Hello, I am looking for a motor manufacturer who sells about a 20,000 rpm, 20HP,inverter duty motor. It can be designed but the cost is an issue. Any clues will be appreciated. I am also curious if I can run any inverter duty motor at 20,000 rpm once I replace the bearings and balance the rotor?
Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations MintJulep on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!
Ultra high speed inverter duty motor, about 20,000 rpm 3
- Thread starter williamoh
- Start date
ozmosis
Electrical
- Oct 12, 2003
- 1,794
I've used the following for high speed operation:
they are specifically designed for this function having low inertia rotors, special bearings and the correct winding suitable for high frequency operation circa 200~400Hz. They also have low inertia fans to facilitate correct cooling.
Modifying a 'standard motor' by replacing bearings would be a dangerous option IMHO.
they are specifically designed for this function having low inertia rotors, special bearings and the correct winding suitable for high frequency operation circa 200~400Hz. They also have low inertia fans to facilitate correct cooling.
Modifying a 'standard motor' by replacing bearings would be a dangerous option IMHO.
Sounds like you are looking for a spindle motor. High speed, 2-pole, probably water-cooled for 20HP, high precision bearings.
TygerDawg
Blue Technik LLC
Virtuoso Robotics Engineering
TygerDawg
Blue Technik LLC
Virtuoso Robotics Engineering
ozmosis
Thanks for that perske motors link. They seem to make some pretty cool stuff.
One thing though that stuck was that these high speed (6000 to 24000 rpm) motors have a poor pf (0.5 to 0.7), which goes against convnetional wisdom that pf goes down with the speed.
Any theories why ?
Thanks for that perske motors link. They seem to make some pretty cool stuff.
One thing though that stuck was that these high speed (6000 to 24000 rpm) motors have a poor pf (0.5 to 0.7), which goes against convnetional wisdom that pf goes down with the speed.
Any theories why ?
Skogsgurra
Electrical
Edison,
It is true that PF goes down with speed - if you obtain lower speed using more poles. But, this is about using, probably, a two-pole motor and higher frequency.
Motors running off 50 or 60 Hz are optimized in every way you can think of. That results in the PF we are used to in standard ASIMs.
High-speed motors (like spindle motors) are usually PM motors and should, therefore, be able to run at PF=1. But that is not very common, I think that a motor running at a load angle around 45 degrees has better dynamic properties than around zero degrees. Just look at the torque vs load angle. It is rather flat at 0 degrees. Load angle influences PF and PF is quite low at 45 degrees. Actually, PF is close to cos(load angle).
C S Wilson should have something to say here.
Gunnar Englund
--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...
It is true that PF goes down with speed - if you obtain lower speed using more poles. But, this is about using, probably, a two-pole motor and higher frequency.
Motors running off 50 or 60 Hz are optimized in every way you can think of. That results in the PF we are used to in standard ASIMs.
High-speed motors (like spindle motors) are usually PM motors and should, therefore, be able to run at PF=1. But that is not very common, I think that a motor running at a load angle around 45 degrees has better dynamic properties than around zero degrees. Just look at the torque vs load angle. It is rather flat at 0 degrees. Load angle influences PF and PF is quite low at 45 degrees. Actually, PF is close to cos(load angle).
C S Wilson should have something to say here.
Gunnar Englund
--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...
If you are lookin for a North American mfg, check out Reuland Electric.
I agree on Reuland, macmckim beat me to it. I have used them on large router spindles, excellent products.
"If I had eight hours to chop down a tree, I'd spend six sharpening my axe." -- Abraham Lincoln
For the best use of Eng-Tips, please click here -> faq731-376
"If I had eight hours to chop down a tree, I'd spend six sharpening my axe." -- Abraham Lincoln
For the best use of Eng-Tips, please click here -> faq731-376
-
2
- #11
skogs -- When you are talking about power factor/torque angle in running synchronous motors, it is important that you distinguish open-loop from closed-loop operation.
