3 Phase Generator question

[tcmtech]

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.

That slip that induction motors have is the result of the stator inducing a magnetic field into the rotor which in fact does have windings but not as most would think of them. That heavy solid copper or aluminum squirrel cage molded into the laminated iron rotor core is nothing more than a set of dead shorted windings that get a magnetic field induced into them by the stator in order to create a set of magnetic poles that will follow the rotating field of the stator.
Being a dead short the slip difference from ideal synchronous speed to actual speed is the result of having to induce that magnetic field into the rotor which will only polarize in reference to the stators fields due to a small amount of speed slip.

If that makes any sense.

 

[rumpfy]

So it looks like one needs a residual magnetism in the rotor to get it going. If so, then we are talking about a synchronous machine and not a squirrel cage machine.
With auto alternators, the magnetic characteristics of the rotor are very soft. It is IMPOSSIBLE to start one of these alternators unless a dc current is applied to the rotor winding. The reason for this is to allow the alternator to be shutdown if no charging current is required.
In Mabs case, it is not clear what is meant by 'capacitor excitation', but presumably, it is using a capacitor to phase shift the output current from one phase by 120 degrees to another phase winding. This is often done in order to operate a 3 phase motor from a single phase supply. But to generate DC for battery charging it is not clear what is the point of this capacitor.
I just dont understand why a PMDC motor is not the way to go.
In post 14, reference is made to 'wound rotor'.
My understanding of squirrel cage motors with wound rotors goes back to the idea of repulsion motors. At startup of a squirrel cage motor, the rotor bar currents are at mains frequency and the impedance of the rotor is high. Consequently, the rotor current is limited by the rotor impedance. As the speed of the rotor increases, the induced rotor current frequency falls and reduces the rotor impedance, so the rotor current increases, along with the motor torque. To improve the starting torque of a squirrel cage motor, increasing the rotor resistance improves the power factor of the rotor current and improves the motor torque. When the motor comes up to speed, the rotor winding is short circuited. The rotor winding is either brought out to a starting resistance (adjustable), or controlled by a centrifugal switch. So how this is modified to make an alternator is up to the possessor of the actual machine.

 

[localbroadcast]

rumpfy... you are confusing the 3 phase squirrel cage motors we are talking about with a single pole capacitor start / run squirrel cage induction motor. With a 3 phase squirrel cage motor, there is no need for a capacitor to offset the phases as the phase shifts are inherently built into the 3 phase power being fed into the motor...

The Capacitor excitation being talked about before was being used to start power generation in a stand alone 3 phase induction squirrel cage motor... In order for a squirrel cage motor to generate power, the STATOR needs to be energized in order to produce a magnetic field. When the bars of the squirrel cage in the rotor rotate past this magnetic field, it induces a current in the bars, creating a magnetic field. The rotors magnetic field reacts with the magnetic field in the stator and since the rotor is moving FASTER than the field of the stator, it ADDS to the current already in the stator, and this is the excess energy that can be collected as generated power. With the capacitors connected, the capacitor absorbs this energy, and releases it in order to repeat the process. There will be excess energy in addition to the energy absorbed by the capacitors, and that can be fed into a battery bank or to the grid or used by a load etc.

If, however, the generator is not in a stand alone system, then there is no need for the capacitors. The stator can simply be energized by the grid, or by a battery bank. Here is a graphic that will be worth a thousand words.



FIGURE (a) is a self excited induction generator with a gearbox attached.

As for the guy who said that batteries cannot be charged by a voltage lower than their output voltage.. THANKS. That's the exact type of answere i've been looking for from you guys. THANKS. I don't think I'll need the dump load resistors with my VFD though... If the batteries are full, I'll simply shut off any power going to the stator, and the motor will stop generating power. problem solved.

 

[shortbus=]

Thanks for the answers about generating with out magnets. Of all the motors I've taken apart over the years, I never found one with enough residual magnetism to show. Learn something new every day.

 

[tcmtech]

Seems like this design is getting overly complicated rather fast in order to keep from having to use anything but a cheap induction type motor.

I am still not sure how you figure you will successfully design a high powered VFD unit to do what you want when you are having difficulties understanding basic AC power and DC battery charging principles.

If you are after a grid tie type wind turbine system this is a very complicated and ultimately expensive and inefficient way to do it.

