Discussion LRK (SPS) Wind

 #1 Skylar Aug 10, 2011 03:40 PM

LRK (SPS) Wind

Looking at an old thread, I realized that I agree with James, who, against all odds, got his motor working.

To summarize: James had an 18-tooth 24-pole motor and insisted that he saw a similar motor run with an SPS winding. After warnings from everyone, including Christian Lucas, he followed his nose and wound it AaBbCcAaBbCc and he said that the motor ran very well - "a lot better than ABC".

I think it should work, because it's very similar to the 18N12P configuration, just double the magnet poles. Look at this table. It shows AaBbCcAaBbCcAaBbCc and also the SPS (LRK) winding A-B-C-A-B-C-A-B-C for that configuration.

BTW. SPS is short for Split Phase Sector, also known as LRK.

Christo

 #2 z-matrix Aug 11, 2011 07:03 AM

Quote:
 Originally Posted by Skylar (Post 19011267) Looking at an old thread, I realized that I agree with James, who, against all odds, got his motor working. To summarize: James had an 18-tooth 24-pole motor and insisted that he saw a similar motor run with an SPS winding. After warnings from everyone, including Christian Lucas, he followed his nose and wound it AaBbCcAaBbCc and he said that the motor ran very well - "a lot better than ABC". I think it should work, because it's very similar to the 18N12P configuration, just double the magnet poles. Look at this table. It shows AaBbCcAaBbCcAaBbCc and also the SPS (LRK) winding A-B-C-A-B-C-A-B-C for that configuration. Christo
What makes you think it runs: "a lot better than ABC"?

That configuration is an exact equivalent of the 12N16P by simple division.

Z

 #3 Skylar Aug 11, 2011 05:11 PM

I can't comment on how well it runs, because I haven't seen it run. All I'm saying is that I'm impressed that James persevered while everybody advised him not to try a wind different from ABC.

Yes, it's similar to 12N16P (which is also ABC), but if you look at 18N12P in Dr Okon's table, you'll notice that it can also be wound LRK.

BTW, I have a 12N16P motor and I just love the sound it makes.

Christo

 #4 Fourdan Aug 13, 2011 11:04 AM

Quote:
 Originally Posted by Skylar (Post 19011267) Looking at an old thread, I realized that I agree with James, who, against all odds, got his motor working. To summarize: James had an 18-tooth 24-pole motor and insisted that he saw a similar motor run with an SPS winding. After warnings from everyone, including Christian Lucas, he followed his nose and wound it AaBbCcAaBbCc and he said that the motor ran very well - "a lot better than ABC". I think it should work, because it's very similar to the 18N12P configuration, just double the magnet poles. Look at this table. It shows AaBbCcAaBbCcAaBbCc and also the SPS (LRK) winding A-B-C-A-B-C-A-B-C for that configuration. Christo
Hi Christo
Following my brute force personal simulations
For 18 slots 18 coils 24 poles
1) Config (ABC)x6 works, suppose Kv = 1000.0 rpm/V for some turns/slot
2) Config (AaBbCc)x3 works, BUT Kv = 1194.6 rpm/V for the same turns/slot

Conclusion 1) Config (ABC)x6 is better than 2)
(Lower Kv for the same Rm is always better)
Louis

 #5 Skylar Aug 17, 2011 01:49 AM

Hi Louis
Thanks for your input. It's interesting that ABC windings give lower Kv. Have you tested this in practice?
Recently, I became more interested in LRK winding again - for the simple reason of "easy of winding", which is very attractive with high number of slot motors/generators.
Following the usual "LRK-rules":
k=1,2,3,4,...
Bewickelte-Nuten = 6 * k
Statornuten (alle) = 12 * k
Magneten-Polpaare = 6 * k +/- 1
The following windings are possible:
K=1 (12-6-14) : A-b-C-a-B-c
K=2 (24-12-26) : A-A-b-b-C-C-a-a-B-B-c-c
K=3 (36-18-38) : A-A-A-b-b-b-C-C-C-a-a-a-B-B-B-c-c-c
In your opinion, would these LRK examples also have higher Kv?

