TSW1400EVM Trigger Issue

Hi Thayne,

There are occasional errors that I see, but 75% seems quite low.  First, this is a CMOS input, so your signal levels seem ok.  In HSDCPro, first turn off the trigger mode in the Data Capture Options > Trigger Option. Then verify that regular capture is working consistently.  Assuming that the trigger is going to J11 of the TSW1400 directly with no T connections, re-enable the trigger mode and also choose "arm on next capture button press" .  Then, press the capture button and it should capture data.  If it says "No trigger occured", then this is most likely the Hw trigger signal.

By the way, Which ADC is this?

Thanks,

Chuck Smyth

Hi Chuck,

The ADC I am using is the ADS5562.

I have tried many different debug options and have ruled out hardware/cabling or issues with the trigger signal itself (have also tried this on three different boards and seen the same inconsistencies on all of them).  Upon further testing using a function generator to directly drive the trigger input I determined that if I make the rise and fall times of the trigger pulse 1ms each, the data capture works every time.  If I shorten either the rise or fall times to 100us or less I start to see errors (I need to do further testing to characterize exactly where the threshold is).

For now I can place and RC filter on the trigger signal to slow down the transition times, but would like to find a better solution to this issue for further testing down the road in which I will need precise trigger times between multiple boards...

Do you have any further suggestions?  Or do you see a similar dependency on rise and fall times?

Thank you,

Thayne

Thayne,

I have never had an issue with R/F times.  What is the original trigger generator?  Perhaps the edge is so fast that it is ringing and then double triggering?

Thanks,

Chuck

Hi Chuck,

The issue is not double triggering, because there is no trigger.  If the software triggers once it will not trigger again until the data stored in memory is read off.  And when I press the "Read DDR Memory" button I get the no data stored in memory error message.

The original trigger generator was a PTC901 timer from Omega, but for testing purposes I used a Tek AFG3022 function generator with the output impedance set to High Z.  I  tried R/F times ranging from 18ns all the way to 1ms.  I had some number of errors in all R/F time combinations until I made each of them 1ms.  For further clarity on the test situations, a few of the notable configuration I tried with a 3.3V trigger were:

1. R/F time 18ns, R/F time 100ns, R/F time 1us, R/F time 10us, R/F time 100us, R/F time 1ms,

2. R times of 18ns, 100ns, 1us, 10us, 100us, 1ms with F time of 18ns

3. R time of 18ns with F times of 18ns, 100ns, 1us, 10us, 100us, 1ms

I also zoomed in closely on each of these pulses to make sure there was no extra ringing or artifacts that would lead to a false trigger and tried the same tests on three different TSW1400's.

You said in your original post that you have seen issues with the boards not triggering.  How often does that occur and do you have an idea what the root cause might be?  Does that issue go away if you increase the R/F times of the trigger pulse?

Thank you,

Thayne

Hi Kang,

Sorry for the delay in reply, I was waiting to get these boards back from an on-going system build.

The updated firmware seems to work flawlessly.  I executed ~50 triggers on two boards simultaneously and they triggered every time.

Thank you for your help with this issue,

Thayne

Thayne, 

thanks for the update. We will go ahead and integrate this on the next version of HSDCPRO.

-Kang