Believe me I know DIY level grid tie tech for AE power. I practically wrote the book here on it.

 

[localbroadcast]

who said I had any difficulties understanding basic AC power? I've never built a circuit for charging DC batteries, so I needed to ask what was required to do so.. That means I don't know anything? hmm.. strange way of making people feel welcome.

I worked at allen bradley building large 5000 HP VFD's for years. Don't worry about me.. I know VFD's inside and out.

 

[tcmtech]

Okay? Then from that you should know how induction motors interact with them as well along with the whats and hows of feeding power back to their DC bus power side of things during regen braking.

What exactly do you need us for then?

 

[localbroadcast]

Wanted to know if synchronous speed could be changed making a squirrel cage induction motor work as a generator at more than one speed... and also what was required to charge a DC battery bank. that's why I asked those questions.

 

[localbroadcast]

Wanted to know if synchronous speed could be changed making a squirrel cage induction motor work as a generator at more than one speed... and also what was required to charge a DC battery bank. that's why I asked those questions.

 

[rumpfy]

Hi Local,
You originally asked about the use of 3 phase induction motors as alternators for use as energy generation driven by windpower.
The rotational speeds you talked about were 25 rpm.
You also seemed to recognise that slip is an important factor in producing rotor current.
You then asked for advice about feasibility of a scheme you proposed.
tcmtech has fiddled with a lot of this stuff for years and has committed his plans/ideas/designs to this forum.
The bottom line seems to be that it may be possible to generate energy with a 3 phase motor; but the common thread of comment is that it is unclear as to how the motor can generate given the typical rotating field is NOT present and that much of the discussion is 'how to generate a rotating magnetic field'.
In your post 23, you show the general arrangement for a wind generator using a SEIG and in (b) you have presumably a permanent magnet IG.
My understanding of ALL alternators is that they have a rotating magnetic field and that this is supplied from a permanent magnet OR a wound field.
My understanding is that a SEIG actually is constructed using a rotating field which is generated by transformer action from the main stator windings, to a secondary which is attached to the rotor. By this means, the machine can dispense with the use of slip rings. The AC current induced into this secondary also has attached rectifiers to produce the DC current for the field winding.
In your concept, you say this is all un-necessary. You may be correct, but I do not believe that the general form of electricity generation using rotatating machines has ignored the overall problem of efficiency. There have been posts on this web site, I think, that describe the fitting of permanent magnets to the rotor of a squirrel cage machine. The rotor bar regions are machined out to produce a slot into which the magnets are glued.
In your concept, the varying slip of a motor used as an alternator, seems to be irrelevant.
My view on this is that while you may get something, most of the energy you generate will dissipate itself in the excitation process.
The general thrust of the comments is that your ideas are going to give you a lot of work for almost no result.
However, it is also possible that if you build a device based on your principles, that it will work and it will be such a significant development that you will have invented a new way to produce cheap electricity.
So while I have no wish to denigrate anything you say or believe, the fact that you asked a question originally about motors used as generators suggests you are wondering about certain aspects of the problem you set for your self.
As an aside, I saw yesterday some suggestion that Nicola Tesla invented an electricity generator that produced 10 kilowatt of output for only 1 kilowatt of input. These kinds of suggestions pop up all the time and generally they can all be characterised that the proposer is not bound by the basic facts of physics, and the target audience is of the extroverted faith healer class. Unfortunately with engineering of mechanical and electrical devices and systems, we all find out who the frauds are when the start button gets pushed.

 

[localbroadcast]

the 25 RPM reference was just an example. maybe I should have used an example like 1000 RPM instead because all of you seem to be getting hung up on the fact that it was such a low speed. The graphic that I posted in post 23 was not one that I drew, and not a system that I will be using. I was simply posting that to show that a squirrel cage induction generator uses capacitors to self excite the stator without an outside powersource, whereas single phase induction motors use capacitors to offset the phase of the second set of windings to produce a rotating field. The permanent magnet generator in figure B was not supposed to be part of this discussion.

