Hi!

I have been given a nice Scorpion HK-5020-450KV.

Unfortunately i got it without any windings on, they have been taken off because it burned out. The rest of the motor and mechanics are just fine

I am planning on rewinding this motor for use on my bicycle. The idea was to make a drive unit attached to the front wheel using the "Velo Solex" principle where a spring actuated roller drives the front tyre. In this case i need as high a KV wind as possible.

I have rewound motors before, but nothing as big as this, and nothing with this much power.

I have been given the old burnt windings, and the magnet wire is approx 0.35mm and the wind is made of 16 or 17 strands of magnet wire.

I have no idea if it was wound in LRK or DLRK.

Now i just need to know, what do i do to make a wind that makes the most KV ?

It has 10 magnets and 12 stator poles.

If anyone have any advice for this wind, please chime in

Thanks,

// Per.

Originally Posted by Zapro

Hi!

I have been given a nice Scorpion HK-5020-450KV.

Unfortunately i got it without any windings on, they have been taken off because it burned out. The rest of the motor and mechanics are just fine

I am planning on rewinding this motor for use on my bicycle. The idea was to make a drive unit attached to the front wheel using the "Velo Solex" principle where a spring actuated roller drives the front tyre. In this case i need as high a KV wind as possible.

I have rewound motors before, but nothing as big as this, and nothing with this much power.

I have been given the old burnt windings, and the magnet wire is approx 0.35mm and the wind is made of 16 or 17 strands of magnet wire.

I have no idea if it was wound in LRK or DLRK.

LRK would only have six of the 12 arms wound, that is easy to spot. Can you look at the old windings and figure out the number of turns that were around one arm? That will be very useful info for rewinding.

If if was DLRK it would have had two arrms side by side wound and than a transit run across the stator to two more side by side arms. Four arms wound for each phase.

If you can determine the length of one of the wires you have from where it started to where it went to the next arm by the bends in it, that will get you close on the number of turns.

Now i just need to know, what do i do to make a wind that makes the most KV ?

It has 10 magnets and 12 stator poles.

If anyone have any advice for this wind, please chime in

One turn would give you the best Kv but that is not what you want. You want something more logical than that. Do you know the diameter of the drive roller? Do you know the diameter of the tire? Do you know what a realistic expectation of the top speed would be?

For a 12 arm 10 pole motor you can do a LRK or DLRK wind, you can see the practical winding schemes here:

http://www.powerditto.de/Kombinationstabelle.html

and get images of them by entering the arms and poles here:

http://www.powerditto.de/bewicklungsrechner.html

You can use the drive roller and tire diameter to tell the rate of rotation of the motor and tire and the speed you can expect.

Jack

Thanks. I have looked up the data on the motor on Scorpion site, and it says this:

Stator Diameter	50 mm (1.56 in)
Stator Thickness	20 mm (0.787 in)
No. of Stator Arms	12
Magnet Poles	10
Motor Wind	15 Turn Delta
Motor Wire	16-Strand 0.33 mm
Motor Kv	450KV RPM / Volt
No-Load Current (Io/10V)	1.21 Amps
Motor Resistance (Rm)	0.018 Ohms
Max Continuous Current	85 Amps
Max Continuous Power	3770 Watts
Weight	468 Grams (16.51 oz)
Outside Diameter	61.0 mm (2.40 in)
Shaft Diameter	5.98 mm (0.24 in)
Body Length	51.0 mm (2.0 in)
Overall Shaft Length	96.50 mm (3.80 in)
Max Lipo Cell	12s
Peak Current	120A (5 seconds)
Peal Power	5320 Watts (5 seconds)
Motor Timing	5deg
Drive Frequency	8kHz

So it's 15 turns with 16 strands of wire.

I can borrow a new one of the exact same motor and see if i can figure out the winding scheme by looking at it.

This motor exsists in a 450 and 520 KV variant. I have not yet made the drive roller for the motor, so i have nothing to calculate on. Speed on the bike would be maximum 30 Km/h so nothing spectacularly fast.

The worst problem i am facing right now, is removing the stator from the bearing tube assembly. I knocked out the bearings, heated the bearing tube untill i could smell the epoxy and tried pulling the stator off - no go. I then made a jig that supported the stator and tried hitting the center of the bearing tube with a 12mm socket driver which just fits into the center, and i managed to knock maybe 5 sheets of the stator off, but the rest won't budge.

