Triple Rewind of Unite 500W Motor

[duivendyk]

Safe,I'm not a glutton for punishment.It does not really make any difference what the current limit on the motor is 20A,30A whatever.I was making a basic presentation about this rewinding business and used the 24 V 20 Amp ,480 W motor as an illustation.The result at 30 or 300 Amp would have been identical.I could also have thrown a bunch of (not all that complicated) equations at you to make the same point,but I have learnt that most people like concrete examples.
My contention is that this rewinding/multiple winding business boills down to increasing the motor speed,and that this is what really makes the improvements in power output &efficiency possible.A motor of this type is power limited by the permissible dissipation (temp) so improving efficiency permits higher output. The only other way you can improve efficiency short of increasing speed is to get more copper on the armature.Multiple parallel windings can sometimes used to advantage to bring this about.If the speed is changed the overall vehicle mechanical design MUST be modified to accommodate this change in motor speed.If not the improvements CANNOT be realized.

 

[safe]

30 Amps Matters

duivendyk, you are not dealing with enough "real world" issues and are relying too much on the abstract. In the real world the controllers current limit is a major factor on the shape of the motors powerband.

Let's remember what is our "perfect configuration":

Peak Power and Peak Efficiency rpm are the SAME.

...now that only happens in one particular configuration and that's what all this exercise has been about. Ultimately if you want to get the best power and the lowest heat you need to get the PP -> PE exactly right on. (which I've done)

You also want to use as many rpms as possible so you need to push the rpms up to around 6000 rpm for the no load.

Can you do this by buying new batteries (overvolting) or buying a new controller?

Yes... but that's not what we want to do here... the whole point is to "tune" the motor perfectly. Most of these motors are waaaaaaaaay out of "tume". ("tune" being the perfected PP -> PE)

So you are "right" in the abstract... there are alternative methods of modification... but rewinding allows you to get it perfected given the battery and controller you ALREADY have.

(I stress the ALREADY part)

There is only one perfect configuration.

Just think of this as a "stealth only" modification. No battery changes or controller changes are allowed. The bike must "appear" stock.

The point is to deliver more performance without making the bike appear modified.

-------------------------------------

Take AussieJesters trike for example... would you ask him to modify his trike so that it might carry 72 volts of SLA batteries in order to achieve the perfect configuration? While it might be possible to do it also might not be all that practical. By doing these rewinds you get all the perfornance of a 72 volt system while only needing a 48 volt system.

High voltage is also more dangerous to deal with and you also have the fact that any of the cells in the battery could fail and that means more reliability problems. (48 volts means 4 SLA's, while 72 volts would require 6 SLA's)

The stock WalMart bike is only 24 volts (of SLA) so trying to get that up to higher voltage might mean you need to buy some expensive stuff.

Anyway... I hope you are getting the "practicality" argument that not everyone wants to spend money on new stuff.

Save your money, just keep everything stock and then rewind to achieve perfection.

 

[duivendyk]

Safe you are something of a romantic,admirable in a way but a something of a pain in the-you- know-where. at times.You sort of proceed to make up you preferred laws of physics as you go along.Validated by copious graphs print-outs and what have you."Papier ist geduldig".Paper is patient as the Krauts put it.It does not talk back.May be cavorting around in cyberspace does that to you.The basic fact is that what you call my "abstractions" although not the total reality are useful.They have predictive value,instead of stumbling around till you finally hit some sort of optimum, and sit scratching your head,thinking ,may bethere's gold in them there other hills and go charging off in a new direction.I try to determine the outlines of the performance envelope and develop an understanding of the design tradeoffs.It has served me well in my work.From analysis to synthesis.For instance rewinding this Currie motor won't do much good in terms of efficiency until we can optimise the drive ratio,that's what these 'abstractions' tell me.They are worth more than those printouts.The key insight is that you have to run in a low current high speed mode for best performance &efficiency.
Please peruse my posts and try to understand it.The math is not that bad,mostly just Ohms law and old fashioned arithmetic.

 

[safe]

Only One Target... Many Paths To Get There...

The target is that perfect configuration where PP -> PE.

