Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

3 Phase Generator question

Status
Not open for further replies.
Hello. I'm new to this forum, but old to electronics / electrical stuff.

I've been playing with the idea of turning a 3 phase AC induction motor into a generator. It would have a squirrel cage rotor, and a synchronous speed of 1800 RPM. From what I've read, in order to get this type of motor generating electricity, the rotor needs to be spun at above synchronous speed, and have the stator field excited externally, or by capacitors. This would result in a generator that only works in a very narrow speed range. That's not what I want. I want to generate good power at all speeds.

Here's the question. Would it be possible to attach the stator windings to a variable frequency inverter fed with external power, thus, making the synchronous speed adjustable? So lets say the machine is attached to a wind turbine spinning 25 RPM. The system would detect this rotor speed and would change the frequency of the stator field to be slightly under this, say 24 RMP, thus creating a slip of 1 RPM, causing the system to generate power?

It seems the only other option that closely resembles this is a doubly fed induction motor / generator.. but this would be more expensive, and 3 PH ac squirrel cage motors are very common.

Thanks for any input, I hope you guys understand the question.
 
Hello there.

One of our members "tcmtech" is experienced in this area.
I don't know if he is on the forum these days, but in his absence let me put forward a few ideas.

An induction motor works because the rotor is turning at slightly less than the synchronous speed and the rotating magnetic field created by the stator inducing currents in the bars of the "squirrel cage" in the rotor, which in turn creates a magnetic field which reacts agaings the field of the stator.

To create a generator, you have to have the rotor create a magnetic field. This is usually done by having windings on the rotor excited by a DC current via a pair of slip rings.
As the rotor turns, the magnetic field cuts the windings on the stator and induces a voltage.
Think automotive alternator.

On that basis, simply turning the rotor of a motor is not going to do much at all.

I have seen somewhere on the interwebs where someone fitted neodymium (spelling?) magnets into the rotor of an old motor to create a generator.
This involved much work with a milling machine.

JimB
 
I'm still around! Just not a lot of questions peak my interests enough to reply. :p

As the VFD method goes sort of yes but your return power will be less than what the VFD take to drive the system at that low of RPM.

What you want for that type of application is a wound rotor three phase motor. With that you can drive the rotors magnetic field at a rotational speed different than the actual mechanical speed is at to get the frequency what you want from the stator.

It's called sub synchronous or super synchronous operation and that's how the majority of the modern commercial scale wind turbines work but it has the same power VS speed limitations as any other type of induction motor has which means that at a 25 RPM input on normally 1800 RPM motor your range of available mechanical power that can be converted back to electrical power is going to still be pretty small which is why the big wind turbine systems still use a gearbox to get their gensets running at a speed close to their ideal synchronous speed.
 
To create a generator, you have to have the rotor create a magnetic field. This is usually done by having windings on the rotor excited by a DC current via a pair of slip rings.
As the rotor turns, the magnetic field cuts the windings on the stator and induces a voltage.
Think automotive alternator.

On that basis, simply turning the rotor of a motor is not going to do much at all.

JimB

You are kind of mistaken here. It is very possible to have a squirrel cage rotor produce power output at the stator windings by rotating the rotor faster than synchronous speed. It's the same principle operation as when the motor is operating as a motor.. the rotating magnetic field of the stator cuts the conducting bars of the squirrel cage, producing an induced current which produces a magnetic field, causing the rotor to be attracted to the rotating magnetic field in the stator. BUT when the rotor is rotating faster than the stators magnetic field, the rotors conducting bars cut the field in the opposite direction, causing excess energy that is output from the stator. From what I've read however, this is generally happening at slightly above synchronous speed, and I haven't been able to find any information about people changing the input frequency to the stator in order to manipulate and change the sync speed in order to have the motor generate power at lower RPM! I understand that 25RPM wouldn't produce much power.. That was simply an example. I should have used 1000 RPM as the example or something like that.

