EEVblog 1446 - Analog Watt Hour Meters are AMAZING! 

We got the basics of this in the first year of university Physics classes, then revisited it two years later to solve the full set of field equations for both ideal and real materials, and then measure the difference in the lab to get the right values in the model for the real materials and see the affect of mechanical design choices. The hardest part, by far, was summing the effects of all the fields in the disc and the topology (shape) of the induced currents. We also had to allow for the thermodynamics. A factor in the size and design of the disc is to dissipate heat, particularly to minimize peak local heating during high loads, and to have consistent behavior during the full range of environmental conditions. Aluminum is chosen for the disc because it has a good balance between conductivity and heat transfer while also minimizing angular inertia. Iron alloys could not be used because they could be magnetized (though we did not consider austenitic stainless steels). A copper alloy could work, but the disc mechanical design would need to be very different, with affects that would ripple through to the rest of the meter design.

Out of passing interest... eons ago, a friend lived in a large apartment complex. All the electric meters were on on a wall in the laundry room... there must have been 50 or more. He was doing laundry one day, & out of mere curiosity found his meter & noticed it rotating quite fast... then noted it speeding/slowing in time various laundry machines cycling on/off! Cheesed him off... but to be sure he ran & turned off his main breaker, & returned to find the same! He called & reported to his DWP; and they were there seemingly within minutes. It seems the landlord had jumpers behind that wall & rotated the laundry and common areas to different meters on a regular basis. He said it turned into a colossal legal S storm that lasted about a year.

I think the grid pattern of divots in the rotor is intended to make the rotor acceptably flat. When you stamp sheet metal, it often won't come out flat if you just smash it between two flat dies. The blank was almost flat to begin with, so the tiny amount it moves during stamping isn't enough to overcome the metal's elastic limit. When the press opens, the stamping springs back to it's original almost-but-not-quite flat shape. But if you make little divots everywhere, you stretch the metal much more. By exceeding the metal's elastic limit, you convince it to take on the shape of the flat dies.

You have to remember that V x I x COS(theta) on the disc creates a TORQUE, not an RPM. That means once the friction is overcome, it would go faster and faster. Those permanent magnets provide a drag proportional to the speed of the disc. Without the drag magnets it just won't record watt-hours. When you add drag proportional to speed, and TORQUE proportional to Watts, you get Watt-Hour.

I love the mechanical solutions of the time just before the digital age, they are so incredibly clever always.

Years ago I had a bundle of wires inside a conduit and needed to figure out which of the 20 or so wires was connected to a particular branch circuit. I made sure every appliance was turned off, then connected a hair dryer on high to the branch circuit I was trying to identify. I then held a 1" x 3/4" neodymium magnet close to the wire bundle and spread the wires apart. It was very easy to identify the correct wire, the magnet was vibrating in my hand just like you showed in the video. A very useful method.

The amount of accumulated knowledge built into these things is quite amazing

That style of plug in meter base can actually have the meter installed rotated 90deg, when in this arrangement none of the electrical connections are made and the meter is just acting as a cover for the bare terminals. This function is used when the electricity company wants to disconnect supply from a property without removing the meter from site as the meter is tied to the address. It is used mainly when disconnecting people for non payment or when a rental property is unoccupied for an extended period and has the advantage that the meter can have the seals fitted to prevent the customer from tampering with the disconnection. Source - former metering tech

Hi Dave .. you did an excellent job explaining all this. Back in the 80s I spent over 10yrs working for a power company calibrating single & 3 phase meters and troubleshooting meter problems. People used to drill holes through the case and poke a bit of wire into the brake magnet (and of course remove it before the meter reader was due) and play around trying to slow them down with an external magnet. Our own experiments with external magnets showed that you are just as likely to speed the meter up ... This particular meter like most later ones has magnetic suspension usng the two repulsion magnets at the bottom combined with an upper and lower centering guide needle .. We typically tested meter calibration at both unity and .5 PF (Cos 60), the power factor adjust is the 2 copper vanes on the rotatable shaft. Full load adj usually involves movement of the brake magnet but this later L&G appears to be using a different form of brake magnet adjustment

We had a mechanical meter for a little while with a solar system. The retailer's systems eventually managed to cope with negative electricity usage each month, although eventually they fitted a non smart digitial meter that only lasted a few months before they replaced it again with one with a radio. It was pretty fun to watch the mechanical meter on a bright day when someone was welding as it would be spinning moderately fast backwards, before spinning very fast forwards for a few seconds, then going back to its usual spinning backwards :).

