Microwave turntable motor RPM slower than specs

[Kinsmana]

I've been using microwave turntable motors to drive art projects for years because they are strong, and easy to find on the curb.  I know that synchronous motors are locked to the frequency of the AC mains and that old clocks were very accurate due to the keepers of the grid making certain they let the right number of cycles escape from the generators each day.  I've been told this isn't as important as it was in the past, but due to interconnections in the grid, it would seem that it must still be very important for the entire country to be pretty close in sync.    So I planned to make a clock from the microwave turntable motor.  I knew they often started in the wrong direction, but once started they continue to run the same direction, so no matter.   I tested one while working out the gear ratios and found that while the label says 60Hz, 3RPM, it is actually 2.9637 RPM, making it difficult to design gears for a clock.   These are clearly synchronous motors.  They run close to the labelled speed, but why wouldn't they gear them to meet the stated RPM?  It is only a mater of the right gear ratios.   Mine is marked 50Hz 2.5RPM, 60Hz 3RPM.   I've tested about 6 microwave ovens.  Put a piece of tape on the turntable, cup of water for load, then get your stopwatch out and count the revs/minute.  None seem to run at the stated RPM.   It is difficult to have a unique idea not discussed on the internet, but no search engine seems to address the issue of why these motors
seem to run slow.  Very curious here.

[alsetalokin4017]

You are able to measure precisely down to the ten-thousandth of an RPM, using tape and a stopwatch?

Where's the "Bow down to a Measurement Boss" emoji when I need one?

[amyk]

It is difficult to have a unique idea not discussed on the internet, but no search engine seems to address the issue of why these motors seem to run slow.  Very curious here.

I remember reading a thread on an engineering forum a while ago about this, but because of the continued degradation of search results, I can't find it either. From what I remember, the motor rotor itself runs at synchronous speed but due to the use of a non-hunting gear ratio, the final ratio ends up inexact and so does the RPM.

[Kinsmana]

Actually if you run a motor through enough reduction gears it is easy to determine quite accurately after several hours of running the RPM to many decimal places.  I ran it for about 5 hours through a 10-1 reduction and plotted the results.   I stand by my numbers out to at least 3 decimal digits.

[Corporate666]

Everything has a tolerance, including your measurement (which isn't 2.9637 exactly, but plus or minus some unknown amount).

2.9637 is off by about 1%.  How accurate are you expecting a microwave turntable motor + gear mechanism to be?  How accurate does it need to be to turn a bowl of food that's being cooked?

[Kinsmana]

I've handled synchronous motors in electric clock repair for decades and found them all to produce the exact output specified on the motor for RPM.  This is the first time I've seen one that isn't accurate as marked.  I'm puzzled as to why it either wasn't designed for the specified output RPM, or what is going on regardless of the end use of this motor.

A related question is how to make it accurate so microwave turntable motors can be used for clocks.   I could design a gear train to correct it gearing up/down to a desired RPM.   This actually involves a lot of teeth and only gets close, not close enough for clock use.   Disassembly, 3D printing a few gears, and reassembly seems to be the only option.  So why wouldn't the manufacturer have chosen a collection of gears that produced the specified output so this motor was accurately marked and available for other applications besides turning our food?  Is it being intentionally made inaccurate so I must pay top dollar for a 'clock' synchronous motor?  Curious.

[phil from seattle]

The grid frequency is stable over a long time (hours?, days?) but it can vary over the course of hours - it's called TEC. wikipedia article on this: https://en.wikipedia.org/wiki/Utility_frequency#Stability

However, it appears that the corrections are several orders of magnitude less that what you are seeing (.01%). Here's an interesting display of regional variations in "real time".  http://fnetpublic.utk.edu/tabledisplay.html

[vindoline]

I'm no expert on motors, but I expect that as it's a synchronous motor it's turning at a synchronous rate. That's the physics of it. As Corporate666 says I believe that its a combination on non-precision mechanical gearing in the motor as well as your measurement setup. Even a perfect motor run through a sloppy gear train will be inaccurate.

[Paul Moir]

It shouldn't matter how precise the gearing is since you're a slave to the number of teeth on each gear. 

I think amyk has it:  it's intentionally slightly off-ratio to prevent wear.  Hard to make non-hunting gears with an even division ratio so they go for 201:1 instead of 200:1.  Interesting, I had never heard of it!