In general, I want to be able to generate power in much the same way that you would with a doubly fed induction motor WITHOUT THE COST of a doubly fed induction motor. (AC induction motor which has both stator windings and rotor windings). The reason for this is because squirrel cage 3 phase induction motors are cheaper, more reliable, and more common. I'm more likely to find them at a demolition site or a garage sale or whatever. Also, I don't want to have to rely on a gearbox or something like that in order to keep the motor spinning at a constant RPM in order to generate electricity. I'd like to be able to generate at ANY speed. Preferably, below the synchronous speed.

I have designed a variable frequency drive for 3 phase AC induction squirrel cage motors already, I would like to know if it would work to generate power if I programmed the VFD to output a sine wave voltage to the stator windings slightly below the actual measured speed of the rotor.

This would also be useful for regenerative breaking in an electric vehicle situation.. coasting downhill etc.
thanks for your replies guys.. It would be nice to hear from someone that's actually done it.
 
Yes trying to spin the motor faster than the speed teh VFD is set at will cause it to do regenerative braking into the VFD.

From there if your VFD has a specific load dump resistor the power from regen braking can be sent someplace else although the main function of the braking resistor is to simply bleed off the power being fed back tot eh DC bus in the VFD unit which if the VFD was being powered by a DC source like a battery pack the regen braking would ust recharge the battery.

Still power return is limited by the speed Vs torque curve of the motor which ultimately in itself is limited by what the motors windings can handle in amps. From that a 1800 RPM motor at full continuous duty regen at 1000 RPM would only be capable of 5/9th's of its stock 1800 RPM power capacity as a generator.
 
Okay so as suspected, the squirrel cage 3 phase induction motor CAN be used as a variable speed generator if the stator field is powered by a variable frequency drive... keeping the RPM of the stator field slightly lower than the actual rotor RPM. Sounds good. Now.. do you have any advice as to how exactly I could capture this power as it is produced?

In the picture that I've attached to this post, the power generated will be absorbed by the capacitor bank... However, I would prefer that it goes to charge a battery bank. How would I do this? What would I put in place of the capacitor bank in order to have it charge a battery bank?? I figure that the power being generated won't be a nice uniform voltage... so yea.. what should I put there in order to get it in shape for hooking up to a battery bank? Let's go with a 48v bank just for discussion purposes.

squirrelcageinductiongenerator-gif.85255

In practical use, I would have the battery bank being charged also be the battery bank being used to power the stator field... But I want to keep things simple for now. Thanks for the help!
 

Attachments

  • squirrelcageinductiongenerator.gif
    squirrelcageinductiongenerator.gif
    7.2 KB · Views: 1,608
But I want to keep things simple for now. Thanks for the help!

Well the whole VFD feedback system pretty much rules out the simple aspect.

Simple would be to either use a permanent magnet rotor type induction motor or a permanent magnet type DC motor as a the generator and just do a bit basic buck/boost voltage conversion to get the voltage you need.
 
Simple would be to either use a permanent magnet rotor type induction motor or a permanent magnet type DC motor as a the generator and just do a bit basic buck/boost voltage conversion to get the voltage you need.

Thanks for the suggestion.. But I was already well aware of the option to use PM DC motor or doubly wound AC induction motor as generators... but that's not the goal of my project. The goal is to make the best use of the numerous 3 phase AC squirrel cage induction motors I have access to at reasonable prices.. and to implement regenerative braking / coasting into my VFD design for 3ph squirrel cage motors.

I suggested keeping things simple in this instance so that we can focus on the issues I want to find solutions to right now... Kind of braking things down into reasonable pieces and solving each piece one by one. The puzzle piece I'm working out right now is how to best absorb the surplus power from the above system into a battery bank that is 48 volts.

If you have a solution to this specific issue, I'd love to hear it!
 
