Analog meters - Dealing with high voltage and current levels - Embedded.com

Analog meters – Dealing with high voltage and current levels

Editor's note: If you're like us and can't bear to part with that crusty old (but magnificent) analog meter, David Ashton offers everything you want to know about these gems in this three-part series:

Shunts and multipliers
A bare meter (if it is of the moving coil type) has a fairly low DC resistance and a full scale deflection (FSD) that will be anywhere from 50 µA to 10mA or higher. At FSD it will typically have a fair bit less than a volt across it. You can make it read a different value, which may be a higher current or a voltage, by using shunts , which are resistances used to divert — or shunt — current away from the meter, and multipliers , which are resistors used to drop voltage, or limit the current to FSD at a certain voltage.

Here are some values I measured on a few meters I had knocking around (don’t take me literally, meters are fragile and you shouldn’t knock them around!). Don’t worry about how I got these measurements, I’ll get to that later.

No. Type FSD current FSD Voltage Resistance
1 Moving coil 50 µA 143 mV 2860O
2 Moving coil 1 mA

190mV

190O

3

Moving iron

5 A

143 mV 0.0285O
4 Moving iron @ (30V) 131mA 6V 45O

@ Note: No. 4 is included for comparison.
See “Multipliers” below for further explanation.

If a meter directly measures the current that you want it to measure, you’re home free. If not, you have to adapt it to do so. You can’t get a meter to measure less at FSD than its basic sensitivity without using some electronics – for example the 5A meter above cannot be adapted to measure 50 mA. You can get a meter to measure more than its base FSD current, or indicate voltage instead of current, by using shunts and multipliers. So let’s look at those.

Shunts
Let’s say we have meter No. 2 in the table above with an FSD of 1 mA. Let's say we want this to give us an FSD of 1 Amp — 1000 times as much. We can do this by putting a shunt across the meter terminals so that 999 mA flows through the shunt and only 1 mA flows through the meter as show below.

 

The shunt is obviously going to have a very low resistance — 1/999 of the meter resistance in fact. With the 1mA meter above, this works out to around 0.19 O. You can probably just use a piece of copper or resistance wire as a shunt, though you’d have to experiment with the length and gauge of the wire. If you are doing this, you can measure a long piece of your shunt wire (a metre or two) and then guesstimate how much you are going to need for your shunt. Another trick is to make your shunt a bit longer than you need, and move one of the meter wires along the shunt until you get the correct reading (again, use a known good DMM to check). You can use a screw connector or a soldered connection to fix the meter wire in place. If your shunt winds up a bit too short, you can often file or saw your shunt a bit to increase its resistance and get the FSD just right. If you end up using just a low value resistor for your shunt, you can put a higher value preset pot across it to trim it to the right value (for example a 100O preset across a 1O resistor if your shunt needs to be about 0.9O), but shunts are generally too low a value to use standard resistors. You can get current sensing resistors down to 0.1O or less and these are sometimes ideal — see below.

You can work out the shunt resistance with the following formulae. You have to know either the meter resistance or its voltage at FSD for this, and how to find this is covered under “Testing Meters” in the final installment of this series.

 

The rule of thumb is to divide your desired FSD sensitivity by the meter FSD sensitivity, then divide your meter resistance by the result minus 1, which will give you the resistance of your shunt. Alternatively, if you know the voltage at FSD of your meter, just use Ohm’s Law to find the shunt resistance, as in the second formula above.  If there is a big difference between your actual and desired FSD — in the above case it is 1000 times — then you can leave out the -1 in the brackets of the first formula without going too far wrong.

Just as an aside: the 50 µA moving coil meter and the 5A Moving Iron meter above (No. 1 and No. 3 in the table) have almost identical voltage across them at FSD, and if you shunted the 50 µA meter to read 5A, the shunt resistance would be almost identical to the resistance of the 5A meter. This is probably coincidence to some extent, but illustrates that most meters, of whatever type, have around 100-200mV across them at FSD, and you don’t gain a lot by using a more sensitive meter than you need.

