multi-device synchronization with retriggerable pulse train for continuous acquisition using external digital trigger

[bluerunner]

Hi guys,

I need to synchronize two PXIe-6358 using external digital trigger for continuous acquisition. At each trigger, the n pulse train generated by the counters will be used to acquire n samples.

I am trying to use the Reference Clock synchronization method. Basically, my code(attached) is based on the examples of "Multi-Function-Ctr Retrigg Pulse Train Generation for AI Sample Clock" and "Multi-Device Synch-Analog Input-Finite Acq-Ext Dig Start".

However, when I run the code, it gives me errors as below:

Possible reason(s):

Specified route cannot be satisfied, because it requires resources that are currently in use by another route.

Property: RefClk.Src
Source Device: Dev3
Source Terminal: None

Required Resources in Use by
Task Name: _unnamedTask<C2>
Source Device: Dev3
Source Terminal: PXIe_Clk100
Destination Device: Dev3
Destination Terminal: RefClockInternal

Task Name: _unnamedTask<C4>

"

It seems that the counter cannot use the Reference clock. 

Please tell me how to solve the problem? 

Besides, I am not sure whether my code could satisfy my application, or are there any other better ways?

Thanks!

[KatieC]

bluerunner, 

Thanks for using NI forums! You may want to look at the Multi-Device Synch-Shared Ext Sample Clk. vi; it seems more applicable to your application than the ones you looked at. 

I would specifically look at how they are doing the sample clock source. 

Let me know how that goes!

Katie

[KatieC]

Also--I had one more thought: try outputting the counter to a PFI line rather than routing internally so that both sample clocks can run from the same counter task.

Katie

[bluerunner]

Hi Katie,

Thank you for your suggestion.

I added another DAQmx Timing properity node to set the RefClk.Src and RefClk.Rate for the counter, using the same source and rate, as in the attached file. It is working now.

You suggestion is based on the Sample Clock Timebase Synchronization, right? It seems easier than the ones I use. I will try it and let you know the result.

I have another question about the phase of the sample clock. Are all the sample clocks in the multi-devices in the same phase? In other words, do they sample the siginal at the same time? In terms of keeping in phase, which method is better, the reference clock synchronization or the sample clock synchronization?

Thanks again for your kindly help.

[KatieC]

Bluerunner, 

The example is based on sample clock synchronization, yes. Using this example, the devices will be in phase and take samples at the same time.

In terms of whether sample clock or reference clock is better, it really depends. The reference clock is coming from the chassis so all the cards will see it at the exact same time. When using the sample clock, you have one master and several slaves. The master will see the the pulse slightly before the slaves, and the slave in the slot right next to the master will see it slightly before the one farther away. The time delay, the propagation delay down the backplane, isn't noticeable for regular DAQ devices. You would need to be over 10MHz to really notice a difference. Our DSA cards, for instance, would be able to tell a difference so for them it would be better to use a reference clock. For your application with the DAQ cards, sample clock synchronization will work just as well as the reference clock. 

Also, here is an article on basic synchronization you may find interesting. There is another article that is particular to DSA cards as they have different requirements. 

Katie