Feb 01, 2011, 01:09 PM
Jack
USA, ME, Ellsworth
Joined May 2008
18,026 Posts
Discussion
Tower Pro BM2410-12D LRK "Y" Rewind

Tower Pro BM2410-12D LRK "Y" Rewind

Background.

Winter in Maine, it has been very cold, flying days have been few, and I've been having plenty of time on my hands. After all, I am a retiree and my motto for life is "Seven Saturdays A Week!", I have the tee-shirt to prove it.

But anyway, nearly all of the motors I have bought were Distributed LRK or DLRK winds and most had Delta terminations.

As I read more and learned more about winding and other winds, the LRK wind became one I wanted to try. I thought that the LRK wind, with windings on only 6 of the 12 stator arms, would be a good way to get both larger wire and more turns into a wind, especially on a smaller stator. I say that because I've found that the physical interference or jamming together of wires on adjacent arms is a problem or at least an important consideration when winding motors.

The original specs on this Tower Pro BM2410-12 Delta motor were:

Kv 1000
11.2A Continuous
12A Maximum
86 Watts
props 9x4.7SF, 8060HD

Choosing a New Wind and Termination.

The motor was originally wound with three strands of 0,24mm or 31 AWG wire, it had 12 turns and a Delta termination. The pros and cons on the Delta and Star (also called "Y" or WYE) terminations are that, for the same number of turns and wire thickness,

- DELTA gives 1.73 higher power and amp draw than STAR

- DELTA Kv (RPM) = 1.73 higher than STAR Kv

- DELTA Kt (Torque) = 1.73 lower than STAR Kt

another way to look at the difference in the two is that, for the same number of turns:

- WYE has only .58 of the power and amp draw of a Delta

- WYE Kv (RPM) = .58 of the RPM of DELTA Kv

- WYE Kt (Torque) = .58 higher than DELTA Kt

The implication to me is that a WYE termination will produce lower RPM and less power. But WYE will also draw less current and give you more torque. The more torque may, at least to some extent, compensate for the drop in power.

I think the lower current draw with the WYE term may let you use smaller or lighter batteries and/or improve the flight duration over a comparable motor with a DELTA termination. In the end it will all come down to how you use the motor and if you are happy with the results.

Since most of us have a propeller size in mind and need a motor of a certain Kv, an important factor to remember is that the turn counts for DELTA and WYE terminated motors will not be anywhere near the same to attain a given Kv. If you have a motor that is already wound, changing it's termination from DELTA to WYE (or vice versa) will produce a motor that is very different as far as the props it will use.

The original motor had a Kv of 1,000, I wanted to have a Kv that was a little higher than that, so I wanted a higher Kv LRK wind with a WYE termination.

To figure out a turn count I entered the turn count and Kv data for the original wind in Manuel_v's Turn Calculator 5 spreadsheet and looked at the resulting Kv's for the LRK STAR winds. That showed that LRK STAR winds in the 10 to 12 turns range would produce Kv's from 1320 down to 1100 or so. The image shows spreadsheet and the all choices I had.

I did some test winds with wire I had and found that 6 turns of 23 AWG would fill one layer on the stator arm. And determined that winding 4 turns back over the first layer, for a total of 10 turns, would give me a nice fill. And it would also be a fairly easy wind as far as fitting the turns on.

The bottom line on the experimenting is that using smaller wire would allow more turns per layer and larger wire fewer turns and I wanted to find the wire size that would give me the desired turn count with the largest wire size.

I chose to do the LRK wind with 10 turns of 23 AWG and terminate it STAR (STAR is also called WYE or "Y") and that should give me a Kv of about 1,320.

More to follow...