Anyways.. if you look up squirrel cage induction generators, you will see that yes it is possible to generate power with one of these motors without magnets and without windings in the rotor. The rotor just has to be driven above synchronous speed. When SCIM are powered from the grid at a stable 60hz, the synchronous speed is constant, and thus, the motor will only generate power at slightly above synchronous speeds. (around 1860RPM for a 1800RPM synchronous speed motor). However, what I am proposing is to feed the motor at a variable frequency, changing the synchronous speed, and making it able to produce power at a wide range of speeds. That's why I made this thread, to ask if anyone has done that before and if it was possible. I also wanted advice on how to capture that excess energy for use in charging a battery bank.

Two simple questions that a lot of people have gotten really confused about. I realize that a wound rotor induction motor is able to produce power at variable speeds, but that's not what I'm interested in doing. Since I'll be programming a VFD to control the speed of an induction motor anyways, it wouldn't make it any more expensive to build to simply add the regenerative code into the microcontroller... so why wouldn't I try and do it?? After I build it I'll be able to run tests and see what range of speeds I'll be able to produce regenerative braking at.

 

[skimask87]

So, if I'm smelling what you're cooking, you're talking about changing the line freq feeding the motor according to how fast the motor is spinning thereby keeping that speed/freq relationship in the range where it can make decent power.
Eg...if said motor is normally driven 60hz & 1860rpm, but this day the motor wants to be driven at 3720rpm, then you would feed said motor with 120hz. Or maybe 930rpm at 30hz.....and so on.
Is that halfway correct?

 

[localbroadcast]

VFD stands for variable frequency drive. The speed of the motor depends on the frequency its being powered by. At line frequency (60hz) the motor only goes one speed. If you provide it with a different frequency using the VFD, then the speed changes. The frequency actually is the speed at which the rotating magnetic field of the stator rotates. The rotor follows it because of induced current in the bars of the squirrel cage, and this induces a magnetic field that is attracted to the stator's magnetic field. There will always be some slip, and it's actually required for the motor to operate. Without the slip, the bars of the cage would not be cut by the magnetic field, and no current / field would be induced in the rotor.

In generator mode, the rotor is spun faster than the stators rotating magnetic field. The bars of the squirrel cage in the rotor still cut the magnetic field of the stator, but in the opposite direction. This causes the magnetic field of the rotor to induce current in the stator that ADDS to the current already in the stator. This is the power being produced. Power is being fed into the stator to create the rotating field, but its being added to, and thus, more current out than what went in.

What I want to do is detect the actual speed that the rotor is travelling when attached to, FOR EXAMPLE, an electric vehicle coasting down a hill. Lets say FOR EXAMPLE that this speed is 900 RPM. FOR EXAMPLE. IT COULD BE MORE, OR LESS, 900 is an EXAMPLE for discussions sake. What my VFD would do when the regenerative brake is applied, it would apply power to the stator of the motor at a frequency that is slightly less than 900 RPM. FOR EXAMPLE, it would feed the motor a frequency that creates a rotating magnetic field in the stator at 580 RPM. In actuallity, it may need to feed it a frequency that equals out to 590RMP or 595 RPM, but for discussion sake I am not concerned with that at this time. Since the new synchronous speed of the motor is now 590 RPM, since the motor is actually spinning 600 RPM, in theory, excess power will be produced. As the brake causes the motor to slow down, the speed of the motor will decrease, and the VFD will detect this and adjust the frequency of the power being applied to the stator.

I hope everyone who reads this will take the time to understand the purpose of the questions I'm trying to ask here.

 

[skimask87]

OK, same thing talked about from opposite direction. Don't see why something like that wouldn't work. Its not like the range where you could generate is really narrow. However, I would wonder if all of the extra losses of the drive and recovery methods might wipe out any gains from regen'ing using this method...

 

[localbroadcast]

Skimask.. Glad to have you understanding the theory and getting on board with the idea. And yes, that is the big question.. Will the losses be more than the gains. I am confident that there will be surplus energy using this method when at high speeds, but as the motor slows down, there will be a cutoff speed where the energy used to energize the stator will be more than what is being recovered. That cutoff speed will have to be determined once this system is actually put together and field tests can be run.

The other things that will need to be field tested and decided are the optimum voltage, current, and slip % to be fed into the stator during regen mode in order to get the most efficient energy recovery rate.

Anyways.. onto the battery charging part of this discussion.. I think I'm going to go with a standard battery charging cicuit implementing a voltage / current regulator. Attached is a sample circuit. This, again, is an EXAMPLE and is not exactly what I plan on using. Voltage and current limits will have to be adjusted according to the battery bank, and the power source is obviously not an 18v 3A transformer, but the output of the stator of the squirrel cage motor. Also, it will be 3 phase.