I wonder if the dimples are to stop the disc warping over temperature changes, reducing surface stresses

Faraday's law: The induced emf in the disk and the induced eddy current in the disk is proportional to the flux rate of change. The torque on the disk is proportional to the eddy current times the instantaneous magnetic field. When voltage is maximum (positive or negative), it provides a background magnetic field centered on the coil assembly so that the offset eddy current induced by the rapidly changing current is attracted or repelled. When the current is maximum (positive or negative), it provides a background magnetic field offset from the center that attracts or repels the centered eddy current generated by the quickly changing voltage. All four torques are in the same direction. Torque provides angular acceleration except that the permanent magnets' eddy currents provide opposite torque proportional to disk speed. Together the disk's speed is in equilibrium at some speed proportional to the IV's torque... proportional to power. The counter integrates the disk's speed (power) wrt time to yield Watt-hours. Very sharp spikes are filtered out by the disk's inertia.

Had a funny one for you. Had an analogue meter on a remote site that was permanently shut down; site uses like $150/month in power. After the site was shut-down, 3 months went by, utility company swapped the utility meter, and then another billing cycle went by and they billed us $15,000 because they thought the meter rolled over. What a great day at the office. 🤙

This was really fascinating. I always wondered how those old meters work. Thanks Dave for the great explanation. Please do more of those educational videos.

My uncle worked for the local utility company. He saw every trick imaginable to alter the meter. He made sure his house had the "best" one

3 Phase meters are interesting too, they just have 3 of those disks on the same spindle. Being connected on one spindle it will measure true resulting power. These measure true power regeardless of the noise on the powerlines due to switching powersupplies, led lamps and other odd consumers. Digital meters need all kinds of clever engineering to get the same accurate result. Those mechanical meters do not need any of it and are just accurate, even 50 year old ones.

The meters in my 1970's block are still the clock face dials. None of this "odometer rubbish". 😀😀😀

The dimples on the disc can be scanned with a photo electric cell during calibration. It speeds up the testing process when a line of meters is on the test bench. The readers are mounted above the meter and read the 10 dimples as the reflect the light back from the opto LED in the photocel. So for each revolution of the disc, 10 pulses are 'generated'. The Landis & Gyr CL 147 has similar patterns on the disc, which was a predecessor of this meter, made in the UK.

The temperature compensation blew my fucking mind. It's amazing how much engineering goes into these simple-seeming little electromechanical devices.

I worked on meters for 4 years, and that was the most concise lecture I have yet seen.

The trick is to get an old low voltage/high current filament and wire the secondary across the meter, one lead on an AC outlet the other on the mains. Then put a Variac on the primary of the filament transformer. as you raise the Variac voltage the disc slows down and starts the disc running backward. This is of course purely theoretical and is for educational purposes only.

It is truly amazing how accurate these meters can measure, not to forget how little they cost. It is like the predecessor of hard drives in these regards.

I wondered for years how these things work, thank you Dave!

We still have analog power meters at my home. It's fun to see it, spinning like a motor when a lot of appliences are running inside house!

Decades ago as a young pre-teen technology enthusiast I got one of these kWh meters from my grandmother's house renovation and of course took it apart. I don't think I then understood how it works at all, it was just an interesting gadget to dismantle :) I vaguely remember wondering how the simple metal disc could spin when there is no contact with anything but the counter.

I was an electrician in the states for many years and one of my hobbies was collecting vintage A-base watthour meters. I have meters going back to the early 1900's and a few more modern ones. I made table lamps out of a few of the modern ones.. These meters are fascinating and beautiful in their simplicity. A voltage coil and an amperage coil and a magnet.

It’s so simple, yet complex at the same time! Stuff like this is amazing to me. Anything that computes via mechanics is so neat, like those Friedman calculators. Amazing thought is put into stuff like this

I love the analog meter. When I was 6/7, I broke one of these old school meter. Used a dc motor to spin the dials with gear mechanism. It was fascinating for me at that age to use kinetic energy to show numerical values. Now after all these years, watching this video brings back memories. As always, great explaination Dave, my Aussie Teacher. Respect ++

This teaches and demonstrates the basic principles with a practical example as clear and efficient as possible. Another classic video.

