Getting my Wavetek 752a filter to work with Labview? Possible bug?

[optoelectro]

Hi, so I have a Wavetek 752A Brickwall Filter. It is basically just a combination low pass/high pass filter, and it is GPIB programmable. Here is the PDF manual for it.

There are no NI instrument drivers that I can find for it.

For reference, I *have* gotten it working with my labview program -- what I'm doing is pretty simple, I basically just set the threshold frequency either high or low with a single command.

My problem is connecting to it with LV to begin with.

Some background: I'm using an Agilent 82357A USB/GPIB interface, and I know it works very well (I have 3 other machines connected to it that all behave very well). I have all the drivers for the Agilent interface installed and working correctly; I'm almost positive nothing is missing on that end.

And just to emphasize again: I have gotten a labview program to communicate and program this filter, using this same interface.

Here's the kind of weird part. Normally when I start my experiment, I go to NI MAX and open the VISA test panel for each instrument and just query something to check that it's communicating, like *IDN? or some equivalent. Currently, I have the GPIB address for my Wavetek set to 18, set by some switches on the back panel.

Now often, when I plug in my GPIB/USB interface, go to NI MAX, and refresh the devices and interfaces tab, the other instruments are listed there, but the Wavetek isn't. Sometimes it is, but if I select it and try to open a VISA test panel, it says: Error opening resource: GPIB0::18::INSTR VISA:  (Hex 0xBFFF0011) Insufficient location information or the device or resource is not present in the system.

There are two LED lights on front panel of the Wavetek: a "local/listen" one and a "remote" one. At this point only the remote one is lit.

Okay, now, I also have some Agilent (Keysight, now...) software installed that seems to be their equivalent of NI MAX. It's called Keysight Connection Expert, and it seems to be better at communicating with GPIB instruments sometimes. Strangely, when I "rescan" it (basically, like refreshing in NIMAX), it doesn't find my Wavetek filter. But it does have this "manual configuration" tab, and I can manually enter the GPIB address (18). If I ask it to "test connection", that fails:

But, as soon as I accept, it sends me over to the list of devices, and it's listed there. Furthermore, if I click "send commands to this instrument" (basically their version of the Basic I/O tab in the VISA test panel of NIMAX), I can communicate with the machine! At this point the "remote" LED is also lit.

I don't know why the VISA address is X'd out there. I know I'm communicating with it because I just have a 100Hz signal going to the filter, and the output going to a scope. First, the signal is displayed on the scope (the threshold of the low pass filter is 1kHz or something), and then I send CAF2M1G0 to it, and the output dies (sets the threshold to 20Hz). I can then switch the threshold around, so I'm sure it's working. (An important detail is that according the manual, you need a carriage return at the end of the command to make it work, so if I send these commands without doing anything, they don't work, so I have to select \r under "EOL sequence" in the options, and then it works.)

Alright, so remember how before, NIMAX wouldn't open the VISA test panel for the Wavetek? Well, now, after doing the manual configuration in Keysight, it does. However, the same commands don't work in it, so I'm pretty confused.

For example, I set "enable termination character" and set it to "carriage return \r", hit accept changes, and then try entering my same command as before. The panel at the right says "Write Operation (CAF2M1G0) No Error", but clearly nothing is changing in the scope output.

I really don't know what I did to get it working before with my program. I think it was some magic combination of restarting and un/plugging things in. I also think it may have something to do with the "assert REN" option (under the line control tab), but I don't really understand it to be honest.

Okay, I've actually figured out one more thing about this: I just got it to work with NI MAX, but I may have found a bug. I suspected that it wasn't actually entering the crucial EOL carriage return character so I used NI I/O Trace, first entering my command without the "enable termination character" checked and then with it checked. They had the same value in the NI trace. However, if I manually entered "\r" after my command, now it does what it's supposed to at the filter, and you can see a "." after my command in the trace:

So, that really seems like a bug, but at least I found a solution. However, I still can't get it to work in my Labview program (that was running before!):

Any advice would be appreciated, thank you!

[nyc_(is_out_of_here)]

http://digital.ni.com/public.nsf/allkb/F7C187DBF09EBE1186256F550065BD32

http://cp.literature.agilent.com/litweb/pdf/5990-3731EN.pdf?cmpid=1273CN

[Dennis_Knutson]

The enable termination character sets it for a read, not a write so there is no bug.

[optoelectro]

Thank you for the response. However, I used both of those documents long ago when I installed Labview and the drivers for my Agilent USB/GPIB interface, so those don't really help here. Like I said, my Agilent interface clearly works, so it doesn't seem like an installation issue.

[optoelectro]

I'm sorry, I'm not sure what you mean -- do you mean that if you have the "enable termination character" box checked, it only adds the termination character for a Read command? If so, why?

Also, any idea for a solution to my real problem of connecting to the device reliably?

thank you!