Why You Shouldn’t Quite Forget The Moving Coil Multimeter

Did any of you have an AWS multimeter? Was it the best? Radio-electronics magazine, August 1981.
Did any of you have an AWS multimeter? Was it the best? Radio-Electronics magazine, August 1981.

If you were to ask a random Hackaday reader what their most fundamental piece of electronic test equipment was, it’s likely that they would respond with “multimeter”. If you asked them to produce it, out would come a familiar item, a handheld brick with a 7-segment LCD at the top, a chunky rotary selector switch, and a pair of test probes. They can be had with varying quality and features for anything from a few dollars to a few hundred dollars, though they will nearly all share the same basic set of capabilities. Voltage in both AC and DC, DC current, resistance from ohms to mega ohms, and maybe a continuity tester. More expensive models have more features, may be autoranging, and will certainly have better electrical safety than the cheaper ones, but by and large they are a pretty standard item.

If Hackaday had been around forty years ago and you’d asked the same question, you’d have had a completely different set of multimeters pulled out for your inspection. Probably still a handheld brick with the big selector switch, but instead of that LCD you’d have seen a large moving-coil meter with a selection of scales for the different ranges. It would have done substantially the same job as the digital equivalent from today, but in those intervening decades it’s a piece of equipment that’s largely gone. So today I’m going to investigate moving coil multimeters, why you see them a lot less these days than you used to, and why you should still consider having one in your armoury.

Evidently the Triplett Model 60 had hidden depths. Radio-Electronics magazine, August 1981.
Evidently the Triplett Model 60 had hidden depths. Radio-Electronics magazine, August 1981.

There was a period through the 1980s into the 1990s, when the transition between moving coil and digital was in full swing. Everyone wanted a digital meter because they were cool. They had both precision and accuracy, instead of saying “That’s about 5 volts”, you now knew it was exactly 4.98 volts. It didn’t matter that the difference between the two readings had no effect whatsoever on the circuit in question, you had a cool multimeter, and you could measure down to 0.01 of a volt. Engineers walked Tall in their aviator sunglasses to their Ford Sierra/Merkur, and drove off into the sunset knowing that they had Tamed The Last Fraction Of A Volt.

So we can make fun of 1980s fashions, both sartorially and in test equipment, but what’s the real benefit of a digital meter? It lies in the internal resistance. When you hook a moving-coil meter into a circuit, you are doing work, drawing energy from the circuit being measured to move that pointer. Imagine the meter itself in a hypothetical inexpensive moving-coil multimeter, that has a full scale deflection of about half a milliamp.  To take a reading at full scale the meter must then draw 0.5 mA from the circuit being measured. So to give a full scale deflection of 10 V for example the meter must have a resistor in series with it of value 10 KΩ, and measuring a voltage thus involves placing what is in effect a 10 KΩ resistor into the circuit in question. Most decent moving coil multimeters had meters with much lower full scale deflections that required greater series resistances, but the principle was the same. A moving coil multimeter loads the circuit it was testing, altering its characteristics.

High-impedance voltage measurement, 1950s style.
High-impedance voltage measurement, 1950s style.

The problem of meter resistance was one for which there were solutions in the age before digital meters. Many decades ago you could buy a valve (tube) voltmeter, in which the meter was driven through an amplifier with a high impedance input. Even when these were replaced in the solid-state era with FET amplifiers they remained expensive devices, and the average bench would not have had one. When mass-market digital meters arrived in the 1970s they incorporated FET input stages, so immediately there was the possibility of a multimeter with an impedance in the mega ohms rather than the kilo ohms. These meters could be hooked up to a circuit from which they would draw negligible amounts of current, and whose characteristics they wouldn’t affect in the slightest. People bought digital multimeters, and never looked at their moving coil devices again.

Moving Back to the Coil

My Avo 8 isn't as clean as this one. Megger Ltd. [CC BY 3.0]
My Avo 8 isn’t as clean as this one. Megger Ltd. [CC BY 3.0]
So why should you have a moving-coil meter on your bench, if digital meters are so good? Aside from looking a bit retro and never having to worry about the batteries running out, that is. The answer lies in being able to measure voltages that change. If you have ever worked with radio or with analogue circuits that require adjustment you will know something of this, tuning for a peak or a trough is extremely difficult when you have no view of the trend. A typical use for a moving coil meter here is to set up an RF amplifier by tuning a tank coil, there will always be a peak in the current drawn by the amplifier as the coil reaches resonance.

