3 phase motors in series

[strantor]

What? that sounds crazy, impossible. Maybe, maybe not.

Assuming you had (2) 12-lead 3 phase motors like this...

...and assuming:

• they had identical specs
• they were mechanically coupled to a common load
• the applied voltage is double the rated voltage (880V in the case of the 440V motor shown above)

Could you connect them in series like this?:

The windings inside the red circle represent Motor #1 and those inside the blue circle represent motor #2.

Possible yes/no/why?

Thanks
chuck

 

[tcmtech]

As long as the insulation can handle it then yes they would run just fine and work like one bigger higher voltage motor.

 

[strantor]

As long as the insulation can handle it then yes they would run just fine and work like one bigger higher voltage motor.

I figured you'd have something to say bout this. And if you didn't chime in, I would have called your attention. Thanks for replying.

Can you explain your concern about the insulation?
To me it seems the same concept as putting 2 DC motors in series; the voltage is split between them so all that either motor ever sees is 440V.
Is that a bad assumption?
Or are you thinking about VFD spikes? (I didn't say it, but yes, this is/would be a VFD application).

 

[MaxHeadRoom78]

I can see a problem with varying load of each and, the difference in slip frequency is going to unbalance things.
Not something I would tend to do!
Particularly in a VFD application.
Max.

 

[tcmtech]

Consider the voltage differential between the live line and the earth grounded stator cores.

In the Us most 480 VAC three phase systems are 277/480 Wye connected so that no line insulation is seeing more than 277 VAC between it and ground.

Although There are applications where corner grounded 480 VAC is used as well so I personally I don't see it as being much of a concern really given a properly built motor should be built to withstand at least several KV of voltage potential between the windings and frame without breakdown issues.

I can see a problem with varying load of each and, the difference in slip frequency is going to unbalance things.
Not something I would tend to do!
Particularly in a VFD application.
Max.

Mechanically coupled I take to be mechanically locked shaft to shaft so there is no variances in slip ratios just the same as when two motors are wired parallel and sharing a common load with their shafts locked together.

Series or parallel operation the electrons making the magnetic fields could care less.

Where was VFD operation mentioned?

 

[MaxHeadRoom78]

Or are you thinking about VFD spikes? (I didn't say it, but yes, this is/would be a VFD application).

I once had to consider running 2 identical motors, identical loads in parallel off of a VFD, I contacted WEG who supplied the VFD and they confirmed that as long as the load was identical, it should be OK.
Is this a sensorless vector or the more simple variable frequency VFD?
Can you guarantee the identical load for each?
Max.

 

[strantor]

Consider the voltage differential between the live line and the earth grounded stator cores.

In the Us most 480 VAC three phase systems are 277/480 Wye connected so that no line insulation is seeing more than 277 VAC between it and ground.

Although There are applications where corner grounded 480 VAC is used as well so I personally I don't see it as being much of a concern really given a properly built motor should be built to withstand at least several KV of voltage potential between the windings and frame without breakdown issues.



Mechanically coupled I take to be mechanically locked shaft to shaft so there is no variances in slip ratios just the same as when two motors are wired parallel and sharing a common load with their shafts locked together.

Series or parallel operation the electrons making the magnetic fields could care less.

Where was VFD operation mentioned?

VFD wasn't mentioned in the original post. I just mentioned it.

I get what you're saying about line-earth voltage. I didn't consider that because this would be powered by an isolated step up transformer after the VFD.

You understood correctly about coupling shaft-shaft; I intentionally made it sound that way. But that's not how it would be. It would be more like a bandsaw. Picture a bandsaw where both pulleys are powered pulleys. They share the load, and they are coupled by the blade. There is a chance the blade breaks and they come uncoupled. What happens then?

Sorry for slipping details in one at a time and changing things up, but I want to begin the discussion at the most basic level and then increase the complexity until the wheels fall off, so I know where the boundaries are. If I spilled all the beans in post #1 it would be so long that nobody would read it, and if they did they would think I'm retarded.