Jack

# Images

Last edited by jackerbes; Feb 03, 2011 at 06:21 PM.
 Feb 01, 2011, 01:12 PM Jack USA, ME, Ellsworth Joined May 2008 18,026 Posts Comparing Surface Area and Density Comparing Surface Area and Density The surface area of the wire and density of the copper in the windings gets mentioned often when I read about motor rewinding. So I wanted to know what the difference would be in the wind I had chosen to do. As I understand it, more surface area in a wire means less resistance, more current flow, and lower heat in the windings. And increasing the density of the copper in a motor improves the power and efficiency. Improving copper density is one of the basic, if not the most basic, objectives in rewinding motors. To compare the surface area and copper density of this wind to the original wind I looked up the specs on the wire used in the original wind (3 strands of 31 AWG) in the wiki AWG table: original wind - 3 strands 0.24mm (31 AWG or 0.227mm) 31 AWG surface area = 0.0404 mm2, 3 x 0.0404mm2 = 0.121 mm2 The new wind with 23 AWG wire has a surface area of 0.258 mm2. Original Surface Area = 0.121 mm2 New wind Surface Area = 0.258 mm2 So that was a good thing, I would have about 113% more surface area than was in the original wire. So, even though I will have only six arms wound, I should be able to draw at least as much, and probably more, current than the original wind did. As far as copper density, the original windings were 29"/736 mm long. The new 23 AWG windings were later determined to be 24"/609 mm long. So I calculated the volume of copper in the two winds (all figures based on the dimensions on the wiki AWG page) and the results are: 3 strands 31 AWG 0.227 mm dia. x 736 mm long = 0.08936 cc x 12 arms = 1.07232 cc of copper 1 strand 23 AWG 0.573 mm dia. x 609 mm long = 0.15704 cc x 6 arms = 0.94224 cc of copper So there is about 14% less copper, by volume, in the 23 AWG LRK wind than was in the original wind. So there will be a 14% loss in the density of the copper in the windings. I'm not sure what the effect of that is going to be. I'm not a world reknowned mathematician so I don't guarantee all my numbers are right. More to follow... Jack
Feb 01, 2011, 01:25 PM
Jack
USA, ME, Ellsworth
Joined May 2008
18,026 Posts
Winding the Stator

Winding the Stator - 10 Turns LRK Wye

The stator was found to be uncoated, it had the plastic insulating caps on the ends of the stator arms that are commonly found on older Tower Pro stators. The stator diameter was 24mm, it was 10mm high, and had 20 laminations. The laminations were not coated or glued together, they could be rotated out of alignment when the windings and plastic caps were removed.

The stator caps were in good condition and I decided to reuse them. It was quick and easy to do the winds as per the image attached to this post. I think the credit for that image goes to Manuel_v, and I thank him for that.

The image does not show every turn and the actual arrangement of the turns. The reality is that instead of simply starting at E1 (Spanish for Start 1) and finishing at T1 (Spanish for End 1), the wind is actually started in the center of the wire (a 26" piece of wire was about right for this wind) and each arm is wound from the bottom of the "V" out to the hammerhead for 6 turns and then four more turns were wound back towards "V" in layer two.

The turn direction *has* to be right according to the figure or the motor will not run. If you wind it wrong and persist on trying to make it run, you can damage a ESC.

For a better description of the actual winding process you can look at timocharis's "Winding Distributed LRK Using The 24-gram Hexkit" thread. But remember, that thread is not a LRK wind so you want to use the image posted here to decide which arms and what directions to wind. The direction is important, the motor will not work if you don't have it right (I have proved that to myself a number of times).

I checked the winds for shorts to the stator as each arm was wound.

Doing the "Y" termination on a LRK wind is very simple. There are six wire ends, any three alternating ends connect to the three ESC connectors. The other three alternating winding ends are gathered together and soldered together to form the "Y" bundle. The end of the "Y" bundle needs to be insulated to protect the stripped and soldered part on the end from shorting out against the stator or bearing tube.

So E1, E2, and E3 went to the ESC connectors and T1, T2, and T3 formed the "Y" bundle.

I used a Winding Terminator from Strong Motors to make the transition from a single solid 23 AWG strand to 18 AWG super flexible stranded wire for for the ESC connectors.

For one last check I measured the resistance across the three phases, found they were all the same, and also made a final check for shorts to the stator.

The images show the finished motor, a close up of the winds (it just does not look like there is enough wire there, does it?), and a close up of the strands at the winding terminator. The green heat shrink tubing seen in the last image is the "Y" bundle or insulated ends of the three strands that are soldered together. It was simply bent down against the winding will stay there as the motor runs.

I had double and triple checked that I had put the windings on in the right direction (it is pretty easy to get one wrong, I've proven it to myself several times). And I had checked for shorts several times too. So I took a chance, connected the motor to an ESC and powered it up.