The moving coil meter of choice here for the last 30 years or so has been a venerable Avo 8. This is the classic mid-twentieth-century multimeter, a large and heavy Bakelite unit with an extremely high quality meter lurking within it. The full scale deflection of the meter itself is a relatively tiny 50 µA, so its impedance when measuring voltage is better than the cheap meter example quoted above, but in 2017 it’s a bit of a museum piece. Thus this article will now turn into a mini-review of a more accessible moving coil meter, as  I’ll take a readily available model from 2017 and put it through its paces.

5.8V on the MF47 when it should be 6V.
5.8 V on the MF47 when it should be 6 V.

The TianYu MF47 can be had from the usual Chinese suppliers for about £15, or $20. It’s not the most compact of multimeters, having a front panel slightly larger than and being about three times as thick as a DVD case. The upper half of the front panel is occupied by the large meter, while the selector switch takes up the bottom half. There is a handle that folds behind the meter to make a desk stand. It offers the usual multimeter voltage, current, and resistance ranges, plus a transistor tester, a battery tester, and a capacitance meter which I haven’t figured out how to use yet due to an all-Chinese manual.

On the back is a hatch for batteries, a C cell and a PP3, and another for a fuse. A really nice touch is a spare fuse clipped to the inside. The Chinese text below the meter face extols the virtues of its protection circuitry and gold-plated contacts according to Google Translate. If I wanted to describe it in a sentence, I’d put it as a semi-decent budget workhorse multimeter, its relatively low price is a function of its coming direct from Chinese suppliers rather than indicative of deficient quality.

I can demonstrate immediately the effect of the relatively lower internal resistance of the MF47 when compared to a digital multimeter by looking at a simple potential divider. Connecting two 22 kΩ resistors in series across a 12 V supply, you would expect the resulting voltage at the meeting of the two resistors to be 6 V, or half the supply voltage. This is confirmed with a digital multimeter, my run-of-the-mill Uni-T UT33D. However, the MF47 measures 5.8 V, as it has a low enough internal resistance to affect the lower half of the potential divider and pull the voltage down by 0.2 V. The MF47 documentation has a table of specifications in which the internal resistance is quoted as 20 kΩ/V, and indeed measuring the MF47’s resistance on its 10 V range with the Uni-T returns a figure of 200 kΩ. Adding this extra resistor into the voltage divider circuit gets us to 5.8 V.

When you are dealing with lower impedance circuits there should be no issues when using a moving-coil meter such as the MF47. These meters make rugged and cheap instruments for automotive electrics and (subject to the claims of protection circuitry being true!) household wiring, in addition to the previously mentioned use for finding current peaks and troughs in analogue circuits. The MF47 will stay on my bench as a much lighter alternative to the Avo, but it won’t replace my Uni-T by any means.

It’s been an interesting reconnection with the past to refamiliarise myself with a moving-coil multimeter, and to remember why we all use digital meters today. There is a temptation to look back through rose-tinted glasses, but in the case of multimeters they aren’t really justified. Still, it’s not entirely a waste of time to have a moving-coil meter on your bench, so if one comes your way don’t necessarily pass up on it. If nothing else, you’ll never have to worry about whether you have a spare 9 V battery on hand again.

80 thoughts on “Why You Shouldn’t Quite Forget The Moving Coil Multimeter

    1. Yep. A restored/calibrated Simpson 260P right next to the Eico 5MHz scope. Both get used routinely. Just not as much as 20 years ago. The daily driver is the Fluke though. Diode-mode, ftw.

      1. 73 y.o, retired EE. Still have and use my Simpson 260 and Eico VTVM. Simpson 260 production was started before I was born, and still in production. Bunch of model changes over the years.

  1. There’s things that you should probably check out with other types of equipment that you can at least “see happening” on a moving coil meter, that are very hard to read on a digital. Examples might be voltage ripple (needle twitching), back EMF of a coil, the difference between a capacitor that’s failed open circuit and one that’s still capacitating away (Whether the value has drifted or not.)