 

[strantor]

I once had to consider running 2 identical motors, identical loads in parallel off of a VFD, I contacted WEG who supplied the VFD and they confirmed that as long as the load was identical, it should be OK.
Is this a sensorless vector or the more simple variable frequency VFD?
Can you guarantee the identical load for each?
Max.

Yes running parallel motors off a VFD well documented. I have done it myself. But this is a different animal. I can't find even a whisper about it online.

I think it would have to be a V/Hz job. I don't think a VFD can "feel" the motor through a transformer well enough for vector.

In theory the load should be perfectly coupled but no I cannot guarantee it. If one pulley were to wear faster than another (smaller diameter now) then its load would be less. There are probably other scenarios I haven't thought of.

What I can't wrap my head around is if they would naturally try to share the load or if they would naturally try to avalanche.

 

[Diver300]

I didn't consider that because this would be powered by an isolated step up transformer after the VFD

It seems to me that it would be a lot simpler to run the motors in parallel, with no need for a transformer.

Before I saw that you were thinking of using a transformer, I thought that you would be better to step the voltage down with a transformer and run the motors in parallel.

I think that if you ran two motors in series, and one got loaded more, it could not take much more current, so the voltage on the loaded one would reduce, resulting in too much voltage on the other one, possibly saturating it. I would try parallel first, especially if it meant not needing a transformer.

 

[strantor]

It seems to me that it would be a lot simpler to run the motors in parallel, with no need for a transformer.

Before I saw that you were thinking of using a transformer, I thought that you would be better to step the voltage down with a transformer and run the motors in parallel.

I think that if you ran two motors in series, and one got loaded more, it could not take much more current, so the voltage on the loaded one would reduce, resulting in too much voltage on the other one, possibly saturating it. I would try parallel first, especially if it meant not needing a transformer.

here's the thing: I need 60kW worth of motor at the end of a 200m cable. I need to seriously cut back on the amps or else voltage drop will quickly consume all of applied voltage and the we end up with tens of kW dissipated in the cable and a stalled burning motor.

Installing a transformer of 60-75kVA, 20-30Hz (used for VFD applications up to 60Hz) at the source is possible but not at the motor. Right now we use a special order 3000V 15A motor, but it costs more than a brand new truck and takes months to arrive from overseas. I'm trying to find a way to do this with off the shelf local parts.

Using multiple motors gives us other advantages as well.

 

[kinarfi]

It looks good to me, and if the motors have been meggered, I wouldn't worry about the applied voltage, except you may want to check the wire you're using for it's voltage rating. I see the advatage of this being smaller wire due to lower current, and disadvantage of needing wire with a higher voltage rating.
Jeff

 

[tcmtech]

As far as my 3 phase motor theory goes they should work just fine in normal working conditions and should the band that is coupling them break the short term change in load is unlikely to do any harm to the motors no worse than the loaded motor seeing a undervoltage condition while the unloaded one sees a over voltage condition.
As a whole the system should show a drastic drop in current draw letting the operator know something broke and to shut things down.

As for pulley slip or wear a small variance of a few RPM's between the two motors won't cause any significant problems either. At worst one motor will exhibit a slightly higher voltage drop than the other given they will have equal current flowing though them.
I've seen it done on a number of gravel processing machines that ran with two belt coupled motors to a central gear box and they never had problems from slight speed mismatches when belts would slip.

If it was me I would give the whole concept a test run and see what happens.

FWIW if you ever have the motors apart take a close look at the actual wire used for the windings. If its more than a few HP there is a good chance that it was wound with multiple wires on each coils set which although not always recommended would give you extra winding sets that could be stacked in series for each phase over their as built parallel configuration thus allowing your 440 VAC motor to be configured for 880 or possibly higher line voltages.

