02-25-2013 02:48 PM
Solved! Go to Solution.
02-25-2013 04:03 PM
Hi Luis,
You are seeing the expected behavior. As you noted, the 9233 has a max rate of 50kS/s and the 9234 a max rate of 51.2kS/s however that is only what they are capable of using their internal timebase. The 9233 specifies a max rate using an external timebase of 51.3kS/s. When you put 9233s and 9234s in the same task, DAQmx chooses the fastest module timebase (from the 9234 in this case) and shares it with the other modules. This has the effect of running your 9233s at 51.2kS/s and is within the module specs.
I'm not sure why you would want to run run the 9233s at 50kS/s, but if you wanted that you could put them in a separate task without the 9234 and they would use their internal timebase to run at 50kS/s.
The other option is to set the sample clock timebase source and rate to a timebase from one of the 9233s. If all the modules were in the same task, your 9234 would then run at 50kS/s as well.
02-25-2013 06:31 PM
Thank you very much for your clear explanation. I really don't want to use the 9233 at 50 kS/s, instead, I would rather a faster sampling rate like the one of the 9234 card. The higher the sampling rate, the better for us. My doubt rather than problem was that it was actually happening with the whole system and your clear it up very well. Thank you very much for your quick response. Regards, Luis.
04-08-2013 01:43 PM
Hello again, first of all, I would like to clarify that we are using 2 cards 9234 and 1 card 9233. As I said beforee, we want to achieve the highest posible sampling rate and start sampling at the same time. Using the DAQ 9174 with these cards and 12 microphones, we made measuments with all the microphones together and a source at 30 cm in front of all the microphones. This experiments were made with the goal of confirm no delay between the microphones, this means achieve a value of 0 or 1 sample between any pair of the measured signals using the cross-correlation method. Instead, we observed the following:
1. Between the signals measured with the microphones in the 9234 cards no delay were obtained. In other words, any correlation between any pair of signals was in the range [0,1], regardless of whether the microphones were of the same card or not (9234s). (Expected behavior)
2. Between the signals measured with the microphones in the 9233 cards no delay were obtained. (Expected behavior)
3. Any pair of signals measured with different card (one microphone in the 9233 and the other in any of the two 9234) and compared using the cross-correlation method produced a value between [20,30] samples. (Unexpected behavior)
Why is this happening? If the DAQ says that it starts the measurements at the same time and all the cards are using a 51.2 kHz sampling rate, all the cross-correlation values should be in the range [0,1]. Is the sampling rate different between the cards? Isn't the DAQ 9174 syncronizing the cards at the beginning of the measurement?
04-08-2013 01:57 PM - edited 04-08-2013 02:02 PM
Hi Luis,
Yes, DAQmx synchronizes the start of the modules and prevents them from drifting but the data returned is time shifted by different amounts. See this KB for more information: Why Is My Data Delayed When Using DSA Devices?
9233:
<= 25.65kS/s: 12.8/fs + 3us
> 25.65kS/s: 9.8/fs + 3us
@51.2k: 194.4us
9234:
38.4/fs + 3.2us
@51.2k: 753.2us
(753.2us - 194.4us) * 51.2kS/s = 28.6 samples. I think I did that math correctly.
04-08-2013 02:50 PM
Thanks for your fast answer. I appreciate it. I guess everything matches now. Those values seem to fix with the processing results acquired by us. We will check this in deep and compare it with our measurements. Again, many thanks. Regards, Luis.