    1. Yep, I used to have an analog meter for troubleshooting circuits on burglar alarms that would “randomly” trigger; it was hardly ever random and was normally caused by induced voltages from other equipment switching on and off. A digital would just read the nominal voltage even though the alarm was triggered but with the analog you could catch the “twitch”, then the real fun began hunting down the offending equipment.

  2. I was taught by an experienced broadcast engineer to value an analog multimeter. He didn’t like the digital ones because real world voltages often have ripples. While working on some device, I forget what it was he showed me the problem. The voltage was quickly varying between some x.9 and (X + 1).0. For the application that tenth of a volt was completely unimportant we could have done without that precision. The one’s place however was important and also unreadable because it was moving between the two numbers so fast it just looked like all the segments involved were continuously lit!

    The analog meter on the other hand can’t even respond to such fast changes. The needle just rests right into a comfortable in-between spot. He likened this to having a built-in low-pass filter.

    Of course.. if this were an application where we really cared about that ripple it would have been better to know about it but sometimes such things aren’t important. When they are important a multimeter isn’t the right tool anyway, that’s what an oscilloscope is for.

    Unlike him I do still appreciate an auto-ranging digital multimeter too. When one isn’t suffering the ripple problem they are quicker and easier to read. I keep both around and usually use whichever one I happened to have laid down closest to where I happen to be working.

    So.. why doesn’t anyone make a combo unit? I’m imagining a typical digital multi-meter circuit but the microcontroller has a pwm output feeding an analog meter as a second display. Hmmm… sounds like yet another project for my some-day list!

    1. Fluke meters have or had a bar graph at the bottom of the digital display to help solve this problem. It never seemed as useful in real use as it sounded.
      If you sampled the signal fast enough, feeding the meter from the MCU would work. Otherwise, having an analog path to the meter would give better results. You could auto-range the meter through the MCU if you wanted to. Personally, I dislike auto-ranging meters, it makes measurements much slower to get.

      1. Nowadays I don’t mind auto-ranging meters, but back in the 70s and 80s when I worked as a TV tech, I hated them. I could work a lot faster with a non-auto-ranging meter, and time was very much money. For the hobby stuff I do now, speed is not normally important.

      2. Ah yes, the Fluke 77 – it had a bar underneath the digits. I still have a far east clone of the Fluke 77, and it also has the bar. The bar does respond faster than the digits, and it’s analog-y, so it’s still quite handy to see transients.

        One thing I like on meters: a continuity or resistance setting that beeps.

        When you need to concentrate on one range, all auto-ranging meters can usually be forced into manual range selection. Some meters also have min/max functions that will grab the min (or max) value and hold it til reset.

        Anyway, analog meters are still on my bench too. I loved seeing the big ole AVO meter here – the older versions of those were the ones on the bench at my first job.

  3. FET voltmeters weren’t that expensive, even as a poor student from a poor family I could afford a Micronta FET meter. And those old meters had one big advantage, you could watch how the needle would behave over time if you didn’t own an oscilloscope.

    1. watching the needle move over time… wouldn’t an X-t writer be more suitable then?
      But seriously, the needle has it advantages. As mentioned in the article, I also have one around in case the batteries of my digital meter are dead but also when I don’t trust my measurement. Sometimes my digital meter shows values that are simply wrong, showing a higher voltage then it should. Then I rotate the voltage selector and measure again. I know I should clean the contacts of the switch and I will do so when it happens more often.
      In the mean time my analog meter is my trusty backup. Also very handy when I need to measure two different things at the same time.

    2. I had a Micronta FET voltmeter and a Heath Vacuum Tube Voltmeter when I was in high school. They did the job well. The problem with the VTVM was that all measurements had to be referenced to power line ground. If you forgot and clipped the negative lead of the meter to a point that had a potential other than power line ground you suddenly understood the need for an isolation transformer… I was a HAM back then, and the analog meter was much better for peaking or dipping a circuit. When I graduated high school and was in college paying tuition through a day job as a TV Tech, I got myself a Beckman digital meter. It lasted many years after that. IMHO vastly better than an auto-ranging Fluke of the same vintage.

    1. Simpson 260 is still my main troubleshooting meter. I’ve got a calibrated Fluke 8842A on the bench for making precise measurements, but you can’t beat the old 260 for a quick check-out.

      Seemingly little known fact: there’s a small version of the 260, the Simpson 160. It uses a 22.5V battery for high resistance ranges, which is still obtainable from various suppliers (and lasts forever anyway). I’ve got one in my travel toolkit, it’s as nice as the 260 just smaller. Make sure you get probes with it, they’re not standard sized banana jacks!