And lastly: When looking at this table by Ralf Okon, one can see he also has a LRK winding for 18-9-12, which is A-B-C-A-B-C-A-B-C. I wonder how he arrived at that particular winding scheme.

Regards
Christo

 #6 Fourdan Aug 17, 2011 03:03 AM

Quote:
 Originally Posted by Skylar (Post 19066919) Hi Louis Thanks for your input. It's interesting that ABC windings give lower Kv. Have you tested this in practice? Recently, I became more interested in LRK winding again - for the simple reason of "easy of winding", which is very attractive with high number of slot motors/generators. Following the usual "LRK-rules":k=1,2,3,4,... Bewickelte-Nuten = 6 * k Statornuten (alle) = 12 * k Magneten-Polpaare = 6 * k +/- 1The following windings are possible:K=1 (12-6-14) : A-b-C-a-B-c K=2 (24-12-26) : A-A-b-b-C-C-a-a-B-B-c-c K=3 (36-18-38) : A-A-A-b-b-b-C-C-C-a-a-a-B-B-B-c-c-cIn your opinion, would these LRK examples also have higher Kv? And lastly: When looking at this table by Ralf Okon, one can see he also has a LRK winding for 18-9-12, which is A-B-C-A-B-C-A-B-C. I wonder how he arrived at that particular winding scheme. Regards Christo
Hi Christo
My figures are coming from my personal software:
Finite elements simulating the Lorentz forces around 360 deg
** distribution of conductors around a stator
** distribution of magnetic field around a rotor
The result is given in torque per ampere (reverse of Kv rpm/V) with a scaling coeff. I can see the amplitude and phase relationship of torque.
(when exciting A or B or C phases)

When you compare different scheme you have to take into account for a given kit
** the possibility to wind a given number of conductors per slot
If you compare 12 slots versus 24 slots for the same stator diameter, you cannot have the same area per slot and there is a constraint on the number of conductors
** regarding the number of poles (or number of magnets) you have to take into account the constant coverage in %.
If you compare 10 magnets versus 14 magnets (LRK or dLRK) you have to know if you compare the same "volume" (or coverage) OR the same dimensions of every magnet. (results are different)

In my original post #4, I was comparing two schemes exactly with the same nb of slots and poles (so no ambiguity).

How do you want that I simulate your 3 proposed configs ?
Same copper available volume + same magnet volume for the SAME Rm ?
(that is same length of given copper wire per phase) then comparing Kv ?
I don't know if I am clear.
Louis

 #7 Fourdan Aug 17, 2011 10:09 AM

Quote:
 Originally Posted by Skylar (Post 19066919) Hi Louis ... And lastly: When looking at this table by Ralf Okon, one can see he also has a LRK winding for 18-9-12, which is A-B-C-A-B-C-A-B-C. I wonder how he arrived at that particular winding scheme. Regards Christo
Hi Christo
Here under are 5 configs for 18S-9C-?P (18-9- --> every two)
a) 18-9-10P A-a-A-B-b-B-C-c-C-
b) 18-9-12P (A-B-C-)x3
c) 18-9-14P A-B-a-C-A-c-B-C-b-
d) 18-9-16P A-A-b-C-C-a-B-B-c-
e) 18-9-20P A-A-b-C-C-a-B-B-c-
For the same number of turns and same volume-coverage of magnets
The best(optimum) is b) config
a) and c) are lower & close together
d) and e) are much lower, {e) the badest)}

Every config is not equivalent to make copper-magnets to "work & cooperate efficiently"

Louis

 #8 Skylar Aug 17, 2011 05:42 PM

Hi Louis

To answer your question in post #6, I mean different winding schemes with the same magnet coverage and same number of windings.

Actually, what I really would like to know is this: If a motor layout is suitable for both dLRK and LRK winding, e.g. 18N24P or 24N26P, then which scheme will give the lowest Kv, dLRK or LRK?

I'm busy with a project that has 24N26P and I'm looking for the lowest Kv when comparing dLRK and LRK. For ease of winding, I'd prefer LRK, unless the Kv would be much higher than dLRK with the same number of turns. I intend to wind A-A-b-b-C-C-a-a-B-B-c-c-.

