thanks,
Valter.

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 2267 - Posted: 15 Jan 2014 | 11:00:16 UTC - in response to Message 2266.

I think it is some kind of bug in the BOINC manager. There is an error message saying "finish file present too long" which means the output file is still pressent 10s after the application was closed.
Does your host run under heavy load ?

Valter
Send message
Joined: 20 Jan 12
Posts: 19
Credit: 151,433
RAC: 69

Message 2268 - Posted: 15 Jan 2014 | 11:11:24 UTC - in response to Message 2267.
Last modified: 15 Jan 2014 | 11:11:39 UTC

I believe so.

Should I limit CPU use to X?

Valter
Send message
Joined: 20 Jan 12
Posts: 19
Credit: 151,433
RAC: 69

Message 2269 - Posted: 15 Jan 2014 | 13:14:18 UTC - in response to Message 2268.

I think I solved it - and post here the solution in case anyone faces the same problem.

I had defined maximum download/upload rates for my BOINC Manager. I eliminated these maximum rates (Tools - Computing Preferences).

One Radioactive@home WU has finished since then, without problems. So I believe it solved the issue.

Thanks again for all your help.

Valter.

Valter
Send message
Joined: 20 Jan 12
Posts: 19
Credit: 151,433
RAC: 69

Message 2271 - Posted: 15 Jan 2014 | 14:39:40 UTC

=( OK. The second WU showed the same computing error.

Do you think that I should then try to limit CPU use? To how much %?

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 2272 - Posted: 15 Jan 2014 | 14:42:08 UTC - in response to Message 2271.

This project doesn't use a lot of cpu so no limit on that would be required. It's basically just polling the serial port/usb port to check what the current measurement is and logging it. Then occasionally it uploads that information to the server.

What is the output of your logs? Maybe the detector is coming unplugged or losing connection?

Valter
Send message
Joined: 20 Jan 12
Posts: 19
Credit: 151,433
RAC: 69

Message 2273 - Posted: 15 Jan 2014 | 15:08:40 UTC - in response to Message 2272.

Sensor 10458 - As far as I understand, result outputs are OK.

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,102,256
RAC: 372

Message 2274 - Posted: 15 Jan 2014 | 16:47:55 UTC - in response to Message 2273.

Try to disable other project(s) and run just one or two work units.
I've seen situations where other apps (from other projects) somehow stressed BOINC core client/manager and it's repsonse was super slow, causing serious glitches. Similar things can happen also when the system is busy doing something (may not be related to CPU load and BOINC at all).

Valter
Send message
Joined: 20 Jan 12
Posts: 19
Credit: 151,433
RAC: 69

Message 2278 - Posted: 16 Jan 2014 | 9:15:27 UTC - in response to Message 2274.
Last modified: 16 Jan 2014 | 9:15:40 UTC

Many thanks, TJM.

I have disabled various projects, kept only one CPU WU working and closed my too many Firefox windows which remain always open. It worked. The Radioactive WU has finished just fine.

I will now keep on testing and will inform of results.

Thanks again,
Valter.

Valter
Send message
Joined: 20 Jan 12
Posts: 19
Credit: 151,433
RAC: 69

Message 2283 - Posted: 16 Jan 2014 | 14:59:01 UTC

Hi, everyone!

I have just found out that here Radioactive WUs go wrong only when Quake-Catcher Network is running. Apparently the presence of two sensors (GMC-300E and the quake sensor) causes problems to Radioactive WUs.

TJM, can it be true? If there is nothing to be done, I will run Radioactive on one computer and QCN on another one.

Thanks again.

Profile Fabio Tomaselli (Sicilia)
Send message
Joined: 9 Dec 12
Posts: 21
Credit: 182,274
RAC: 0

Message 2291 - Posted: 16 Jan 2014 | 21:21:12 UTC - in response to Message 2283.

Hi, everyone!

I have just found out that here Radioactive WUs go wrong only when Quake-Catcher Network is running. Apparently the presence of two sensors (GMC-300E and the quake sensor) causes problems to Radioactive WUs.