The turtle AMR still has the record of a automatic meter reader transmitting the reading over the power line. there was one in the outback of Australia. It read the disk spinning and counted the power. And then it would send the data over the power at ~120Hz in a narrow bandwidth signal. and it transmitted data at 33 minutes per bit. so that's 1/2000 baud.

It feels nice when you compliment this precision product of my home country, Switzerland. The company name is spoken like „Landes and Ger“ as the „y“ is spoken like a „e“ in English or an short „i“ in Alemannic. Even in my current flat there are still Landis & Gyr devices at work. Indestructible and very reliable.

It's only been a couple of years since my old school meter was removed from my house. The energy company got all confused when I kept on giving them meter readings less than the previous one in the summer months......due to the Solar PV sending it rotating backwards....:-)

I like how on the front it has an explanation of how base 10 works

I freaking love your channel, along with AvE, Mustie1, big Clive it just gives me joy and a good feeling inside!

I’ve always been fascinated by these, maybe a little obsessed, thanks for the breakdown.

Thank you for the in-depth review!

Watt hour meters (the mechanical type) are beautiful works of engineering. With all the precision bearings and jeweled movements, and gear train for the totalizer. The old Westinghouse/GE time delay relays used in sub stations use very similar technology (eddy current motors) with magnetic drag for timing and contact placement. Also very beautiful pieces of equipment. I used to calibrate these watt hour meters and time delay relays for a local utility (DWP) so I have hands on with these and was always amazed with the engineering.

it's so simple, yet really amazing piece of engineering and problem solving.

it might be interesting for a lot to learn more about it like the scale the disc has. The counter has an imprint of 187,5 rev per kWh so you can calculate your CURRENT CONSUMPTION based on the the time it needs to pass 5 marks for example cause the whole disc here is divided in 20 marks, not 360 or 36, just 20 which means each mark represents 1 / 20 or 5% 187,5 revs per kWh means that each rev represents roughly 1 / 200 of a kWh or 1 / 200 x 1000 Wh so it is just 5 Wh. and then you know that 4 marks out of 20 would mean roughly 1 Wh used. If it takes 20 seconds to pass 4 marks then it would mean 1 Wh in in 20 seconds, 3 Wh in a minute and 180 Wh in 1 h which tells us we had a 180 W load running over 1 hour. So you can read the consumption if you stop the time it needs to pass 10 marks or segments which would be half a rev. 3600 s / seconds stopped for half a rev x 1000 Wh / (187,5 x 2) imagine it would take 10 seconds for a half rev, then it would take 20 for a full rev and 3 revs every minute or 180 revs each hour. This would mean that 180 / 187,5 kwh would be used or we had used a load of 0,960 kW Let's assume it takes 20 seconds to achieve 1/4 rev then it would mean 3/4 revs per minute and 180 / 4 revs per hours = 45 revs and that means 45 / 187,5 kW of load or 0,240 kW or a 240 Watt load. That is the part you need to know to get behind your energy consumption. Start the water cattle and see the wheel turning a lot faster or the iron for your shirts. But you can start there at the good old ferraris once you know that the number of revs / kwh is the key to read and understand your current energy consumption. Most people here did not learn that.. For most energy consumption is still a secret and they did not learn the basics how to get behind what device needs what energy in a month or depending on the season of the year.

Great video Dave!

Dave, thank you; that was utterly fascinating. You've got me running down to the cellar to watch Eddy make the disc whizz round every time I put the kettle on!

Back in the 70s when I studied my EM course we used the mnemonic "ELI the ICE man" to remember the phase shift due to Inductors and Capacitors. But "CIVIL" is simpler!

The same principle was used for over 100 years in electric transmission line “distance” relays (device 21) that measure both the fault current in a trash mission line and the distance to the fault. These are used to prevent tripping of multiple circuit breakers on a line when a fault is detected and to only isolate the portion of the transmission line where the fault occurred. The principle of torque angle as a measure of phase angle is the same. The term “torque” is still used today with digital programmable relays that do not use rotating discs. This confuses many young engineers! I’m a utility engineer, retired after 45 years.

