Instrument Control (GPIB, Serial, VISA, IVI)

cancel
Showing results for 
Search instead for 
Did you mean: 

VISA and virtual GPIB on Tektronix oscilloscope

I talked to Tektronix myself and it doesn't appear that they have any support for NI-VISA with their virtual GPIB. However, you should be able to use VXI-11 on the localhost with NI-VISA, so you can give that a try.
Message 11 of 23
(4,798 Views)

Well, it is getting pretty confusing an affair. First, Tek makes an impression that TekVISA is indispensable. Now I found the following (readme-file for TekVISA by Tek):

1) Two distinct implementations of Virtual GPIB are provided: for NI-VISA version 2.5 or newer and for version 2.0 of TekVISA.  In both cases, the Virtual GPIB resource is opened via the resource string “GPIB8::1::INSTR”.

2)  The Virtual GPIB resource for NI-VISA 2.5 is a passport DLL. 

 

And Tek support writes: “TekVISA includes NI Passport.  That is an important component
towards simplifying TekVISA with NI applications and NI-VISA with Tek
utilities, however, it is not the answer to complete inter-changeability
of the VISA systems on all PCs and Windows scopes”.

 

Well, when I use NI VISA 2.6 and type “GPIB8::1::INSTR” into VISA resource, I get error: “The interface type (GPIB8::1::INSTR) is valid but the specified interface number is not configured”. 

 

How do I configure it? Or I've got a wrong "implementation of Virtual GPIB"? And why should I care whether TekVISA includes NI Passport, if I don't want to use TekVISA?

Advise, please.

Thanks,

Maxim

0 Kudos
Message 12 of 23
(4,786 Views)
All right, I've done it: replaced the TekVISA by NI-VISA and still able to interface the scope application via VIRTUAL GPIB. The key point was: not to uninstall the TekVISA, but to overwrite VISA32.DLL with NI-VISA. Now there are no problems with the COM port. Apparently the "passport dll" remains visible to NI-VISA (not to me, though).
Given that the magior part of the posts here are mine, it is rather a Blog than a Discussion Forum. I am very disappointed that NI didn't contribute. Since long time I have the feeling that NI has got a Microsoftish attitude towards their users: difficult to get a transparent help (instead of "open here and click there"). Perhaps, they've got too strong a leading position on the market. Maybe it's better to drift to another producer, to refresh the competition.
Sincerely,
Maxim Morozov
Message 13 of 23
(4,755 Views)
Why do you have the feeling that NI did not contribute?
Message 14 of 23
(4,752 Views)
    I was able to talk with normal visa commands of LV7.1 by using the USB conncection.  I'm communicating with a Tek DPO4000, which has the Window's 2000 operating system.

My only problem is getting data with more than 10k points.  The scope has a setting for up to 10M points, I cannot get more than 10k through this communication , maybe it times out, but its funny that it is always 10k.


0 Kudos
Message 15 of 23
(4,748 Views)

To GPIB Guru: I conclude that NI did not contribute to resolving the given problem since:

1) there hasn't been a replay signed by a "NI engineer" which would give me an unambiguous instructions;

2) I have sent an e-mail to NI support with a request for assistance more than a week ago and haven’t receive a response.

Regards,

Maxim


0 Kudos
Message 16 of 23
(4,742 Views)

XBAND wrote: “I was able to talk with normal visa commands of LV7.1 by using the USB connection.  I'm communicating with a Tek DPO4000, which has the Window's 2000 operating system. My only problem is getting data with more than 10k points.  The scope has a setting for up to 10M points, I cannot get more than 10k through this communication , maybe it times out, but its funny that it is always 10k.”

 

- The limited number of waveform points possible to transfer from a digital scope is a usual problem for Tektronix as well as LeCroy windows-based scopes (perhaps unless classic GPIB is used).

I am afraid the reason is that their VISA (respectively XStream on LeCroy) API implements an object-model for the scope and the Microsoft COM (component object model) does not enable the “variant” variable to accommodate too big an array. Again, the producers should know, however they keep quiet on the details.

Regards,

Maxim

0 Kudos
Message 17 of 23
(4,739 Views)
Not all NI Engineers sign their replies. If you look at the boxes under the user name, if they are colored blue, they are NI employees. If they are colored yellow, they are not NI employees.
0 Kudos
Message 18 of 23
(4,729 Views)
An update from my last post, it seems the 10k point limit was due to an internal software setting in the scope, I was able to retrieve waveforms with 100k points, but have a timeout  communication error when trying the 1M or 10M point waveform download.

I'm not sure if the previous persons scope has the usb port, but the communications over the ethernet connection should work too.  My code uses vi's I downloaded from the ni site for the TDS3054 scope.  Most of the software commands are the same and I used regular visa read write vi's from the standard package.
0 Kudos
Message 19 of 23
(4,724 Views)

I would like to apologise to all the NI staff, who eventually contributed to the actual discussion, for failing to appreciate their valuable assistance.

Regards,

Maxim

0 Kudos
Message 20 of 23
(4,713 Views)