TJM, can it be true? If there is nothing to be done, I will run Radioactive on one computer and QCN on another one.

Thanks again.


Hi. I am running Radioactive (GMC-300E) with Quake-Catcher Network and SETI. It's all ok.

Valter
Send message
Joined: 20 Jan 12
Posts: 19
Credit: 151,433
RAC: 69

Message 2293 - Posted: 16 Jan 2014 | 21:52:57 UTC - in response to Message 2291.



Hi. I am running Radioactive (GMC-300E) with Quake-Catcher Network and SETI. It's all ok.


Here, everytime I start QCN, Radioactive WUs end up with computing error. Something in my system, perhaps?

Profile Fabio Tomaselli (Sicilia)
Send message
Joined: 9 Dec 12
Posts: 21
Credit: 182,274
RAC: 0

Message 2294 - Posted: 16 Jan 2014 | 22:02:10 UTC - in response to Message 2293.



Hi. I am running Radioactive (GMC-300E) with Quake-Catcher Network and SETI. It's all ok.


Here, everytime I start QCN, Radioactive WUs end up with computing error. Something in my system, perhaps?


Try to install latest version of Boinc

Valter
Send message
Joined: 20 Jan 12
Posts: 19
Credit: 151,433
RAC: 69

Message 2299 - Posted: 17 Jan 2014 | 8:16:10 UTC - in response to Message 2294.


Try to install latest version of Boinc


I do have the latest version (7.2.33).

Profile Fabio Tomaselli (Sicilia)
Send message
Joined: 9 Dec 12
Posts: 21
Credit: 182,274
RAC: 0

Message 2300 - Posted: 17 Jan 2014 | 10:13:07 UTC - in response to Message 2299.

Hello, it's really weird. What operating system do you have? I use windows vista, no windows 7 because of the problems I had. Have you tried changing the Warrior sensor port of Quake Network?

Valter
Send message
Joined: 20 Jan 12
Posts: 19
Credit: 151,433
RAC: 69

Message 2302 - Posted: 17 Jan 2014 | 11:30:12 UTC - in response to Message 2300.

Hello, it's really weird. What operating system do you have? I use windows vista, no windows 7 because of the problems I had. Have you tried changing the Warrior sensor port of Quake Network?


Hi, thanks. I have Windows XP. And sorry for the ignorance but... what is a Warrior sensor port and how can I change it?

Profile Fabio Tomaselli (Sicilia)
Send message
Joined: 9 Dec 12
Posts: 21
Credit: 182,274
RAC: 0

Message 2303 - Posted: 17 Jan 2014 | 16:29:19 UTC - in response to Message 2302.

Hello, in order to participate in the project I think you have installed a sensor joy warrior like this:

http://www.boincitaly.org/articoli/guide/132-installazione-del-sensore-joywarrior-per-qcn.html

http://qcn.stanford.edu/join-qcn/manualsinstructions

Valter
Send message
Joined: 20 Jan 12
Posts: 19
Credit: 151,433
RAC: 69

Message 2327 - Posted: 22 Jan 2014 | 9:18:00 UTC - in response to Message 2303.

Hi, Fabio, TJM and everyone.

Made further tests, but no results. I still have to run Radioactive on one computer and QCN on another one.

I also have WUProp@home running here. Could it be that three non-CPU-intensive tasks are causing one (Radioactive) not to work properly?

Thanks,
Valter.

Valter
Send message
Joined: 20 Jan 12
Posts: 19
Credit: 151,433
RAC: 69

Message 2336 - Posted: 27 Jan 2014 | 19:01:19 UTC

Well, it seems to have been solved.

Following Fabio's remarks in this same thread, I have changed the positions of the USB terminals of both the GMC-300 and the quake sensor at the back of my computer. Some experiment was required and, after some trials, the four terminals are now:

[] printer [] quake
[] GMC-300 [] external HD

It is now working properly. Thank you for all your help and patience.