Clever engineering right there. And the single turn on one of the coils on the magnetic ammeter portion acts like a shaded pole, thus saving the company that made it a bit money, and to keep it rather simple than installing a Copper ring on the shaded pole, forcing the disc rotor to favor the pole with three - four turns coil, thus only spinning in one direction generally. (If you reverse the ammeter coil polarity against the voltage coil, or flip the ammeter coil in the other direction, so it could spin the other direction, depending on how the motor is designed - your watt-hour meter could be modified to do that as it's much more simpler.)

5:55 - magnetic brake is the term you're after. For fun take one apart and put that disk into a drill, spin it up then introduce a U or horseshoe magnet

I like how it's simple, yet complex. Everything in there has to be just right for it to be accurate, all down to number of turns in the coil, to how far apart magnets are etc.

I hadn't seen CIVIL before here. Way back when I was learning electronics my instructor related "ELI the ICE man", which I found very easy to remember. He also gave us a PG version for remembering resistor color codes; but for some reason I currently am only able to remember the Navy version.

I've always wondered how these meters work and the comments go deeper. nicely done sir.

Very cool Dave- Analog is good for the soul.

The mundane things in life that we often pay little attention to are often the most fascinating to look at and break down, glad i'm not the only one!

I used to work for Landis & Gyr (pronounced Gear) back in 1997. That meter is called a Plugin meter supplied for Queensland. Made in Mt Waverley Victoria. Most the parts came from Switzerland.

More of them, Dave. That was a good one.

Swiss precision design and manufacturing! Landis & Gyr is a Swiss Company, but they where bought by Siemens a few years ago ☹️

I love how much stuff been thought for these meters :3

Thanks for creating this video, I did my first internship at Siemens at the former Landiy & Gyr premises in Zug, Switzerland and it was interesting to see a part of the company's history

I agree with some of the other comments regarding the dimples being there too keep the disks flat. The same tecnique has been used for many years on parts like motorcycle clutch plates, where they can also act as lube resevoirs .

The holes in the disk stabilize the centers of the eddy current vortices and force the disk to move rather than allowing the vortices to 'slide' across the disk. I'm not sure about the brushed stripes, but I think it might compensate for aluminum's thermal coefficient of conductivity.

I always wandered about these meters ,and didnt found a good explanation until now, still they have a lot of mysteries to reveal ,thanks

I don't remember what it's called, but the dimpling is from a flattening process with a big press to make sure when they're stamping them out of the large sheet they're not warped.

super interesting & thanks for sharing.

This was really neat Dave! I have always been curious how those analog meters worked!

As a fireman, pulling these out of socket is also a quick way to kill power to a residence. Where I live now, this isn't usually a big deal since the utility company is often right behind us, but we used it a lot more where I used to live.

Merry Christmas 🎄 Thanks 👍

I love your Friendly nature 😊 Keep up the vids good sir!!!

Excellent!!!

Your video gave me flashbacks to my time in the Meter & Test section of Prospect Electricity (Seven Hills depot) overhauling and calibrating meters similar to this one. The jewel bearings used to come in an applicator that released one at a time and were tiny. Anyway, enjoyed the video 👍

This has probably already been answered in the hundreds of previous comments, but I believe the dimples are too help make the disc flat and rigid, and the brush marks may have to do with balancing of the disc. Thank you for another great video!

Yes, it is amazing! Plus these things must be engineered for reliability and consistency over a few years or even decades, as it would be so damn impractical to cut off the power in order to swap the meter every year or two. I like the meter-base interface used in Australia - it allows for a very quick replacement (even without cutting the voltage off), way quicker than European meters, because they have screw terminals on the bottom side, covered with a plate that is sealed to prevent tampering. Oh, and retarding the meters with neodymium magnets was an energy companies' nightmare to the point that they really welcomed the modern electronic meters... By the way, I wonder how a double-tariff electromechanical meter works.

So much more to them than you would think.

Wow! Such engineeringing! Much scienceness!

Used to be called buzz boxes here, a small transformer that was wired across the meter to slow or reverse the disc, many of the later meters had flags that popped out if the disc rotated in reverse as a visual display

Vote up, nice video clip, thanks for sharing it :)

Steady Eddy! What a legend.

