Problem with GPIB-USB converter

[cocopa]

Hello, I've read a few threads about possible differences between GPIB-PCI & GPIB-USB but I wasn't able to find an answer. 
I'm trying to control a Keithley 6517 electrometer via GPIB and everything works perfect when I use a GPIB-PCI card from National Instruments and the problems begin when I try to use the GPIB-USB adapter (also from NI).

I can do almost everything (set voltage, read temperature etc) even with the usb adapter, but when I try to read a current measurement I constantly get an abnormal value of 9.91E+36. I've tried several PCs (windows 7 & 10), several versions of labview (15 & 17) and every time it works with the GPIB-PCI but not with usb adapter.

The only difference I saw in the NI MAX VISA test panel attributes, is that on the PC with the GPIB-PCI card the "allow DMA transfers" is set to TRUE while in the PCs with the usb adapter it is set to FALSE. Could this be causing any problems? And if yes how can I change this attribute?

Thank you. 

[RTSLVU]

Your problem makes absolutely no sense at all. I have several test systems I support and they are a mix of PCI and USB GPIB interfaces from NI.

I have no problem running the exact same code on any of these test systems.

We really need to see your code as there has to be something going on in there that is causing this.

[Kyle97330]

"DMA" means "Direct memory access" and just refers to how your PC hardware is accessed via the CPU.  It doesn't affect external hardware like your Keithley.

Are you sure your measurement is 9.91E36?  Is it maybe 9.91E37 or 9.9E37?  Because those are special values to Keithley.  From the manual:

READing: Instrument reading. The resolution of this reading tracks the display resolution of the instrument. An overflow and out-of-limit read as +9.9E37; zero-check reads as +9.91E37; underflow reads as 0.00E00.

Do you have only one USB-GPIB adapter or have you tried many?

The only time I can think that a specific bus connector would cause a problem but others would work is if you've got defective hardware that causes a ground loop somehow on whatever you're measuring.

[cocopa]

Thank you for your answers, I think I know where that value comes from. 

After initializing the instrument, the first read value I get should be discarded. In the PC with the GPIB-PCI  card (which I should note is running windows 32 bits) this value is "9.9E+37" and it is included in the code so a second read is performed if I receive this value. In the PCs with the usb adapter (windows 64) I get a different value after initialization and the code is not programmed to discard it. I don't know why that happens, and unfortunately I have only one usb adapter to test if it is malfunctioning or if it is something unrelated to the adapters, but at least I know how to handle it. 

Again, thank you for your answers.

[billko]

---

@cocopa wrote:

Thank you for your answers, I think I know where that value comes from. 

After initializing the instrument, the first read value I get should be discarded. In the PC with the GPIB-PCI  card (which I should note is running windows 32 bits) this value is "9.9E+37" and it is included in the code so a second read is performed if I receive this value. In the PCs with the usb adapter (windows 64) I get a different value after initialization and the code is not programmed to discard it. I don't know why that happens, and unfortunately I have only one usb adapter to test if it is malfunctioning or if it is something unrelated to the adapters, but at least I know how to handle it. 

Again, thank you for your answers.

---

Something VERY fishy going on here.  If you don't fix it the right way now, it will come back to bite you later.  The original "fix" was a kludge.  Figure out what's really wrong.