bigmoose said:
perish the thought.
dish girlfriend.
My ES torque contest DIY motor build thread.
- Thread starter liveforphysics
- Start date
perish the thought.
dish girlfriend.
LFP, a thought occurred to me that may prove useful for you. From reading about DIY wind-generators, the cross-section of a magnets field has a set area. That area is filled with copper wire. Any extra wraps in the coil that put extra copper outside the magnetic field adds resistance without a commesurate addition of current flow.
There was a method described to me where 2 or three strands of thinner wire are wound into a coil to fill the magnetic field more solidly, but the coil wire ends were soldered together and then fed by a single thick wire. Called "two in hand" or "three in hand". 3 skinny wires can carry the same current as one fat wire, but like a stack of fat logs, fat wire leaves a lot of airspace.
I'm told that in very high RPM windings, Litz wire can be used to reduce coil heat by lessening eddy (sideways) currents. Litz is finely stranded with each strand being enameled for insulation. Kind of like having "16-in-hand" wire. I don't know if I've explained this correctly, and I don't know all the benefits and drawbacks. Just a thought...
http://www.fieldlines.com/comments/2008/1/13/214327/136/11?mode=alone;showrate=1#11
There was a method described to me where 2 or three strands of thinner wire are wound into a coil to fill the magnetic field more solidly, but the coil wire ends were soldered together and then fed by a single thick wire. Called "two in hand" or "three in hand". 3 skinny wires can carry the same current as one fat wire, but like a stack of fat logs, fat wire leaves a lot of airspace.
I'm told that in very high RPM windings, Litz wire can be used to reduce coil heat by lessening eddy (sideways) currents. Litz is finely stranded with each strand being enameled for insulation. Kind of like having "16-in-hand" wire. I don't know if I've explained this correctly, and I don't know all the benefits and drawbacks. Just a thought...
http://www.fieldlines.com/comments/2008/1/13/214327/136/11?mode=alone;showrate=1#11
Oooops! thanks, Miles. Sorry guys....
matt_in_mtl
100 mW
- Joined
- Aug 13, 2008
- Messages
- 49
Hi all,
I just want to start by saying that I think this an awesome thread! I have keenly been watching it since it started.
From what i am reading, it sounds like the consensus is that copper foil lamination allow for good fill, but will cause increased losses due to induced eddy currents at higher rpm. I have been going over this and I'm not sure I understand the cause. Here is what I imagine we are talking about:
So, from what I see, as the copper plate (red) moves through the magnetic field (blue), eddy currents will be induced (small black circles). As long as the thickness (T) of the plate is small enough, the currents should be minimal (same idea as iron laminations). If T is too small so that the resistance is too large you could stack multiple layers of copper insulated from each other, sort like really flat, wide litz wire.
I guess my main thought is the width (z dir in the image) of the copper plate doesn't matter, since the eddy currents are not induced in this dimension.
Does any of that make sense? Am I on crack? It has been a few years since my electricity and magnetism courses.
I have been learning alot from this thread, and I can't wait to see the results!
I just want to start by saying that I think this an awesome thread! I have keenly been watching it since it started.
From what i am reading, it sounds like the consensus is that copper foil lamination allow for good fill, but will cause increased losses due to induced eddy currents at higher rpm. I have been going over this and I'm not sure I understand the cause. Here is what I imagine we are talking about:
So, from what I see, as the copper plate (red) moves through the magnetic field (blue), eddy currents will be induced (small black circles). As long as the thickness (T) of the plate is small enough, the currents should be minimal (same idea as iron laminations). If T is too small so that the resistance is too large you could stack multiple layers of copper insulated from each other, sort like really flat, wide litz wire.
I guess my main thought is the width (z dir in the image) of the copper plate doesn't matter, since the eddy currents are not induced in this dimension.
Does any of that make sense? Am I on crack? It has been a few years since my electricity and magnetism courses.
I have been learning alot from this thread, and I can't wait to see the results!
perhaps a cursory review on rectangular waveguides might inspire some joy.
That's what I was thinking when I suggested it, but....matt_in_mtl said:
We haven't quantified the maximum reasonable thickness yet but, it looks as if it will be a lot thinner than if it were a core lamination. The thinner we go, the proportionately greater the insulation layer becomes....
Theoretically, it could have an advantage over (bundled round) Litz wire but I suspect that the difficulties of implementation will mean it's not worth it.
That's why I said:
Miles said:
So coils kind of like a roll-cake slice? Wide but very thin?
kenkad
100 W
Hello,
I would like to recommend this (http://focus.ti.com/lit/ml/slup197/slup197.pdf) for reading regarding eddy losses, skin effect, etc. It is not a simple problem. I believe ultimately that if the motor coil is removable, then, different types of windings (within reason) would need to be tried (cross sections of round wire, square wire, or foil and optimum number of turns) to determine what is the best solution based on current level, voltage, PWM frequency, etc. It also will depend on whether you are driving with a block wave or a sine drive (increased PWM frequency). This is my 2 cents.
kenkad
I would like to recommend this (http://focus.ti.com/lit/ml/slup197/slup197.pdf) for reading regarding eddy losses, skin effect, etc. It is not a simple problem. I believe ultimately that if the motor coil is removable, then, different types of windings (within reason) would need to be tried (cross sections of round wire, square wire, or foil and optimum number of turns) to determine what is the best solution based on current level, voltage, PWM frequency, etc. It also will depend on whether you are driving with a block wave or a sine drive (increased PWM frequency). This is my 2 cents.
kenkad
bigmoose
1 MW
Yes, eddy currents in complex geometries is a difficult thing to get the noggin' around!
