DMM 4070 External Trigger Issue

[groz]

I have an acquisition question regarding the PXI-4070 DMM.

I'm trying to use the external trigger input that comes in on the Aux I/O connector (pin 9). Note: My signal source ground is tied to pin 2.

Current Setup Steps (niDMM):

- Init - Configure Waveform Acquisition

- Configure Waveform Coupling=DC

- Configure Trigger=external

- Read Waveform

- Output Waveform To Graph

- Close

The trigger works in that it will timeout without it and does in deed capture and present the waveform with it. The problem I'm having is the waveform data is not consistent based on the trigger edge. That is, the waveform data captured does not start on the selected edge. I've played with both range and edge but cannot get repeatable timing.

I've been using the DMM to capture and present waveform data using trig=immediate, followed by a producer consumer loop (works fine for periodic waveforms, all edges are software derived). The new measurement requires me to trigger only after a current (signal pulse from LEM measure circuit) has reached a certain value. This type of measurement is very typical for o-scopes.

Is the 4070 capable of performing this type of measurement?

[Mr.F]

Hi Groz,

Unfortunately, the PXI 4070 does not natively support that type of triggering while in digitizer mode. I’d encourage you to file a product suggestion at ni.com, select Contact NI and then Feedback for any features you would like to see in our products or future products you would like to see.

Connecting your signal in that manner may cause your acquisition to trigger at indeterminate times, if it does at all. Your DMM is expecting a TLL pulse and won’t get consistent perform triggering with signals that are not TTL compliant. I’d recommend you use an external comparator and reference the level you want and then trigger from the output of that comparator. Also, make sure the Trigger delay value is specified to 0 if you want it to begin immediately.

[groz]

Matt,

The signal I'll be feeding into the DMM Trigger input is a TTL pulse. The pulse comes from a comparator circuit (0 to 5v) that compares a D/A level with the LEM output. Note: The D/A comes from a buffered 6723 output that allows me to set this threshold value. When it meets the threshold being looked for, the trigger pulses from 5v to 0v which then should start the DMM acquisition.

Looking at the trigger specs for the DMM card, it's suppose to allow TTL levels AND have a latency of no more than -1.8uS from trigger edge. The spec also says this can be compensated via the trigger delay parameter. In my application, the slight skew is tolerable. What is not acceptable, is the shifting in the mS region from trigger edge which I'm seeing.

The setup I'm using to test the code is operating within the same TTL levels i.e. 0 to 5v 1KHz square wave being fed into both the measurement and trigger lines. This setup makes it ideal to see skew which is how I caught this error. From what I can see, I'm not doing anything outside the hardware spec.

[Mr.F]

Hi Groz,

That doesn’t sound fun. Have you taken a look at the example: “Triggered Waveform.VI” in the example finder? This uses the “Initiate” and the “Fetch” Vis so the DMM is actually acquiring samples when it gets the trigger. Try this out and see if you have better results.

[groz]

Matt,

The problem I found was the DMM is queuing the Trigger during the setup/init phase. I would think the DMM would gate off the trigger until ready, not so. When the card leaves this phase, it sees the trigger and immediately performs an acquisition. Since this time shift can vary based on setup, the acquired waveform will not be correct in the time domain.

I've asked for a documentation change at the very least to point this out.

Regards,
Groz

[Mr.F]

Hi Groz,

I see your point. I also see you are working with Adam on this issue. I’m going to sync up with him as to why this is.

[groz]

One last point should be made when using the external trigger option. The settling time property must be set to 0 if you are trying to capture a signal at the triggers edge.

Groz

[enderWiggin10]

Hi Groz,

Thank you for documenting the issue.  I want to link the KnowledgeBase created to document the behavior as well.  It is being put onto the server soon, so give it a try in a couple days if it does not show up.

Best,