AnsweredAssumed Answered

problem with the 86100a scope driver in labview

Question asked by xavt on Aug 17, 2015
Latest reply on Aug 30, 2015 by xavt

Like • Show 0 Likes0
Comment • 5

Hello,

I installed the agilent 86100a driver through labview2015 and im running the examples but all return error1, as can be seen in the attached pics. Im specifically trying to run the "take continuous measurement" VI but all other examples have the same error. Running a different VI (Like autoscale.vi) works fine.

The hardware works fine and is apparently properly installed since it responds to *IDN? and AUToscale commands through MAX as can be seen in another attached pic.

Im running it on win xp s3 using a gpib card connected to the scope.

Please help me get this to work!

Thanks in advance,
Vitali

Attachments

error pics.zip233.6 KB

No one else has this question

Outcomes

5 Replies

JohnDorighi Aug 19, 2015 5:44 PM
Mark Correct
Correct Answer
Hello Vitali,

Thanks for your questions, I think our best option here would be to use the direct SCPI commands which you have verified work through NI Max. I typically find this to be the most efficient way to automate the 86100A/B/C/D DCA. Another benefit to using the direct SCPI commands is that your code would also work with the new 86100D mainframe using the 'Classic GUI' if you need to migrate to the current mainframe platform at some point in the future. (Although there are many benefits to using the new FlexDCA GUI on the 86100D if/when this is of interest) In the short term, here are the commands we can use with the Classic Gui which runs on the 86100A (and the 'Classic GUI' which can run on the 86100D).

http://www.keysight.com/upload/cmc_upload/All/86100_Programming_Guide.pdf

I have a number of examples written for sending SCPI commands in Labview to the 86100 DCA's. Attached is an example which transfer trace and screen images from the DCA. Let me know if this helps and we can determine the next steps.

Regards,

John
Attachments
- 86100 Trace_Screen Capture.llb198.0 KB
Like • Show 0 Likes0
Actions
- xavt @ JohnDorighi on Aug 20, 2015 6:22 AM
  Mark Correct
  Correct Answer
  Thank you for your reply, John.
  
  I have done as you suggested and I'm somewhat successful at pulling the data from the scope "continuously". Before I attempt to do anything further perhaps I should ask you the question that lead me to automating the measurements in the first place.
  
  I have a 4 GHZ sine wave with slow amplitude modulation coming into channel 1 and I'm triggering it with itself by splitting the cable. I want to have at least a full second of the signal in order to be able to observe the slow AM envelope that rides the 4GHZ carrier.
  
  Trouble is that if i set the time scale to a second or so, the sinusoidal nature of the signal is lost. I understand that the reason is that the scope can measure up to about 4500 points regardless of the sampling timescale. Therefore I get 4500 points for the 1 sec duration which is not sufficient to capture the signal (due to Nyquist sampling theorem and so on).
  
  This set me on the idea of sampling a short time and dumping the data on the computer where it can be assembled into a full 1 second of well sampled data.
  
  I hope I managed to explain myself above if not then i would be more than happy to provide pics and further explanation.
  
  Am I missing something? Is there a very simple solution to do what i need and Im just overcomplicating things by using labview?
  
  Thanks a lot in advance, I much appreciate it!
  Vitali.
  Like • Show 0 Likes0
  Actions
JohnDorighi Aug 21, 2015 8:37 AM
Mark Correct
Correct Answer
Hello Vitali,

Thanks for th follow-up; there are a few options here we can explore. If it is possible to trigger the scope on the modulation envelope (and if the modulation remains constant over the aquisition time) we could use the min/max math functions to display the modulation envelope (Measure -> Math -> Operation -> Min/Max). This allows us to adjust the timescale of the scope to the modulation timespan rather than the carrier timespan (see attached Sync). A point to think about is the 86100 is a sampling scope, we acquire 1x data point for each trigger event and bulid up a waveform over multiple trigger events. If the modulation is asynchronous to the trigger (i.e. if we trigger on the carrier) then we can view the min/max of the modulation envelope on the carrier (see attached Async) but we are not able to see how it changes over time (see attached Extended time). If the modulation is not repetative and we need to see how it changes over time we might need to look at using a real time scope. Let me know if this helps and we can determine the next steps.

Regards,

John
Attachments
- 86100_Modulation_Sync_Min_Max.png84.0 KB
- 86100_Modulation_Async_Min_Max.png78.8 KB
- 86100_Modulation_Async_Extended time.png88.1 KB
Like • Show 0 Likes0
Actions
- xavt @ JohnDorighi on Aug 25, 2015 10:45 AM
  Mark Correct
  Correct Answer
  Thanks again John,
  
  Weirdly I don't seem to have min/max under math. I only have 'magnify','invert', 'subtract' and 'versus' options under measure->math->operator. It might have been a good solution but since the channel is expected to be potentially very noisy the min/max might not be the best approach anyway.
  
  I was planing on implementing a lock-in amp in labview that will detect the envelope from the channel despite noise.
  
  I have written a similar VI to the one you attached but with GPIB, since for some reason VISA is not always working properly. As you can see there is a timed loop to get the data from the scope. Trouble is, as shown in an attached pic, that the measurement isn't continuous and the sine wave is cut. I tried changing the time of the loop (you can change it during runtime using a nob) but below a certain time the data isn't displayed at all. Do you know how I could get it to display any arbitrarily long signal smoothly?
  
  Another issue which occurs (even in your Trace_Xfer.vi) is that the time is seemingly incorrect. I've zoomed in on a single sine cycle which I know to be 10MHz, but the period shown seems to be closer to 100MHz, do you know why that might be?
  
  As before, your help is very much appreciated.
  
  Regards,
  Vitali.
  
  Edited by: XAVT on Aug 25, 2015 11:45 AM
  Attachments
  - Untitled 1.vi21.5 KB
  - skidding samples.png100.0 KB
  - wrong time.png11.2 KB
  Like • Show 0 Likes0
  Actions
xavt Aug 30, 2015 8:44 AM
Mark Correct
Correct Answer
Another possible solution is if I could get the current time from (say) the first sample in the batch of samples that are returned by 'WAV:DATA?' and then the time for the first sample in the second batch and so on. That way I could have them ordered properly in time and I would be ok with missing some samples in between the batches as long as the signal remains in phase.

Is there a way I could get the current time from the scope with nanosecond precision?

Thanks!
Like • Show 0 Likes0
Actions

problem with the 86100a scope driver in labview

Attachments

Outcomes

Attachments

Attachments

Attachments

Related Content

Recommended Content