Valter.

Profile Fabio Tomaselli (Sicilia)
Send message
Joined: 9 Dec 12
Posts: 21
Credit: 182,274
RAC: 0

Message 2337 - Posted: 28 Jan 2014 | 13:03:58 UTC - in response to Message 2336.
Last modified: 28 Jan 2014 | 13:06:06 UTC

"Well, it seems to have been solved.

Following Fabio's remarks in this same thread, I have changed the positions of the USB terminals of both the GMC-300 and the quake sensor at the back of my computer. Some experiment was required and, after some trials, the four terminals are now:

[] printer [] quake
[] GMC-300 [] external HD

It is now working properly. Thank you for all your help and patience.

Valter."


Hello Valter, I'm glad to have been of help and that everything is working

JP_Austin
Send message
Joined: 27 Jun 13
Posts: 35
Credit: 56,630
RAC: 0

Message 2339 - Posted: 1 Feb 2014 | 7:23:09 UTC

Is it possible to share the source for the app. I'd like to get a BeagleBone or RPi to read the GMC300. I see there is a Pi posting already but think it's setup for the standard sensor. Would like to see what this app would take to run on another platform.

Profile krzyszp
Project administrator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 16 Apr 11
Posts: 373
Credit: 461,847
RAC: 146

Message 2341 - Posted: 2 Feb 2014 | 18:34:18 UTC - in response to Message 2339.
Last modified: 2 Feb 2014 | 18:35:46 UTC

Is it possible to share the source for the app. I'd like to get a BeagleBone or RPi to read the GMC300. I see there is a Pi posting already but think it's setup for the standard sensor. Would like to see what this app would take to run on another platform.

radac_1.77_rev585.7z are latest sources.

Edit:
Sources are published in this link: http://radioactiveathome.org/boinc/forum_thread.php?id=99#925
____________
Regards,
Krzysztof 'krzyszp' Piszczek
Android Radioactive@Home Map
Android Radioactive@Home Map - donated
My Workplace

Valter
Send message
Joined: 20 Jan 12
Posts: 19
Credit: 151,433
RAC: 69

Message 2382 - Posted: 27 Feb 2014 | 8:38:01 UTC
Last modified: 27 Feb 2014 | 8:39:29 UTC

Hi, everyone!

Again, I am facing difficulties in attaching my GMC-300e to Radioactive@home. =(

Everything was OK when, all of a sudden, trickles stopped to be sent. I received an email indicating that I should check if my sensor was disconnected and, a few hours later, it was excluded from the map.

First I checked if the sensor was connected to the computer, and I found it is. The software that came with the sensor can receive data from GMC-300e normally.

Then I started aborting WUs and getting new ones, and noticed that the BOINC manager showed progress 0.000% for all Radioactive WUs, and didn't move at all. Apparently they also never ended (I have aborted some of them after some hours).

I then reset Radioactive@home at the BOINC manager, with no results. Later, I removed Radioactive, re-attached, re-installed the application, installed the latest version of the BOINC manager, re-started the computer and got new WUs. Now WUs show progress and go to an end, but no trickles are sent. BOINC manager event logs show no attempt to send the trickles.

Now I am changing the sensor from one USB port to another to see if I find one that works. Until now, no results.

You can see below in this thread that I had problems to get GMC-300e and the Quake Catcher sensor working on the same computer. This had been solved well before.

Also, for information, on the same day trickles stopped to be sent, I changed my WiFi router (the old one had crashed). Now I have a D-Link DIR-905L. I wonder if something in the router configuration could have caused trickles to stop being sent (it was installed just before the problems started). I tried to check configuration settings, but could find nothing (I am a non-technical person, by the way).

Can anyone give a hint?

Many thanks,
Valter.

Eric Damon Walters
Avatar
Send message
Joined: 27 May 12
Posts: 1
Credit: 0
RAC: 0
Message 2547 - Posted: 16 Apr 2014 | 4:33:59 UTC

I am looking to add a GMC-300E+ as well.