In open-loop operation, it is vital that you not get too close to a torque angle of 90 degrees (PF of 1.0) because, as you note, the torque-vs-angle curve is flat. You no longer get the automatic stabilizing negative feedback of additional load increasing the generated torque, so you can "fall off the cliff" and lose synchronization.
In closed-loop operation, you can run at a torque angle of 90 degrees, and therefore a PF of 1.0, all day long, because you use the feedback to modify the angle of the output waveforms. So if you get a momentary additional load which causes deceleration, you don't advance the angle of the waveforms as much, so you never get on the side of the torque curve where torque decreases with increasing angle. Also, your servo algorithm will increase the magnitude of the waveform in response to the error in velocity/rotor angle to try to bring things back into full synchronization. Operation at a torque angle of 90 degrees is desirable because it produces the maximum torque per unit current.
If the OP selects a PM (synchronous) motor, he should definitely use closed-loop feedback control.
Also, with regard to power factor of the motor operation: When operated through an inverter, the motor PF is quite well hidden from the line due to the DC bus in between.
Curt Wilson
Delta Tau Data Systems
In open-loop operation, it is vital that you not get too close to a torque angle of 90 degrees (PF of 1.0) because, as you note, the torque-vs-angle curve is flat. You no longer get the automatic stabilizing negative feedback of additional load increasing the generated torque, so you can "fall off the cliff" and lose synchronization.
In closed-loop operation, you can run at a torque angle of 90 degrees, and therefore a PF of 1.0, all day long, because you use the feedback to modify the angle of the output waveforms. So if you get a momentary additional load which causes deceleration, you don't advance the angle of the waveforms as much, so you never get on the side of the torque curve where torque decreases with increasing angle. Also, your servo algorithm will increase the magnitude of the waveform in response to the error in velocity/rotor angle to try to bring things back into full synchronization. Operation at a torque angle of 90 degrees is desirable because it produces the maximum torque per unit current.
If the OP selects a PM (synchronous) motor, he should definitely use closed-loop feedback control.
Also, with regard to power factor of the motor operation: When operated through an inverter, the motor PF is quite well hidden from the line due to the DC bus in between.
Curt Wilson
Delta Tau Data Systems
Skogsgurra
Electrical
Thanks Curt, much needed!
Gunnar Englund
--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...
Gunnar Englund
--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...
whitehendrix
Electrical
could someone please explain the torque-angle you've mentioned? thats a new one to me..
i'm assuming it deals with wave form maniplulation to gain efficiency?
again.. guessing.. that is something i didn't learn when i secured my degree
i'm assuming it deals with wave form maniplulation to gain efficiency?
again.. guessing.. that is something i didn't learn when i secured my degree
-
1
- #14
Yes, thanks Curt. Nice description.
whitehendrix; Talking synchronous here only.
The torque angle referred to is the phasor, (rotating vector), between the phasor of the power system and the phasor of the rotor chasing the system. As the angle gets greater you are loading the motor more and more. At a specific point you will get pull-out and the rotor will lose synchronizm with the system. When you aren't crowding 90 degrees the motor is on a curve that will increase the torque and provide self regulation of the torque. As you start to crowd 90 degrees you've gotten everything you can from the motor and the curve has no more torque to offer. Further load increase will cause pull-out.
This works the same way with generators. My prof always complained that as the power grid gets more and more loaded every machine on the network has this angle getting closer and closer to 90 degrees. Any subsequent brief overload will cause a machine on the fine edge to pull-out. The machine must be instantly removed from the network or something will be destroyed. The networks loss of this machine means all the other machines are pushed closer to their pull-outs. If another machine pulls-out you can get a rapid cascade of all the machines on the network doing it and have it all go black.
Keith Cress
kcress -
whitehendrix; Talking synchronous here only.
The torque angle referred to is the phasor, (rotating vector), between the phasor of the power system and the phasor of the rotor chasing the system. As the angle gets greater you are loading the motor more and more. At a specific point you will get pull-out and the rotor will lose synchronizm with the system. When you aren't crowding 90 degrees the motor is on a curve that will increase the torque and provide self regulation of the torque. As you start to crowd 90 degrees you've gotten everything you can from the motor and the curve has no more torque to offer. Further load increase will cause pull-out.
