Logic Analyzer - Data Acquisition on Trigger - Files saved show no triggered data

[Slawek]

Using Logic Analyzer on an AD2, I am reading six digital inputs, and one I2C bus and have enabled logging to CSV file upon a "triggered acquisition".  I have gotten five triggered acquisitions over the course of two days but the CSV files show no digital transitions or I2C packets at all.

1) Why is this happening?
2) Can I somehow preview the the prior captures in Waveforms?

This is very important for the project I'm working on.

Version 3.8.2 64-bit Qt5.6.3 Windows 7

acq2019-04-04 10.07.21.4.csv

[attila]

Hi @Slawek

The trigger positions in your capture is at 10ms. You are capturing from 7 to 12 ms after the trigger event and I see no activity on the I2C lines.

You can look back the last 10 captures (by default) using the Buffer option. The highest number is the last capture.

[Slawek]

Hi @attila. I'm pretty confident I had "Position" field at 0s prior to arming during my previous log attempts. It just happened to be at 10.12ms in my screenshot due to me searching around for the transition edge. My I2C bus is quiet.

The csv I attached shows differently, but I've had csv log files that go from -0.0007 to +0.004415 seconds which suggests that the trigger point (0.0000 seconds) is where an edge would be present. If your output file saves times relative to the "Position" setting, then that is very misleading, it should be absolute from the time of trigger.

I stopped the run and went down through the buffer by using the down arrow. The capture on the screen only decrements by 10ms (at 500us/div, my window is 5ms). You seem to suggest it'll show previous captures that have occurred in the past which in my case can be  many hours. How do you define capture?

[attila]

Hi @Slawek

The samples are relative to the trigger position, zero time.
The capture date-time is low resolution software stamp, the device has no RTC to identify at sample level the absolute time at nano-second level.

[Slawek]

If Time="0" is present in a CSV capture, how could there be no 0->1 or 1->0 transitions recorded on any channels?

EDIT: Assuming that capture was for a "triggered acquisition".

[attila]

Hi @Slawek

In your attached CSV the trigger T0 is out of the capture domain. You have captured from 3ms to 8ms.

Time (s),DIO #,DIO #,DIO #,DIO #,DIO #,DIO #,I2C
0.00300125,1,0,1,1,0,1,X
...
0.00812,1,0,1,1,0,1,X

[Slawek]

I know, that's what I said in my prior edited comment.

On 4/5/2019 at 8:29 AM, Slawek said:

The csv I attached shows differently, but I've had csv log files that go from -0.0007 to +0.004415 seconds which suggests that the trigger point (0.0000 seconds) is where an edge would be present.

If in the CSV, there is no 0->1 or 1->0 transition at time = 0, then this is a bug of some sort.

Is the low level hardware telling the software there is a trigger, but it cannot appropriately send the data upstream to the software?

[Slawek]

@attila This keeps happening. I think this is a bug in your software or hardware.

I keep getting triggered captures and at T=0 there are no transitions present!

Very annoying!