Channel crosstalk on PCI4472 card

[holtj]

I am looking for a standardized method of verifying channel crosstalk specifications on the PCI-4472 measurement card. My rough estimates (using approximately +/-10V of input signal) show that I am seeing somewhere around -63dB crosstalk between adjacent channels. Product specs state -80dB to -90dB maximum crosstalk depending on resistive load. I want to make sure I have my troubleshooting process robust before I flag any major concern about the hardware.

My "hack" method of verification includes generating approximately +/-10V of input excitation (using a shaker-to-accel or modal impact hammer), and comparing it to the unloaded response I am seeing on neighboring channels. It appears to be phase-matched, and using the input level as a reference, I am seeing about -63dB crosstalk.

Thanks in advance for any ideas to help ID if this is indeed an issue.
John

[chericks1]

Hi John,

The way that we spec crosstalk is with a tone (between 0-51.2kHz) with an amplitude of +/-10V on one channel and a we short the neighboring channel or terminate it with a 1kOhm load (we spec both cases). If you are attempting to reproduce our spec, that's what you should do.

An impact hammer is used to excite a structure for a vibration test. This way of testing is not as controllable as the way we did it since there is the human factor involved. But, at the same time, it would work to some degree. I just can't tell you how close to the specs you'll get. I think the biggest thing is to make sure that the other inputs are either shorted or terminated.

Let me know if you have further questions or this does not resolve your issue!

Chad AE
Applications Engineer