error -200452

[dmourer]

I am trying to synchronize a 4472 analog input card with a 6731 output card. I am getting the error:

Error -200452 occurred at Property Node DAQmx Timing (arg 1) in Gen Mult Volt Updates.vi

Possible reason(s):

Specified property is not supported by the device or is not applicable to the task.

Property: MasterTimebase.Src

Task Name: _unnamedTask<1>

If this operation is not supported by the 6731, then how do I synchronize these two boards?

I am attaching my vi.

[dmourer]

I am trying to synchronize a 4472 analog input card with a 6731 output card. I am getting the error:

Error -200452 occurred at Property Node DAQmx Timing (arg 1) in Gen Mult Volt Updates.vi

Possible reason(s):

Specified property is not supported by the device or is not applicable to the task.

Property: MasterTimebase.Src

Task Name: _unnamedTask<1>

If this operation is not supported by the 6731, then how do I synchronize these two boards?

I am attaching my vi.

[reddog]

The problem here isn't the 6731 but the 4472. The 4472 uses a different timing/synchronization model than the 6731 and doesn't support the master timebase property. Instead, it uses a sample timebase signal which is roughly the equivalent of the master timebase on the 6731. To get at the sample timebase signal, you normally use the SampClk.Timebase.Rate and SampClk.Timebase.Src properties in the Timing Property node. However, the problem here is the sample timebase signal for the 4472 can only be imported/exported on RTSI 8, and the master timebase signal on the 6731 can only be imported/exported on RTSI 7. You can verify this by looking at the routing tables in MAX. This means you can't share these signals between the two devices on PCI. I'm assuming you're using PCI because you mentioned a RTSI cable on your other post. If you're using PXI, there are a couple of other options available for sharing these signals by placing the 4472 in slot 2 of the chassis. If this is the case, let me know and we can explore this option further.

With all of that being said, there are still some options for synchronization that are illustrated in the attached example (the example is based off of the VI you originally provided). First, if you want to start the generation and acquisition at the same time and have them sample at the same rate, set the Synchronized Sample Clocks boolean control to true on the front panel. Using this method, the two tasks will share the same sample clock and start trigger. With this synchronization method, there may be some small phase offset between when the two tasks start, but the offset should remain constant throughout time. If you want to have different sample rates for the AI and AO tasks, you can set the Synchronized Sample Clocks boolean control to false, and the tasks will only share a start trigger. Using this method, the two tasks may have a small phase offset when they start, and the tasks will gradually fall farther out of phase as the task runs. The rate at which the tasks fall out of phase depends on the accuracy and drift of the oscillators used on the devices. The 20 MHz oscillator is typically accurate up to +/- 0.01% and the DDS on the 4472 is accurate up to +/- 25 ppm. If you're only performing finite operations and your sample set is small, this may not even become an issue. However, if you're running long continuous generations/acquisitions, you will likely notice the phase drift over time. To run this example on PCI, don't forget to first create a RTSI cable in MAX and add the two devices to the RTSI cable. Without this step, the driver won't know that it's possible to route signals between the two devices.

Finally, I should mention there is an intrinsic filter delay in the ADC's used on the 4472. This filter delay is 38.8 samples for the 4472. If this delay is causing problems with the synchronization, you should be able to use a counter on the 6731 to delay the start trigger and compensate for it. However, I won't go any farther into this unless you find it's a problem. Anyway, give the VI a try and let me know if there are any other problems. Good luck!

[reddog]

Apparently the attachment didn't take the first time.

[dmourer]

Reddog,

Thanks for your help. I need to have precise timing between the AO and AI. The sample rates are different. The 4472 is scanning at 5kHz and the 6731 sample rate is 1/15 Seconds. The test duration is 180 seconds. I need the AI and the AO to start and end at exactly the same time.

Dave

[George M.]

Hi,

You can use start trigger and stop(reference) trigger for an AI task. You can use start trigger for an AO, but there is no reference trigger for AO. The AO will have to be stopped through software. I hope this helps.

George

[reddog]

I just thought of another way you might be able to synchronize these devices as shown in the attached VI. This VI uses the sample clock from the 4472 as the sample clock timebase for the 6731. The 6731 then divides down this timebase by an integer amount for use as its sample clock. To calculate this divisor, the VI divides the AI sample clock rate by the desired AO sample clock rate. For example, to achieve an AI sample rate of 5 kHz and a desired AO sample rate of 1 sample every 15 seconds, a divisor of 75,000 (5,000 / (1/15)) is used. I hope this helps.