I've read the forum here and realize I need to download the generic app - Rev 195/196 in order to replace the xml and exe files in BOINC/modify the port number, etc.

But the Rev 195/196 download link appears to be a broken link now. "Nie ma/nie bylo tu takiego pliku, lub zostal juz usuniety" Is there a new location/revision of this download?

JP_Austin
Send message
Joined: 27 Jun 13
Posts: 35
Credit: 56,630
RAC: 0

Message 2746 - Posted: 8 May 2014 | 4:10:01 UTC

I'd repost it but there does not appear to be an upload from this forum. I think I still have the PC application if the original source is gone.

JP_Austin
Send message
Joined: 27 Jun 13
Posts: 35
Credit: 56,630
RAC: 0

Message 2783 - Posted: 21 May 2014 | 13:20:17 UTC - in response to Message 2746.
Last modified: 21 May 2014 | 13:20:33 UTC

Send me a private message on this site with your email and I will email you the files.

sebastien pernecker
Send message
Joined: 2 Jul 14
Posts: 1
Credit: 0
RAC: 0
Message 2810 - Posted: 2 Jul 2014 | 16:58:03 UTC - in response to Message 2547.

Same issue, I got a GMC-300Eplus and want to link to BOINC project on my linux workstation.

JP_Austin
Send message
Joined: 27 Jun 13
Posts: 35
Credit: 56,630
RAC: 0

Message 2821 - Posted: 9 Jul 2014 | 5:35:41 UTC - in response to Message 1394.

The code that was distributed here was the PC application. You will need to find that source code to build it for Linux.

bwgray
Send message
Joined: 16 Aug 14
Posts: 2
Credit: 0
RAC: 0
Message 2911 - Posted: 30 Aug 2014 | 21:50:40 UTC

Hi, I just bought a GMC-320plus but it's supposed to be GMC-300 backwards compatible. Please send me the software for R@H. Thanx. But a worse problem is the usual Microsoft screw-up. If I am reading stuff correctly the latest drivers for the 320 are not compatible with Win 8 nor Win 8.1. Why is Microsoft always trying to break hardware drivers????

DanUntied
Send message
Joined: 20 Sep 14
Posts: 1
Credit: 0
RAC: 0
Message 2945 - Posted: 28 Oct 2014 | 1:46:21 UTC
Last modified: 28 Oct 2014 | 1:49:04 UTC

I recently purchased a GMC-300Eplus and want to use it with the project. Is there still files available to allow it to work with Radioactive@Home?
Thanks,
Dan

gtippitt
Send message
Joined: 23 Nov 14
Posts: 4
Credit: 0
RAC: 0
Message 2963 - Posted: 26 Nov 2014 | 22:27:36 UTC

I live a short ways from Oak Ridge, Tennessee USA, where they have had reactors running longer than anywhere. The old X10 reactor (also known as the Clinton Pile but had nothing to do whatsoever with Bill) was the second reactor ever built after Fermi's Chicago Pile. Besides the old spills and the reactors still running, they still build and store the bombs there as well. They prefer to talk about having the second fastest supercomputer instead of the radiation.

I bought one of the GMC-300E+ detectors off eBay to monitor radiations levels rather than ordering one of the kits, because I'm better at taking things apart than putting them together. I'm a lot better with software than hardware, so I thought perhaps I could get it to work with RadioActive@Home under Linux.

I crunch for lots of other BOINC projects and would like to add your project to my list.

If somebody could send me a copy of whatever code they have as a starting place, I'll try to see what I can get working.

Thanks,
Greg Tippitt
greg@tippitt.net

Profile krzyszp
Project administrator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 16 Apr 11
Posts: 373
Credit: 461,847
RAC: 146

Message 2964 - Posted: 26 Nov 2014 | 22:54:05 UTC - in response to Message 2963.