You can get there from all sorts of different angles and the first time it was by accident for me. (with my first rewind I had no idea what to expect)

I've in a sense set the criteria as:

Rule One: The battery must remain stock. (no more $$$ allowed)

Rule Two: The controller must remain stock. (no more $$$ allowed)

...if you start with this criteria and go forward the only way to achieve the PP -> PE perfection is with a rewind.

The "lazy" way to approach a better configuration is by spending more money for more batteries and then having to add on top of that spending spree the cost of a new controller. Depending on how much correction is needed you might be out some serious $$$.

The rewinding only costs about $20.

---------------------------------------

Or are you arguing that throwing money at the problem is the best action?

---------------------------------------

My present 500 Watt Rewind started out like this:

...and because it now runs more lean it's like this:

...heat is normally "waste" which I'm now using for more power.

This is assuming that nothing else changes. (battery or controller)

You get the "behavior" of a higher voltage without needing the higher voltage.

Notice how the "lean" motor has that "perfect" PP -> PE configuration. The "rich" running motor has a spread of about 500 rpm between the PP and the PE. You are understanding these charts right? I mean the efficiency curves are really what matters. The goal is that perfect curve with no dents in it.

 

[AussieJester]

But you used 21 AWG. I found that the 22 AWG was painful and the 20 AWG was not so bad and it does depend a lot on how tight you wind things.

11x2=22 21 AWG

That sounds like it will work well. Should be strong and run cool.


(mine was the same configuration, but 20 AWG)

For i think @ least the 4th time i have posted this

I AM USING AWG 20 WIRE AGGGGGGH

I have ordered 250grams of shellac (will make 1 liter when mixed with methylated spirits) to dip the motor in its coming from eastern states so with a liiiiittle luck it will be here this week. I'm also optimistic Franks motor he has sent me will also arrive this week...Whichever is together first gets the first run...either way both will be tried and reports of which works better IN REAL world terms ie. in the trike NOT on paper will be reported back here.

Here's a few before and after pics of my rewound Unite 1000watt motor with AWG20

Few before shots

http://members.ii.net/~aussiejester/index/pictures/tn_motor_rewind_pic1.JPG

http://members.ii.net/~aussiejester/index/pictures/tn_motor_rewind_pic2.JPG

http://members.ii.net/~aussiejester/index/pictures/tn_motor_rewind_pic3.JPG

Stripped and ready for some paint

http://members.ii.net/~aussiejester/index/pictures/tn_motor_rewind_pic7.JPG

End plates stock on left enlarged vent holes on right

http://members.ii.net/~aussiejester/index/pictures/tn_motor_rewind_pic8.JPG

Both End caps with enlarged vent holes and additional holes added

http://members.ii.net/~aussiejester/index/pictures/tn_motor_rewind_pic112.JPG

Read to rewind

http://members.ii.net/~aussiejester/index/pictures/tn_motor_rewind_pic113.JPG

THE START!

http://members.ii.net/~aussiejester/index/pictures/tn_motor_rewind_pic114.JPG

Progress shots of rewind including "secret weapon"

http://members.ii.net/~aussiejester/index/pictures/tn_motor_rewind_pic119.JPG

Full 20AWG 11 x 2 == 22 Rewind pic FRONT

http://members.ii.net/~aussiejester/index/pictures/tn_motor_rewind_pic123.JPG


Full 20AWG 11 x 2 == 22 Rewind pic FRONT

http://members.ii.net/~aussiejester/index/pictures/tn_motor_rewind_pic122.JPG

ONCE AGAIN though i will stress to anyone thinking about taking this on IT IS NOT a quick task, i started work on this at ~10am and didnt finish till 7PM in the evening Granted there was several 'breaks' including a trip to the shops for lunch and to 'freshen' the beverage, if this was done in a start to finish affair it would still consume most of your day. Even if it does produces a better motor i would hands done prefer buying another motor than rewinding again, its a mind numbingly boring job i hopefully will never have to do again, shall be keeping an eye on the amps more closely this time round LoL..

You sort of proceed to make up you preferred laws of physics as you go along.Validated by copious graphs print-outs and what have you

I have heard this alot before but not put so politely LoL...

KiM

 

[duivendyk]

AJ, I'm getting to be pretty good at this politenes game,have the bad rep. of being overly scornful& now attempting to mend my evil insensitive ways. Admire your rewinding fortitude&hope it pays off.I'll stick to over-volting, if I take the plunge,but not sold on this cheap WM Currie job,due to lack of variable speed drive.