As promised some resistances of materials:
So our copper windings are about an order of magnitude more conductive than cold rolled steel, and 25 times more conductive than SiFe transformer/motor laminations. Hence compared to a normal motor, if you do set up eddy currents in your windings... well, your toast, or more rightly, your motor will cook itself to death.
Some tips:
As promised some resistances of materials:
- Cold Rolled steel 15.9 microOhms per cubic centimeter
Silicon Steel 45 microOhms/cm**3
Copper 1.72 microOhms/cm**3
Aluminum 2.82 microOhms/cm**3
Carbon 3.8 microOhms/cm**3
So our copper windings are about an order of magnitude more conductive than cold rolled steel, and 25 times more conductive than SiFe transformer/motor laminations. Hence compared to a normal motor, if you do set up eddy currents in your windings... well, your toast, or more rightly, your motor will cook itself to death.
Some tips:
- Eddy currents always flow in planes perpendicular to the magnetic flux
Eddy currents always create a magnetic field that opposes the main field flux
Thud
1 MW
I'll continute with some content orented drift here,
Regarding Laminations & the fight with the puke awfull eddy currents,I assume it is important to insulate the lamination stack segments from each other.
The reason for the question is, I recently tore down a bunch of stators from a stash of old AC motors (nice 12 tooth ones) with the idea of repourpousing them "someday". It seemed there was no coating of any kind on these between the lams. counter intuitive from my recent research.
Thanks. T
Regarding Laminations & the fight with the puke awfull eddy currents,I assume it is important to insulate the lamination stack segments from each other.
The reason for the question is, I recently tore down a bunch of stators from a stash of old AC motors (nice 12 tooth ones) with the idea of repourpousing them "someday". It seemed there was no coating of any kind on these between the lams. counter intuitive from my recent research.
Thanks. T
bigmoose
1 MW
More on reducing eddy current losses, but this is for a conventional BLDC motor. Perhaps we can learn a bit by similarity.
Using knowledge of the fundamental principles that cause core losses, they can be reduced by:
Using knowledge of the fundamental principles that cause core losses, they can be reduced by:
- • Reducing the lamination thickness. Ideally, eddy current losses are directly proportional to the square of the lamination thickness. Therefore, if lamination thickness is reduced by a factor of two, eddy current losses decrease by a factor of four.
• Increasing the resistivity of the lamination material. Eddy current losses are directly proportional to material resistivity. Adding silicon to lamination steel is the most commonly adopted approach to increasing material resistivity.
• Annealing laminations after they have been stamped or cut. This eliminates the influence of mechanical stress on core loss.
• Reducing the amplitude of the magnetic field within the material. Hysteresis losses are directly proportional to the amplitude of the magnetic field raised to a power between 1.5 and 2.5. Eddy current losses are directly proportional to the square of the magnetic field amplitude. Using this property to reduce core loss is in direct conflict with maximizing torque production. As a result, other techniques for minimizing core losses are often implemented first.
• Reducing the number of magnet poles Nm. Hysteresis losses are directly proportional to the fundamental electrical frequency. Eddy current losses are directly proportional to the square of the fundamental electrical frequency. Since the fundamental electrical frequency is Nm/2 times greater than the motor shaft speed, reducing the magnet pole count allows one to reduce core losses significantly without lowering the motor shaft speed.
Thanks bigmoose.
I have a question...
In relation to an axial flux motor:
To minimise eddy currents induced in the core, the laminations need to run around the axis (like tree rings). To maximise the force generated, the copper needs to run radially (as much as is possible). For a coreless axial flux motor, is this an additional factor to be considered (when making comparisons with eddy currents in cores)?
I have a question...
In relation to an axial flux motor:
To minimise eddy currents induced in the core, the laminations need to run around the axis (like tree rings). To maximise the force generated, the copper needs to run radially (as much as is possible). For a coreless axial flux motor, is this an additional factor to be considered (when making comparisons with eddy currents in cores)?
bigmoose
1 MW
Well I guess my wifey would say I am addicted to learning... what can I say? I got a couple of quotes in today on magnetic design software for magnetic circuits... i.e > motors.
First was from Integrated Engineering Software:
2D Magneto think like 3/4 of a new Honda Civic
3D Amperes think like 7/5ths of a new Honda Civic
Ansys 3D and 2D... don't even ask! like a new Chevy Tahoe.
Vector Fields 2D Opera think like 2/3 to 3/4 of a new Honda Civic.