Years ago I picked up a 7.5 HP three phase 200 VAC 3 wire motor and made a rotary phase converter out of it. It worked on 240 VAC well enough but would get way to hot when loaded do to being over driven.
I was going to scrap it but when I took it apart to pull the copper out I found that it was wired as a 200 VAC wye unit and each phase had 2 parallel wires in it.
From there I tried reconfiguring it into having each phases pair of windings in series and then put everything back together as a Delta configured motor figuring at worse it just burns up. Well it worked great and ran much cooler on the load test. Its been years now but as far as I know they guy who has it still uses it all the time.

 

[strantor]

It looks good to me, and if the motors have been meggered, I wouldn't worry about the applied voltage, except you may want to check the wire you're using for it's voltage rating. I see the advatage of this being smaller wire due to lower current, and disadvantage of needing wire with a higher voltage rating.
Jeff

The wire is rated 4.5kV. It's a subsea power umbilical. The application is a subsea saw tool. Pretty much exactly like the bandsaw analogy but with a few twists.

 

[strantor]

Thanks tcmtech I never thought about there being parallel windings internally but it makes perfect sense. I wonder if manufacturers will disclose winding details before purchasing; I'd hate to have to make several "test" purchases just to look up skirts for compatibility.

There's got to be a limit to how much you can get away with in doing this though. I suspect that since they wound the motor never intending for the windings to be placed in series, the windings are not designed for the elevated potential between windings. Your example worked by what sounds like <2x the designed applied voltage. But say I have a motor with 3 parallels, and reconfigure it in 3 series and apply 3x the rated voltage. I think some areas of winding-to-winding insulation are going to be very unhappy about that.

Would you agree?

 

[strantor]

Since it sounds like my scenario is evolving in a promising direction, let's see if I can't kill it with disclosure of a couple more details...

• These motors are going to need modification to work under water at a depth of 500ft. At a minimum the mods will consist of replacing the sealed bearings with open bearings and placing the motors inside an oil filled enclosure. The rotor(s) will be spinning in pressurized oil.
• The "bandsaw" actually has 4 pulleys. a "square" bandsaw. I'd like to make all 4 pulleys powered pulleys for several reasons, most importantly that it provides exponentially more traction (Eytelwein formula). So we're talking about 4 motors now, not just two. Since the theory seems to work with two motors in series, how about 4 in series? is that "pushing it"? if so, 2 parallel sets of 2 in series is always an option, but adding up all 4 in series would be ideal if possible.

Have the wheels fallen off yet?

 

[tcmtech]

Sounds reasonable to me but then I am one of the few who doesn't mind practicing the 'experiment until it fails' method of learning either.

I was not aware of the running under pressurized oil though. In my books that gains you a considerably wider operating range on everything. Having pressurized oil in between the windings and saturating everything it can soak into greatly reduces the likelihood of arc over.

Just look at a typical oil filled tank type utility transformers insulation design Vs an open air one and the safe overload time each can handle. The one submerged in oil has substantially less insulation clearances yet also has a substantial overload rating compared to the open air one of equal continuous KVA capacity.

I'm pretty sure that extended overload capacity carries over to oil submerged motors as well being the main direct hands on example I just worked on was an old cast iron sump pump that's been in my used oil bulk tank for a few years now that I use as a tank stirring system.
I never thought much about what type of motor it has in it other than it was a damn good one being there were many times I ran that thing in the dead of winter to get the heavy oil that settles to the bottom remixed with the lighter oil on top.
Starting dead cold trying to get upto speed with an impeller full of near grease fluid could take that thing 5 - 10 minutes to get up to speed. Occasionally it would overheat and trip out for a minute or two but most of the time it just sat there turning as best it could with the start windings fully active pulling 16 - 18 amps on a 6 amp rated motor until it got enough of the lighter oil down to the bottom that it could get up to speed.

I never thought much about it and figured it was a typical shaded pole motor but the other week I ran the tank empty and decided to pull the pump up and have a look at it. The old vinyl jacketed cord was brittle junk from being in contact with oil for years so I figured I would pull it apart and put a good one on it while I had it out.

I got it open and to my surprize it has a standard 1/3 hp centrifugal switch start type open frame resilient mount motor just like any old belt drive furnace blower uses. Those ones that can take maybe 30 seconds of overload in start mode before they let the smoke out type.