Christo

 #9 Fourdan Aug 18, 2011 03:11 AM

Quote:
 Originally Posted by Skylar (Post 19072491) Hi Louis To answer your question in post #6, I mean different winding schemes with the same magnet coverage and same number of windings. Actually, what I really would like to know is this: If a motor layout is suitable for both dLRK and LRK winding, e.g. 18N24P or 24N26P, then which scheme will give the lowest Kv, dLRK or LRK? I'm busy with a project that has 24N26P and I'm looking for the lowest Kv when comparing dLRK and LRK. For ease of winding, I'd prefer LRK, unless the Kv would be much higher than dLRK with the same number of turns. I intend to wind A-A-b-b-C-C-a-a-B-B-c-c-. Your opinion would be appreciated. Christo
Hi Christo
For 24N-12C-26P A-A-b-b-C-C-a-a-B-B-c-c-
Torque/A = 1014.9
For 24N-24C-26P AaAabBbBCcCcaAaABbBbcCcC
Torque/A = 1007.2
Difference is very very small (brute force simulation)
note : for the same number of wires per slot (or same turns per pair of teeth)

For windings the useless heads of copper wires could be slightly different
My suggestion := dLRK with 24coils (every tooth)
Louis

 #10 Ron van Sommeren Aug 18, 2011 02:04 PM

/* Moved */

 #11 Skylar Aug 18, 2011 02:07 PM

Quote:
 Originally Posted by Fourdan (Post 19076112) Hi Christo For 24N-12C-26P A-A-b-b-C-C-a-a-B-B-c-c- Torque/A = 1014.9 For 24N-24C-26P AaAabBbBCcCcaAaABbBbcCcC Torque/A = 1007.2 Difference is very very small (brute force simulation) note : for the same number of wires per slot (or same turns per pair of teeth) For windings the useless heads of copper wires could be slightly different My suggestion := dLRK with 24coils (every tooth) Louis
Thank you Louis.
Since this is a prototype, I might try both dLRK and LRK and compare.
Christo

 #12 Ron van Sommeren Aug 18, 2011 02:15 PM

Christo, for the original LRK motors Lucas Retzbach found that dLRK gave 10% higher Kv for same number of winds (wires per slot).

Have you seen this new winding diagram 'calculator' for larger numbers of magnet/statorpoles?
http://www.rcgroups.com/forums/showt...ght=calculator

Vriendelijke groeten ;) Ron
diy motor tipsDrive Calculator
diy motor groupCumulus MFC

 #13 Fourdan Aug 18, 2011 02:27 PM

Ron
I measured 3% .. 3.5% more for 12N dLRK (my simulation gives +3.5%)
For 24N it is slightly less
Louis

 #14 z-matrix Aug 19, 2011 05:01 PM

Quote:
 Originally Posted by Fourdan (Post 19069196) Hi Christo Here under are 5 configs for 18S-9C-?P (18-9- --> every two) a) 18-9-10P A-a-A-B-b-B-C-c-C- b) 18-9-12P (A-B-C-)x3 c) 18-9-14P A-B-a-C-A-c-B-C-b- d) 18-9-16P A-A-b-C-C-a-B-B-c- e) 18-9-20P A-A-b-C-C-a-B-B-c- For the same number of turns and same volume-coverage of magnets The best(optimum) is b) config a) and c) are lower & close together d) and e) are much lower Every config is not equivalent to make copper-magnets to "work & cooperate efficiently" Louis
Louis,

So winding a 12N16P motor as dLRK will result in a 10% increased Kv, reduced, but evenly distributed torque and less noise/vibration?

Z

 #15 Fourdan Aug 19, 2011 05:11 PM

Quote:
 Originally Posted by z-matrix (Post 19090899) Louis, So winding a 12N16P motor as dLRK will result in a 10% increased Kv, reduced, but evenly distributed torque and less noise/vibration? Z
Hi Z-Matrix
12N-12C-16P config is not called dLRK (which is 12N-12C -10P or -14P)
A lot of "classical" configs are evenly distributed
Noise and vibrations could come from imperfect windings of perfect configs
Only some exotic configs are unbalanced (not evenly distributed)
Louis

All times are GMT -5. The time now is 12:28 AM.