Think what would come of it if a dozen of us could have a copy for a year! ... alas, it isn't going to happen... Just too blasted expensive. You have to be in the business already to justify these costs. Here is a nice compilation of magnetic circuit software vendors http://www.arnoldmagnetics.com/mtc/fea_bea_software.htm
First was from Integrated Engineering Software:
2D Magneto think like 3/4 of a new Honda Civic
3D Amperes think like 7/5ths of a new Honda Civic
Ansys 3D and 2D... don't even ask! like a new Chevy Tahoe.
Vector Fields 2D Opera think like 2/3 to 3/4 of a new Honda Civic.
Think what would come of it if a dozen of us could have a copy for a year! ... alas, it isn't going to happen... Just too blasted expensive. You have to be in the business already to justify these costs. Here is a nice compilation of magnetic circuit software vendors http://www.arnoldmagnetics.com/mtc/fea_bea_software.htm
kenkad
100 W
Hello Bigmoose,
I knew that 7 months ago when I had the same thought so I am very glad you confirmed my conclusions. So, I had a even better idea! How about going to a major university and getting a graduate student (or advanced undergraduates) in the EE department (where you would think that they already have such software, right?) and sponsor this/these student/s to help with such a project. The short story is 'good luck'. I contacted three very large university EE departments (refuse to name them on this forum) and found out that 'no one' is interested in studying motors and even worse, there is no departmental staff expertise (probably read 75 resumes). So, undaunted, I contacted the 'US expert' in BLDC, a Dr Ramu, then at Virgina Tech (check out his books on the subject). He is no longer in the teaching profession (quit as of 1/1/2010) and is now in private business. He said almost all of the new technology research is being done in Europe and the Far East. He could not address any of my questions because the 'company' considers such information proprietary. The bottom line, a very very big dead end. I finally decided that I am too old to have to learn all of the capabilities of such software. So, I contacted MagNet and gave them a list of questions that I would want such software to answer. Their comment was that most of my questions have never been addressed in any simulation they have tried. The bottom bottom line is 'use your common sense'. Where have I gotten some really good information? Shane Coltons work. I have said this to Miles:
1. AFM dual rotor design
2. Maximize copper fill (foil copper)
3. Minimize core iron (decrease inductance)
4. Individual replaceable coils with properly shaped flux director ends
5. Increase the number of 3phase groups (surely more than 1 3phase group)
6. Increase the voltage and decrease the current per 3phase group
7. Modular design that allows differing output drive shaft configurations
8. Skew rotor magnet placement to strive for a bEMF sine waveform
9. Motor controller drive must be sine drive, not block or trap drive
10. Optical positional sensing (forget Hall sensing) for rotor/stator positioning.
There are probably more important points that these 10, but this is my opinion on the subject. It is good to see many different approaches being discussed on this forum. I am always eager to learn from what others are doing.
kenkad
I knew that 7 months ago when I had the same thought so I am very glad you confirmed my conclusions. So, I had a even better idea! How about going to a major university and getting a graduate student (or advanced undergraduates) in the EE department (where you would think that they already have such software, right?) and sponsor this/these student/s to help with such a project. The short story is 'good luck'. I contacted three very large university EE departments (refuse to name them on this forum) and found out that 'no one' is interested in studying motors and even worse, there is no departmental staff expertise (probably read 75 resumes). So, undaunted, I contacted the 'US expert' in BLDC, a Dr Ramu, then at Virgina Tech (check out his books on the subject). He is no longer in the teaching profession (quit as of 1/1/2010) and is now in private business. He said almost all of the new technology research is being done in Europe and the Far East. He could not address any of my questions because the 'company' considers such information proprietary. The bottom line, a very very big dead end. I finally decided that I am too old to have to learn all of the capabilities of such software. So, I contacted MagNet and gave them a list of questions that I would want such software to answer. Their comment was that most of my questions have never been addressed in any simulation they have tried. The bottom bottom line is 'use your common sense'. Where have I gotten some really good information? Shane Coltons work. I have said this to Miles:
1. AFM dual rotor design
2. Maximize copper fill (foil copper)
3. Minimize core iron (decrease inductance)
4. Individual replaceable coils with properly shaped flux director ends
5. Increase the number of 3phase groups (surely more than 1 3phase group)
6. Increase the voltage and decrease the current per 3phase group
7. Modular design that allows differing output drive shaft configurations
8. Skew rotor magnet placement to strive for a bEMF sine waveform
9. Motor controller drive must be sine drive, not block or trap drive
10. Optical positional sensing (forget Hall sensing) for rotor/stator positioning.
There are probably more important points that these 10, but this is my opinion on the subject. It is good to see many different approaches being discussed on this forum. I am always eager to learn from what others are doing.
kenkad
j3tch1u
10 kW
bigmoose said:
bad news, motor r&d is all but dead in taiwan as well according to a EE prof friend of mine.
good news, a quick search revealed most of that s/w is on bt. i certainly don't advocate piracy, but for self-educational, not-for-profit, evaluation purposes..
the guilt is soon met by relief that i didn't have to sell my car, use my kids tuition and take out a second mortgage on the house
