Three phase induction motor efficiency

[Alex Cramphorn]

Hi,

Is there a difference in efficiency between a three phase motor and a single phase motor, in terms of power in against power out. If so, could you please explain why one is more efficient than the other.

The reason I ask is because I'm researching the development of the induction motor for a project at university and I am currently on the introduction of multiphase motors. Just curious if this would have influenced that development.

Thanks in advance

Alex

 

[zoki85]

Hi,

Is there a difference in efficiency between a three phase motor and a single phase motor, in terms of power in against power out. If so, could you please explain why one is more efficient than the other.

3-phase induction motors have better "watt/kg" ratio. IOW, a 3-phase motor is smaller, cheaper and better than a single phase motor of same power.

 

[Mike_In_Plano]

The simple answer is yes. Three phase motors consistently beat out their single phase counter parts for roughly the same size / load. With excellent construction, you can sneak up towards 90% with a single phase, but a three for the same construction will be hovering at 93% - at least 4 points ahead.
If you study the magnetic vector in the single phase machine, you find that it's really just bouncing back and forth along a line of action. The starter winding can help it produce a sort of ellipsoid, but the field is still reaching high peaks and decaying.
With a three phase motor, the magnetic vector is smoothly rotating in a circle, so the peak excitation is significantly less.

Beyond the limitations in physics, single phase motors are popular for "cost-sensitive" applications which means that many of then run with efficiencies in the 20's.

When driving a motor with a drive, the efficiency can be improved by making heavier shunts and placing them towards the outside on the armature. This would be a no-go for an offline motor because it would look like a terrible short circuit upon start up. But, drives can start with low frequencies and voltages. Then, the drive/motor pair can benefit from the reduced slip.

 

[Alex Cramphorn]

Thank you for your responses.

If you study the magnetic vector in the single phase machine, you find that it's really just bouncing back and forth along a line of action. The starter winding can help it produce a sort of ellipsoid, but the field is still reaching high peaks and decaying.
With a three phase motor, the magnetic vector is smoothly rotating in a circle, so the peak excitation is significantly less.

Mike, I might be misunderstanding this but when you say the magnetic vector, is that what is illustrated by the sinusoidal curve? If this is the case, is the three phase motor not using the same total energy as the single phase but producing it in a shorter period of time?

 

[Mike_In_Plano]

The best way to understand the vector is to imagine the rotor having been removed, and replaced it with a super agile magnetic compass that sits in the center.

Now, as a test of the basics, you send a DC current through the run winding of a single phase machine. You'll see the compass align with an induced magnetic field. This field has a direction, and strength. Thus it is vector.

Now, let's put aside the single phase machine for while and place a three phase machine on the bench. Again, it just has the compass inside.
We could attach this machine to a three phase power outlet, but for this experiment, I want you to observe what happens. So, we'll attach the motor to a low frequency three phase source (say a motor turning a three phase generator, but through gears to reduce it's speed).
Once again, you can see the compass pointing out the direction of a magnetic field, but now, the compass is rotating. The magnetic vector is now smoothly rotating .
(Historic note - Nikola Tesla claimed to have imagined this while sitting on a bench. Thus he found a solution to make motors that ran on AC!)
If we speed up our three phase generator, the compass will spin correspondingly faster as it follows the field.

Now, let's go back to our single phase motor and apply a low frequency single phase supply to the run winding. We see that the compass points for awhile in one direction than very quickly flips to the opposite direction, and then back... Now, our magnetic field is not rotating. It is simply building in one direction, collapsing, and then building in the other direction. It's very difficult to imagine how this could turn anything...
But, we see that the single phase motor has an additional winding, the start winding. If we supply current to this winding through a capacitor, the current will have a phase lead.
When we watch the compass, we see that it now turns.
But, we are missing something! Our compass tells direction, but fails to tell us the strength of the field. Suppose we pull the compass out, place thumbtack a spring in the middle with a magnet at the end of the spring. Now, we can see the magnet being pulled in an ellipse because the strong run winding only pulls hard along one line, and the weaker, start winding, pulls a little bit off to the sides.