  4. The moving coil meters were rated in Ohms/Volt to indicate loading. 20,000 Ohms/Volt was a common value, so on the 3V scale, the resistance across the probes would be 60K Ohms, while on the 12V scale, it would be 240K Ohms. The AC setting was generally different. My old Tripplett meter shows 5000 Ohms/Volt for AC scales. Just for grins, I checked these results with another meter, and they are close, but not exact. The Tripplett meter has these ratings written on the meter face.

  5. Ah, the 260 – the mainstay in the house full of engineers I grew up in. 20k/volt, not half a ma per…ten times less load. Everyone who could laughed at the lesser meters. Early (and most modern) digitals really stank with trending voltages, and didn’t show ripple correctly via the pointer twitch. As a DEC field service guy, checking some early switching power supplies a digital would show goodness, but a scope would show an open output capacitor – a matter of some importance when that many $ worth of logic was going to smoke. Digitals are just now catching up with an ohms-type setting to test diode forward drop (or get the polarity on a diode or LED).
    Thing is, while I have some really sweet old D’arsonval meters and movements – I have to be super careful with them around my shop, as I machine iron things and the resulting fine filings have a way of getting inside meter movements. While you can get them out – it’s not a simple easy job at all.

  6. I’d definitely advise folks buying a used Simpson/Triplett/Avo to look beyond whatever was the most popular model, there are some other models that may do everything that you want but have less name recognition on ebay and thus go for a lot less money :)

    1. I have one of the older Micronta (Radio Shack) 20kohm/volt meters (no mirrored scale), which I used before I acquired the Simpson 260. It worked well, and seemed fairly accurate for the 39.95(?) I paid. Still have it.

      The probes are crap though…lots of cheap plastic and only a few strands of highly non-flexible wire. Simpson probes are MUCH better, and used banana plugs instead of the pin plugs and crap sockets on the Micronta.

  7. “So why should you have a moving-coil meter on your bench, if digital meters are so good? […] The answer lies in being able to measure voltages that change.”
    Bar Graph in better multimeters.

    1. Bar graph often doesn’t have the resolution needed to see interesting changes, and the response depends on the low pass filtering implementation. One less expensive one at work (I don’t recall the make) seems to have no meaningful low pass for the bar graph, and when looking for a change in the DC level with a superimposed AC signal, it is often invisible, even when the change is of sufficient magnitude to change the number of bars. The ones with decent low pass are still too low resolution for many uses.

          1. I got used ‘scopes at hamefests. I’m not sure that’s a good option anymore, I haven’t been to one in years, but I’ve been told that they aren’t as good a source as they used to be. Perhaps eBay? It might be possible for a serious young hobbyist to get a low-end new ‘scope for $200 or less, in 2017 dollars that’s probably about what I paid years ago in ’70s dollars.

          2. I just noticed this right here on Hackaday for young people who want ‘scopes: https://hackaday.com/2017/11/09/review-jye-tech-dso150-oscilloscope-kit/ It’s really inexpensive, $21, but it is a single channel unit with a very modest bandwidth of 200KHz. You can get a dual channel 10MHz ‘scope with probes from pico technology. I’ve used their 2204A and it works very well. It does require a computer running Linux or Windows, no display of its own.

  8. I worked in electronics repair for 30 years. (damn I’m old) Used the analog meters till the end. You can find a bad transistor in curcuit with a moveing-coil, digital forget it. Use as a tracer in a pa system , or for testing speakers in general, analog every time. Seeing the change in a 70 volt (PA) line when you forgot your headset in the van. Digital don’t work.
    True, on the bench I had a VTVM and in the field I had a FET. But any analog meter would do in a pinch.
    We had digital meters avalable, sometimes it was important to know a voltage ( amp. ohm) exactly. But to trace a problem , not very often.

      1. To find a bad transistor in curcuit, you check the forward current in the diodes. It takes a bit of practice because it is in curcuit. But with the tansistor you are talking open or short. (check both ways, just try to find the diode) you could see this very quickly with a needle, the digital ones (because you where in circuit ) where always ambiguous.

  9. Bar graph would need a lot of divisions to provide the information a MC meter does. One example being rate of change, bar graphs will not indicate that the way a moving needle does.

