Unable to get synchronization between PXI devices

[Damill]

We´re running a new PXI system with NI-DAQmx 7.4 and LabviewRT. We´re programming in LabView 7.1. The system has a 6254 DAQ-card and a 6602 counter card. We´re trying to send the "sampleclock" from 6254 to the gate of the 6602 ctr0 by using DAQmx routing on the PXI trigger line 1. This does not work. When we instead test to route "sampleclock" to PFI0 locally on the 6254 connector there is no signal available on the output pin although a buffered measurement is running. No warning/errors are encountered. Doing the same routing with the signal "reference clock" seems to work fine ! We have done this thing several times using RTSI in PCI based systems and it works fine. Now we can't even find any traditional NI-DAQ configurations in the preinstalled NI-DAQ on the PXI-system so we seems to be forced to use the DAQmx environment.

Anyone out there who can give us a hint on what to do???

Regards, Dan Larsson/Damill AB Sweden 

[Jimmie_A.]

Hello Dan!

Sorry for the delay but I have been looking for a suitable connector block so I could do some further tests! I have now been able to create an application where the sample clock from a E-series device has been routed to a counter timer board so they now are synchronized. 

Hope this helps!

Regards,

Jimmie A.

Applications Engineer, National Instruments

[Damill]

Thanks Jimmie,

The code you gave us does not work with the environment we have, an PXI-1052 chassi, a 6251 A/D card and a 6602 counter-card. On the other hand it works fine when use it for routing internally on the 6251 (to its built-in counters) or if routing the sampleclock signal to the external connector pin PFI0 of the 6251 where we connect a cable to the connector pin PFI0 of the 6602. Maybe we have some defect on either of the components making backplane-connections unusable.

Regards, Dan

[Jimmie_A.]

Hello!

I used the very same code I attached before in a PXI-1000B chassis with a PXI-6289 and PXI-6602 and it works flawless just like I expected since I just use the backplane of the PXI to route the signals. I am not sure what might be wrong with your setup or if you have faulty HW. Please try to isolate the problems if possible since the code, the driver, the boards and the chassis works fine here. 

I am really sorry that you are experiencing these problems and that it has caused you lots of extra work.

Regards,

Jimmie A.

Applications Engineer, National Instruments

[Damill]

Hello again,

Just to end this discussion properly for you other readers:

After further tests the problem has turned out to be a faulty chassi (PXI-1052).

Regards,

Dan Larsson, Damill AB, Sweden     

[Damill]

WARNING !!!!

This story is not yet over.

The replacement (a new PXI-1052 chassi) did arrive today.

The problem is still there!

Now we are both disappointed and irritated with the support given. 6 weeks of recommendations to test that and that and the problem is still left. 

Dan Larsson

Manager, Damill AB, Sweden    

[Adam_U]

Damill,

I'm sorry you have been having these problems with your PXI system. This is a very unusual problem and we'll continue to work with you on solving it.

The 1052 chassis does not need any trigger routing (like is necessary in the 1045 or old 1006 chassis) since it has a contiguous trigger bus. The identification of the chassis (in MAX) is used so trigger reservations and slot identification is possible. It is highly improbable that you received a single defective 1052, and even more improbable that you received two 1052 chassis defective in the same way. I think that we are likely either seeing a problem with the boards in the chassis or with the driver software for your setup.

To narrow down the possibility we are seeing a software problem, I would like you to try and identify the 1052 as another chassis that does not require trigger routing (such as the 1042) to see if your problem persists.

Additionally, do you have another PXI-6602 or PXI-6289 that you can try to help remove the board's hardware from the list of possible culprits.

-Adam U.

[Damill]

Dear Adam,

Results of tests so far:

PXI-1000B with PXI-8184 RT: Sync ok

PXI-1000B with PXI-8176 RT: Sync ok

PXI-1052 with PXI-8184 RT: Times out and errors

PXI-1052 with PXI-8176 RT: Times out and errors

PXI-1052 with PXI-8176 Win2000: Times out and errors

The problems seems not related to the 6602, 6251 or the controller.

Your colleague Eric Lee has in email dated 18 August told us that he has tested PXI-8186RT + PXI-1052 with success ?

Regards,

Dan Larsson 

[Roger Isaksson]

Adam thanks for the suggestion, this message is on behalf of Damill,

I have now tested the chassis (running as a 1042 in MAX) with the 8184 controller, with the same negative result! See the attached image for the MAX setup.

Is there any possibility you could set up an identical system for test? A 8184 controller in a 1052 chassis?

Regards

Roger Isaksson, Damill AB

[Adam_U]

I will set up an identical system first thing Monday morning. I was not able to reproduce the failure under Windows XP, DAQmx 7.5 with a 1031 identified as a 1052 last night.

Have you tried DAQmx 7.5 on any systems yet to see if they exhibit the same behavior? If not, I would encourage you to do so to help parallelize our efforts.

Talk to you soon,

-Adam