The oil was black and smelly but the motor windings themselves looked just fine and even passed a 1000 volt megger test I did just for curiousity sake. Well it got a new cord and new oil and back in my bulk tank it went!

Personally I won't guarantee your motors wired in series running a higher than stock voltage will work but everything I know about transformers and motors and the added benefits of running them submerged in oil would have me trying a test run that way.
To be honest I would be half tempted to try running all four in series as well with the pressurized oil containment system in place. Especially so if they were VFD rated motors that have the higher than standard winding insulation built in.

That's my thoughts.

 

[strantor]

Sounds reasonable to me but then I am one of the few who doesn't mind practicing the 'experiment until it fails' method of learning either.
[...]
Personally I won't guarantee your motors wired in series running a higher than stock voltage will work but everything I know about transformers and motors and the added benefits of running them submerged in oil would have me trying a test run that way.
To be honest I would be half tempted to try running all four in series as well with the pressurized oil containment system in place. Especially so if they were VFD rated motors that have the higher than standard winding insulation built in.

That's my thoughts.

Sounds like we have similar thoughts on test methods and it sounds like we're on the same page; the page I was on when I decided to post this. Thanks for humoring me.
I just wanted to see if I could get at least one person (one person who knows what they're talking about) to not call me crazy before I open my mouth about it in the real world.
I got that, I got your "blessing" - or at least your withholding of condemnation, so that's good enough for me. If it fails miserably I'll tell everybody it was all your idea.
I am going to ask for money to test this wacky idea (w/ fractional hp motors first) and if I get the green light I'll keep you in the loop.

Again sorry for disclosing everything in baby steps; I know that must have been annoying. I think we can agree that this is way way "outside the box" and when you compile all the ways in which it flies in the face of typical, it probably sounds preposterous. But examining each aspect alone, they all seem to have their own merit.

Thank you for your time and experience.

 

[strantor]

wtf why this doublepost thing again?

OK I posted, 5sec later I get a "there have been new replies " message, I refresh, I see my post twice. I only clicked once.

So I change the text of my 2nd reply to "wtf why this doublepost thing again?" and then refresh and my first post is gone and all that remains is my 2nd post which I've now overwritten.

Looks like the forum software just juked me into spending 10min typing up a blank post. Silly forum, it didn't know that I sometimes type replies in notepad and then copy over.

 

[tcmtech]

Sounds like we have similar thoughts on test methods and it sounds like we're on the same page; the page I was on when I decided to post this. Thanks for humoring me.
I just wanted to see if I could get at least one person (one person who knows what they're talking about) to not call me crazy before I open my mouth about it in the real world.
I got that, I got your "blessing" - or at least your withholding of condemnation, so that's good enough for me. If it fails miserably I'll tell everybody it was all your idea.
I am going to ask for money to test this wacky idea (w/ fractional hp motors first) and if I get the green light I'll keep you in the loop.

Again sorry for disclosing everything in baby steps; I know that must have been annoying. I think we can agree that this is way way "outside the box" and when you compile all the ways in which it flies in the face of typical, it probably sounds preposterous. But examining each aspect alone, they all seem to have their own merit.

Thank you for your time and experience.

Not a problem! To be honest I wish there were more threads like this where someone wants to push the limits of common knowledge and application to see what they can get away with.

I for one can relate very well to the concept of trying to find a better cheaper way to do something opposed to having to rely on a single outrageously over priced component that's almost impossible to get.

Right now I am rather wishing I had kept my collection of odd metric electric motors I got years ago being I had at least 5 identical 220/380 volt 1 KW 1170 RPM oddball face mount motors that would have been perfect for me to do some testing on to see how well they would work wired in series.
Unfortunately I never found a practical use for them so they got scrapped for the copper.

 

[shortbus=]

Not an expert on this by any means. But running them in oil won't they need to be over sized in HP rating? Due to the added drag of running in oil instead of air.