    I have both, but I prefer my Simpson.

  10. Good article; but it missed or glossed over a few point:

    – The standard name for this type of meter was VOM (Volt Ohm Meter).
    – The VOM is “always ready”. It never has a low battery that produced false readings, or shut itself off just when you need it.
    – Most VOMs are 20K ohms/volt (50uA full scale). Cheap ones were worse; good ones even better.
    – VOM circuit loading changes with the scale. Modern meters present a fixed load (10 meg etc.) So on higher-voltage scales (common in the vacuum tube days), the VOM had *less* circuit loading than a VTVM or digital meter.
    – VOMs usually have a *lower* burden when measuring current. Digital meters typically have a 400mV drop at full scale when measuring current. VOMs were generally less (the Simpson 260 is 250mV).
    – A VOM is less likely to “lie” to you than a digital meter. With oddball waveforms, near RF, TV flybacks, and transformers with big magnetic fields, the digital meters were more likely to display incorrect readings (especially the cheap ones).
    – The VTVM (Vacuum Tube Volt Meter) was just as common, and priced about the same. The Heath IM-10, IM-11, and V-6 were common examples. The VTVM needed AC power, but didn’t load the circuit; so each had its place.

    The article’s conclusion is correct; you really need *both* on your bench. After all… with only one meter, how do you know it isn’t lying to you!

      1. Unlike a digital meter, a VOM does not give false readings with a weak battery. To measure resistance on a VOM, you set it to “ohms”, short the probes, and then set the Zero knob to show 0 ohms. Then connect the unknown resistance, and read its value. If that works, it’s accurate even with a low battery.

        Of course, it doesn’t need a battery at all to measure voltages.

        A VOM battery lasts so long that it dies of old age rather than from use. The battery in my Heath VOM finally failed. When I opened it up to replace it, I saw that I had last replaced it 20 years ago!

      1. Interesting. I’ve been a ham and used moving-pointer meters for over 50 years, and never heard of this. Can you provide references? The d’Arsonval movement only responds to DC, and there is a huge amount of resistance in series with the inductance of the coil, so I don’t see how it could happen.

        There are of course many other types of moving-pointer meters than the D’Arsonval. Some are intrinsically AC. Maybe one of these types could have false readings due to frequency?

  11. When I ran an electrical test lab a few years ago, I had an old analog HP 4329A high resistance meter with its own test cell from the 60’s or 70’s. I could gain more knowledge about the material I was testing by watching how the needle moved than I ever could watching the scrolling and flashing numbers digital meter I bought to replace it when it died. Watching the needle was great to be able to get an accurate average on measurements that were so sensitive the the needle moved when I approached the sample or even when I breathed with high enough resistances. I mourned it the day it died and couldn’t be kept in calibration.

  12. Should someone forget their moving coil multimeter, just give it to me; the analog scale is invaluable to appreciate small changes in a measurement.
    We could fake that function by implementing dials on very high resolution screens, but that would cost a bit. In the meantime, an opamp, a bunch of diodes, some precision resistors and a switch can turn any “old” meter into a true-RMS very accurate high input impedance instrument.
    http://www.hqew.net/files/Images/Article/Circuit_Diagram/DC-RMS-PEAK-TO-PEAK-High-Impedance-Voltmeter.gif

    1. Nice circuit; but it doesn’t really display true-RMS (or peak or peak-peak, either). It always displays the average value; but the switch changes the calibration to show the “textbook” relationships between average, RMS, and peak for the special case of a sinewave.

      This is a very common practice. *Most* meters on their AC ranges only measure the average voltage; but their scales are calibrated as if they are displaying RMS. This only works for the special case of sinewaves.

      1. You’re right of course. Non sinewave signals would need a different approach, but it should be possible to keep it very simple as well. A fast enough ucontroller sampling the signal to get the real average value would give more accurate measurements, probably even using less parts since AC rectifying would be done in software.

  13. Another reason to use a traditional analog meter: Schematics for antique radios are sometimes labeled with voltages measured with such a meter. If you use a digital meter, you need a load resistor to match their measurement. Of course, you still have to make sure the ohms/volt rating of the meter matches the one they used.

  14. @Ostracus Good analog meters have *never* been cheap.

