WinXP - 488 driver doesn't work with old software

[Dnmeeks]

This is a Win XP machine, with Labview 7 installed and running just fine.

My company developed software in-house for controlling some HP test equipment (8505A network analzyer). The author of the software passed away several years ago.

The program runs, but on loading it says that it can't find DEV16, which is the HP8505A.

I have used MAX to verify that this device is visible, and the address is correct. When I "communicate" from within MAX, the REM light on the instrument comes on.

Can anyone help me figure out what might be wrong? The software was written between about 1988 and 1993, which was back in the Win 3.1 / Win 95 era.

I tried loading a new driver, but it didn't change anything. Is there something else I can try to figure this out?

Thanks in advance!

Dan

[JÞB]

You are using an instrument that was introduced 37 years ago.  Really, some progress in test and measuremant instrumentation has been made in the last quarter centuary or so, consider upgrading your hardware - I'm positive that the device engineers never fully appreciated the processing power or communications speeds available today.  If you have a working 8505A contact your local historical society they will be thrilled.

LabVIEW 7 is also a bit long in the tooth and did not handle VISA sessions well when non IEEE488.2 compliant equipment was hung on a GPIB bus.  Later VISA versions helped but cannot completely make up for the fact that in 1976 HPIB was still evolving and not standardized to modern levels.

Your company found it beneficial to upgrade the PC - Why? Upgrade the software and hardware too- and for the same reasons.

[nyc_(is_out_of_here)]

---

@Dnmeeks wrote:

This is a Win XP machine, with Labview 7 installed and running just fine.

My company developed software in-house for controlling some HP test equipment (8505A network analzyer). The author of the software passed away several years ago.

The program runs, but on loading it says that it can't find DEV16, which is the HP8505A.

I have used MAX to verify that this device is visible, and the address is correct. When I "communicate" from within MAX, the REM light on the instrument comes on.

Can anyone help me figure out what might be wrong? The software was written between about 1988 and 1993, which was back in the Win 3.1 / Win 95 era.

I tried loading a new driver, but it didn't change anything. Is there something else I can try to figure this out?

Thanks in advance!

Dan

---

If the software is that old, it most likely wasn't making use of VISA sessions. In 2003, I had XP and LabVIEW 5 on test PCs so XP and LabVIEW 7 is doable.

I am wondering if the original programmer was using GPIB.ini in conjunction with ibfind(). http://digital.ni.com/public.nsf/allkb/20111CA8C1BB0E6D86256F2E0064FEE0

Worse comes to worse, you will need to run the program in development and see what it is actually doing.

[Curt_C]

Hi,

You might need to configure some settings to talk to the old instrument.

Start the 'GpibConf.exe' application.

It should be in C:\Program Files\National Instruments\NI-488.2\Bin or wherever you installed 488.2

Mess around with some of the termination settings on the form.

Also, try slowing down the Bus Timing on the gpib board. Try 2usec.

That can be found in MAX in GPIB Interface Properties.

Hope this helps

Good luck

Curt

[Dnmeeks]

Thanks -

I ran that little utility, and it looks like the settings were different than when I look using MAX. What I saw was that "DEV19" said that it was on interface "GPIB1", but I only have one GPIB interface, and it's GPIB0. In Max, DEV19 is listed as part of GPIB0. Same thing with DEV20. But DEV16 shows up on the correct GPIB0.

I changed them, but is there something else I need to do, or does that update something different than MAX?

I also found some other instructions for madifying gpib-nt.com file to enable DOS support, but the mods were already made there.

Thanks again -

Dan

[Curt_C]

Hi,

Did you try changing the bus timing? and the termination settings.

Have you used ni-spy or ni trace? Open up a communications session with the device in max and run ni-spy.

Send some common gpib commands like *RST, *CLS and see if you can catch the error.

Curt

[JÞB]

manual here

As I feared,  This is one of those devices that predates modern GPIB protocol by a few decades.  VISA will not be able to work with this device.  MAX will not be able to find it.

Why?  Simply put the instrument talks and listens on different addresses.  It will need to be addressed to talk and addressed to listen and unaddressed after either talking or listening.  The low level 488 calls will need to be utilized and no other equipment should be allowed on the same bus.  Read the manual- there are some unique EOI characteristics for the test set as well since it uses "E" to terminate messages rather than hardware lines.

Good luck!

[Dnmeeks]

Yes, thanks, I should have mentioned that I already did that, to verify that I could talk to the instruments. I was able to send commands from the programming guide and it worked.

And I do have an update - it is working. After making the mods to the gpib-nt.com and changing the bus addresses using GpibConf.exe, it started working. I have some other issues, but they seem to be unrelated to the GPIB comms. Thanks again for the help!

Dan

[JÞB]

I did not say you could not get it to work...(Kudos) only that it would involve some challenges that would not be necessary with technology available today  Can you post an example of a driver for that dinosaur?