email from Mr. John Miles with his KE5FX GPIB toolkits

[Cheh]

We send one piece of S82357 to John Miles for his evaluation with his great tools called KE5FX GPIB toolkits. Which is a collection of free Windows utilities that will help you make and record research-quality measurements with GPIB-based electronic test equipment. (For more information about this software, please looks here: http://www.ke5fx.com/gpib/readme.htm)

Here is the feedback from John Miles.

John Miles	May 1
to me

I had some time to try out your adapter today, and it seems to work very well.  The biggest problem is that I use National Instruments and Prologix GPIB interfaces, not Agilent 82357 interfaces, so I can't be sure if any given issue is due to your hardware or to the Agilent I/O Libraries package itself.  However, I can say that nothing really serious seems to be wrong! 

 

The adapter's compatibility with the Agilent 82357 hardware was good enough to fool the Agilent Connection Expert utility.  (I wasn't able to test this in depth, because all of my own applications use only NI488.2 mode.)

 

I was able to run all of my programs (SSM.EXE, PN.EXE, and 7470.EXE in host-requested plot mode) on the HP 8568A, HP 8566B, and Tektronix 494AP with no problems.  I also tested SSM on the Agilent E4406A, where your adapter was able to exceed 30 frames per second like the NI PCI-GPIB I normally use.

 

I was able to obtain host-requested plots on the 8568A, 8566B, 494AP, and 8753A VNA in 7470.EXE, as well as frequency readings from the HP 5370B counter in TimeLab.

 

There were a total of three issues:

 

1) Device-initiated plots, where the adapter presents itself to the instrument as a "plotter" and 7470.EXE is used in its Acquire->Wait for device-initiated plot mode, did not work on the 8566B and 8753A.  This equipment sends query strings to the "plotter" (to obtain page size, etc.), and 7470.EXE has to respond to these queries.  Somehow this does not seem to work; the 8753A shows an error indicating the plotter is turned off or at the wrong address, while the 8566B doesn't generate an error message but fails to send any data. 

 

Device-initiated plots did work on the Tektronix 494AP, which does not attempt to send any queries to the plotter. 

 

I have heard that the HP 82357 adapters also fail this test, so it is probably not your fault, and that there is nothing you can do about it.

 

I was also unable to get PrintCapture (another popular plotter emulator) to work... but again, this is probably an Agilent 488 compatibility issue.

 

2)A proprietary (closed source) NI488.2-based phase noise application that I tried did not work properly; the HP 3561A received some corrupted commands.   It will not be possible to say who's responsible for this, so I wouldn't worry about it.

 

3) It seems that certain error conditions, including a bad instrument address, will lock up the application for much more than the 10-second timeout period that the default Agilent 488 settings call for.   Whenever I mistakenly used the wrong GPIB address with one of my programs, I always had to physically unplug the USB cable before I could force the program to terminate or recover.  I'm not sure if there is anything you can do about this, but if so, it would be nice.

 

Thanks again for sending the unit!  I will update my GPIB Toolkit's FAQ page to indicate that the adapter works well.

 

-- john