NI DAQCard 6062E

[MI guy]

I've found the most bizzard problem with the NI DAQ card in use for throttle testing. Say, if I collect data while a throttle goes to fully open from close at a rate of 1KHz, the time difference between the 2 positions is ~1.8 to 2 second. If I collect with a rate of 10KHz and higher (20KHz) the time difference shorten significantly to ~0.1 to 0.2 second.
The min scan rate for this type of testing is 100samples/sec. Does anyone experience similar problem?
My OS is win2000, NIDAQ driver 6.9.3. The card is a new
out of a box. In my VB6 program, I collect 4 channels simultanuously. Any help or suggestion is greatly appreciated.

Thanks

[JustinBritten]

Greetings,

Are you saying that you are scanning at a rate of 1Khz or 10Khz with your DAQ card and measuring the time for the throttle to go from fully open to closed? Or are you saying that the throttle is going from open to closed 1,000 or 10,000 times per second and you are scanning at a rate of 100samples/second?

Regards,
Justin Britten

Applications Engineer
National Instruments

[MI guy]

Hello Justin,

Thank you. I've been waiting to see if anyone could give me some pointer.
The throttle goes from open to close (and close to open) in less than 100ms. To capture this movement I
calculated out that 1KHZ scan rate (1000 samples/sec)
should be more than adequate, but in reality I had
to use scan rates greater 10K because the error is
huge at lower scan rate.

[JustinBritten]

If the throttle position changes in 100ms, and you are scanning at 1KHz, then 100 samples will be taken of your throttle going from open to closed. Increasing your scan rate to 10KHz (10,000samples/sec) means that you will have 1,000 samples of your throttle going from open to closed. Depending on the voltage levels that correspond to open and closed throttle, 1000 may not only be recommended, but necessary, in order to accurately capture your throttle transition in the 1/10 of a second period. As always, increasing the scan rate leads to a better representation of the signal being acquired/analyzed.

Now, having increased the scan rate, are you satisfied with the result you are getting? I'm not sure if there is another question here so
mewhere.

Regards,
Justin Britten