My idea was to customize a circuit such as this charging circuit to work on a 48 volt battery bank, not 12v. The power source for the charger circuit would be where the capacitor bank is in the VFD system I posted above. What do you think?
 
I think I can't make any sense of the whole concept you are proposing. I can not find the gains of doing a power generation system as such being not one single component justifies it by cost simplicity or efficiency.

Off the shelf stock three phase induction motors run on 208 - 460 VAC in most cases which relates to VFD units that have DC bus voltages of ~290 - 340 VDC or 680 VDC of which neither is directly compatible with a 48 VDC battery system.

On top of all of that at low RPM's VFD units are putting out a proportionally low voltage and when doing dynamic braking at those speeds most tend to simply inject a small amount of DC voltage into the three phase motor to create the braking effect without having to absorb power themselves.
 
OK... I guess I have to go into detail to explain the reason why I'm asking the question... I wanted to keep the question simple in order to get straight to the answer without confusing people.. But since you asked.. Here we go.

I have designed and will be building a VFD for a 3 phase squirrel cage AC induction motor that will be used to power an electric vehicle. The power source will be either 48V, 72V, 96V, or 108V DC battery bank.. Haven't decided yet because I haven't purchased the batteries yet. In any case, the DC source will be converted to an AC frequency by using mosfets, IGBTs, or SCRs. (Also undecided) The AC frequency will be delivered by pulse width modulation on a carrier frequency of 8 khz in order to avoid audible noise and harmonics. The voltage amplitude will be controlled independently from the voltage frequency. That is to say, this will NOT be a volts per hertz (v/hz) type drive. Thus, the torque will not vary directly with frequency. Thus, full torque and full voltage will be available at low speeds when needed or wanted. The actual rotor speed will be available to the microcontroller via a speed sensing quadrature encoder. If this is not already built into the motor, a similar type sensor will be retrofitted to suit the same purpose.

With all that being said... I want to build into the VFD a function that will allow regenerative braking in a very straight forward manner and without any extra hardware. The way I want to get regenerative braking to work is that when the brake pedal is pressed, or a special button is pressed, the VFD switches to a function where instead of providing a frequency to the motor that will maintain current speed or increase current speed, it will provide a frequency that is slightly LOWER than actual rotor speed. Since the voltage amplitude is controlled independantly from frequency, an appropriate voltage can be provided that will in fact produce excess power at the stator because the motor is now functioning as a generator that can be fed back into the battery bank. Since the rotor is spinning faster than the magnetic field in the stator, it is now a generator. Whatever voltage is needed at the stator to do this, since voltage is controlled independently from speed, the most efficient voltage to do the job will be calculated and applied.

I also want this VFD to be able to function as a generator controller for a wind turbine or any other generator application without needing to be modified or any additional hardware to be added. As I've said before, 3ph AC induction squirrel cage motors are one of the most common motors and cheapest out there.. and also one of the most reliable with the least amount of maintenance needed. They are also very efficient, which is very important in electric vehicle application.

I hope this explains the purpose to my question. Is there anything else you'd like me to clarify?
Thanks again for the replies
 
I still cant figure how, from a squirrel cage motor, you are getting the magnetism to generate an output. The rotor is only magnetic when the stator is energized. An alternator has either a DC coil and slip rings or permanent magnets in its rotor.

In an EV, even though the can do regenerative braking, there is only so much of the regenerated current that can be used to recharge the battery. Most of the regen voltage is burned off in power resistors. A battery can output a much higher amount per time period than it can take as an input/charge. This is why the charge time is always longer than the run time on say a electric forklift.

You should join up at DIYelectriccar, https://www.diyelectriccar.com/forums/ And read about what is being done by the guys there.
 
The phrase in post 11 ;

"bank. Since the rotor is spinning faster than the magnetic field in the stator, it is now a generator. Whatever voltage is needed at the stator to do this, since voltage is controlled independently from speed, the most efficient voltage to do the job will be calculated and applied."