Interesting, an *anti* creep hole! Most holes attract creeps.

That multimeter has your logo on it (6:49) - OMG! THumbs up.

Here in the states, the box the meter is installed into is put in by the electrician, then once the electrical inspection is passed, the power company comes out and installs the meter. That's why the meter just plugs into the box.

I remember someone who had a small hole in the bottom of his meter, with a paperclip up through the hole and caught through that disc hole to freeze the meter.

I might have missed it in the explanation, but the magnet doesn’t just keep the disc from spinning wildly, it’s actually part of the calculation itself(!) The disc multiplies the I and V components, resulting in a torque, and you’re right, if the magnet wasn’t there, the disc sound just keep accelerating. The drag induced by the magnet is *proportional to the speed of the disc* though, so a given amount of torque will translate into a specific speed of rotation. Voila, speed is proportional to V x I ! 😁

Ah yes, our old meter. When we got a quote to get a smart meter (which I was able to hack - good old software backdoors), the engineer said “Jesus, you didn’t tell me it was an old analog meter you had!”. That meter was installed when mainstream electricity first became available in our region and it had never been changed since then. We’ve still got that meter (not connected to mains, we’ve got it in a custom acrylic display case because of its history) because I love anything electrical and old/vintage.

What you could do is make a small diameter hole through the plastic casing with a dentist drill or heated wire, place a small metal rod through that makes contact with the disc and slow it down that way. Just remember to remove it before they come to measure it and do not to get too greedy. For good measure, cover the meter in some dust so the whole is less visible ;)

I was a Meter reader until mid 2021 (Perth, Western Australia) and I never encountered that particular Meter.

I love the transparency layer update to DaveCAD. You're really an analog guy, aren't ya? ;)

Hey Dave, you actually don’t even have to disassemble the unit to pull it out of the receptacle. There’s a puller tool linemen use to pull it out as these things need replaced often enough to warrant it.

Yep. Changed 4 & 8 GHz twats. Traveling Wave Tubes. The Reflex Klystron , Magnetrons and TWaT were part of my USAF career. Most gain is from antennas. Broke a 4GHz TWT in a Philco transceiver. A delicate procedure. FELDBERG GERMANY 1984-88

This is peak Dave over here!

Awww…I was expecting you to test the accuracy of that meter. See if it was ripping people off in its past life. 😯

I remember when they said electric power from nuclear would be plentiful and too cheap to meter.

These things don't miss a beat.

Decades ago, when I was much younger, I tried to see if I could get that disc to spin fast enough to break the sound barrier. Unfortunately the only thing that broke the sound barrier was my father's temper, after seeing the electric bill that month. 😁 ⚡️

Dude the amazing amount of balance you has to control ! I have one of those here on ARG, now we are using those digital rubbish.

Thank you for tearing that nice piece down for us. I remember we had a three phase one, much bigger and uglier:)

"Made in Australia" by Siemens Interesting artefact.

I have one with an indoor socket that the power company left in my grandmas basement in the 90s. I wired it up for use on 120v and still works.

Did you install back into your electrical box without any trouble?

Your meter can spin in reverse direction with readings decreasing due to it( you started with 79686 units, which fell to 79683 units). Didn't expect that from Siemens. In India, our electromechanical meters come with a mechanism which prevents reverse rotation of the disk

Steady-Eddie current, so fun! ;)

My favorite line was the part about the steady eddy current. :) It was truly fascinating to watch you take it apart. Could the dimples be there to reduce drag in the air flow, like on a golf ball?

You would love some old electric grid protection relays. All mechanical relays that protect lines, buses, transformers, etc. Distance, differential, over current and more. Many are still in service and still accurate to +-5%. Very similar to mechanical meters except for 50x more stuff packet into one.

The dimples are there because the plate can be pressed flatter with dimples than without. I don't know the proper technical term but it has to do with spring back when you bend a piece of metal into shape. I guess with dimples you get easier into plastic deformation.

The holes in the disc are for testing, a laser is passed through the disk and you'll get a test pulse every time the laser passes through that hole. In the states, we perform three tests on meter. A full load test, a light load test, and a power factor test. Adjustments are made to the magnet in order to speed up and slow down the disc to calibrate the meter.