Let's but the amateur back in the single phase winding, disconnect the start winding and apply power. It just sits there buzzing. There is no circular path to rotate the amateur. If we reconnect the start winding, the motor starts turning. That's not surprising. Even though the start winding is weak, it does describe some sort of rotating vector (an ellipse) that gets the armature rotating.

Now for a really fun experiment. While the motor is rotating, we disconnect the start winding. And, it keeps turning! (Tesla didn't anticipate this!)

To understand what's happening, we have to examine the armature. It has copper or aluminum conductors all along the outside, while the middle is all iron plates stacked one atop another. The armature is essentially an inductor with a shorted winding. If you have current in an inductor that's in series with a resistor, you will see that the current will decay by i = i_beginning s e^(tR/L)
So, a really low resistance in the winding, R along with a really large inductance, L, will give you a cylindrical lump that will respond to changes in it's magnetic environment slowly. Slowly enough that it resembles a magnet after being exposed to the magnetic vector, but doesn't have time to stop being a magnet by the time the vector has been reversed.

In the end, you have an armature that has it's own magnetic field that's continually charged by the stator's field and yet always lagging behind it being physically dragged by the difference in angle between the two fields.

Now for the bonus material -
Suppose you put a handle or something on the armature and physically crank it as fast as the rotating magnetic vector spins. Then, the fields align, and the motor is not producing any torque (that's your job, the man with the crank, lol). The motor will now be spinning at it's synchronous speed.

Suppose you're feeling your Wheaties and choose to turn the motor even faster? Now, the weirdness begins. This motor that has now magnets, field coils, etc, is now a generator.

 

[Mike_In_Plano]

Ah, I see that I gave a long winded explanation, but failed to answer the question.
Essentially, if losses were proportional to excitation, you could drive something with short pulses and expect to get the performance of a smoothly driven design. In actuality, loses are nonlinear. The copper losses vary to the square of the current and iron losses increase to B^1.5-2. Thus the peak currents and fields have a disproportionate loss associated with them.

 

[zoki85]

Let's but the amateur back in the single phase winding, disconnect the start winding and apply power. It just sits there buzzing. There is no circular path to rotate the amateur. If we reconnect the start winding, the motor starts turning. That's not surprising. Even though the start winding is weak, it does describe some sort of rotating vector (an ellipse) that gets the armature rotating.

Now for a really fun experiment. While the motor is rotating, we disconnect the start winding. And, it keeps turning! (Tesla didn't anticipate this!)

Hmm, I wouldn't bet on that. His patents on induction motors originally described a 2-phase motor, associated principles and construction. In order to start the motor (to electromagneticaly generate rotating field and some initial torque), at least 2 phase windings are needed, geometricaly and "electricaly" shifted with respect to each other. The simplest case is an induction motor having 2 pair of coils, displaced by 90° on the stator, and fed by currents shifted by 90° in time. Accordingly, a 2-phase supply network supply would be a best choice for such motor. If 2-phase supply isn't available, but just a single phase supply, second phase can be artificially emulated by connecting a capacitor of adequate size in series with one the winding. This is a principle of a single phase capacitor run (and capacitor start-capacitor run) motors. It is correct that once a single phase motor runs, one of the windings can be disconnected, and motor still runs, but this degrades motor power capability and performance. True 2-phase motor have a better overall characteristics than described two cases, but it still generates a pulsating rotating field. Tesla knew that, so to overcome this drawback he constructed and patented 4-phase induction motors and 4-phase alternators, discussed their advantage and operation. All of his early motor designs were of a salient pole type, while design we're most familiar with today is with windings placed in slots cut into the stator laminations. What Tesla didn't anticipate (at least didn't patent it) is a 3-phase induction motor! Three phase system is already theoretically enough to have nonpulsating rotating field. Somebody else invented a 3-phase induction motor later, but I can't remember who*. IMHO, main advantage of a polyphase motor over single phase one isn't a nonpulsating magnetic field/torque but, simpler construction, more power and higher efficiency for the same magnetic circuit/material. To illustrate, here is comparison of two motors of one manufacturer regarding this matter