There are additional checks that can be taken to check your windings before you run a re-wound motor. If you wind a motor wrong you can destroy an ESC when you try to run the motor. As I said, in this case I decided to take a chance and simply try it. But if the motor did not start immediately and run smoothly I would not persist at all, I would quit immediately to save the ESC.

But a few clicks of throttle had the motor turning smoothly and nothing was getting warm or hot or smoking after several short duration and lower RPM test runs. Whoopee! I got lucky again!

I connected my Eagle Tree eLogger V3 and ran the motor at medium RPM for 10-15 seconds and checked for heat again. Then I ran it up to full throttle for a few seconds to get a no load RPM reading. I use a 3S A123 pack ( 10V or so) to do that as the lower voltage keeps the RPM down a little bit. When I graphed the data from that it showed:

RPM = 13,856
Voltage = 9.74V
Current = 1.86A
Kv = 1,423 (13,856 / 9.74 = 1,423)

That is called a "raw Kv", it is an approximate value but is close enough for this. I've read that the true Kv, one that is taken in a more accurate manner, would be a little bit higher.

The fourth image below shows the data from the no load testing. The fifth image is a close up of about three seconds of time with the motor at full throttle. The values seen there for average RPM, voltage, and current at what were used to get the raw Kv value.

A question: What happened to my estimated 1320 Kv winding?

I need to start counting my turns differently. I have been counting each strand that crosses the front of the stator (the end away from the terminations) as one turn. So my wind has 10 strands crossing there in one layer of six and a second layer of four. But my winds usually end up being of a little higher Kv than the spreadsheet predicted. And, since a higher Kv means fewer turns, I probably need to add one more turn to get the count right.

So from now on when I want to do a 10 turn wind I will look for 11 crossing strands. I think it has to do with the fact that the half turns on the start and end of each arm are not effective and should not be counted.

So I think I have actually have a 9 turn wind here. If you look at the image of the Turn Calculator 5 spreadsheet for this motor that I posted up above, it shows that a 9 turn LRK STAR wind would have a Kv of 1466. And if you consider that the 1,423 raw Kv I got is going to be a little bit lower than the real Kv would be, it looks almost certain that this is a 9 turn motor. And it certainly shows how accurate and useful that spreadsheet can be too. My Kv was only 3% different than the prediction. Thanks to Manuel_v!

But the motor is a keeper for now and I'll do some more prop testing with it.

More to follow...

Jack

# Images

Last edited by jackerbes; Apr 11, 2012 at 07:36 PM.
Feb 01, 2011, 01:32 PM
Jack
USA, ME, Ellsworth
Joined May 2008
18,026 Posts
Prop Testing...

Prop Testing...

I did my first test with a Great Planes 10 x 4.5 Slow Fly prop because it is a prop I use a lot. And it is strong and will handle being over revved should that happen. It is actually too large for this motor I watched RPM to keep the prop RPM down to safe limits. And the motor temp was monitored as the test proceeded too. I measure the temperature with a probe that is in contact with the windings. My normal limit on temperatures is to not exceed 130F/54C. That is a safe or conservative cut off temp, no damage will occur at that temp.

I use a 12V 7Ah sealed lead acid battery for testing. That simulates 3S LiPO voltage and has enough capacity to minimize voltage drops under load and can be used for a long time. And it saves wear and tear on my flight packs too.

With the eLogger V3 I can see all the readings on the LCD Power Panel as I test. I can monitor the RPM, temperature, Amps, and Volts as the motor is running.

I brought the motor up to about 3,700 RPM and the temperature started to rise from the 65F/18C ambient. And at that time the motor was drawing about 7A. I increased the RPM until the motor was pulling 12-13A and the temp (the magenta line) started rising quicker. After about 25 seconds at about 4,000 RPM, I closed the throttle. All of that can be seen in the first image.

Note that the temperature continued to increase after the motor was stopped, that is the cooling air flow is stopped and the temp sensor (in contact with the windings) is feeling more heat in the windings. But the temp inside the windings will always be higher than where the sensor is. In this case I think the windings got to or very close to my 130F/54C peak.