A professional shunt will be made of material with a lowresistance/temperature constant, so the reading will not vary withtemperature, but home-made ones are almost as good unless you’re reallyfinicky. You can often buy shunts for the standard meter brands andsizes, and they can be mounted externally or sometimes on the meterterminals.  Current sensing resistors often make good shunts, butcurrent sensing usually involves low voltages (10 mV or so) so checkthat your resistor can take the power of dropping 100-200mV for yourmeter.

A word of warning: Where possible, connect the meter andthe main current to the shunt separately. If they use the same screwsfor connection and a screw comes loose, the main current may try to flowthrough the meter. If this happens, exit one meter, stage left. Shuntswith separate connections like this are called “4-terminal shunts” forobvious reasons (2 terminals for the current, 2 for the meter). Using aconventional resistor on a PCB, you can do this: Mount your shuntresistor high off the board, which is good practice anyway, and take themeter connections off the resistor wires, not off the PCB. So if youget a dry joint and no current flows in the shunt, it won’t go throughthe meter.

Here is a typical 4-terminal meter shunt. It is ratedat 25A and will, with 25A going through it, generate 100mV for themeter. The current to be measured is attached via the large bolts, andthe meter via the smaller screws. The grid squares are one inch. Thisone is from Galco,which offers a lot of shunts and all sorts of other industrialelectronics stuff. For currents under about 10 or 20A a shunt will oftenjust be a length of suitable wire mounted in the back of the meter, butabove that something like this would probably be used. 

 

Icame across some rather nice 4-terminal resistors recently that wouldbe ideal for meter shunts. Available in a range from 1Ω to 1mΩ, thesePCB- and heatsink-mountable resistors are made by Rhopoint and willdissipate up to 10W. Intended for precision current sensing, they arehigh precision resistors (1%) but would make great meter shunts. Theywould probably need a resistor (or preset) in series with the meter toget the right meter FSD. Here’s a link to the datasheet


Rhopoint 4-terminal resistor

Multipliers
Ifyou want to use a meter to measure voltage, the trick is to put aresistor in series with the meter coil and arrange it so that when thedesired voltage is put across the combination, a current will flow tomake the meter indicate FSD. You would usually try to keep this currentas small as possible to limit the loading on what you are trying tomeasure. So I’d use my 50mA meter from the table above.

Sotake the 50 mA meter (No. 1 in the table above). If we want this meterto indicate 25 volts FSD, the entire resistance (i.e., resistor plusmeter coil) will have to be 25V / 50 μA = 500 KΩ. Now the meterresistance is 2860Ω – which is about 0.57 % of 500 KΩ — so we could justuse a 500 KΩ resistor and our readings would only be 0.57 % out — theprecise resistance needed is 497.14 KΩ. If you are finicky, or yourmeter resistance is more than 1% of the total resistance you need, youcan use a preset resistor — preferably a multi-turn one — or betterstill a combination of a fixed resistor (use the nearest preferred valuelower that what you calculated — say 470 KΩ here) and a preset (say 50KΩ) in series to get to the exact value required. Your calculatedresistor will often, as here, turn out to be a non-standard value, sothis is a good way to go. The average DMM these days is more accuratethan most analog meters, so you can use your DMM to compare andcalibrate your meter. Here is the formula you should use to work outyour multiplier:

 

Knowingthe meter resistance is not as important for calculating multipliers asit is for shunts, as the meter resistance is often a pretty negligibleterm in this formula and can often be omitted. And if you do drop thisterm, RMULT is directly proportional to the desired VFSD ,leading us to the Ω/V metric that was essential to know in the days ofanalog multimeters. Using a 50 μA movement, you will need a totalresistance of 20 KΩ for every volt of FSD. This is good for “quick anddirty” guesstimates of the resistor you will need, you can then use apreferred value a bit less than this and use a preset to make up thedifference and get your FSD exactly right. For your convenience I haveincluded a table below of the Ω/V for the most common metersensitivities.

