Voltage Measurements in Noisy AC Environments

[Geoff201]

I've read a few of the posts here regarding multimeters, and am confident someone can answer my question regarding voltage measurements in noisy ac environments.  Firstly, I'm not a professional in the electrical sciences, but I possess a good general knowledge of ac and dc systems, so I shouldn't be perturbed by the language of your trade.

I'm trying to measure voltage of a power supply board, in operation, in a (relatively small) 150 KVA, 1000 Hz induction furnace (yes, I've taken safety precautions).  This board provides power to the logic board that produces the pulses for the scrs.

This is all just background information.  The problem I'm having is that when I move the multimeter leads near the test point, the digits on the multimeter's display change wildly, such that the reading is meaningless.  In fact, moving the leads anywhere within 2 feet of the cabinet produces this effect.  The effect is the same, whether the selector switch is set to dc or ac voltage, or any range, though it only happens while the leads are plugged in.  Upon touching the leads to the test points, the reading doesn't settle.

I know that I could remove the dc supply board, power it remotely, and measure its output with a dummy load, but I've posted this question here, only to find out what's happening with the multimeter.  The model I'm using is an MTP-2850 with CAT III, 10A Mastech leads, nothing stellar, but seems to have been alright for general purpose use in the past.

I've also observed this effect when measuring voltage from an alternator on a forklift with this very same meter.  I have a "general" idea of what's happening, but I'd like to be more confident in my knowledge of the problem, and would deeply appreciate any insight you have towards a solution.  Do I need a better multimeter?  Better leads?  Or is a multimeter even the right tool for this kind of environment?

Thank you.

[Lurch]

I think you'll find part of the problem is that digital meters are low impedance devices so will wander about if they are sat on a desk let alone put near electrical equipment. What you need is an old AVO!

[Geoff201]

Do you mean high impedance?  The meter has an input impedance of 10 megaohms.  We do have an old Weston DC analogue voltmeter stashed away somewhere.  It seems kind of low tech though.

[SeanB]

Try twisting the leads together and measure again. Open circuit you will still get a varying reading but connected to a lower impedance bus it should settle. If it still varies you need a multimeter with better shielding, which means you will be going out and buying a Fluke 113, 117 or 87.

[Lurch]

Do you mean high impedance?

Yep.

The meter has an input impedance of 10 megaohms.  We do have an old Weston DC analogue voltmeter stashed away somewhere.  It seems kind of low tech though.

Low tech = reliable!

[Geoff201]

Thank you very much for this advice.  I'll try these suggestions tomorrow, and post back the results.

[Geoff201]

I tried again by twisting the leads, unfortunately the reading still fluctuated, and never settled upon touching the leads to +12v and gnd.  The passive Weston voltmeter did register a voltage that seemed to be around 12v, but the scale is from 0 to 300v dc so the resolution isn’t very good.

The Weston’s resistance is 300 ohms per volt, so at 12v that’s a lot lower than the dmm’s 10 megaohms.  The galvanometer portion of the meter is housed in a steel case, so it’s well shielded.  I think I’ll take SeanB’s advice, and recommend to my employer that we invest in a fluke 117, but I’d still like to know, if the problem with the current dmm is one of impedance or shielding.