Reading Buffer data Keithley 2701 through TCPIP

[Batphreak]

I am using Labview 7.1, DAQmx 8.3, IVI package 3.0, NI VISA 4.1, ke27xx v1.2, and the latest firmware update on a 2701 DMM. The problem I have is when I using the *Read Buffer* vi to download data stored in the buffer when the 2701 is running a scan. If the 2701hasn't finished all of its progammed scans the data returned by the vi is incorrect, the channel and time-stamp data seems to get mixed into the actual scanned data. Once the scanning is complete the vi reads the buffer data without a problem. With older versions of the driver I was able to read the buffer just fine while a scan was active. Thank you for any help you can offer.  The code I'm working with is attached, the "2701 buffer to file.vi" is my main code.

-Bryan West

[Rimasaurus]

Hi Bryan,

Welcome to the NI Forums!  Do you remember which version of the driver worked?  There may be a difference in the drivers that causes the problem, so it will be helpful if we can compare the two driver versions.  Thanks!

Regards,

Rima

[Batphreak]

I believe the driver version that was working was NI-DAQ 7.3 and KE27xx version 1.2.  With new versions of NI-DAQ the older version of the KE27xx software doesn't work either.

Thanks,

Bryan

[Dennis_Knutson]

NI-DAQ or DAQmx has absolutely nothing to do with communication with your instrument. You can remove both and the communication will still work. The only driver involved is NI-VISA. Try to determine what the previous version of that was. You also mention you have the latest firmware from Keithley. Does that mean you also upgraded that?

I don't have experience with that model of Keithely but is doing a read before the scan is complete a valid thing to do? Normally, you check a status bytes or do an OPC? query before doing a read. Have you asked Keithley about the issue?

[Batphreak]

I upgraded the Firmware on the Keithley before I upgraded VISA and Driver software.  The old VISA version that works is 3.0.1, the new version is 4.1.  I have contacted Keithley about this issue, they have had an engineer work with it some, but suggested that I get in touch with NI because Keithley claims to have no responsibility for the driver from NI.

I'm guessing doing a read while a scan is active isn't a "best practice" but the tests we monitor with this unit are long tests (150 hr +) failure can be seen partway through these tests, allowing us to cut the test off without wasting the remaining time.  Since the check had been working in the past it needs to continue working.  Right now we have 1 computer with the old software setup on it, that works for checking the data, since new computers get upgraded with new software this needs to continue working with newer versions so that the read program can be used on any computer.

Thanks,

Bryan

[Dennis_Knutson]

So, if you uninstall NI-VISA 4.1 and then install 3.0.1, it works?

[Batphreak]

I just tried going back to the older NI-VISA version, it does not work with the new Keithley driver.  So far the only combination I can get to work the way I'd like it to is the old Keithley driver and the old NI-VISA version.  I hope that isn't what I'm stuck with since all the new NI-DAQ versions automatically install newer NI_VISA versions.

[Dennis_Knutson]

Newer NI-DAQ or DAQmx installations do not install NI-VISA at all. They only thing they will update is MAX and different versions of MAX should have no affect at all. You should have no problem at all using the older LabVIEW driver and the older NI-VISA.

It's possible that the newer NI-VISA does Ethernet communication a little differently. Also, since the only driver available for download is the 1.2 version, if you post the older version, maybe someone can look at the code to see what was changed.

[Batphreak]

This is the old version that works for me.

[Dennis_Knutson]

So the old driver is IVI and now you are trying to get the LabVIEW driver to work? That will make a comparison pretty hard.

I'm afraid I don't have an answer. I don't have that model and I don't have any Keithley with an Ethernet connection. Looking at the native LabVIEW driver, are you using the Data Buffer Status and either the Read Data Buffer or Read Data Buffer subset function? If so, I would recomend you look at the raw VISA Read string. There may be a simple mistake in the data parsing routine.