0
\$\begingroup\$

I have a 12 GHz oscilloscope with a 40 GSa/s sampling rate from Keysight, and I have a National Instruments GPIB-USB-HS with it. Sampling for 30 seconds with no averaging of data, it only gives 20 waveforms or sets of data. This doesn't seem right.

Is it due to the GPIB speed not matching the sampling rate of the scope? If so, what do you recommend to use? How can I get more data sets with this scope within 30 seconds? I assume that there should be more data sets due to the high sampling rate.

\$\endgroup\$
4
  • \$\begingroup\$ You'll have to gie more details about what you're attempting to do. If you're transferring each dataset over HPIB immediately after it is captured, clearly that is going to be the bottleneck. \$\endgroup\$
    – Dave Tweed
    Commented Jun 9 at 21:22
  • \$\begingroup\$ I’d expect a high end oscilloscope these days would have ethernet built in with a webserver and modern instrument protocols like VISA. Whilst I wouldn’t expect to be able to stream 40 Gs/s waveform data in real time, I’d expect to be able to get waveform captures fairly briskly. GPIB/HPIB is ancient but is what most systems support at a minimum. \$\endgroup\$
    – Kartman
    Commented Jun 9 at 22:11
  • \$\begingroup\$ The GPIB-USB-HS datasheet says it can transfer, at maximum, 7230 kbyte/s. Whether this is the bottleneck in your system depends on what condition the scope is triggering on, what the acquisition gate time is, etc. You might be able to get faster acquisition using the "DIGItize" command to make acquisitions without updating the screen. (But 0.7 waveforms/second is probably limited by something other than screen updates) \$\endgroup\$
    – The Photon
    Commented Jun 9 at 23:04
  • \$\begingroup\$ thank you for all the comments! what I need to do is to get a waveform as fast as possible. I have a pulsed laser of 1 ns duration and 10 MHz repetition rate that I detect with a fast photodetector connected to the scope. I want to do postprocessing so I need the raw waveforms (non-averaged) so I need to get the data streamlined to my computer. do you recommend using a DAQ instead? if so, what kind of DAQ would be useful? \$\endgroup\$
    – science enthusiast
    Commented Jun 12 at 10:56

1 Answer 1

Reset to default
2
\$\begingroup\$

GPIB is OLD bordering on prehistoric, it is not a high bandwidth connection in any sense so your scope is having to buffer captures in its RAM before transferring them.

Reduce the amount of data per waveform, subsample if appropriate (turn down the acquisition rate and work in a higher Nyquist zone), use a much faster bus to pull the data off (The thing probably has an ethernet interface that will probably be quicker, but TBH while this will speed up your workflow, it is unlikely to be fast enough to win much here), or use a scope with more acquisition memory (EXPENSIVE).

\$\endgroup\$
1
  • \$\begingroup\$ thank you! what I need to do is to get a waveform as fast as possible. I have a pulsed laser of 1 ns duration and 10 MHz repetition rate that I detect with a fast photodetector connected to the scope. I want to do postprocessing so I need the raw waveforms (non-averaged) so I need to get the data streamlined to my computer. do you recommend using a DAQ instead? if so, what kind of DAQ would be useful? \$\endgroup\$
    – science enthusiast
    Commented Jun 12 at 10:54

Not the answer you're looking for? Browse other questions tagged or ask your own question.