Re : Re: Using Non-DaqMX module as master to 4472s

[wired]

Hi Brian,

It's interesting that you had no problem with the 4472s.  Could there be some difference among HW revisions?  I am running the latest version of LabVIEW, DaqMX, etc.  Are you sure you were running the boards at 100KHz each?  Were any of them in Slot 2?

I am not able to run more than one 4472 at 100KHz rate, even when combining them into one task, as I did in my latest post.

With respect to the streaming rate, I realize that we will be BW limited by the hard drive.  I will probably have to come up with a BW calculator to warn the user of potential problems.  For instance, if all 6 slots use the 6120, the higher data rates will not be allowed.  If all size slots used the 4472, the highest data rate should be able to be achieved.  I'll call you today to discuss this further.

Kevin 

[wired]

Brian - I modified the PXI-4472 Collect Data to take data at 5KHz.  The Collect Data vi runs both instances successfully at that rate.

[wildcat_600]

Kevin,

I set up a test system using your same hardware.  I used a PXI-1042Q and then placed two 4472 cards in slots 7&8.  I then ran your Collect All Data.vi that you posted earlier and I did not run into any problems running both cards at the same time.  I was running the VI with the default values (sample rate of 100kHz).  As we discussed on the phone we are using the same software versions (LabVIEW 8.20, DAQmx 8.3.0f03).  Both of your DSA cards are revision G and the 3 different DSA cards that I tried are revisions C, D, & G.  So it is strange that two very similar systems are not producing the same results.

Since our systems are about identical we might want to try a few tests just on the 4472's.  One thing you might try is to remove all of the other cards except the 4472's.  So just try to run the program with just the 8186 in slot 1 and the two 4472's in slots 7 & 8.  This will remove all of the other cards from the bus and eliminate some other variables from your system.

A second test that we might want to run is to move the 4472's from slots 7 & 8 to slots 2 & 3 or two other slots.  This will make sure that it isn't the PXI slots that are causing the problem.

Please let me know if you are able to run these tests.

Brian P.

[wired]

Brian - I reorganized the devices so that the two 4472s were in slots 3&4, and the four 6120s were in slots 5-8.  I ran just the two 4472s at 100KHz with no problems.  I added the 6120 block so that all six devices were collecting (6120s at 500K, 4472s at 100K), and everything ran fine.  Do you have any idea what this could mean?

[wildcat_600]

Kevin,

That's great that all of the cards are now running at their max rates at the same time!  It might be that there wasn't a good connection before when the 4472 cards were in slots 7&8 and reinserting them into another slot fixed the problem.  So you might try placing them back in slots 7&8 and see if you have any luck after reseating them.  When you place them back in 7&8 you might try switching the order of them as well.  So the card that was in slot 8 put in 7 and the same for the other card.  If this works then put the cards back in their original slots.  If it works in all cases then I would say it was just a bad connection.  Please let me know if this works.

Brian

[amineS.]

hello Brian,

I am trying to do an acquisition of acceleration usion a pxi 1042 .I'm workin g with 2 pxi 4472.i've tried to understand this discussion.I am acquiring a result from he slave but the master return the error 200361 it's concerning teh busy state of the buffer used.I've tryed to define a new one with a higher size the same problem, increasing the samples rate...always the same problem.

Thanks

Amine

[amineS.]

my RT is a ni PXI 8184.

Thanks

here you are an simplified example of my vi

[ManLongAtNI]

Hello Amine,

Can you clairfy the intended purpose of your application. It appears you are trying to acquire the signal from your accelerometer? Can you also further elaborate, or maybe include screenshot of the rror you are receiving? 

Regards,

Glenn

[amineS.]

Hello Glenn,

Thanks a lot for your answer.

I'm trying to acquire acceleration using 2 Pxi 4472.I configure the first one located in slot 2 as master and the second one as slave.The problem is that the buffer used by the master is busy and return the error 200361.

I've tryed to use a new buffer with different size but the probelme still appearing.

You'll find a simplified vi doing exactly what i'm asking for.

Thanks a lot.

Regards,

Amine

[wired]

First suggestion: Put each of those acquisitions in their own loop.  I can't see enough of the VI to tell if both are at the same sample rate, but if they're not, this will never work.  Even if they are, it's better to put each in its own loop for multi-threading purposes.

If both boards are at the same rate, you should consider setting up a single multi-device acquisition task. There are examples from NI on how to do that.

If you want both boards to start acquiring at exactly the same time, you can have your slot 2 board generate a start trigger, and both boards can wait on it. You do not need to perform the sequencing on the task start then.

Good luck!