All of that happened at less than half throttle, I quickly realized that the higher Kv was not going to allow full throttle operation with that prop due to both RPM limits and heat in the windings. Both the RPM and temperature would be excessive at full throttle with this prop and motor.

I let the motor cool down to about 85F/29C and ran it up again two more times and the results were repeated as seen in the second image.

It looks like I had created a motor that would require very judicious use of the throttle and/or smaller props.

In it's original wind the 2410-12 motor was recommended for use with GWS 8060HD props so I tested the motor with that prop. I ramped the motor up to about 8,200 RPM in a one minute run and kept it there (about 3/4 throttle) for over 20 seconds. The temp was stabilizing at about 125F/52C and the averages of the readings for the last 20 seconds were:

RPM = 8,230
Amps = 10.2A
Volts = 11.1V
Watts = 113W
temp 125F/52C and stabilizing at that temp

So I found that encouraging as I use 8060 props on one of my planes. The test indicated to me that I was getting about 27 Watts more than the original wind and 1A less than the original wind's 11.2A max rating. So I can say the motor is better for having been rewound I think. But if I used this motor with that prop, I would not run the motor at full throttle. But the motor is a candidate for use with 2S batteries also that more testing will be needed to see what result that produces.

But I still had not run this motor at full throttle on the 3S equivalent test setup. I only got through 18 of the 26 "clicks" on my throttle stick (about 70% throttle) on the last test. The GWS HD 8060 is generally considered safe at up to 11,000 RPM or so (using the generally accepted 90,000 / Diameter in inches rule). But I didn't run it up to full throttle as I didn't want to do that in static testing and be in the same room with it.

More to follow...

Jack

# Images

Feb 01, 2011, 01:37 PM
Jack
USA, ME, Ellsworth
Joined May 2008
18,026 Posts
More Prop Testing...

More Prop Testing...

After sleeping on it for a night and thinking about it a little, I had a startling revelation. It dawned on me that I am trying to use this motor wrong. It is a higher Kv motor and I need to test it with smaller props! Duh!

I don't have any planes that use high Kv motor and small props but I went through my prop box and found a couple that looked like they were worth testing. The first was a GWS HD 5030 (I don't even know why I have it).

As soon as I fired that up, it was obvious that smaller props were going to be the key to using this motor at it's full power potential. The first image shows almost one minute of running time at half and then full throttle. During that run the peak readings were:

Peak RPM = 15,036
Peak Amps = 8.03A
Minimum Volts = 11.3V
Peak Watts = 91W

The temperature rise during the run only went from 67F/19C to 72F/22C during the run and there was very little temp rise after the motor was stopped. The windings were barely warm to the touch. That is the run seen in the first image

I had another prop that was about this size, a Master Airscrew 5.5 x 4.5, and tested that too as it was a little bit larger in both diameter and pitch. I ran that prop at partial and then full throttle for a little less than a minute and a half and the peak readings were:

Peak RPM = 13,869
Peak Amps = 11.0A
Minimum Volts = 11.5V
Peak Watts = 121W

So at the end of it all I'm certain I have better motor even though it is not suitable as a direct replacement for a 1,000 Kv motor. It will be considered a 1,500 Kv motor and and just needs to be propped right to be useful.

The next time I get over to the LHS I'll pick up a couple of 6" and 7" props and with this motor, those props, and a little more testing I'll be inspired to find a plane on my "foamies waiting to be built" list that needs this motor.

And I still have the 2S testing to do on this motor, I think it will work nicely with some of my larger props and my A123 3S packs as they are equivalent to 2S LiPO as far as pack voltages under load.

Now I need to do another one of these, pack a few more turns on it, and get that Kv back down a little...

Jack

# Images

 Feb 01, 2011, 03:27 PM Curiouser and curiouser Rochester, NY, USA Joined Oct 2005 2,066 Posts What a great job of clearly communicating your thought process to us - reads very easily. This is a keeper - I'm gonna join the thread; bookmark it; and copy it to a file on my server. At one point you said: "It looks like I had created a motor that would require very judicious use of the throttle and/or smaller props." I wonder if winding only half of the poles has caused you to lose some ability to dissipate heat? Also, I am curious as to what the final weight is. Thanks for sharing, Walt
Feb 01, 2011, 06:09 PM
Jack
USA, ME, Ellsworth
Joined May 2008
18,026 Posts
Quote:
 Originally Posted by wjbite What a great job of clearly communicating your thought process to us - reads very easily. This is a keeper - I'm gonna join the thread; bookmark it; and copy it to a file on my server. At one point you said: "It looks like I had created a motor that would require very judicious use of the throttle and/or smaller props." I wonder if winding only half of the poles has caused you to lose some ability to dissipate heat? Also, I am curious as to what the final weight is. Thanks for sharing, Walt
Thanks for the kind words, Walt.

The primary problem was that I had just not thought it through or learned enough about the motor yet. The smaller prop worked for the 3S pack, I'll test it some more with 2S packs too and the lower voltage will let the prop sizes go up some again.

Getting the source voltage and load in sync motor's ratings has always been the balancing act for using motors.

And I don't think it was a heat dissipation problem, I just put too much of a load on the motor. If I had pressed on or not been paying attention, I would have smoked the motor. It's that easy to do.

LRK wound motors have quite a history and it is just another way to do it I think. I spent a lot of time looking for an answer to the "which is better, LRK or DLRK" and if there is an clear answer I didn't find it. The discussion get very technical on stuff I didn't understand and had math in it that really made my head hurt. So when I found where one guy says:

"I'd say the differences are pretty small overall. Choose which one you want and go for it."

At that point I decided to just try a LRK wind. Maybe someone can post a simple (layman's terms please) explanation as to what the difference is between LRK and DLRK. Or maybe the pros and cons on using either one.

I do not see any clear drawbacks to a LRK wind yet. And it was easier to do on a small stator.

Jack
 Feb 02, 2011, 10:20 AM Registered User United States, WA, Seattle Joined Jan 2007 5,461 Posts That is an extremely useful thread. I've wondered about trying LRK for some of the small motors but have not yet. Think I vaguely remember reading in one thread that it's more useful on larger motors than small ones? I only deal with small motors currently so that may have put me off. I may have to give it a try. Thank you for the detailed work. fly
 Feb 02, 2011, 12:11 PM Jack USA, ME, Ellsworth Joined May 2008 18,026 Posts I played with it a little bit more this morning using my 3S A123 packs. Those start out at about 10.2V charged and rested a little but they pull down pretty quickly and stabilize at voltages that are similar to a 2S LiPO. And I already like this motor better, it is going to be a good motor for those packs. I still haven't looked at the data real close but it looks like 12-15A continuous is going to work well as a continuous or near continuous limit with 2S packs. That will put it up around 100W max or so. I would encourage anyone to try an LRK wind, I really can't see any disadvantages to them. I'm definitely going to do another one and get the Kv down around 1,000 or so. That would be 13 turns or so I think I can get that on there with the 23 AWG again. Jack
 Feb 03, 2011, 01:06 AM Registered User Mexico, BC, Mexicali Joined Aug 2004 5,651 Posts Jack. very good thread. Congratulations. turn on the calculator, possibly some of their errors in predicting that this LRK is calculated with a simple algorithm that the number of turns in half star is for the same KV LRK. I have never been tested with this type of winding, so I have no way to adjust. may be time to experience it and make adjustments to excel. Another reason may be: What is the original turns and Kv?. according to the manufacturer is 12 laps and 1000 kv. But perhaps there are differences. kv sure that is not exactly 1000. and probably some are not fully turns 12 in all teeth. The other option that you mention is the way of counting turns. which creates a lot of controversy, since one never finishes agree. Some manufacturers only count wires that are in a cavity or nuten. Axi as Scorpion or that when they score 16. usually one says 8 turn in each tooth. but in reality they are counting 16 wires in the cavity between two teeth. this criterion, it can be considerate that a turn is when the wire passes through the two cavities adjacent to the tooth. and no matter if you cross or not cross at the end of the lap. For me. the winding diagram have 4 complete turns. Manuel V. Last edited by manuel v; Feb 03, 2011 at 01:25 AM.
Feb 03, 2011, 02:43 AM
Plane and multirotor's - FPV
Moss, Norway
Joined Jun 2007
859 Posts
Quote:
Agree on the "easier on a small stator"
What i have picked up is that a motor with high magnetcowering 85-90 +% a LRK is less desierable and that with equal #wind's on each fhase a dLRK has a little less KV, all els equal.

I'm not that teknical/scientiesic in my aproch on winding, but i have winded all cises from a 5g upp to a 55-50 LRK and dLRK, whatever suits my needs

Kai
Feb 03, 2011, 08:05 PM
Jack
USA, ME, Ellsworth
Joined May 2008
18,026 Posts
Quote:
 Originally Posted by manuel v Jack. very good thread. Congratulations.
Thanks! I have learned a lot here in a short time and you have been a major contributor to what I've learned.

Quote:
 turn on the calculator, possibly some of their errors in predicting that this LRK is calculated with a simple algorithm that the number of turns in half star is for the same KV LRK.
I have the same thing happen when I did a Half Parallel DLRK and also another 2410-12. So I think if I just add one turn it will work. At least for that spreadsheet anyway.

Quote:
 I have never been tested with this type of winding, so I have no way to adjust. may be time to experience it and make adjustments to excel.
If it is working OK for everyone else it may just be something I am doing. If you think you'd like to test with this motor I'll send it to you.

Quote:
 Another reason may be: What is the original turns and Kv?. according to the manufacturer is 12 laps and 1000 kv. But perhaps there are differences. kv sure that is not exactly 1000. and probably some are not fully turns 12 in all teeth.
As I stripped it I counted the turns on several arms and came up with 12 turns. But I didn't test it before I stripped it because it had a short of something. But I see what you mean, there are several things that can make the Kv come out different.

Quote:
 The other option that you mention is the way of counting turns. which creates a lot of controversy, since one never finishes agree. Some manufacturers only count wires that are in a cavity or nuten. Axi as Scorpion or that when they score 16. usually one says 8 turn in each tooth. but in reality they are counting 16 wires in the cavity between two teeth. this criterion, it can be considerate that a turn is when the wire passes through the two cavities adjacent to the tooth. and no matter if you cross or not cross at the end of the lap. For me. the winding diagram have 4 complete turns.
So people don't all agree on how to count it? This is getting complicated.

I see your figure as having 1/2 turn at the start and 1/2 turn at the end and 3 full turns in the middle. And that is how you get a total of four turns, right?

I have another one I am going to do and I'll test it first and also count turns on every arm if I can. Sometimes if they are glued I can't get a count.

Jack
 Feb 04, 2011, 12:27 AM Registered User Mexico, BC, Mexicali Joined Aug 2004 5,651 Posts Jack I very much appreciate the offer. no need, because I have some motors pending winding. even so thank you very much. now I count the wires that pass through the cavity .. if you pass one on each side is a return. No matter if this is crossed or not crossed. I think the most important thing here is to maintain the same standard for all motors. independently as they do others. Possibly never standardize the way we count. And we discussed a lot. A good way to get a more realistic kv motor is rewind and then feed the Turn calculator with this data. Manuel V.
Feb 04, 2011, 05:28 AM
Will work for foam
United States, WI, Muskego
Joined Apr 2005
1,502 Posts
Jack - Kudos on a very well thought out post. I wish i took the time with each of my rewinds. There is always something I forget to test and have to reset my test stand.

I agree with Manuel, it's the number of wires in the cavities between teeth. When winding, I count the wires crossing the front (bell) side of the teeth. This goes for all but the last turn of each phase (all types) or when crossing between teeth in a pair (dLRK). I drew it out to convince myself and I crudely duplicated it below.

In my experience, there is no difference in kV between LRK and dLRK as long as the same number of winds per tooth (LRK) is the same as the number of winds per pair of teeth (dLRK). Winding LRK is nice change of pace, but be careful of how much wire stacks up on the bell side of the stator. You can run out of room with a tight tolerance motor.

I rewound one of these motors a couple of years ago for a Slow Stick. I used 14 turns per pair (dLRK) of 24 ga wire which ended up with motor in the 950 - 1000kV range. The biggest prop i run on it is a GWS 10x4.7EP. It draws about 12.5A on 3S and pulls the 21oz Slow Stick (flapperons and a big, heavy battery) straight up as long as you want. The motor does get a bit warm, but nothing that will hurt it at all.

Keep having fun with the rewinds and let us know how things work out.