High Resistance Meter Recommendations

[KungFuJosh]

Hi,

I'm looking for a recommendation for a resistance meter or DMM than can read up to 1 or 2G ohms.

Currently my meters don't go above 50M ohm, and with projects containing parts between 150M to 1G, it bothers me not to be able to test them before putting them in the circuit.

Thanks,
Josh

[Gyro]

At that sort of level, you're unlikely to find a DMM that's up to the job, too many issues with low source voltage, test lead pickup etc.

For any sort of accuracy your best bet it is voltage source and Picoammeter type setup, possibly in a screened jig. A source measurement unit (SMU) is the official 'tool' for the job.

[bob91343]

For a one time measurement there is no simple solution other than a power supply and microammeter.  I have a Megohm bridge that goes into the teraohms so that would work.  But today's gear generally won't measure that high a resistance.

There are other considerations as well when you get up there.  Voltage coefficient becomes important, which is due to the high voltage gradient causing constriction of current path.  You must keep watch on temperature, as these components are often sensitive to small changes.  It can be dangerous, as the voltaage needed to reach measurable current may be at high levels.  A mistake in hookup can cause big sparks and damage.

To measure 1 Gigohm with 1 microampere current flow you need 1000 Volts.  If your meter can measure 1 microampere accurately and you have the high voltage available you are all set.  Keep one hand in your pocket.

[kj7e]

If your lucky you can find a HP 4329a used;

https://www.eevblog.com/forum/testgear/test-equipment-anonymous-(tea)-group-therapy-thread/msg2473674/?topicseen#msg2473674

Otherwise look into Insulation resistance meters, like the Uni-T UT505B or UT512.

[Gyro]

For a one time measurement there is no simple solution other than a power supply and microammeter.  I have a Megohm bridge that goes into the teraohms so that would work.  But today's gear generally won't measure that high a resistance.

There are other considerations as well when you get up there.  Voltage coefficient becomes important, which is due to the high voltage gradient causing constriction of current path.  You must keep watch on temperature, as these components are often sensitive to small changes.  It can be dangerous, as the voltaage needed to reach measurable current may be at high levels.  A mistake in hookup can cause big sparks and damage.

To measure 1 Gigohm with 1 microampere current flow you need 1000 Volts.  If your meter can measure 1 microampere accurately and you have the high voltage available you are all set.  Keep one hand in your pocket.

You need to be well clear of the uA region for measuring resistors that high, I'd say nA at most.

Of course, it's not difficult or expensive to make a workable picoammeter  ....    https://www.eevblog.com/forum/projects/picoammeter-design/msg790045/#msg790045

[Performa01]

What accuracy do you need?

Measurements up to Gigaohms is usually a task for Electrometers - maybe you can get a Kethley 610C, these are very useful and don't need expensive cables, since they still use a SO239 connector instead of Triax.

If you can't find one and accuracy requirements aren't too high, an insulation tester could do the job. But these devices work with high voltages (usually up to 1000V) and still not many go as high as 1 Gohm.

EDIT: seems like either my memory wasn't serving me well or times have changed since I've last looked

Finally there's also the cheap yet accurate and safe method: Get your DMM and measure its input impedance with another (accurate) DMM. It should be close to 10Mohm, but it doesn't really matter if you know the exact value for each of the ranges you're going to use. Put the unknown resistor in series to the meter input and measure e.g. a 10V DC source (could be a function generator or a lab PSU). Then measure the voltage and do the math...

For 10V and 1Gohm and precisely 10Mohm input impedance, you should measure something close to 100mV (99.9mV exactly).

EDIT: especially newer DMMs might not have the very low input current that is required for this. So you should check the accuracy of that method with a known high resistor value beforehand.

[KungFuJosh]

I don't need fantastic accuracy, only to know I'm not significantly far off or putting in a defective resistor.

Unless I'm reading it wrong, it looks like a lot of insulation testers go well beyond what I need. Even the Uni-T UT501A says it goes to 5.5G, which is really not expensive.

The UT501A claims it can read up to 5.5G at lower voltages, which is good because the resistors I'm using in that range are usually 1/4 or 1/2 watt, and this one for example is only rated for 750V: https://www.mouser.com/ProductDetail/279-RGP0207CHK1G0

Does the UT501A sound like it should be sufficient for what I want? Or am I missing something?

Thanks,
Josh

[Performa01]

Yes, the UT501A looks nice and usable.

I happen to have the UT510 (bought many years ago) and this wouldn't be what you want. It has deepley recessed sockets, so you can only use the supplied probes, which are unusable for electronics.

[KungFuJosh]

Thanks, I ordered the 501A. Hopefully it doesn't take forever to get here.

[bob91343]

I just saw a GR1644-A for not much over $100 on ebay.  This is a great instrument and I value mine highly.  It goes way beyond what you need here.  A plus is that you can test the unknown with any voltage up to 1000V.  It also is good to measure capacitor leakage at rated voltage.

[joeqsmith]

A bit late, but many handhelds have the conductance feature.  Several meters I looked at can read down to 0.01nS.   Even the old Fluke 189 isn't too bad. 

[David Hess]

When I have needed to make a high resistance measurement without the proper instrument, I have resorted to placing the resistor in series with a standard 10 megohm input meter, applying a known voltage, and measuring the voltage.  The external resistance can then be calculated based on the measured voltage across the meter's 10 megohm input resistance.

Otherwise a dedicated insulation resistance tester is probably the least expensive option.  Some multimeters which support conductance can also make high resistance measurements but they are not easy to find.

[joeqsmith]

Here are a few datasheets for meters I have looked at that support conductance.  The Summit / TPI 194II and Fluke 289 also support it.   Oddly, the Gossen Ultra and Dave's 121GW do not support it.   

[KungFuJosh]

I love my 2450 SMU when work with rear Triax cables. With good test conditions that SMU can precise measure up to several Teraohms.

Wow! That thing is $5500!!! I got the Uni-T thing for $57 shipped and it works fine. If I had $5500 to blow on test equipment, I have no idea what I would get. Probably another guitar. 

[KungFuJosh]

Dear KungFuJosh, 2450 SMU is very cheap solutions....
I also love my B2985A on Petaohms range at least 1-6 time per week, but it more-more expensive...

All Uni-T meters is rubbish!
UT501A +-3% hmmm... pack of gigaohms deviders CDHVAF1G00G5000GET from my last project have better precise.

Lol! At least the B2985A is only $4500 used.

It's okay though, I don't need that kind of accuracy. I'm mostly looking to make sure things aren't defective.

[bob91343]

I recommend the GR 1644-A as a reasonably priced teraohmmeter.

Works for me.

[horo]

I usually take my Fluke 87 for high resistance measurement, the conductance range (max 40 nS) gives 1% resolution at 1 GΩ (reading 01.00 nS) while the last digit (00.01 nS) equates 100 GΩ. Very helpful to detect leakage in switches and connectors. Also my famous old Fluke 8060 from the 70s allows the detection of up to 1GΩ (reading 0001 nS) in the 2000 nS range. (thx drtaylor!)

Martin

[GerryR]

An older Keithley 160B will get you to 1999 M , all be it with +/- 30% "accuracy," but can be had in the $50.00 range on Ebay.  Very easy to calibrate, and if you have some known good resistors you can use as standards, you can keep a check on the testing.

[bluey]

Even a 1990 Fluke 87 does conductance.