Meter Current Ohms Per Volt
50 μA 20 KΩ
100 μA 10 KΩ
200 μA 5 KΩ
500 μA 2 KΩ
1 mA 1 KΩ
2 mA 500 Ω
5 mA 200 Ω
10 mA 100 Ω

Onething to note: If you’re adapting your meter to read a very highvoltage, make sure your multiplier resistance can take that voltage asit will have most of the voltage across it. You can get special largeresistors for this purpose. Take care with your insulation andclearances too! Here is a typical resistor for this sort of thing — youcan get resistors like this in values up to 500 GΩ or so. This one isabout 2 inches long, rated at 20kV:

Asan illustration of the insensitivity of Moving Iron meters, considerthe last one (#4) in the table of my four meters above. It is a 30 VMoving Iron meter. The coil resistance is 45 Ω and at FSD the coil has 6V across it, and a current of around 131 mA in it. It has an 180Ωmultiplier resistance, which drops the additional 24 V. That’s 7.5 Ohmsper volt! The whole assembly at 30 V dissipates 4 watts! My employeruses a 15 V version of this meter in our radio sites, and we found thatafter a few years the plastic faceplates would go brown. Now I know why.See below (under “Commercially Bought Meters”) for some pics of thismeter’s guts.

Measuring AC
Ifyou want your meter to measure AC voltage, all you have to do is put arectifier — preferably a full bridge rectifier – in series with it likethis:

 

Howeverthere are some considerations here: The diodes you use will have acertain voltage under which they will not conduct (around 0.6V forsilicon and 0.2V for germanium), so in the diagram above, if you usesilicon diodes you will need at least 1.2 volts in your measured voltage— or about 0.85 V RMS — before your meter will indicate anything. Withgermanium diodes it will only be around 0.4V, or 0.28 V RMS — so thereis a use for those old germanium diodes in your spares box!  Because ofthis, the bottom end of the scale will be decidedly non-linear. On ameter reading 300 VAC FSD this will not be a huge problem, but if the FSD is 10 VAC the effect on the scale will be very marked. Some circuits I have seen —notably AC millivoltmeters — include the meter and diodes in thefeedback loop of the final amplifier, so that will keep the scale linearwithin the limitations of the amplifier’s slew rate.

If you onlyuse one diode as below, you get less voltage drop, but this halves thesensitivity of your meter. You should put another diode in the oppositepolarity across both the meter and the diode, unless you know the meterdiode can take the full reverse voltage of the AC. But diodes are cheap,so why not go full-wave?

Bearin mind that a simple rectifier like this will not give you a true RMSAC voltage reading but an average value. Depending on your waveshape youmay have to recalibrate it, so making your RMULT partly a preset is a good idea, as is using an RMS DMM to calibrate it.

Ifyou have a meter reading current, with a shunt your options are torectify all the measured current — which if it is reading 300A willinvolve buying some very expensive diodes — or just to rectify the smallcurrent through the meter — which will mean that your shunt has to haveconsiderably more voltage drop than for a DC shunt. There are other,better ways around this: You can use a Moving Iron meter, and/or usecurrent transformers for high currents, or you can use electronics tomake a precision rectifier — but this is an article about meters, so Iam not going to go any further there.

Center-Zero meters
Youcan get center-zero meters with the needle at rest in the center of thescale — which obviously indicate both positive and negative values.They are useful for things like battery backup or solar power systems,so you can see whether your battery is charging or discharging and byhow much. They were also common in the old teleprinter days, asteleprinters used either 20 mA “double current” — negative for a markand positive for a space, or 40 or 60 mA “single current” which waseither on (mark) or off (space). So old teleprinter distribution systemshad meters that read +/- 60 mA. I looked after a few of these, and gotsome meters out of them when the Internet came in and they went out.This is a really nice one that I have adapted using the above techniques- to read +/- 60 mA, 600 mA and 6 A — very useful for doingcharge/discharge tests on batteries.

