High-Speed Digitizers

cancel
Showing results for 
Search instead for 
Did you mean: 

Calibration of 5102 with Fluke 5520A in VB6sp5 with niscope2.0; Can't get Verifying Timing to work.

I'm using the calibration procedure for NI5102 doc # 370327B-01 dated Dec 2001. The second part of the test requires 1.8Vpp @10Mhz that is intentionally undersampled. Then it's measured with (scalar) niscope_val_Average_Period. The measured period is approx. .028sec (no where near 1/10M). Another issue is the formula. For the formula to work (equal sample rate) the measured period of 10Mhz needs to be 1.
0 Kudos
Message 1 of 2
(6,156 Views)
Sam,

It seems that you need to send your NI 5102 board for repair.

The 5102 "Verifying Timing" procedure is broken up in two stages (see the 5102 Calibration Manual😞

FIRST STAGE
The first stage asks you to configure a signal generator for a 1 MOhm load and generate a 10 kHz, 1.8 Vpp sine wave. Then you need to connect this signal to the NI 5102 input, and configure the 5102 to acquire 100k samples at 1MHz.
You should be able to measure the frequency of this signal without a problem (freq=1/average period). If the NI 5112 is in good conditions, you should be getting a frequency value that falls between 9999 and 10001 Hz. If not, then you need to send the board for repair.

SECOND STAGE
Now configure the signal generator for a 10MHz, 1.8 Vpp sine wave. Initiate the acquisition of the 5112 with the same parameters that you were using before. Now you should be trying to measure the average period of the signal (freq=1/average period). You should not be able to get this measurement. It is expected that you will receive the error NISCOPE_ERROR_UNABLE_TO_PERFORM_MEASUREMENT.
If you receive a different error, then you need to perform steps 14 and 15 of the procedure.

In the second stage you are not getting the expected error, so we need to go ahead and perform steps 14 and 15. You are getting a period measurement of 0.28. It is OK that this value is nothing close to 1/10M. Remember that you are undersampling your signal.
So if you enter this value to the equation mentioned in the User Manual, the actual sample rate (x) will be:
x =(specified sample rate X source frequency X period)/(source frequency X period - 1)
= (10^6 X 10^7 X 0.28)/(10^7 X 0.28 -1)
= (10^13 X 0.28)/(10^7 X 0.28 -1)
= (28,000,000,000,000)/(2,799,999)= 10,000,003.57

This value is outside of the success condition (999,950 < x < 1,000,050 Hz), so you need to send your NI 5102 for repair.

To do this, please create a service request over the web at :http://www.ni.com/ask and contact the Applications Engineering department.

Best regards,

Claudia Lorente
Applications Engineer
National Instruments
http://www.ni.com/ask
0 Kudos
Message 2 of 2
(6,156 Views)