Question about a panel meter. / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Question about a panel meter.

previous 1 2

Move Thread

LAN_403

Steve Conner

Sat Aug 29 2009, 09:26AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

And while we're getting pedantic, don't forget that regular 1/4 watt size resistors are only rated to withstand 350V. If you put 1kV across one, it'll arc over internally. I found this out the hard way.

Proud Mary

Sat Aug 29 2009, 10:55AM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

Don't forget the very severe limitations of some of these basic voltage measurement techniques.

To give an easy-peasy picture of the biggest of these limitations, think of a simple case like this:

You want to measure the voltage drop across a 1 meg resistor.

You connect your moving coil meter and series resistor across the 10 meg and make your measurement, and discover a voltage far, far lower than what you had been expecting. Why is this?

Suppose the resistance of your meter movement and the series resistor is 1K to make things nice and easy.

What you have done is this:

You have connected a resistance of 1K (your meter and multiplying resistor) in parallel with the 1 meg resistor. So you are not measuring the voltage drop across 1 meg at all, as you thought, but across a resistance of 999R - the parallel values of your 1 meg plus your 1K meter!

A second limitation which will often apply to very small HV supplies is that the amount of current drawn by the meter may well be sufficient to seriously lower the output voltage you are trying to measure.

To try to measure a source of 3kV @ just 1mA with a simple meter and multiplying resistor arrangement is a rotten idea. The sensible way of measuring voltages of this order is with a voltage divider that places only the lightest possible load on the voltage to be measured, which is also a constant load.

To measure 3kV, a voltage divider of 100:1 would enable you to measure your 3kV on an ordinary digital multimeter, which would now represent your 3kV as 30V.

Your 100:1 potential divider is no more than two resistors in series, the one 100 times larger than the other (for rough work) - the large resistance we shall call the dropper, and the smaller one across which you measure your voltage, is naturally enough called "the measurement resistor."

The higher the ohmic value of your potential divider, the less it will load the circuit to be measured, and so the more accurate it will be.

Selection of component values;

Most consumer DVMs have an input impedance of 10M. So if you use one of these to measure the voltage across a 10M resistor, you will in fact be measuring the voltage across the 10M resistor in parallel with the 10M of your meter impedance. So you are measuring the voltage across only 5M, and not 10M as you thought.

So in our potential divider, you should make the measurement resistor at least 10 times smaller than your meter impedance, which will work out as 1 meg for most meters, so your meter reading with only be pulled down to ~90% of the true value. (When you measure the voltage across a 1M resistor with a 10M input meter, you are now measuring the voltage across 1M and 10M in parallel, (909.091k) and not across 1M as first appeared.

In your particular case, and using practical component values, I'd suggest a measurement resistor of 1M in series with a dropper of 100M, so your DVM will read 30V to represent 3kV, and will be accurate to 5 percent or so in practice.

Selection of moving coil meter

From the foregoing it will be apparent that the higher the resistance of of a moving coil voltmeter, the better it will be. (This property is known as ohms-per-volt, and the higher the better) Trashy meters of the kind used in consumer products often have resistances of less than 100R, whilst a good quality instrument will have a resistance of 3K or so).

A 'wattless' measuring instrument which becomes useful around the 3kV level is the electrostatic voltmeter, which has at its heart the attraction or repulsion of metal plates by the charges placed upon them. In a model theory of the electrostatic voltmeter, the DC impedance of the instrument is infinite so no current is drawn from the circuit to be measured, beyond that tiny amount one-off amount needed to charge the meter's deflector plates in the first place. These meters, are very costly new, but I have a number of ex-military types bought on ebay for very little - cheap because few folk have need of them, or understand their properties.

What is to be done? (as Lenin famously has it)

You can make almost any meter into quite a good measuring instrument, simply by sticking a high impedance buffer in front of it. A simple hybrid source follower using the famous (and very cheap) FET 2N3819 and a couple of cheap general purpose bipolars will increase your DC input impedance to 500 megohm or so. The drain current of these FETs tends to wander with temperature like an electronic thermometer, so its best to use a pair of them in a differential amplifier so that thermal drift in one is cancelled by the drift in the other. A few dollars worth of parts will make you a very useful DC voltmeter with an input impedance of 500M - while you can spend $100 on a meter with all sorts of amazing features, but which still has as an input impedance (Zin) of the bog standard 10M.

Hope this helps,

Harry.

previous 1 2

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob