Single Vs. Dual layer winds

[1BOHO]

Letting the neutral bundles protrude through the vent holes made for much more assembly clearance and shorter bundles. Note the two 6mm id washers use to space the bell so that the radial fan can clear the end turns.All the original bearings were replaced with ceramic hybrids.
Turnigy never intended to run dual 13 Ga. wire.

[1BOHO]

Same wind 12N8P 6 turns A-B-C-/A-B-C- but 2 strands of 1.22428mm vs one of 1.91516mm. Noted only one more turn can be done in 1.9. but in the 2 strands of 1.22428 it appears many more could be completed. At a head to head glance the multi strand wind is much more compact. It is known that it would be less prone to losses associated with skin effect.

[1BOHO]

From Rapid Tables charts

13 ga. is 5.1785 kcmils

2x 17ga. is 4.09 kcmils

So likely with wire from same manufacturer is also more flow able area with larger mono wire in this case but it appears from photos in real slot more use of actual space with smaller multi strand.It seems in 6 turns here another strand of 17 ga could easily be added for 3x's and increased the cross sectional area of in kilo circular mills to 6.1382.

[1BOHO]

More photos.

[1BOHO]

IEEE TRANSACTIONS ON ENERGY CONVERSION,VOL.21,NO.1,MARCH 200695

Comparison of PM Brushless Motors,Having Either All Teeth or AlternateTeethWound

DahamanIshak,Z.Q.Zhu,SeniorMember,IEEE,and David Howe

One of the reference papers used in Gerling's findings...

It seems Dahamanlshak, Z.Q. Zhu and David Howe suggest through finite element predictions and real practice a few conclusive comparisons of DL and SL motor winds.

*SL's winds have more trapazoidal BEMF that DL's.
*SL's winds have more self inductance than DL's.
* SL's showed less cogging torque when driven by zero cross algorithms(BLDC) than DL's which would be lower if driven in BLAC.
*Static and output torque output is higher with SL's
*SL's have higher torque density.

And lastly not reported in this paper but should be noted SL's have more copper fill potential.

The BEMF measurements taken on an oscilloscope played a major role in the real time confirmation of theory in this Dajaku Gerling reference.

Paraphrased conclusion:

The winding factor significantly effects the BEMF waveform as well as the self and mutual inductance's while the motors exhibit low cogging because of the fractional number of slots per pole.

[1BOHO]

This idea is further confirmed in figures 10, 11, and 12, of Gerling's

Eddy Current Loss Minimization in Rotor Magnets of PM Machines using
High-Efficiency 12-teeth/10-poles Winding Topology

Really amazing the significant differences in stator loss and magnet loss between the two configurations in fig. 11 and 12 that continues even with the new stator core structures in fig. 8

Fig 3 may explain the 12n10p Gerling wind better for some than what has been offered thus far. I'm still trying to grasp it before an attempt with no copper to waste.

The SL and DL stator core structures are very also prototype worthy Id think.

[Baedarlboo]

This has been a very cool read...although a lot of it is over my head. Thanks for sharing and the updates.

[1BOHO]

Thanks. I have wound a single layer 5020 24 turns parallel wye. 1.23mm it added only addded,6 grams,more copper. I'm wondering if a nice F3A or pattern can load it in the,field? No chopper and definitely can't afford one 800 class.

[1BOHO]

Thanks. I have wound a single layer 5020 24 turns parallel wye in 1.23mm it added only 6 grams more copper. I'm wondering if a nice F3A or pattern airplane would be a way to load it in the,field No chopper and definitely can't afford one 800 class. Maybe a beta tester is needed. To go along with the with a load bank I also plan to build a thrust rig with a series of propellers to make even more motor comparisons.My rc boats load really well but all race props run vastly different rpms than the heli setup motors.

[1BOHO]

Photo of alternate tooth 5020. It seem like helis probably experience heavy radial loading on the bearings. I was curious and took a peep inside factory bearings and wonder about testing full compliment deep groove radial bearings with ceramic balls free of the retainer. Based on one of my ME friends previous bearing work I think it may be possible to rebuild a standard steel bearing with retainer to full compliment or into a hybrid.

[1BOHO]

Hi to all interested,
Here is the beginning of my prototype stator with relevant slot and dimensional numbers help from CL of LRK. It seems through investigative reading of several thesis concerning the subject axial pole pairing,special windings, alternating teeth, magnet segmenting, multi stranding, and use of flux barriers may all work together to drastically reduce the stator/magnet loss and THD produced by a FSCW motor. My other curiousity is the 4 pole 12 slot lapped wound motor with a KW of 1. It seems in an 8 pole motor if 4 magnets are removed and re glued in the correct place a highest torque producer segmented 4 pole'r could be created. Then to wind it....

[1BOHO]

SL 5020 Bemf sine and spectrum with main at 173.6hz.

If I believe the drill was 2500 rpm and the scopes vrms at 4.87 then the kv is 382.Looks like 30 turns or more might fit Should be interesting to compare drill method to binary pickup of QTI or the quasi TTL frequency of an optical break tach.

[powercroco]

old experiment from 2007 without the promissed win:




with more (3 instead of 2 magnets) it worked better.
cogging is unbelievable strong. hard work for esc to start it.
al lot of these experiment have been done, before sorting it out as practically useless.
12N4P works very well in inrunner, but not efficient in outrunner.

[1BOHO]

Beautiful wind but the magnets are not in the right place for my experiment. As shown those poles will not cover 3 teeth. I heard about incredible cogging from CL.

[powercroco]

3 magnet version:



we also did 4 magnets (almost 100% coverage), but I can't find back the photo.
the result was also not good.

[1BOHO]

That's very beautiful. You should start a thread showing how to lap wind and inrunners. That would be useful. Patience will yield my actual attempts and magnet topology here in my experiments. I will post it when I arrive as it is none of the above. Difficult to glue all those I'm sure.

Hows the Gerling coming along?

[powercroco]

I think, you will try the flyware variation of magnetic system with spaces between the single stripes.
also this we just did.
success?

halbach?
or something really new?

hm, I will wait for your findings about it now.

[1BOHO]

Nice guess but no cigar. Ill share it when I arrive. It is not the flyware topology in the photo I posted a while back.. Not a good practice just to copy from a photo don't ever do that. Really no surprise to your test as the stators are not optimized for distributed winds with large mono wire. FSCW motors were introduced partly as a way of getting rid of the torque ripple associated with distributed winds as I understand.

Make sure to post your Gerling and lap windings here on helifreaks.

Back to testing for now thanks for your help Ralph.

[powercroco]

phase A,
A and B and
A; B and C

[1BOHO]

Nice. Can I see the backside terminations?