The stator sheets are not covered by any green epoxy, so it should be fairly safe to heat them. Later today i will try to heat it even more untill the epoxy starts smoking a bit, maybe that can break the bond.

Twisting off the stator is not a solution, since there is a counter-rotating pin installed, and it sits deep, i cannot reach it to pull it out

// Per.

It was probably DLRK Delta then if all 12 arms are wound and it has 10 magnets.

You can use a turn calculator like Manuel_V's Turn Calculator 5 or the Simple Turn Calculator and it will show you the turn counts you can consider.

The Scorpions are tough to work on, they have a snap ring on the top of the stator that has to be removed before the stator will come off. And then you do the heating thing. You can see the details on some of the Scorpion kits here:

http://www.gobrushless.com/shop/inde...brushless+kits

Look for the assembly instructions and stuff there.

Also, a search of this forum using Advanced search, setting an Any Date range, and looking for Scorpion in the thread title will bring up some info on rewinding them.

The issue you will run into is that that you want to use fewer and larger wires to rewiind with and as the wires get larger the work gets much harder. That German web site (powerditto or powercroco) has a lot of photos of some of the best work done on the big Scorpions.

To equal or increase the current handling capacity of the motor you want to have at least as much or more copper in the windings as is there now. If you look at the wiki AWG table you can calculate the surface area of the 167 strands of 0.35mm wire and then look at the larger wire gauges to see might work to replace the windings with an equal amount of wire or more wire.

If you go to the Winding Calculator and enter 12 slots and 10 magnetpoles, choose D, and it will show you a figure for the winding.

An option for you would be to use the LRK wind. That wind, using only 6 of the 12 arms, will have more room for wire on each arm. To see that one go to the Advanced mode, enter A-b-C-a-B-c- for the winding scheme, 10 poles, and D, and it will show you the LRK winding scheme. I copied the A-b-C-a-B-c- scheme from the 12N10P section of the Winding Scheme table.

Is your head starting to hurt a little? If so, welcome to motor rewinding!

Jack

this scorpion 5020 might be a little bit too small to drive a bicycle, and if the motor is running at high speed with gear system, the noise may be huge. But anyway, with good gear system and proper winding/Kv, it may worth a shot.
For bicycle use, you are looking at long time continuous running of the motor, as well as possible damage from the dust on the road. it maybe a good idea that you solve these practical problems first: gearing system, motor heat dispersion and dust protection, before you move on to your project.

I do not know why you want higher Kv on this motor. The possible actual rotation speed of the bicycle wheel maybe, let's say, around 3000 rpm. Most electric bicycles use specially designed wheel motor to directly drive the wheel, which is the most efficient way because it does not use a gear system. The bigger the difference between the speed of the motor and the output speed of the gear system, the more energy is lost on the gearing system. So if i were you, i may first try to find a proper gearing system with appropriate ratio and is strong enough for bicycle use. The based on the desired speed of the wheel and the ratio of the gear system, as well as the battery (voltage), you may calculate the desired Kv of the motor. Then like Jack posted, you will find a proper winding scheme for that Kv.

It is better to use high-temperature single thick wire winding for better copper coverage and cooling of the coils, and you need to choose the wire based on the number of turns as well. For example, with 1.5mm wire, you may be able to wind 6 or 7 turns on each of the teeth; while with 1.1mm wire, you maybe able to wind 10 turns on each of the teeth.

again, scorpion motor are of the best quality, but it is still designed for RC use: short operation time, good cooling, better working environment (no need to consider dust damaging the motor, and etc. Scorpion has released a 7050 sized motor, which is a very good candidate for bicycle use.

You will also need to find a proper ESC, and the ESC also require extra cooling to make sure it will stay cool during long time operation.

I am not trying to discourage you, but there are a lot of thing that you need to think over before you start winding the motor.

Originally Posted by jackerbes

The Scorpions are tough to work on, they have a snap ring on the top of the stator that has to be removed before the stator will come off. And then you do the heating thing.

