No response to *IDN? when NI Spy is turned off

[Nick_F]

Hi, I am using a software made by a third party to communicate with a Fluke 81 instrument. The software initializes the instrument successfully and it is able to send commands to it (i.e *RST, *CLS) and the instrument executes the commands.

But any queries sent to the instrument time out (such as *IDN?, FRQ?). Now, if I turn on NI Spy, to monitor what is going on, the queries are executed! At this moment I can run the software only if NI Spy is turned on. Did anybody experience something similar? It appears NI Spy somewhat does something to the commands sent. I received the source code for the driver file and I was able to enter some delays and it appears there is nothing related to delays between commands. If I leave NI Spy turned on until the first query is executed, subsequent queries work even after turning NI Spy off.

What can I do to solve this problem? The developer has a similar system and they do not experience the same problems. If I use MAX, the instrument responds to *IDN? every time.

I attach a screenshot of the communication that takes place with Fluke instrument. Obviously, with NI Spy on, everything works fine. Maybe somebody can advise about some changes to be done in the software.

Regards,

Nick

[Justin__D]

Hi Nick,

It sounds like something to do with the timing, although I know you said you added some delays.

I think NI-Spy slows down the commands are sent to the instrument.

I'm not sure if there as anyting else I will be able to help with since you are using third party software.

Are you using NI software for anything else?  I'm curious to know if it works on a machine without NI-spy or MAX installed.

[Nick_F]

Dear Justin,

Thank you for your reply. In the meantime I realised the problem could be in a piece of software for which I don't have yet the source code. It is responsible with executing queries for any instruments. I am not sure if NI libraries provide a function to run queries, or if the query function is built by users from combining a Write +Read functions. If the later case, maybe I could add a delay in between the write and read commands. I will keep the thread updated if I have new info.

Best regards,

Nick

[gagi1]

You may be able to resolve the timing problems, either by introducing delays between GPIB commands, or by changing the GPIB bus timing to the slowest setting (in the GPIB Configuration Utility).

http://digital.ni.com/public.nsf/allkb/862567530005F09C852563D5007598C4