Inst. won't respond to 'read' commands

[obm]

I have a Coherent Labmaster power meter that responds correctly to 'write' and 'read' commands when run from the PC on my desk, but when I try it on a newer pc in the lab (using the same cable and same power meter) the instrument will not respond to 'read' commands.

On my desktop machine I can query *IDN? (in MAX) and get a good response. In the lab *IDN? times out.

In the lab only 'write' commands work. Any suggestions?

[Albert.Geven]

GPIB or Serial communication?

I expect GPIB and maybe the card is simply too fast for the machine. Try to see the properties of the gpib card in max.

[obm]

When I use RS232 it works fine on both machines. The machine it doesn't work on is 3 or 4 years newer than my desktop machine. I'll check the specs on both. Thanks for the reply.

[obm]

Here's what I have:

desktop: Intel Pentium 4,               2.4 GHz, 3.25 GB RAM

lab:         Intel Core 2 DUo E6850, 3.0 GHz, 3.48 GB RAM

The GPIB bus timing is set to 500 nsec, but I tried all the options and it didn't help.  The choices are 2 usec, 500 nsec, 350 nsec.

I have my application running fine with RS232 but I'd like to know what's wrong with the GPIB on this system. I even swapped cards and had the same problem. In addition, I have another similar pc in the lab and it behaves the same, ie it doesn't respond to read commands.

In searching the web I found a reference to "bus mastering" but that didn't lead to a solution.

[Albert.Geven]

Also try 2 microseconds. The machine is anyhow slow.

And try also to switch autopolling to off.

sorry I don't have a real good idea.

But once I had an instrument that was so slow that reading the statusbyte with a fast machine blocked the instrument from sending real data.

[obm]

Thanks, Albert. I'll try those suggestions.

[Dennis_Knutson]

I would also suggest you try swapping the GPIB controller between the 2 pc's. It's pretty unusuall that you could do a write but not a read (and are you absolutely sure the writes are working?) but a bad controller could cause that type of error.

[obm]

I tried all combinations of bus timing, parallel-poll suration, time-out, system controller, send EOI, set EOI and the power meter didn't respond to *idn?. Then I brought in a function generator and it responded correctly with the default GPIB settings (which is how the power meter was working at my desk). 

So it looks like it's not a card issue but something peculiar to the Ultima Labmaster. I'm going to use it with RS232 since that's working. I'll let Coherent know about the problem in case it comes up for someone else.

Thanks to all for the suggestions.