Author Topic: 34401A 10Mohm range (Read 4973 times)

electrolust · « **on:** May 19, 2016, 05:11:18 am »

I recently acquired a 34401A. It is in most excellent cosmetic condition. It passes self-test of course.

I have a decade resistance box. With that configured for 10M, my Fluke 28II (which was just calibrated in Jan) reads 9.99M. The 34401A reads 24xxM (09 or 19 or something, don't recall exactly) with 2 wire measurement. Way, way off. I havent tried other ranges or 4 wire.

Do folks think this is simply in need of calibration, or is it more likely to need a repair of some sort? I'm a beginner and I am not confident I can repair such a precision instrument. Of course I only discovered this problem as part of a project I'm working on, and so I also can't afford to take a month to learn how to debug this thing.

As this sounds like a pretty popular piece of equipment I'm hoping someone has an intuition on the complexity of the problem, and some guidance for me.

bitseeker · « **Reply #1 on:** May 19, 2016, 05:29:10 am »

First try it in 4-wire mode. Then, try it both ways from the rear terminals (if you initially used the front ones) to see if the problem is mechanical (oxidation in jacks, front/rear switch). If it's similarly off for all of these, also try engaging/disengaging the front/rear switch several times. If that makes a difference, cleaning the switch will likely take care of it.

If none of that rectified the situation, it's possible that the current drive for that range has a problem. However, I think the self test should pick up on source current problems. So, it might be something in between.

Since you have a decade box, do test with other values to see if it's just one range that's having issues.

Anyway, start with the easy stuff.

Dr. Frank · « **Reply #2 on:** May 19, 2016, 06:06:36 am »

Yes, start with the easy stuff..

Grab the service manual, page 98 lists all the Ohm test currents; it's 0.5µA for 10M range.

In 4W Ohm mode, you may measure this current (for the different ranges) on the Input jacks(rightmost), separately from the sense jacks.
The sense jacks might show excess leakage currents, e.g. by a broken input amplifier. This would be the next test.

Frank

electrolust · « **Reply #3 on:** May 19, 2016, 06:13:46 am »

This is awesome, ok I will start digging into the debug tomorrow.

electrolust · « **Reply #4 on:** May 19, 2016, 09:42:01 pm »

It was me. There was another resistor in parallel. I must have probed it differently when I used the Fluke, although I tried Fluke vs HP several times to verify.

Dr. Frank · « **Reply #5 on:** May 19, 2016, 09:49:13 pm »

Things happen.. very brave, that you admitted that booboo, so that I can sleep a lot better, now... the 34401A is an extremely rugged instrument, in the end.

Frank

electrolust · « **Reply #6 on:** May 19, 2016, 11:25:19 pm »

I'm very happy with the 34401A. Today I used it for freq measurement, something I hadn't anticipated using since my func gen has a counter. But you can't use the counter and generator simultaneously! It was one of those "glad I have this, I must be brilliant to have foreseen the unforeseen" moments.

bitseeker · « **Reply #7 on:** May 20, 2016, 02:16:44 am »

Phew! Enjoy the meter. I still like mine too.

electrolust · « **Reply #8 on:** May 20, 2016, 06:22:50 am »

I sure wouldn't mind having the graphical "trendplot" / whatever-its-called graph for DCV readings. What's the easiest way to get my computer (mac, ideally) to talk to the 34401a? I guess I also need the hardware to interface to the 34401. I am not trying to become an expert in interfacing to lab gear, just want to take the path of least resistance (lol).

Dr. Frank · « **Reply #9 on:** May 20, 2016, 07:18:52 am »

Quote from: electrolust on May 20, 2016, 06:22:50 am

I sure wouldn't mind having the graphical "trendplot" / whatever-its-called graph for DCV readings. What's the easiest way to get my computer (mac, ideally) to talk to the 34401a? I guess I also need the hardware to interface to the 34401. I am not trying to become an expert in interfacing to lab gear, just want to take the path of least resistance (lol).

The 34401A has RS232 and GPIB.
So you might interface nearly cost free over RS232, if your PC is still equipped with one. (You may find an RS232 interface on most mainboards, anyhow)
Keysight offers BenchView for free, which would automatically use the implemented interface.

Then you also would have trend chart capability.

I doubt that this program is offered for MAC also, but have a look on their site.

Other free solutions may be found here, KE5FX is member of the time-/volt-nuts community:
http://www.thegleam.com/ke5fx/

Frank

bitseeker · « **Reply #10 on:** May 20, 2016, 05:04:59 pm »

One option is to use a USB-to-RS232 interface. Make sure it's RS232, not just serial (aka, serial TTL). The voltages are completely different. If one end is USB and the other is DB9 or DB25, it's likely the right one.

Depending on the interface chip used and the corresponding OS, there might be an issue with the USB compatibility, at least that's what I've heard. So, if at first you don't succeed, try a different one. Also, check through the forums for related discussions.

If you want to try GPIB, I've seen a couple of projects to build a USB-to-GPIB interface. They're quite pricey otherwise. One project used only an Arduino Uno and a GPIB connector. Without the line driver chips, it won't work with really long cables or many devices on the GPIB bus, but that likely won't be an issue for your use case.

Andreas · « **Reply #11 on:** May 21, 2016, 04:19:16 am »

Quote from: bitseeker on May 20, 2016, 05:04:59 pm

One option is to use a USB-to-RS232 interface.

Hello,

Caution: the HP34401A requires a "null modem" cable (with crossed pins) when using the RS-232 interface.
(see users guide page 150 (Chapter 4)).
So a USB to RS-232 interface (or serial PC port) cannot be used directly.

With best regards

Andreas

bitseeker · « **Reply #12 on:** May 21, 2016, 05:42:50 am »

Thanks for that clarification, Andreas.

USB-to-RS232 interface plus a null-modem adapter (or wire up your own).

electrolust · « **Reply #13 on:** May 21, 2016, 06:05:42 am »

Would a product like https://www.sparkfun.com/products/8841 work with BenchVue? It's especially interesting because I see there's a mobile (ipad) version of BV. I guess for $50 it's cheaper to get a USB-serial (+ null modem) adapter.

HighVoltage · « **Reply #14 on:** May 21, 2016, 06:36:16 am »

The GPIB to USB Agilent adapter also works very well with the 34401A in combination with BenchVue on a PC and iPad.

electrolust · « **Reply #15 on:** May 21, 2016, 06:37:56 am »

cool, half the price of the prologix as well

electrolust · « **Reply #16 on:** May 28, 2016, 08:03:12 am »

Just got it (agilent GPIB->USB). I should have looked more carefully before buying. It's great and all but wow it's massive. The DMM itself is already eating all available room on the bench plus a little bit. With the GPIB adapter there's no hope of it fitting.

HighVoltage · « **Reply #17 on:** May 28, 2016, 05:36:11 pm »

You can use a short GPIB cable and then plug de adapter in to the cable to make it fit on your bench.

electrolust · « **Reply #18 on:** May 29, 2016, 09:16:22 am »

yup that should do the trick! thanks


EEVblog Main Site	EEVblog on Youtube	EEVblog on Twitter	EEVblog on Facebook	EEVblog on Odysee

Author Topic: 34401A 10Mohm range (Read 4973 times)

Share me