Single phase capacitor run motor:
Power: 2.2 kW
Number of poles: 2p=4
Full load efficiency: 77%
Mass: 27 kg

3-phase squirrel cage motor:
Power: 2.2 kW
Number of poles: 2p=4
Full load efficiency: 81%
Mass: 14.5 kg

And this is just a standard efficiency (IE1) 3-phase induction motor.
Premium efficiency (IE3) motor of nominal power 2.2 kW has η ≈ 87% and mass still doesn't exceed 27 kg.
200 kW, IE3, 2p=4 squirrel cage induction motor has η≈96%.

edit*: Mikhail Dobrovolsky

 

[Mike_In_Plano]

My recollection was based upon his memoirs. He described sitting on a bench, envisioning a rotating field and becoming so excited about the matter that he sent a transatlantic message home. I believe he was digging ditches at the time, so this was a huge expense.

Since reading of his history and acquiring my degree, I realized that there were any number of conceptual errors in his pursuits. The greatest of these was his lack of understand of damping and mode shifting in resonant systems. That being what it may, there is a sect of pseudo-scientist enthusiasts that follow his every conjecture as absolute.

Regarding your motor numbers, I am not surprised. There are any number of machine tools that use cheap three phase motors and have forced air cooling or fins.

As to a 96% efficient motor, that's an extreme. If you tear it down, you'll likely find copper in the squirrel cage with the shorts close to the surface. That and exceptional laminations.

Cast-in-place is preferred for forming squirrel cages. Unfortunately, copper has a high melting point, so must be cast outside of the cage and installed afterwards. So, squeezing an additional 1.25-1.5% by using copper is an expensive undertaking.

An additional bump in efficiency can be obtained by placing the squirrel cage shorts closer to the surface. However, this presents a near short at start-up, a high locked rotor load. As a mitigation, the motor can be started in a delta configuration and switched to wye. Alternately, an invertor can serve to control start up.

 

[zoki85]

Tesla invented the motor in Europe and even constructed small demo model and showed it to certain industrial hot shots there. But they didn't have guts to invest. He was also very adicted to gambling at the time. He lost practically all his savings and then left to USA where he worked for Edison who didn't wan't to hear about AC and the motor too. Then he left and digged the ditches for some time. When he earned enough he constructed the motor, showed it in public and Westinghouse saw it.Westinghouse had guts.They made a contract and the rest is history.
As concerns premium efficiency, use of copper in squirrel cage rotors, and other techniques of loss reduction, this is becoming a standard :
http://cdn.intechopen.com/pdfs-wm/14083.pdf

 

[Mike_In_Plano]

May or may not be. My references were from books I acquired in or about 1983 - before the misinformation highway was free to propagate revisionist stories.

If you wish to convince anyone that copper in rotors is the prevalent technology, please exhibit some order forms with pricing. Research papers do not a product make. As of 2014, I read that the difficulties of separately casting copper was still keeping these motors priced out of the market. However, Tesla Motors uses them, so you should sleep well knowing that they have found a niche.

 

[zoki85]

Wiki page has info as regards both present and future in Europe: New minimum energy performance standards in EU
And I know that local manufacturer doesn't make big motors of standard efficiency (IE1) anymore.
About situation and trends in Asia and America(s) I can't say anything .

 

[jim hardy]

Wiki page has info as regards both present and future in Europe: New minimum energy performance standards in EU

OMG gov't is doing to the old fashioned induction motor what they did to the automobile.
Of course if EU mandates copper our bureaucrats will have to one-up them and demand silver.

Home Hobbyists - be advised to lay in a supply of old washing machine motors for those workshop tools you build.
else that simple buffer wheel will soon require a three phase field oriented controller.
http://en.wikipedia.org/wiki/Vector_control_(motor)

sigh.

Perhaps educators will rise to the task and these complex devices will become child's play.
Science and language evolve together.
In 1800's a motor was cutting edge and Steinmetz was just thinking up phasor notation.
In my day we studied both in high school but a phase locked loop was exotic.