 

[localbroadcast]

Here's a great video for anyone still confused about how AC motors / generators work. It details the operation of the majority of AC motors. The type of motor I'm talking about using is an ASYNCHRONOUS AC induction squirrel cage motor.

 

[skimask87]

Anyways.. onto the battery charging part of this discussion.. I think I'm going to go with a standard battery charging cicuit implementing a voltage / current regulator. Attached is a sample circuit. This, again, is an EXAMPLE and is not exactly what I plan on using. Voltage and current limits will have to be adjusted according to the battery bank, and the power source is obviously not an 18v 3A transformer, but the output of the stator of the squirrel cage motor. Also, it will be 3 phase.

And if you're concerned with efficiency, this is exactly the circuit you would NOT want to use...a linear regulator. Among other things, you'll want a switch mode regulator.
Basically, a linear regulator does it's regulating by dumping excess energy as heat. Conversely, basically, a switch mode regulator does it's regulating by 'topping off' an output periodically.
Throwing the old electricity vs. water analogy at it...if you're trying to keep a bucket full, a linear regulator lets the extra water flow onto the ground. A switch mode regulator turns the water on and off keeping the bucket just barely at the point of overflowing.

 

[tcmtech]

I followed the concept of the using the VFD unit to change the operating frequency but my concern is that as the RPMs drop the total available power that can be taken back also drops.

At normal speed the induction slip may take 5% of the total energy going into the motor so say you are using a 10 KW motor 500 watts are being used to keep the rotor inductively energized.

At half speed you still need 500 watts to keep the rotor energized but now only have 5000 watts to work with but being your stator field energy is half the slip ratio will need to double to keep the 500 watts induced into the rotor.

That's the problem. You get a rapidly diminishing rate of power return with this type of process the further you drop away from the stock operating frequency and speed. Add in the control system power requirements and losses and things will get worse even faster.

As I said in an earlier post this is an overly complicated and inefficient way of ding what should be a simple process.

Nobody here is trying to stop you from building anything but we are giving you a educated warning that what you may end up with will be a vary complicated and expensive system that does not perform anywhere close to expectations.

With AE and wind power in particular simple efficient and robust is what works. Maximum returns for minimal input of energy and materials is what makes a system practical.

 

[mab2]

I think part of the problem is that you're being rather vague about the actual application:- if it's for an EV then the VFD with regen makes perfect sense to me; if it's for hydro power (like my system) it obviously wouldn't.

Also the application determines whether charger efficiency is a priority or not - I'm sure the VFD/regen system will work but as to whether it's worth the effort depends on where/how it's going to be used.

As for your battery charging system, again without knowing more about the application it's difficult to offer advice; the charger you posted would work well for a small battery such as you might find in a house burglar alarm, but would be useless for an off-grid household battery or an EV traction battery. The charging system depends on battery chemistry, size of the system and on the nature of the source of charging power (long duration low power / short high power bursts). For e.g.: if your VFD/regen system is in an EV you are likely to get short bursts of energy when braking and your battery is likely to be partially charged so you can (and probably would want to) dump most/all the power you get in a few seconds of braking into the battery; if your system were being driven by a sustained power source you would likely opt for a longer slower charge even if it meant you were throwing away the unused power - particularly if the battery is anywhere near full charge.

if it's an EV application your best option would be a lithium type, but for a stationary then lead-acid is still (in my opinion) a better choice. Choosing the battery chemistry would be a good staring point before looking too deeply into charger circuits.

 

[localbroadcast]

The usage is for an electric vehicle. If it was solely for a windmill or for a water fed generator, then I would not be using this type of motor. Since, however, it is for an electric vehicle, the generator function of the motor is a SECONDARY function. The main function is to drive the vehicle forward. If I can absorb SOME of that energy back as regenerative braking, that will be a bonus. The purpose of this thread is just to discuss that option, because from my research, it doesn't seem like anyone has exploited this option with this type of motor yet. maybe for good reason, but I'd like to examine the option anyways.

thanks, ski, for the tip about not using a linear regulator. I will have to look into using a switch mode regulator. I've never heard of this before, so I'll do some reading. If you have any useful links, I'd be glad to take a look.