Nope - Not on this motor, there is just a counter rotating pin, no snap ring. I got the stator off in 5 pieces today, heated the assembly using a big blow torch untill i choked on the smoke - let it cool, and threw it into the freezer to shrink the bearing tube - the stator parts came of without too much problems. The stator and bearing tube are quite discolored by my torching, but that doesn't matter for me.

Originally Posted by jackerbes

The issue you will run into is that that you want to use fewer and larger wires to rewiind with and as the wires get larger the work gets much harder. That German web site (powerditto or powercroco) has a lot of photos of some of the best work done on the big Scorpions.

To equal or increase the current handling capacity of the motor you want to have at least as much or more copper in the windings as is there now. If you look at the wiki AWG table you can calculate the surface area of the 167 strands of 0.35mm wire and then look at the larger wire gauges to see might work to replace the windings with an equal amount of wire or more wire.
Jack

Interesting! - I thought that it was better to use more strands of thinner wire to fill the stator better, since the thinner wires can be packed tighter, but you'll learn every day :-)

Originally Posted by jackerbes

Is your head starting to hurt a little? If so, welcome to motor rewinding!
Jack

Yeah, it's not as simple as i first thought. I will have a look on what kind of drive roller i can make on the lathe and try to do some calculations of what KV i need.

You may think this motor is too small for this project, but my expectations isn't as big as you may think. This project is for fun, and it should for the most part be free, so go buing that $1000 Scorpion motor isn't an option (or for that sake, the monster Turnigy 80-100) - I even use a frankenstein ESC i have built with parts from two broken ones ;-)

In Denmark where i live, there is a restriction on E-bikes on 250 Watts and the motor must only be able to power up when the rider is pedaling. These laws are ridiculous!. So 250W or more for this "for fun" project is fine with me.
Cooling of the motor should not be a problem, i can always figure something out.

// Per.

"..Interesting! - I thought that it was better to use more strands of thinner wire to fill the stator better, since the thinner wires can be packed tighter, but you'll learn every day :-)..."

It is interesting and does seem contradictory. I think we wind up getting the turn on in perfect layers with the turns side by side and not crossing and that contributes more than any small advantage there may be in the smaller strands forming a denser bundle.

I have gotten increases in the surface area that were as much as 40-50% of the original. The DLRK wind, with each phase being on four arms, is a long piece of wire. Sometimes if we do LRK or Half Parallel DLRK wind each wire is shorter and that drop the resistance in each run a litle and that helps too.

Interesting details on the bike, I don't know much about them. What is the size of the tires on the bike? It would be interesting to see what the rotational speed is to go 30 kmh. And that could lead you to getting a practical Kv too.

Jack

If you are trying to save money, i think the best way is to sell the scorpion to fund your project. You can easily find much cheaper stuff with the money you can get from the scorpion. I personally, is willing to pay for your kit for 100 bucks. I guess it will not be very difficult to sell it for this amount of money.

Most of the e-bike has actually very low power consumption, comparing with RC world. In my country, the most powerful e-bike that is legal will be somewhere around 500 watts, or speed is limited below 30 mph (normally 25 mph).

I think your major concern is how you can "convert" the power of the motor to the wheels of your e-bike. you will also need to understand the amount of energy lost in the "conversion" system.

For e-bike use, they have specially designed speed controller with many functions that is required for an e-bike, for safety reasons. For example, soft braking, locking, and smooth and mild speed-up/down adjustment. It maybe very difficult or unsafe to use an RC esc to control the e-bike.

Anyway, it will be fun for sure, during the assembling process. But be safe during the "bench test" on the road, my friend.

By the way, i just got a cheaper motor, which is a Chinese copy of AXI 4120. I finished rewinding it this afternoon, and here is a thread i will be posting regarding the rewinding process. I hope my experience can give you some help. To note that the motor is 4120 size, 12N14P, which has the same winding scheme as 12N10P, both dLRK or LRK.

https://www.rcgroups.com/forums/show...6#post20719805

Hi guys, Sorry for bumping this rather old thread, but as i see no purpose on making a new one, since i'm still going to rewind the same motor, i just have the time now.
Just got some 0.14 Nomex for insulation, and i'm trying to find out what to do with the motor.