I envy you young guys for the progress you are going to see.

old jim

 

[Mike_In_Plano]

Jim,
I've been through the appliance efficiency crunches. They don't mandate anything regarding motors, they just mandate that the system does the same job with less energy. For air conditioners, we started by putting electronic time delays on the thermostat so that the fan was on an additional 60-80 seconds. Later, we upgraded to brushless motors in the air handlers because it was the least costly means of obtaining the required numbers. The motors worked well. Too well. So we removed copper and aluminum from the coils until we met the requirement at minimal cost :-)

In Japan, people were paying so much per kW-hr that our machines were unacceptable. So Japan uses variable speed compressors and blowers. We tried that experiment as well, adding variable speed compressors to the mix. It was quite a thrill building power factor correction and motor drives for up to 8hp for consumer grade (i.e. priced) equipment. Anyway, the marketing didn't support the notion. I suspect Americans would have bought it if it had an NFL logo...

The mainstay of American ACs turned to pole-switching or dual compressors designs and though I cried foul at the time, I'm now older and much prefer the latter design. Why? Well, if a microcontroller hangs up while controlling the drive, then you have about 10 us of shoot through before your IGBTs go to silicon heaven and take the rest with them. Anyway, drives and their motors are not as robust as your good old fashioned Induction motor. Many an AC is still slaving after 20 years. Most drives are history within 10.

As for refrigerators, GE made great inroads by replacing the IIM (infinitely inefficient motor) used for air circulation with a miniature brushless design. Less heat delivered into the cooling space translates to less energy to remove :-) Next up, perhaps vacuum panels. These are essentially foamed silica sealed in an aluminum envelop with the air evacuated. Unfortunately, when they say vacuum, they mean a serious vacuum. Pumping to 1 torr never made much difference in the lab. 1 millitorr started to be useful. Thus slow leakage into the panel can destroy it's usefulness 7-10 years out (Just outside the warranty - darn).

The Germans have begun to use vacuum panels in home construction (Eeek!) I would like the same construction, but only if I had access to replace them easily. With them used in home construction, I'm sure that somebody has already chosen them as the answer to refrigerator energy mandates.

 

[zoki85]

OMG gov't is doing to the old fashioned induction motor what they did to the automobile.
Of course if EU mandates copper our bureaucrats will have to one-up them and demand silver.

Home Hobbyists - be advised to lay in a supply of old washing machine motors for those workshop tools you build.
else that simple buffer wheel will soon require a three phase field oriented controller.

World of technology and engineering is changing faster than ever nowdays. On average, it takes about 10 years that new advances & improvements become visible. In some areas that period is even shorter (for ITs → 3-4 years)...Regarding washing machines I have a funny story to share. When I studied EE, professor mentioned we could find induction motors in washing machines. That's great news I thought, becouse we had just bought new washing machine for our home and I wanted to see if I could take out the motor from the old one. Imagine my disappointment when I opened that thing and realized a commutated AC universal motor was there lol.

 

[jim hardy]

Wow . Thanks Dave.

Imagine my disappointment when I opened that thing and realized a commutated AC universal motor was there lol.

Now THAT"S a first...for me at least.

one wonders if it might have been a repulsion-start. Now there was a contraption, they had a centrifugal mechanism to lift the brushes after start.

I not long ago picked up a $3 flea market "modern" appliance motor just to see what it was. Judging by the pulley i'd guess it's probably from a dryer.
Looks brand new. Three phase 150 volt 1.2kw two pole. No idea what to do with it, but it's $190 on ebay...

 

[zoki85]

Wow . Thanks Dave.

Now THAT"S a first...for me at least.

one wonders if it might have been a repulsion-start. Now there was a contraption, they had a centrifugal mechanism to lift the brushes after start.

Commutated AC/DC series wound universal motors are/were not so uncommon in washing machines here. Sometimes they are called just AC series motors. Brushes always "on" during run. Electronic boards and thyristors for motor speed control.