Keithley 192 GPIB control

[The Soulman]

To not derail other threats further I've started this one.

My goal is to be able to control my Keithley 192 dmm and read back the data with Testcontroller, I'm using the lan/gpib adapter by Kofen (ar488/prologix compatible?).

Some links:
Keithley 192 user,programming and service manual: https://xdevs.com/doc/Keithley/192/
GPIB adapter: https://github.com/Kofen/PoE_Ethernet_GPIB_Adapter https://www.eevblog.com/forum/projects/poe-ethernet-gpib-adapter-open-source-project-now-public/
Testcontroller: https://lygte-info.dk/project/TestControllerIntro%20UK.html https://www.eevblog.com/forum/testgear/program-that-can-log-from-many-multimeters/msg5887268/#msg5887268

I have minimal programming experience so obvious errors in my logic (pun intended) are to expected, please point them out if you spot any.
I also cannot rule out any hardware errors in this 40yo multimeter, I don't have other gpib controllers or devices to pin-point any faults.

Before starting on a device definition to work with Testcontroller I'm manually testing all the commands as listed in the 192 manual with Putty.

With the 192 in talk only mode (dip switch setting) ++read returns the currently displayed value.

With talk mode only disabled the only command that works is ++dcl it restores the device to default settings as it should.
All other commands don't produce a result/do anything.
Also the "Remote" and "Listen" leds will not turn on (they do during start-up as part of the display test).

Likely it has something to do with initializing the bus/handshake this is where I'm stuck.
Sending ++ren 1 and ++ ren 0 toggles the Remote ENable line and can trace that to pin 22 of the 68488 ic inside the 192, but generates no response at all.

Is this a hardware fault or does it require more commands to enable remote control?

 
 

[Gertjan]

Good idea to check the communication first with a terminal program. This is a good way to weed out the basic problems.

Did you find in the the 192 manual that you need to add an "X" (for "execute") to the end of each command line? Without the "X" the meter will just wait for more commands, without executing anything.

Also, check what the default handshake method is of your GPIB adapter, and send the commands to the 192 so it uses the same method. (or the other way around)

Really, the devil is in the details... So carefully studying the documentation (GPIB basics, meter, Keithley 102 adapter, TestController) is a must to get working correctly.

regards & good luck, Gertjan.

[MiroS]

I use talk mode wih self-developed python code.  GPIB commands work, all is stated in manual, but  think thse commands are not usefull for easy driver for TC.

I think the other mode. other than talk does not have any sense as range/mode is selected by physical switches on front panel of the instrument. Returned string contain value and unit which is simple to parse. I think you would need to use talk mode and parse string like I am doing in python code.
 

[The Soulman]

Thanks Gertjan, no I had not read about the X yet, is was a few pages further in the 192 manual.

MiroS, Talk mode only would be plan B, I would like to be able to control the 192 from TC, in a same way Gertjan did with his 199.
 

[The Soulman]

Although the meter does respond to
++dcl (resets all devices on the gpib to default)

It does not respond to:
++addr 8 (sets the gpib adapter to address the device that has gpib address 8 )
++clr (resets the device that is addressed, in this case 8 )


I've found out the dip-switches on the rear of my 192 that set the gpib-address are not reliable anymore (measure differently than actual set), likely caused by decades of not being exercised.
Will remove the switches and replace by hard wire for maximum future reliability and the minimum amount of  .

[Hydron]

I've got a partially done Keithley 617 driver written (still needs some work before I contribute it to HKJ) - there's a bit of mucking around with string manipulation to get range info etc, but it does work with these "R2D2 language" instruments. There's also another DMM (199?) with a driver already written and included.

[The Soulman]

I've got a partially done Keithley 617 driver written (still needs some work before I contribute it to HKJ) - there's a bit of mucking around with string manipulation to get range info etc, but it does work with these "R2D2 language" instruments.

Cool, the 617 looks like it is a bit closer to the 192 gpib wise than the 199.