Sniffer Not Working In Windows

[DougALug]

Hey,

I purchased my Logic Sniffer about 3 weeks ago and am unable to get it to reliably read any data. I've tried multiply 'signals' on different pins and about the only thing I will see is a pin go high and stay high. I've downloaded and run SUMP and used OLS 9.6.1 and OLS 9.4. All three of them behave identically.

For testing I used an arduino nano, and an arduino unu that I had laying around. I wrote a program the will toggle bits every couple miliseconds.

I connected the arduino ground to pin 0 on the connector on the right . D7 and D9 from the arduino to pins 1, 2 (channel 0 and 1) on the Logic Sniffer board. Using my scope, I verified that there is 0 to 5V swing on the pins.

1) I've tried multiple pins and channels.
2) I've tried lowering the sample rate and buffer sizes
3) I've tried using the trigger method for acquiring data
4) I've tried two different arduinos
5) I tried waving a magic wand and some mystic incantations.

None of these have netted anything positive.

The connectors I am on using is the 90-degree one on the right hand side of the board. I assume this is the right one.

Am I missing something or is the board defective? Any help would be greatly appreciated.

-Doug

[tayken]

Just to be sure: Which number scheme did you use? (OLS, Capture, under acquisition)

Also did you update to the latest firmware (for uC) and bitstream (for FPGA)?

[DougALug]

I don't understand the number scheme question totally. I am using channel 0 and 1, which I assumed was 0 and 1 on the lower right hand-side of the board on the 90-degree header. The number definitely seem to follow. It is actually like it is taking 1 sample and no more.

I haven't updated the firmware, I will try to do that right now.

[tayken]

I don't understand the number scheme question totally. I am using channel 0 and 1, which I assumed was 0 and 1 on the lower right hand-side of the board on the 90-degree header. The number definitely seem to follow. It is actually like it is taking 1 sample and no more.

Check out this page for the numbering scheme. If you look on your board, there are 2 numbers for each pin, one inside, the other outside. What I understand from "Lower right hand side channels 0 and 1" is that you are fine there, you are using the inside numbering scheme, which should be the default one. If you follow the path in my previous post (Capture menu -> Acquisition tab) you can see which one you are using.

I haven't updated the firmware, I will try to do that right now.

Please do, I believe that'll solve the problem. If not, try running the test program (it is inside the update package, named as selftest.bat), that can give us some clues but if there was a problem, you shouldn't be able to start capture.

[DougALug]

Hey,

Okay I updated the firmware on the board.

I tried to run the self-test and it gives me the following error:

C:\Program Files\ols-0308\OLS_Upgrader>selftest.bat\..\winbin\ols-loader -selftest -p:COM6
Logic Sniffer ROM loader v0.3 (November 9, 2010)
Opening serial port 'COM6' @ 921600 ... OK
Error reading OLS id

COM6 is currently where the sniffer resides.

Playing with the trigger's I was able to get some data. I had to go back to SUMP because the nicer ones only give me about 24k of data... not really enough for what I am doing. I set the trigger to 50/50, simple, serial and masked channel 0 and 1. I now see some data, but only part of it.

Am I misunderstanding something, can't I just sample data for a time-slice and be done? I don't need a trigger for what I am doing do I?

[tayken]

Hey,

Okay I updated the firmware on the board.

Nice! That helps a lot with debugging.

I tried to run the self-test and it gives me the following error:

C:\Program Files\ols-0308\OLS_Upgrader>selftest.bat\..\winbin\ols-loader -selftest -p:COM6
Logic Sniffer ROM loader v0.3 (November 9, 2010)
Opening serial port 'COM6' @ 921600 ... OK
Error reading OLS id

COM6 is currently where the sniffer resides.

Is it in update mode (ACT LED lit)? The program can only talk with it in update mode.

Playing with the trigger's I was able to get some data. I had to go back to SUMP because the nicer ones only give me about 24k of data... not really enough for what I am doing. I set the trigger to 50/50, simple, serial and masked channel 0 and 1. I now see some data, but only part of it.

Am I misunderstanding something, can't I just sample data for a time-slice and be done? I don't need a trigger for what I am doing do I?

I don't know anything about SUMP, I only used the OLS client by Jawi, so I can give examples about that. Did you check out the trigger tutorial? It has tons of data. 50/50 means that you'll see 50% of the data before the trigger and 50% after it. I usually use 10/90 as I am interested about the data after the trigger. Serial is usually not good, I only use parallel. If you are continuously toggling pins, you don't need triggers, it is used for capturing a specific event (like the beginning of an SPI transmission).

It is nice that you don't have a brick, that is a relief on my end. :) If you have more questions, read the wiki, shoot them here and I'll be in the IRC for a while (not much, getting awfully late here).

[DougALug]

Hey Thanks so much for the help. It all seems to be in working order now.The trigger tutorial was a great help. I believe it was a combination of the firmware update and turning off filtering, and re calibrating the user a little that did the trick. I was monitoring an i2c line and I can now see data like I would expect. The i2c doesn't seem to recognize data, it only recognizes starts and stops, but I can figure that one out.

BTW SUMP is like the plain-jane of OLS, but it has a 200k buffer as compared to OLS's 24k. It's interface is cruder, but it will work with the sniffer.

Thanks again for the replies and help! :)
-Doug

[jawi]

FWIW: the OLS client does support the original SUMP hardware; just select "Original SUMP device" as device type...

[tayken]

I am happy that you managed to get it running. :)

BTW SUMP is like the plain-jane of OLS, but it has a 200k buffer as compared to OLS's 24k. It's interface is cruder, but it will work with the sniffer.

Meeh, OLS was out when I started using it, I am kinda used to it. Old habits die hard, you know. :)