Can I use the 2 digital triggers on the PCI-6023E to trigger 2 seperate Analog inputs on the same board?

[dimentia]

I want to monitor 2 analog DC signals (single point measurements, not a waveform) and I want to use the digital triggers to start the acquisition, 1 trigger for 1 AI. the measurements need to be made at 1Hz, maybe slower. Is this possible with the PCI-6023E?
If not what hardware should I be looking at? I have a total of 6 triggers and 6 AI.

[aasenjo]

Dimentia,

I would need confirmation on what you want to do:

1) I want to monitor 2 analog DC signals

This means that you will have one Analog Input operation monitoring two channels. Please confirm.

2) single point measurements, not a waveform

Ok, no buffers.

3) I want to use the digital triggers to start the acquisition, 1 trigger for 1 AI

This would mean that a single digital pulse or edge on a single line would start your 2-channel analog input operation. Please confirm.

4) the measurements need to be made at 1Hz, maybe slower.

Hardware-timed or software-timed? 1 Hz could be easily implemented with software timing. But the triggering could force you to use hardware timing. It will not be a problem but it would be good to know.

5) Is
this possible with the PCI-6023E?
Everything I mention above is possible with the PCI-6023E.

6) I have a total of 6 triggers and 6 AI.

This last phrase is what really confuses me and seems to contradict the previous sentences. Does this mean that you will want to perform 6 independent Analog Input operations, with its own triggers (one each)? Will every Analog Input operation have 2 channels?

I'll clarify why this is important:

- Each Analog Input operation can only have one hardware digital start trigger assigned to it.

- Only one Analog Input operation (multiple channels allowed) can be performed per board at any time.

Since the rates are so slow I would recommend that you perform sample all channels within a continuous acquisition and then discard the data that does not meet a certain condition. You could sample the triggers as well and discard those channels for which the trigger has not been received.

These are just some ideas based on my assumptions on the
problem. Assumptions are never a good thing so please confirm on the point s above and we'll try to help.

Thanks

Alejandro Asenjo
Applications Engineering
National Instruments

[dimentia]

aasenjo,

thanks for getting back to me. let me clarify my application. I am upgradeing a test validation system. a total of 6 stations for validating ferrite cores used in ground fault interrupts. each station provides a trigger and an anolog test voltage. I have several 6023E cards which I can use but I want to use the minimum amount of hardware. Each station runs independant of one another so there is a trigger associated with each analog DC measurement. each station makes 1 measurement per second and runs independent of one another. The trigger is not a TTL signal so I though about just doing the triggering in software as an anolog input but in my experience I find that can bog the CPU down and there may be timing issues. If I use several ca
rds (I have 3 available PCI slots) I would like to use hardware triggering if possible. 2 triggers, 2 DC measurements per card with each running independently. does this clarify enough?

[aasenjo]

Thanks for clarifying,

Here are some of my thoughts and suggestions:

- Theoretically, just like I mentioned before, it is not really possible to have multiple independent Analog Input operations within the same board. Only one operation can be performed at a time and only one trigger can be assigned to it. Still, based on your application requirements, I believe you can accomplish it with one or two boards.

-Given the relatively slow sample rates you could sample the trigger lines with a single point, software timed acquisition and then discard or keep the data from your test channels based on the trigger condition.

- If you need your application to be more responsive to the trigger then you could sample even faster (10 times the requir
ed rate) and then discard the samples for the channels that don't meet the trigger condition. The extra samples for the test channels could be averaged to give you the DC measurement.

- If your application can be accomplished with single ended channels then you could sample your triggers and your test channels (for a total of 12 single ended channels). If you need differential channels then you could use 2 boards.

- If the triggers are not TTL compatible but are some other logic level (like 24V) you could still connect them to one of the digital lines through a voltage divider (a couple of resistors to lower the logic voltage). You could then poll the digital lines in your application and watch for the trigger condition and then scan the respective channels. This is assuming that the logic signals don't have a different voltage reference.

These are just some initial suggestions. I really don't think you need to use more than two boards for this application. Let me know if ther
e are any other concerns.

I hope this helps,

Alejandro