    There was an EEVblog thread about worst multimeter and someone posted a picture of a $2 2K/V meter. Those can actually be OK if you add an FET to the voltage circuit. And very handy on field day if you have an RF probe to go with it.

  15. A bunch of comments…

    1) If you really want to be retro, you get a potentiometer like this: https://img0.etsystatic.com/110/0/9415172/il_570xN.1014780738_k5g0.jpg

    2) “That’s about 5 volts”, you now knew it was exactly 4.98 volts. ” You have to be really careful with that one. You get into that entire accuracy vs precision in measurement thing. And when you get even more digits you get things like thermocouple effects to deal with.

    3) If you want the “cool” vintage DVM you want the non linear systems one. I finally got one of these http://www.stevenjohnson.com/pics/nls-lm-4-front.jpg 30 years after the fact.. I lusted for one of them as a kid.

  16. One small disadvantage was the angle of the moving coil. If the meter lies flat on the table or stands vertical, gives a rather large difference. On the scale of the meter there’s a special icon to show the angle it is designed for.

    To me the biggest advantage if digital meters is it’s almost shockproof. Fluke is/was exceptional with that regard, alsmost unbreakable.
    MC meters will bend or break when dropped or mistreated.

    1. I second every word you wrote! Transportability, orientation (btw. I could not find the symbol on any picture of Simpson 260 I glanced at).

      Anyway: I do have AND use both meter types :-)
      Like just 2 hours ago while checking some “DC wall warts”: DVM showing anything but reasonable values, da trusty Metravo 1H helped me sorting out the business.

      And do never forget: “One always measures a bit off, one just has to know by how much.” (D. Packard, as far as I can tell)

  17. Many schematics of organs thru those classic decades usually specified “all voltages measured on a Simpson 260 voltmeter unless noted”, that’s a standard for sure, never heard of the one featured. The other thing it had is a reversing switch for the ohms. Clip two leads on to a transistor or diode, the junction is tested just by switching not lead fumbling. Gave meaningful results with germanium stuff too.

    Telcos loved the meter (50volts scale) and switch, as the peak reading when switching back and forth across a capacitor on high ohms world give a ballpark reading of capacitance and hence how many ringers were on a house line. An illegal extension phone, oh those times!

    1. Back in those days, I had set up several illegal phone extensions at my parent’s house. I used a homemade optocoupler with an NE-2 neon light and a phototransistor. The neon light in series with a resistor was connected instead of the ringer, a battery pack, a Mallory sonalert and a few other components finished it. I did this when I was maybe 14 years old, and we never got caught. Our neighbours who just plugged in extra phones did…

  18. Working on 60’s Hondas (points) a VOM wins for ease. The jitterry needle indicates any iffy connection. My personal fav was Michronta 1Kohm/v (methinks) for $3.98. Size of a cig pack, went into my ashtray w firecrackers beside to deter pushy smoker friends… had it since 11 yo. Never failed me. Now use a $100 water resistant, 6 ft drop Klein (China.) Frustrating time lag. Push a button for DC, instead of just setting it to begin w. Like a power mixer and a whisk, each excel, differently. VOMs kept one thinki g, so you didn’t torch your neter. Autoranging SUCKS… It’s just ICs are cheaper than a physical switch!!! Bean counters… ruin a lot of good things.

  19. I had a VOM, briefly. It was 1972, someone gave me one, I think he’d gotten it from someone. It was decent, 20,000 ohms per volt, and an actual range/function switch, cheaper ones sometimes required you to shift the leads to different jacks. The back was missing too.

    It didn’t last too long, I’d forget to change the switch, so the needle slammed against the stop frequently. Soon it was bent, and then it was terminal.

    I had the use of an HP-410b VTVM after that, better since the AC probe was good at RF. But it was on loan, so eventually replaced by an RCA VTVM. I still have the RCA, and another one I found on the sidewalk about five years ago, but also got my own HP-410B about 1990.

    So my experience with a VOM was limited and short.

    In 1978 I came into fifty dollars and spent it on an Intersil DVM kit. This is when they introduced the 7107 DVM IC, so they offered a pair of kits, but the 7106 LCD was more expensive. They were simpler than previous DVM ICs.