I'm not sure this is right.

I am of the view that over driving a squirrel cage motor simply modifies the power factor of the mains input. This comes from my lecturers statement of 50 years ago that; "a squirrel cage motor is simply a rotating transformer".

A squirrel cage motor requires a rotating magnetic field, and this comes from induction into the rotor bars from the stator. To regenerate, you HAVE TO HAVE an independent rotating field and this comes from amp turns provided by the rotor and not amp turns in the rotor derived from the stator.
I think most of the replies are saying the same thing.
 
Last edited:
Actually overspeeding and induction motor make them work as a generator works just fine. All that is required is the that it be ran proportionally over its synchronous speed as it normally runs under it to make it back feed power to the source.

To get a 10 Hp (7.5 KW) 1740 RPM motor to backfeed at 10 Hp (7.5 KW) is to over run it at around 1860 RPM or a bit more to overcome its inherent efficiency losses. I've done it countless times and it works well.

Now that said trying to do that with a VFD unit working as the electrical power source may not work so well. Most VFD units don't care for continuous or extended duty backfeed of power plus as the motor speed and frequency drop the return of power drastically falls off due to the induction slip effects from trying to keep the rotor energized.

As the stator field strength drops the rotors induced field strength goes with it and the ratio of slip increases. Same with slowing down the rotor. Less rotating field speed in the stator and less induced currents in the rotor so the slip ratios has to go up more to compensate and at some point all most all of the mechanical power being put in goes into slip losses along with magnetizing losses leaving very little to none being backfeed as usable electrical power.

A wound rotor motor may work better in this sort of application to a point but still there is a input/output torque limit that will determine the total amount of power it can convert to or from electrical to mechanical or back.

Simply put you cant take a 10 KW 1800 RPM wound rotor motor and run it at 25 RPM and get 10 KW's of power back. You would only get 25/1800 th's of the power it can produce due to the amp limits of the windings which ultimately determine the torque that can be placed on the unit as a motor or as a generator.
That's why the huge wind turbines have a limited range of rotor speed they can work with to produce full power in regen mode.
Below that range the capacity to work as a generator falls off proportionally to the ratio of speed. Half speed half power 1/10th speed 1/10th power.
 
Hold on, wait a minute...I did not know this...but I'm not remotely close to a motor expert either.
Are you saying that all other things being equal (and I know they're not, but go with it for arguments sake), you could take your average everyday $20 walmart window box fan, turn it on, spin it (either by hand or by sticking it outside in some of these crazy ND winds) some percent faster than it would normally go, and you'd be putting power back into the grid? (again, all other things being equal, for arguments sake, assuming everything in the world is perfect, the planets are lined up just right, and you're holding your tongue at the correct angle, etc)
 
skimask87.. you would have to make the power generated sinewave matched with a gridtie inverter etc... but for arguements sake.. yes. You will produce power.

I am not expecting to get full power back out of the motor when in regen mode at low speed. Of course it's going to produce less power at 250 rpm than it does at 1800 rpm. That's not the question I've been asking, and that's not my goal. and you keep refering to "most vfds"... You seem to not understand that I am designing, programming, and building my own custom VFD for this very purpose! So I don't really care if "most VFDs" won't do the job. Mine will. I will be programming into the system everything it needs to deal with situations where the speed is too low to make regen worthwhile. If the power being put into the stator is less than what it's getting back out of it, then it won't put energy into the stator at that speed.

For the guy that said most of the regen power is burned off in power resistors... My VFD will have something other than power resistors to deal with that. No wasted energy here.

My question is simply what do I need to recharge the batteries... Can I simply have the power coming off the generator at random frequencies and voltages straight into the battery bank, or do I need to first clean up the signal so that it's a constant DC voltage / current? I do have a lot of experience with motors, but lack that experience with batteries. Thankyou.

