Synchronization of 2 chassis cDAQ9174 and cDAQ9171

[jfmauger]

Hi there,

I have been working perfectly with a cDAQ9174 with 4 different modules in it (9211,9217+2x9239) and now, I am trying to add a new cDAQ9171 (w/ 1 module 9211) in my measurement setup (I need 4 extra TC measurements).

For getting my measurements, it seems that I need to synchronize those 2 chassis but I don't know how to do it!

I have seen that a specific module for synchronization exists (NI9479) but this module only works with compactRIO and not cDAQ chassis.

Any idea on how to work with my current setup?

Thanks for your help.

Jeff

[Cavarval]

Hi Jeff,

If you mean the NI 9469, it is compatible with CompactDAQ, but it is still not a solution for you because it would require the cDAQ-9171 to have a second slot anyway (there's gotta be one NI 9469 in each chassis you want to sync).

Multi-chassis synchronization like you need is possible as long as you either have room for that module or at least, a digital module, in order to share the clock and trigger signals between the chassis. Since the cDAQ-9171 only has one slot, it unfortunately cannot be synchronized.

If you really need to synchronize the measurements you will probably need a second 9174 and two digital modules able to use counters to share the signals (two of these, except the NI 9361, the TTL ones will do fine), but probably, while a little bit more expensive, upgrading the cDAQ-9174 to a cDAQ-9178 will be better for the extra room and synchronization simpleness.

Temperature measurements are probably going to be considerably slower than the NI 9239s. Is there a specific reason you need the measurements to be synced?

Regards,

[jfmauger]

Hi Camilo,

Thanks very much for your detailed answer!

Actually, I thought that I had to synchronize them if I wanted to get measurements from both chassis but you are right, I don't really need it.

So then, I am facing a second issue....

I have an error message saying that my specified sample rate (20k) is higher than what can be supported by the device.

Any idea again?

Thanks a lot.

Jeff

[Cavarval]

Hi Jeff,

Well, from all your modules, only the NI 9239 can go that fast. Probably you are setting that sample rate to the RTD or TC modules.

The NI 9211 can update 14 times a second, divided by the number of channels you are using. If you tell it to go faster, it will, but will just give you duplicate samples.

NI 9217 can go up to 100 S/s per channel, so probably that's the one that is throwing the error. If you have all the modules from the cDAQ-9174 in a single task, you can probably separate them into one task for the higher speed NI 9239s and another for the lower speed temperature modules.

Hope it helps,

[bstreiff]

I'm assuming that you now have:

• An AI task ("Task A") with your 9174 that contains your 9211+9217+2x9239 at 20kS/s, and that works.
• Another AI task ("Task B") with your 9171 that has only a 9211, and you get a 20kS/s-is-higher-than-supported error.

A sample rate of 20 kS/s works in Task A because you have a module (the 9239, max rate 50 kS/s) that is capable of a rate of 20 kS/s. In this case, for a module with a slower sample rate (such as the 9211, max rate 14 S/s), the chassis will return repeated samples. See the section "Slow Sample Rate Modules" in the NI cDAQ-9171/9174/9178 user manual.

In the case of Task B, there is no fast module; you only have the 9211, because it occupies the only available slot on the 9171. In this case, you're limited to the maximum rate of the 9211, which is 14 S/s.

[jfmauger]

Hi Brandon,

Thanks for your reply.

Your description is absolutely right, I have 2 different tasks, one with the 9174 and its for modules, the other one with the 9171 and its 9211 module.

I have fixed my problem of sample rate but I can't even go to 14S/s, the max is 2.8S/s...not a big deal.

Sorry about my poor skills in Labview programming but now, I am facing another issue...

It seems that the data acquisition for my second task never starts.

I have a 30s timeout set with my DAQ assistant and then, after 30s, I receive an error message saying basically that no sample was acquired.

Again, any idea?

Thank you so much!

[martinandre]

Hi Camilo,

I have two cDAQ 9178 with eight NI 9234 each one. I read in this link (http://www.ni.com/tutorial/5376/en/) that I have the option to syncronize the two cDAQs without using a sync module, but just connecting two cables between CDAQ1 and CDAQ2 using their PFI 1 and PFI0 BNC conectors, one for the trigger and the other one for the sample clock.

But in the block diagram an error ocurred, it says: 

ERRor 200414 occurred at DAQmx Start Task.vi: 7220004 Possible reason: Requested sample clock source is invalid Property: SampClk.Src Corresponding Value:/cDAQ2/PFI1.

Could you help me please? I dont know what is going on.

I'd appreciate your answer

[Cavarval]

Hello,

The problem here is that Delta Sigma C Series Modules do not support sample clock synchronization, just as mentioned here.

You cannot set the SampClk.Src attribute/property to an external source when a C Series Delta-Sigma device is in the task.

I thought of a possible workaround exporting / importing the sample clock timebase instead. I am not sure you can do that either with this hardware and I am not currently in a place where I can test it. You can try it and let us know if it works.

Now, if it works, there should be no clock drift among the chassis, but there will be a skew that cannot be easily removed just because the timebase has to go through the cable from one chassis to the other. If you need ulta-tight synchronization, the 9469 is the way (DAQmx implements methods to remove that skew if the synchronization module is used).

So, you cannot use the DAQmx Export Signals.vi to export the Sample Clock Timebase from the 9234 through the 9178's PFIs. You need to use the DAQmx Connect Terminals.vi (I suggest, before creating the task) and DAQmx Disconnect Terminals.vi (I suggest, after clearing the task). The connection that vi creates is not tied to the task, that is why clearing the task is not enough; you must either disconnect the terminals manually using the VI or reset the device to clear the connection.

Also, the Sample Clock Timebase from the 9234 will only show up as a source for the DAQmx Connect Terminals.vi if you include the advanced terminals in the constant.

Then, you can import the timebase in the task from the second cDAQ using the Sample Clock » Timebase » Source property. You also need to specify the Sample Clock » Timebase » Rate, just as shown in the picture below (the snippet is from LV 2016).

The 13.1072 MHz value comes from here.

I am not completely sure this will work and if it doesn't, you'll need the 9469s. I can test this later this week if you haven't been able to do it before. If you do, please us know if it runs with no error. (I am actually eager to know ).

Edit: As a side note, if this happens to work and because this thread was originally about a slightly different thing, it would be good to create a new thread about this. Something like "Multi-Chassis 9234 Synchronization Without the NI 9469" or something like that so it is easier to find for other users. 

Regards,

[martinandre]

Hi Camilo,

I tried to do that but unfortunately I continue having no syncrhonization on my signals.

I am surprised because i was expecting a constant phase but as you can see below, the phase is moving out.
Now I am sure that I can not syncrohize signals in my case without using the sync module.

It was a nice try. I learned a bit more.

Thank you, and best Regards!

[martinandre]

Dear Camilo,

I was editing some things on the software just for fun and for look what I get.

And I took out the sample clock trigger, and putted the two DAQmx Read VIs in just one while loop, and my results got better.

I leave some images attached.