 
Commercially bought meters
Ifyou go to electronics suppliers' websites, you will find that many ofthem still supply analog meters with a lot of different current andvoltage FSDs. If you buy a voltmeter that reads 20V FSD, say, inpractice this meter will usually have a multiplier resistor built intothe back of the meter. If you take the meter apart you will see it,probably mounted on one of the terminals inside the case. The same goesfor ammeters. If you buy a 5A moving coil meter, it will probably have ashunt mounted between the terminals inside the case at the back. Thissaves you the trouble of doing it yourself, but it is useful to know,because if you have some spare meters in your “junk box” you can oftenmodify one to do what you want, even if it does not have the FSD youwanted. As an example, I got a couple of meters out of a dead powersupply some time ago. One of the meters read 30 Amps FSD, not of muchimmediate use to me, but the basic meter sensitivity was 4 mA FSD, whichis a lot more useful. Moving Iron meters are a different matter — theyare generally built to their range, for example a 5A Moving Iron meterwill have a few turns of thick wire as the coil. Apart from rewinding ityou can’t do much about that, and you’re stuck with the 5A sensitivity.A 30V Moving Iron meter will have a coil of many turns of thin wire andmaybe a dropper resistor as well. Here are pictures of two Moving Ironmeters with exactly these sensitivities.


Moving Iron 5A meter coil (the back of the one whose face is above) — few turns of thick wire, 0.028 Ω  resistance.Note the high quality terminals :-). Cheap as chips and notsuper-accurate, but very rugged and reliable. The coil bobbin is about1¼ inches in diameter.



Moving Iron30V meter coil and multiplier resistor. More turns of thinner wire, anddue to the dropper resistor they have used a PCB with half-decentterminals. But with 4 W dissipation at FSD, it takes 131 mA from yourpower source before you’ve even begun to power your load!

Besure to check back for the final article in this series where Iconclude with a discussion on several practical aspects associated withusing these measurement wonders.

What are your experienceswith analog meters? Do you love ‘em, hate ‘em? Have you any storiesabout analog meters? Please comment below.

Be sure to read the other installments in this series:


Join over 2,000 technical professionals and embedded systems hardware, software, and firmware developers at ESC BostonMay 6-7, 2015, and learn about the latest techniques and tips forreducing time, cost, and complexity in the development process.

Passes for the ESC Boston 2015 Technical Conferenceare available at the conference's official site, with discountedadvance pricing until May 1, 2015. Make sure to follow updates about ESCBoston's other talks, programs, and announcements via the Destination ESC blog on Embedded.com and social media accounts Twitter, Facebook, LinkedIn, and Google+.

The Embedded Systems Conference, EE Times, and Embedded.com are owned by UBM Canon.

5 thoughts on “Analog meters – Dealing with high voltage and current levels

  1. “Davidnn”Available in a range from 1u03a9 to 1mu03a9, these PCB- and heatsink-mountable resistors are made by Rhopoint and will dissipate up to 10W. Intended for precision current sensing, they are high precision resistors (1%) but would make great me

    Log in to Reply
  2. “David,nBrilliant article.nAre you touching on the electrostatic meter. nBefore the days of solid state charging circuits for car batteries and when cars became main stream they dropped off the charging meter. I used to make my own charge / discharge m

    Log in to Reply
  3. “Thanks Crusty. Electrostatic meters – alas no, the nearest I have ever seen to one was the gold leaf electrometer in physics at school. I wanted to keep to practical topics and the article was getting out of hand length-wise. You should write one though

    Log in to Reply
  4. “@Antedeluvian…it's the weekend and I had a bit of time to fossick around….here's an examplenhttp://www.isotekcorp.com/products/precision-and-power-resistors/A-HnTasty goodies indeed. Thanks for that.”

    Log in to Reply

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.