Unreliable captures - need to reset board each time

Topic: Unreliable captures - need to reset board each time (Read 2516 times) previous topic - next topic

Unreliable captures - need to reset board each time

April 17, 2013, 11:30:53 pm

Hi,
I received an OLS board yesterday, needing to debug an Arduino project I'm working on. I've found that I have to press the reset button on the board prior to each capture otherwise no data is displayed.

So the details:
- OLS installed and (somewhat) operational, using the Logic Sniffer software from lxtreme.nl, version 0.9.6 SP1.
- Arduino Mega transmitting a single byte '0x42' every 100 ms @ 9600 baud on the first auxiliary port.
- Channel 0 physically connected to the port's TX line, with all other channels grounded.
- Channel 0 set up in the software with a 2 stage serial trigger to capture the first rising edge.
- Capture is performed at 50 kHz with no noise filter nor RLE.

And everything works fine. That is If I reset the board before capture.

If I perform a second capture run by pressing the "Repeat capture..." button - *without* first pressing the reset button - the system is triggered, data is transferred, but the display is completely blank, i.e., all zeros. Given that the Arduino is continuously generating the single byte, it should be there (and indeed it is - I've even checked with a CRO).

I'm now quite confused. Have I missed something really basic? Any help would be appreciated - thanks!

Colin

(BTW - None of the device type / firmware version / protocol / ancillary is shown in the Logic Sniffer application. And when I go to check the status with pump-loader or similar, it returns an unknown Flash type. I can only imagine the latest of everything is on the board given that it's a new one.)

Re: Unreliable captures - need to reset board each time

Reply #1 – April 19, 2013, 05:19:55 pm

I'm not sure whether the latest boards also ship with the latest-greatest firmware. There's some confusion about whether or not this should be the case. If no device metadata is returned to the client, it might well be that your OBLS has older firmware in it. Are you able to update it to the latest firmware, for example, the DemonCore 3.0 firmware? If so, does it resolve your issue?

Re: Unreliable captures - need to reset board each time

Reply #2 – April 19, 2013, 10:44:10 pm

Hi jawi,
Thanks for your recommendation.

So I downloaded v3.08 of the Open Logic Sniffer package from http://www.gadgetfactory.net/logicsniff ... r.Download. And then I tried to run the OLS_Upgrader.bat script.

No luck: it bails at line 59 of ols-upgrader.sh (after a couple of errors for not finding the dumb application) because there is no clear command. I tried running the bash supplied with the distribution myself and indeed clear does not work.

I guess I'm further away from where I want to be than before. What's the best method for upgrading the board - FPGA and firmware - when the recommended methods don't work? Should I be installing my own separate version of cygwin?

Re: Unreliable captures - need to reset board each time

Reply #3 – April 19, 2013, 11:03:19 pm

OK - so it was a problem with the firmward/FPGA image.

I finally found the the posting New Windows OLS FPGA (& PIC firmware!) image loader..., updated everything and I think my problems is solved.

Saying this, why isn't the documentation clearer? I'm not so critical here as I realise that all this is the hard work of people unlikely to be getting paid. But things should work "out-of-the-box", or the instructions should state how to get things working "out-of-the-box".

With this in mind, where do I sign up to rewrite the documentation, at least for the Windows platform for which I have access to?

Re: Unreliable captures - need to reset board each time

Reply #4 – April 23, 2013, 01:07:06 pm

By all means, go ahead and fix the documentation. IIRC, you could simply register for an account at the wiki of dangerousprototypes.org to have editing karma.