Synchronizing Data Collection with Agilent N330X Load

[diarmaede]

Hi All,

I am planning to plot current vs. voltage curves using a USB-6229 and a variety of Agilent N330X resistive loads.

I need to choose one of 30 AI channels, then plot the current seen on that channel vs. the voltage.

Originally, I planned on choosing the desired channel via LabVIEW code, receiving voltage data from all 30 input channels, then sending data from the chosen channel out to the load.  Then I remembered that cumbersome serial connection which would make my original plan impossible due to timing issues.  DUH!

My new plan is to capture the current data via the load through the GPIB bus with a GPIB-to-USB converter and the voltage data as a waveform or array via the chosen AI channel, and then to plot the one against the other.  Does this sound reasonable?  If so, any clues about how to configure the triggering mechanism so that I can most accurately synch the voltage and current data would be much appreciated.

If there is a better way for me to accomplish my goal, please let me know.  I am VERY open to suggestions.

Thanks!

FB

Message Edited by diarmaede on 06-17-2009 04:04 PM

[JÞB]

OK Dia- (if I may call you that for short)

You need I-V curves and you aren't afraid to spend the boss's money to obtain good data.  At least, as long as the data is reliable.   I do this type of thing for fun (and a paycheck) every day BUT, I'm insomniatic right now so I'll try this pro-bono.

Voltage and current ranges and required accuracies?

Test cycle time for the Device Under Test?

Have you looked at a SMU? If so, why did you chose your set-up and how will you address phase shifts between I and V waveforms due to lack of common timing between the 6229 and the load?  There are great solutions for most common test applications available off-the-shelf and geeks like me to develop the uncommon ones or find one that is available.  

Essentially, my general feeling for your proposed materials and methods is that, because of the timing, you may not get any bang for your buck.  You chose great instruments but, the time issues cannot really make them play together without a penalty in either accuracy or time per DUT per throughput years under warranty (T/D/TPY) and that's going to cost in operator hours.

[diarmaede]

---

Jeff Bohrer wrote:

OK Dia- (if I may call you that for short)

You need I-V curves and you aren't afraid to spend the boss's money to obtain good data.  At least, as long as the data is reliable.   I do this type of thing for fun (and a paycheck) every day BUT, I'm insomniatic right now so I'll try this pro-bono.

Voltage and current ranges and required accuracies?

Test cycle time for the Device Under Test?

Have you looked at a SMU? If so, why did you chose your set-up and how will you address phase shifts between I and V waveforms due to lack of common timing between the 6229 and the load?  There are great solutions for most common test applications available off-the-shelf and geeks like me to develop the uncommon ones or find one that is available.  

Essentially, my general feeling for your proposed materials and methods is that, because of the timing, you may not get any bang for your buck.  You chose great instruments but, the time issues cannot really make them play together without a penalty in either accuracy or time per DUT per throughput years under warranty (T/D/TPY) and that's going to cost in operator hours.

---

Thanks for responding, Jeff!

Voltage range from 0 - 120VDC

Current range from 0 -  6ADC

Test cycle time is not well defined, but it does not have to be very fast.  I'd like to capture all I and V points in one sweep, but the slew rate can be relatively slow.

I don't know what an SMU is.  I will research.

Any further suggestions would be greatly appreciated.

- FB

[diarmaede]

I THINK I can send out a digital signal from one of the digital lines on the 6229, with the digital output line wired to the load and a digital input line on another digital terminal of the 6229 in parallel.  Then I use the digital That should set a reference point for the voltage and current curves, shouldn't it?  I can use the the trigger point as a time reference for both curves, so I should be able to match up the voltage and current data in the IV curve.

Or would the trigger point not be accurate enough, time-wise, to work as a good reference point?

- FB

[JÞB]

http://www.keithley.com/products/currentvoltage/?mn=2400

the 2400 SMU (Source Measure Unit) is about the best you can get for the money.  You'll save the cost in programming and debug time alone.  It has sweeps and step functionallity built in and of course, the source and measuerment functions are internally timed.   I-V curves on a platter!