I will push TJM to add link for GMC app on main page or somewhere easy to find...
____________
Regards,
Krzysztof 'krzyszp' Piszczek
Android Radioactive@Home Map
Android Radioactive@Home Map - donated
My Workplace

gtippitt
Send message
Joined: 23 Nov 14
Posts: 4
Credit: 0
RAC: 0
Message 2965 - Posted: 27 Nov 2014 | 6:50:07 UTC - in response to Message 2964.

The link in TJM's earlier post referred to a ZIP file on the Enigma@Home forums that cannot be found,

"Nie ma/nie bylo tu takiego pliku, lub zostal juz usuniety"
.


I appreciate whatever you guys can send to me. As I get mine working, I will try to put together a simple "How To", which could be "stickied" in the forum.

This sensor cost me (US) $100, which is a lot more than your kit, but if we could put together some simple instructions for using it on Radioactive@Home, there are probably lots more BOINC crunchers like me that are a bit paranoid about radiation, but not good with hardware. I realize it is silly of me, since the sensors themselves are not radioactive, but the words "hardware tinkering" and "radioactive" used together don't fill me with confidence regarding my competence with a soldering iron.

Dziękuję

Greg
BOINC Team Leader for
STARFLEET - Star Trek Fan Association

Holger Weiss
Send message
Joined: 20 May 14
Posts: 12
Credit: 0
RAC: 0
Message 3181 - Posted: 25 Mar 2015 | 9:46:34 UTC

Any update here?

I also would like to connect my gmc-320 to the project.

Thanks

gtippitt
Send message
Joined: 23 Nov 14
Posts: 4
Credit: 0
RAC: 0
Message 3190 - Posted: 30 Mar 2015 | 6:52:04 UTC - in response to Message 3181.

I have never gotten any help with information regarding the GMC-300 detector, other than the posts in this thread that reference the files that can no longer be found in the Enigma forum.

I've given up being able to share my device's readings with this project, which is a shame given my close proximity to the US Oak Ridge National Laboratories. The ORNL site is arguably the 4th most infamous reactor site, although they have not admitted to any radiation accidents since 1958.

Greg

foovax
Send message
Joined: 15 Mar 13
Posts: 41
Credit: 47,598
RAC: 0

Message 3191 - Posted: 30 Mar 2015 | 14:52:14 UTC

I have a copy of linux source for this but unsure of how current it is.

radac_1.77_rev585

It had bugs and I also had to have a bash script that would check and restart things but when it was working it seemed to do what was needed.

gtippitt
Send message
Joined: 23 Nov 14
Posts: 4
Credit: 0
RAC: 0
Message 3192 - Posted: 30 Mar 2015 | 15:38:33 UTC - in response to Message 3191.

Could you email it to me or tell me where I could download it from. I would like to get this working and am willing to work on the coding and documentation. I just had nowhere to start from. With something that shows me how the project needs the data, I can work on it from there.

My email address is gtippitt at gmail dot com

Thanks,
Greg

Profile KarmannGaz
Avatar
Send message
Joined: 12 Mar 13
Posts: 100
Credit: 129,472
RAC: 60

Message 3193 - Posted: 30 Mar 2015 | 22:13:08 UTC
Last modified: 30 Mar 2015 | 22:18:18 UTC

If you don't get anywhere with that, you could try message TJM too perhaps? Looking at his sensor number 914 on the map, it appears to be using a GMC-300 v3.12.
Also Antti with sensor number 10464, Fabio Tomaselli with sensor number 12190, and Valter with sensor number 12457, all say GMC-300E v3.14.

Holger Weiss
Send message
Joined: 20 May 14
Posts: 12
Credit: 0
RAC: 0
Message 3198 - Posted: 1 Apr 2015 | 5:35:45 UTC

I'm using Windows so I guess I can't do anything with the linux source :-(

Rod Frary
Send message
Joined: 9 Mar 12
Posts: 2
Credit: 0
RAC: 0
Message 3419 - Posted: 27 Jan 2016 | 20:53:45 UTC - in response to Message 3181.

Any update here?

I also would like to connect my gmc-320 to the project.