This works the same way with generators. My prof always complained that as the power grid gets more and more loaded every machine on the network has this angle getting closer and closer to 90 degrees. Any subsequent brief overload will cause a machine on the fine edge to pull-out. The machine must be instantly removed from the network or something will be destroyed. The networks loss of this machine means all the other machines are pushed closer to their pull-outs. If another machine pulls-out you can get a rapid cascade of all the machines on the network doing it and have it all go black.
Keith Cress
kcress -
- Thread starter
- #15
Thanks very much for the valuable information! With the key word "high frequency motors" I was able to find many manufacturers although most of them were machine spindle manufacturers for milling or grinding and CNC routers.
I will contact them to discuss the application.
One curious question: What are the major differences in the windings between 3600 rpm AC induction motor and 24,000 rpm AC induction motor (high frequency motor)? Insulation? Method of winding?
I will contact them to discuss the application.
One curious question: What are the major differences in the windings between 3600 rpm AC induction motor and 24,000 rpm AC induction motor (high frequency motor)? Insulation? Method of winding?
TurbineGen
Electrical
If we are talking about the differences between an AC induction 2 pole machine (3600 rpm) and a synchonous high speed (high frequency) permanant magnet machine, there are many differences. Obviously, the bearings in the higher speed motor are rated for more speed. Also, the rotor in the synchronous motor contains permanent magnets instead of rotor bars. You will also notice that the stator diameter is often reduced, but is longer in the high speed machine. This limits the tangential velocity of the rotor and therefore the centrifugal forces. The winding methods are similar and so is the insulation, however for efficiency, the high speed motors may have thinner stator laminations to reduce eddy losses in the iron.
------------------------------------------------------------------------
If it is broken, fix it. If it isn't broken, I'll soon fix that.
------------------------------------------------------------------------
If it is broken, fix it. If it isn't broken, I'll soon fix that.
Keith -- A small clarification on your entry. Technically, you have the direction of causality backwards. You say, "As the angle gets greater you are loading the motor more and more." This would be better phrased, "As you load the motor more, the angle gets greater."
The torque angle is the angle between the orientation of the stator magnetic field and the rotor magnetic field. The torque generated is proportional to the sine of this angle, and so reaches a maximum at 90 degrees. (The maximum torque on a compass needle is when it is pointing east-west; there is no torque on it when it is pointing north-south.)
In the "open-loop" case, operational stability comes from the change in torque with respect to torque angle. The gain in this electromagnetic feedback loop is proportional to the slope (derivative) of the torque-angle curve. Since this is a sine curve, the derivative is a cosine curve. As you approach a 90 degree torque angle, this gain goes to zero, and as you pass it, the gain changes sign, so you go unstable.
Curt Wilson
Delta Tau Data Systems
The torque angle is the angle between the orientation of the stator magnetic field and the rotor magnetic field. The torque generated is proportional to the sine of this angle, and so reaches a maximum at 90 degrees. (The maximum torque on a compass needle is when it is pointing east-west; there is no torque on it when it is pointing north-south.)
In the "open-loop" case, operational stability comes from the change in torque with respect to torque angle. The gain in this electromagnetic feedback loop is proportional to the slope (derivative) of the torque-angle curve. Since this is a sine curve, the derivative is a cosine curve. As you approach a 90 degree torque angle, this gain goes to zero, and as you pass it, the gain changes sign, so you go unstable.
Curt Wilson
Delta Tau Data Systems
Just a point of clarification: There is no need for this motor to be of synchronous design. The company I has cited above builds induction motors to this speed and frequency and the results, at least for me, have been quite acceptable.
Unless I've missed something, the original poster hasn't mentioned anything that would require a synchronous machine.
Unless I've missed something, the original poster hasn't mentioned anything that would require a synchronous machine.
- Thread starter
- #20
- Status
Similar threads
- Locked
- Question
- Locked
- Question
- Locked
- Question