[Kinsmana]

I do live in an area of the country where our electricity is very cheap.   3.5cents/KWH due to a very old contract between our town and probably Nikola Tesla himself when generators were being installed in Niagara Falls.   I did consider that I was not only getting a discount on power, but on cycles too.   I quick check with a zero crossing detection circuit and a micro to count for a few hours indicated my power was doing just fine cycle wise.  That concern was dismissed weeks ago.  Puzzle remains.

[Kinsmana]

I'm glad somebody else posted this.  Crappy gears do work just fine and must preserve the shaft rotation ratios.  Check out the Antikythera mechanism.  They think it has teeth made with a triangular file.   The ratios must be maintained unless teeth rip out.  It isn't my gears or those in the motor, or "armature hunting" since I don't think this motor has ever missed a cycle.

[rstofer]

Are you sure the motor is synchronous?  I would expect it to be a shaded pole induction motor as they are cheaper to produce and commonly used for things like fans and appliances.

The shaded pole induction motor 'slips' as a function of torque.  In fact, all induction motors slip.  That's why the nameplates are often 1750 RPM when synchronous speed is 1800 RPM.

[bw2341]

I think the hunting ratio idea makes sense. I found a page with a few explanations (and a diagram):

https://engineering.stackexchange.com/questions/11394/why-are-spur-gear-ratios-not-always-whole-numbers-in-practice

If two gears have tooth counts that are multiples of each other, a bad tooth on the small gear will quickly wear down the same few pairs of teeth on the big gear. Without a common multiple, the bad tooth will touch every other pair of teeth before touching the first pair again.

This implies that all clockwork gears will be non-hunting ratio and subject to more wear and tear. Of course, the design goals are different. A clockwork has very little torque on the mechanism. With a microwave oven turntable motor, you could make a clock with bricks or lead pipes as the hands but it will keep terrible time.

If you want to verify this hypothesis, just tear down one of the gearboxes and count the teeth.

[cstratton]

If you are finding these surplus and puzzled by this, then tearing one down does seem the most sensible.

In terms of fixing it, you're unlikely 3d print (at least via an FDM machine) gears which would fit, but you might be able to make some that would correct the ratio.

It may be easier however to build a circuit to synthesize AC at a slightly different frequency.

Or just buy a motor with a specified ratio. High pole count stepper motors are a close cousin and might also be something to consider.

[CopperCone]

I kind of want to make a ultra high RPM motor connected to a microwave so the food has to be strapped down with a harness.

You can call it experimenting with thermal vortexes in a microwave field and get a ton of youtube views lolol. like a woodworking router connected to a microwave.

perhaps the gyroscopic forces would put tension on ham tissue so that it cooks differently.

[amyk]

If you want to verify this hypothesis, just tear down one of the gearboxes and count the teeth.

Someone actually did do that on a forum thread I read about the same thing, and posted the exact ratio (which was very odd, so perhaps searching for it might discover more results), but I can't seem to find it now because Google's search results have gone to shit.

Here is the closest I could find, showing a 2.5/3RPM model has 8 poles (900RPM for 60Hz) and the final gear ratio is 8:23(!) but it doesn't show the other teeth in an easily countable way:
http://rmdavidson.blogspot.com/2014/10/suh-der-sd83-ac-synchronous-motor.html
http://rmdavidson.blogspot.com/2014/09/tyc-synchronous-ac-motor-teardown.html

Related, geared stepper motors with an inexact ratio:
https://forums.adafruit.com/viewtopic.php?f=31&t=41608
https://grahamwideman.wikispaces.com/Motors-+28BYJ-48+Stepper+motor+notes

[Corporate666]

It shouldn't matter how precise the gearing is since you're a slave to the number of teeth on each gear. 

I think amyk has it:  it's intentionally slightly off-ratio to prevent wear.  Hard to make non-hunting gears with an even division ratio so they go for 201:1 instead of 200:1.  Interesting, I had never heard of it!

It makes sense to use a non-divisible (by each other) ratio so that the same teeth aren't always contacting each other, but the OP claims he is confident in his measurement to 1/10,000th of an RPM, which would mean either a multi-stage gearbox with some pretty serious reduction going on, or a pretty high speed initial speed with a big gear and - again - some pretty serious reduction going on to get a reduction that's so close to the desired amount while being so slightly off.