    I had big plans for it, but never got the precision resistors needed for the input attenuator. I had a surplus meter intended for signal strength in a stereo receiver, and a then new RCA ca3130 or maybe it was a 3140 mosfet input opamp. Very high input impedance. I was going to use that op-amp as a buffer, in parallel with the input of the 7107, and then feed the analog meter. Thus an indicator for trends. But it never got that far.

    I spent about $100 Canadian on a Radio Shack DMM in 1984, and that was cheaper than what came before.

    Yes, bargraphs on meters are updated too slowly. There were some DMMs in the seventies with an analog meter, but they weren’t common. I know I got the idea from seeing an ad for one. But none of the DMM project in the magazines included an analog meter.

    It’s handier in one box, but since the analog meter is for trends, one could just find a nice bare meter, and add an attenuator and opamp. You don’t need to calibrate it, you don’t need a precision attenuator, just need to get the input voltage in range, and an arbitrary scale on the meter. There was a period when this sort of thing was common as a magazine project, except intended to replace the VOM so precision and calibration were important.

    Michael

  20. Shown but not mentioned are the antiparallax mirrors under the pointer to reduce reading errors by your aligning the knife edge pointer with its reflection (e.g. moving your head and using one eye)

  21. As an electrician the no load effect of a digital multimeter makes it harder to troubleshoot occasionally- if there is a bad service lice or connection in a circuit the dmm might read 120v (normal North America house voltage) but the circuit won’t actually be able to drive a load. So then I take a pigtail Lamp holder with a 40w bulb and start bouncing around until I can isolate the bad connection

  22. Many moons ago when I was in high school and doing vacation work in a radio repair shop, my mentor regularly did a load test on D size ‘torch’ batteries with an AVO 7 set to its 10 Amp range! My best effort to blow the AVO up was however met with failure as it was built to last and to survive all manner of misuse with a built in trip switch. I did however manage to spook every body in the building who thought we were being bombed by connecting a large loudspeaker to 220 Volt mains. My excuse was that there was a terminal block with white ripcord and brown rip cord so without checking or using the AVO, I connected brown to brown and white to white. Needles to say, my school vacation vocation ended rather abruptly. My current collection of meters consists of an AVO-8 Mk 5, AVO-8 Mk 7, a few AVO-7s. The other lesser brands did not survive.

    73 de Pine, ZS6GST

  23. Most of “advantages” of analog mm in comments are pointless. Obserwing changes can be achieved by min/max/avg functions and bargraph. Bargraph in my fluke 117 is refreshed 40 times per second and I doubt that any analog will let you notice changes that fast. Because of inertia any fast spike may be observed but not measured (again min/max function). There is also impedance measurement on 117 that let you eliminate ghost voltages. Of course I realise that those functins will not bring ergonomics of analog mm and will not totally cover all fields. I also convinced that every workshop should have one amm. It’s like calculators – you do have quadcore desktop for complex computing and calc app on your mobile but in many cases ergonomy of calculator can’t be replaced.

  24. Proud owner of two AVO8’s , a Yokogawa (YEW) Voltmeter and an Ammeter. Most trustworthy set of instruments sitting on my bench. Owned by my Dad since the early ’70’s and passed on to me. My Kiethley KTH2000 is already showing signs of age, but the old faithful’s are still going strong.

    1. Maybe these work best on technology that is old, like smal cc motorbikes from the 60’s. A digital is a pain. And seeing the quality of a vibrating needle surpasses a bargraph’s representation, just an an analog speedo or tach imparts general info in a format quicker for our brain to process than a digial speedo or tach. They make wonderful digital info bits though if beside analog couterparts.

  25. Although I started my career in Flukes handheld DMM business years ago, I still appreciate analog meters for many of the reasons previously mentioned. Additionally, you can use an analog ohmmeter to detect a shorted turn in a 50-60Hz power transformer. Although the resistance difference of a single shorted turn is not practical to measure from a 100-300 turn winding (assuming the short is in the primary winding) you can however, see the loss of inductance from a shorted turn by continually reversing the test leads to re-magnetize the transformer’s core oppositely. When the transformer is intact you will see a slight hesitation as the needle moves upscale towards the stable winding resistance reading. If you don’t swap the test leads between measurements, you will NOT see the hesitation in the needle’s ballistics. So if you don’t see a difference between swapped and un-swapped readings, then you have a shorted turn in a winding somewhere (or you are not on a winding with enough inductance to detect). You can’t do that with a digital meter.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

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