I'll ask it again just so we're all clear the purpose of this thread:

Can a battery bank be charged by any random frequency of voltage and current (as long as it's kept below the battery bank voltage) or does the battery bank have to be charged with a constant DC source voltage and current?
 
You can't charge a battery below it's own voltage! (eg. can't charge a 12v battery with 9v)
As far as being constant anything, SLA, NiCad, NiMH, those types can be charged with practically anything within reason, as long as that voltage/current doesn't violate the battery's spec's. LiPo's are a bit of a different animal, but I don't think we're talking Lithium anything here.
What type of batteries are you dealing with?
 
OK, I ask again, where does the magnetism in the S/cage rotor come from ? There is no inherent magnetism in a rotor, the shorted windings of it take care of it, as the motor powers down.

Local, do you have some 'secret' that allows you to charge a battery at a high amperage , that no one else has figured out? Not trying to be a smarta$$, just wonder why no one else can do it.
 
I've come rather late to this thread; to answer the battery charging question (without knowing what type of battery):
you can take any frequency of AC, rectify to DC and feed it into the battery. Most traction batteries (whatever the chemistry) will take all the power your motor/generator can produce whilst generating - assuming the battery is big enough to power the motor in the first place, and assuming the battery is partially discharged by driving your vehicle; then you are unlikely to over charge it by braking (unless you happen to start at the top of a hill with full batteries). Some batteries like lead acid and nicd/nife will tolerate a fair amount of overcharge anyway; lithium types would need a battery management system which for your system should include a power dump resistor in case of surplus charge the battery can't absorb.

having speed-read the earlier posts there seems to be some confusion about using induction motors as generators; I'm not an expert but I've done some research into this as I'm using a 1425rpm, (1500 synchronous at 50Hz), 240v (delta), as a stand alone hydro-generator using capacitor excitation. It's running at (very approx) 800rpm - optimum turbine speed - and generating ~110v at 26Hz. I use it to charge my 24v lead acid battery and dump the surplus power into an immersion heater.

You can - as previously stated - connect to rid A.C. and et them to backfeed the grid by pushing them above syncronous speed.

I assume your VFD will work similar to a modern 'reversible' inverter - i.e. the type that you can back-feed with a.c. to charge your batteries (I'm currently building one of these - slowly). In theory that should work, but I'm not sure if it will work at low rpm as the emf generated may be lower than battery voltage.

OK, I ask again, where does the magnetism in the S/cage rotor come from ? There is no inherent magnetism in a rotor, the shorted windings of it take care of it, as the motor powers down.

there's always a little bit of residual magnetism in the iron of the rotor - my hydro will self-excite and start generating within 1/2 a second of turning on the water so I can tell you it does work - the only exception is if you stop the generator (or try to start) with a load on - this can effectively demagnetise the rotor. If this happens I have to pass some d.c. through the stationary motor windings for a sec or two before I go and restart, otherwise it can just spin free without generating.
 
Hold on, wait a minute...I did not know this...but I'm not remotely close to a motor expert either.
Are you saying that all other things being equal (and I know they're not, but go with it for arguments sake), you could take your average everyday $20 walmart window box fan, turn it on, spin it (either by hand or by sticking it outside in some of these crazy ND winds) some percent faster than it would normally go, and you'd be putting power back into the grid? (again, all other things being equal, for arguments sake, assuming everything in the world is perfect, the planets are lined up just right, and you're holding your tongue at the correct angle, etc)


Basically yes. Granted those fans are cheap shaded pole motors so they require a considerable overspeed to get them to backfeed but yes they will do so when connected directly to the line.
No special anything required to get most any common induction motors to backfeed power. Just speed them up until you reach the feedback power you want or hit to the motors peak working power capacity. The nice part with doing a pirate grid tie power feedback that way is if the grid goes down the motor itself has zero stand alone power generating ability. Grid power goes off and it just freewheels until reconnected again. ;)
 
Last edited:
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top