I want to end up with roughly 250 KV on the motor. The original wind was 15 turns for a KV for 450 - can i just calculate from that how many turns i need to put on, or is the ratio not 1:1 ?

Modisc already linked to winding diagrams, so i think i have that covered, just need to know approx how many turns i need to fit.

Any thoughts on this ?

Regards,
Per.

winds_new = winds_old * Kv_old / Kv_new

15 turns = 450 Kv

15 * 450 / 250 = 27

That only works when you are doing the same wind and termination of course.

Because you are almost doubling the turn count you might want to consider doing an LRK wind. With that you only wind six of the 12 arms so it takes the adjacent arm interference out of the equation. Not sure if you can use a wire size that is larger than the original wire but if you can it will increase the current handling capacity.

To get the turn count for going from a dLRK wind to a LRK wind you can use Turn Calculator 6 and enter the turn count, termination and Kv from the original dLRK wind and it will give you the turn counts for several different winds including the LRK wind.

Jack

Originally Posted by Zapro

Hi guys, Sorry for bumping this rather old thread, but as i see no purpose on making a new one, since i'm still going to rewind the same motor, i just have the time now.
Just got some 0.14 Nomex for insulation, and i'm trying to find out what to do with the motor.

I want to end up with roughly 250 KV on the motor. The original wind was 15 turns for a KV for 450 - can i just calculate from that how many turns i need to put on, or is the ratio not 1:1 ?

Modisc already linked to winding diagrams, so i think i have that covered, just need to know approx how many turns i need to fit.

Any thoughts on this ?

Regards,
Per.

Jack has got you the answer, but just one more special thing to note on the scorpions.

My experience and one of my friend shares the same experience: original scorpion windings should be counted as (X+2) when you do the calculation. For example, if the original winding is "15T", you may need to see it as "17T", when you do the calculation. This is because the original winding "hide" the jumping wires between two distant teeth in those teeth in between...

I am confident about this for 12N8P motors, they surely should be counted like above. For 12N14P windings, my friend shares the same experience, but i did not test it out. He is major in electric engineering and flys a DIYs a lot of plane and motor, I would say i trust his results.

I ran Turn Calculator 6 on the HK5020-450 and it looks like I'm wrong about the LRK being a good wind for that motor. It is probably not, at least at the 250 Kv you want.

It would take 29 turns LRK Star to get the Kv whereas you could get 244 with only 16 turns dLRK Star. In fact, if you re-terminated the wind that is on it now from Delta to Star it you give you a 260 Kv. So it you wanted to do some initial testing you could do that.

You might not be able to make a significant increase in the surface area of the wind because you will be using almost the same number of turns. But you'll probably find that going with a single larger strand will give you some increase in surface area and you neat, perfectly arranged, windings will also give you some better running.

Some motors where you are simply not able to get more copper into them will still show a lot of benefit from a rewind. You just don't know until you try it...

Jack

Originally Posted by jackerbes

n fact, if you re-terminated the wind that is on it now from Delta to Star it you give you a 260 Kv. So it you wanted to do some initial testing you could do that.

Remember - it has NO wind now. I got it in hand, all taken apart. I just glued Nomex paper to the top and bottom of the stator pack, and i'm going to cut sheets to put into the slots. I have the original wire too, but it's pretty well cooked, so it's scrap.

I could do a test wind, only wind a single strand as star and test out the KV value - as long as i do not put any load on the motor, it shouldn't matter if it wound with a single thinner strand (and the fill is very bad) or if i wound it with several strands and put more copper into it... right ?

// Per.

Originally Posted by Zapro

Remember - it has NO wind now. I got it in hand, all taken apart. I just glued Nomex paper to the top and bottom of the stator pack, and i'm going to cut sheets to put into the slots. I have the original wire too, but it's pretty well cooked, so it's scrap.

I could do a test wind, only wind a single strand as star and test out the KV value - as long as i do not put any load on the motor, it shouldn't matter if it wound with a single thinner strand (and the fill is very bad) or if i wound it with several strands and put more copper into it... right ?

// Per.

Yes, but try to wind to a "not too bad " copper fill, since there will be "some" load. What I mean is do not use single 0.33mm wire and only wind 2 turns.

Let me know what the Kv will be and associated number of turns. I am still curious about the scorpion Kv problem. Thanks.