Thanks

I HAVE EXACTLY THE SAME QUESTION! I've been hoping someone would do the work to make all the different detectors more compatible with each other or at least with R@H. I have a "GMC-320+" that I would like to use with this project. Can anyone point me in the right direction for making this happen.

Thanks
Rod

Rod Frary
Send message
Joined: 9 Mar 12
Posts: 2
Credit: 0
RAC: 0
Message 3420 - Posted: 27 Jan 2016 | 21:08:23 UTC - in response to Message 2911.

Hi, I just bought a GMC-320plus but it's supposed to be GMC-300 backwards compatible. Please send me the software for R@H. Thanx. But a worse problem is the usual Microsoft screw-up. If I am reading stuff correctly the latest drivers for the 320 are not compatible with Win 8 nor Win 8.1. Why is Microsoft always trying to break hardware drivers????


Did you ever get your GMC-320+ to work with R@H ? I have the same GMC-320+ but my solution should be a piece of cake, because I refuse to give up Windows XP as my OS. I've found it is true (in my case) that you can't teach an old dog new tricks!
Thanks for any help you can give,
Rod

--Norbert
Send message
Joined: 6 Jan 15
Posts: 69
Credit: 0
RAC: 0
Message 3421 - Posted: 28 Jan 2016 | 3:54:45 UTC

There are other options for you to contribute with your GMC-320+, if this site proves too difficult. The following alternate sites are compatible with that detector:

www.radmon.org
www.netc.com
www.gmcmap.com

Lars Mehnen
Send message
Joined: 18 Jul 16
Posts: 1
Credit: 0
RAC: 0
Message 3462 - Posted: 18 Jul 2016 | 11:09:52 UTC

I've also got a GMC-300R+ Logger ... runing Ubuntu Linux 16.04
The Logger works fine with it's clientsoftware unter linux ...
what do I have to do to connect it to the Radioactive@home network Boinc client?

Thanx
Lars

Greg Clemmer
Send message
Joined: 22 Jan 17
Posts: 6
Credit: 0
RAC: 0
Message 3507 - Posted: 25 Jan 2017 | 22:53:02 UTC - in response to Message 1396.

I am new to the group and I also am getting a GMC-300. It was all I was capable of obtaining within my price range, age,set skills, and physical ability. The last word on this post was"if the project admins don't approve the results I won't release my work." Has this happened, or has any more work been done on this?

Greg Clemmer
Send message
Joined: 22 Jan 17
Posts: 6
Credit: 0
RAC: 0
Message 3508 - Posted: 25 Jan 2017 | 22:53:18 UTC - in response to Message 1396.

I am new to the group and I also am getting a GMC-300. It was all I was capable of obtaining within my price range, age,set skills, and physical ability. The last word on this post was"if the project admins don't approve the results I won't release my work." Has this happened, or has any more work been done on this? Greg

apotheca
Send message
Joined: 7 Apr 12
Posts: 5
Credit: 243,321
RAC: 86

Message 3669 - Posted: 11 Apr 2018 | 4:12:52 UTC
Last modified: 11 Apr 2018 | 4:29:23 UTC

So I've recently I got a couple of request for the GMC300 files for the Radioactive@home. So I relived how I did it so many years ago and gave the files to them. They reported back to me it worked for them too. So here is a link for the files I have and the instructions on how to do it.

https://drive.google.com/open?id=1VKDHaN4-cvKWEy-CnC9LQ1LhsKlFfQRZ

The only annoyance I have with the application is does not automatically open the com port on the GMC300. It has to be done manually with is outlined in the how to file. If someone who knows coding could fix that would be my hero.

My Sensor ID is 1974 on the Radioactive@home map. It comes up as unknown sensor on the map where other GMC units show they are GMC300. So I don't know if I have an older version of the application. So there maybe later versions out there i'm not aware of.

Post to thread

Message boards : Number crunching : using a GMC-300 for a project detector?


Main page · Your account · Message boards


Copyright © 2019 BOINC@Poland | Open Science for the future