I would be surprised if that's the case - but I would be curious to know.  I'd love to see the inside of the gearbox.  I wonder if it truly is a synchronous motor - because it wouldn't seem to be necessary to be so for a microwave turntable.

[amyk]

It makes sense to use a non-divisible (by each other) ratio so that the same teeth aren't always contacting each other, but the OP claims he is confident in his measurement to 1/10,000th of an RPM, which would mean either a multi-stage gearbox with some pretty serious reduction going on, or a pretty high speed initial speed with a big gear and - again - some pretty serious reduction going on to get a reduction that's so close to the desired amount while being so slightly off.

The actual ratio is likely to be a nonterminating decimal, and he just rounded it off at 4 decimal places for convenience.

In my links above you can see the inside of these types of motors. They certainly have several stages of reduction, in that case a 900/750 RPM (8-pole at 60/50Hz) synchronous motor goes into a (approximately?) 300:1 ratio to give 3/2.5RPM at the output.

[Paul Moir]

I assumed it was like one of those synchronous motors that drove (thank you google) "Flip Tile" alarm clocks from the 70s.  I remembered the motors in them don't spin very fast.  A little googling suggests 600rpm.  I used 201 as an example because that would be 2.985 rpm.  'Course that's not likely exactly that since it would be a one tooth gear running a 201 tooth gear.  Rather it's going to be something like a 10:41, running another reduction stage, etc (probably 4 stages total).  But you can see how you could end up with something really close to but not quite on 3 RPM.
For example, three 7:41 stages gets you 2.986 rpm from 600.
EDIT: These things:

[Kinsmana]

Hey I think we are getting somewhere.  Below is a photo of my motor. 

Here is a link to a youtube teardown... not the exact motor, but clearly synchronous
8 pole with a pile of plastic gears for reduction...


I like the posting that suggested non-exact gear ratios to extend gear train life.   I remember 40 years ago slaving over a split transmission when my dad walked in to impart wisdom.   He had a PHD  in gear design and explained why helical not spur, and why odd teeth, not even particularly when the output RPM was not critical.   Helical promoted quiet running, odd teeth  (prime numbers best) promoted even wear across all teeth, which also reduced noise. 

I bet in this application with output RPM not being critical that priority was given to reduction ratios that promoted product longevity.   No autopsy yet.  I have several similar motors, but none that are identical to one in use.  I can't dissect it at this point.

May have to buy an identical motor in order to take it apart to count gear teeth.   I truly appreciate all of your comments.   

[floobydust]

They have magnets inside, perhaps high-slip to prevent damage if the turntable jams or gets moved?
https://www.eevblog.com/forum/chat/workings-of-a-microwave-oven-turntable-motor/

[Kinsmana]

Decided to do a tear down of one of my motors.   Months ago it was tested as faster than the 60Hz 3 RPM stated on the label.   So this one isn't slower, but actually faster.  I measured it at about 3.025 RPM.  Gear math says it is really 3.0232558 RPM.  There must be a reason for this being the design.   Somebody has a story to share. 

I've included a drawing of the drive train.  It is a 6 pole motor, meaning the rotor spins at 1200 RPM in my 60Hz world.   This motor was clearly designed to run fast, not at 3 RPM.   Here are all the pictures, drawings, and Excel spreadsheet showing the math.   Gears have been completely photographed and counted, over and over for certainty.   Why is that 13-43 tooth reduction used?  Sure wish it wasn't.
There is reasonable and cost effective fix... clue provided by the consistent diameter of the gear shafts and their spacing.  Cutting apart two gears and gluing them together would allow you to reconfigure the drive train with one empty shaft, giving a 10 RPM output. (I think)

Possible reasons for why it isn't 3 RPM as specified...
1) Don't want us to have cheap hobby synchronous clock motors?
2) Prime tooth counts promote long product life in systems where a specific RPM is not required?
3) Mistake in design process?

I discount #3, since these motors are made by experts for a reason.   Conjecture #1 seems a bit paranoid and requires too much forethought. I'm leaning towards #2 due to those two prime numbers
in the gear train.

Here is the documentation.  It pretty completely defines this motor.  Enjoy.