 

[AussieJester]

No i wouldn't bother with the Walmart bike period. If your interested n electric save the dollars for a decent setup would be my advise to anyone. Yes its cheap and yes a rewind will probably make it better but at the end of the days its a walmart bike with a cheap motor on it not geared running SLAS. And as said before 99% of the people WON'T want to sit there for hours unwinding and rewinding a motor to save a few bucks contrary to what safe believes...Just won't happen, majority now buy hub motors as they dont have the knowledge/tools/room to install a non hub kit GOODLUCK finding someone to by a s h i tty wallmart bike so he can rewind the motor hehehe

In regards to my rewind...it was more out of necessity as can be seen in the initial pics if it wasnt burnt i wouldnt have bothered seeing i have Franks 1200 on the way (im using a new pic resizing app and made them a lil smaller than i should have) In the full size pics you can see the insulation lifted on the wires, it wasn't even possible to 'unwind' the original wire, the wires had melted together in clumps, i had to use an angle grinder to cut the wires at each end of the armature and pull them out with pliers, was actually prolly quicker than unwrapping it all TBH LoL... DEFINITELY be keeping an eye on temps and amps ALOT closer this time around. If i fry this one again i wont rewind it, ill use Franks motor (most likely be using that either way it has MONSTER torque, the motor mounts needed to be replaced as did the bolts holding the main gear the motor was driving it snapped the originals LoL) gear that sucker and she will FLOG be like a MOTOGP bike, they have that much torque they have to be electronically controlled they are un-rideable without .. I think this double rewind wlll be super gutless like a two stroke and need revs to get anywhere. I'm not even sure what the fascination is with high rpms to be honest, but my area of expertise is in motorcycle ICE motors having raced them competitively for many years this electric is new to me... you can rev a motor as high as you wish (within its structural limits obviously) but if it don't have the torque it aint pushing nothing far fast. What motors were n the 24 hour LeMans winning cars the last few years ...DIESELS! If you have the Torque to turn the gear you have the speed :: wink ::

Now we will have a 19AWG single wind Unite V's a 20AWG Double wind Unite to compare SIDE BY SIDE No more of these graphs (that as safe knows i care little for i think they're ridiculous) with made up data to show what might happen. I shall Video the results of both motors performance on same course and report back my findings. Either way i guess i have two motors now to last me a few years if needed. After seeing Matt (recumpences) latest video of his twin RC PK Ripper go wont be long before 2-3 rc motors find themselves my way i give u the tip check this out-->

http://www.youtube.com/watch?v=DtcT...?f=28&t=9534&p=157446&feature=player_embedded

Sooner i get my CNC mill finished sooner i get the drive units made ;-)

KiM

 

[safe]

I follow a very scientific process:

Step One: Wild Theories. (with graphical representations)

Step Two: Creation of something that Tests the Wild Theories. (real world)

Step Three: Confirmation and Validation. (final charts that display what really happened)

...sometimes what is a theory and what is the final reality are different, but the process does tend to get to the answer. It is the "Scientific Method" after all.

If we never had "Wild Theories" to test then we would be stuck in the Dark Ages where the church just always had all the answers.

------------------------------------

I'm still optimistic about a WalMart bike rewind despite peoples distain for such "low budget" adventures. The volunteer needs to be someone like I was as a kid... very adventurous, not all that rich (can't just buy my way out of a problem) and needing a "stealth solution" to upgrade their ebike.

Stealth means no one can tell the bike is modified.

If you overvolt you've lost any chance at Stealth.

-------------------------------------

Great job with the Rewind AussieJester... very well documented...

I'm looking forward to see your "real world" testing.

My bet is that the 11x2=22 20 AWG beats FrankG's motor in top end power by a significant amount. Maybe you sacrifice a slight bit of the low end, but the increased top end will be so much more that you will forget about it. The 11x2=22 motor can also pull amps all day long without overheating because it has significantly lower resistance. But beware of the low, low, low rpms because they can actually make the motor rather hot. You are correct that the low end torque is not that great. Be sure to gear your trike down as low as possible so that you can scream the motor all the way to redline or you will be wasting it's potential.

The 11x2=22 will even sound different... I'm eager to hear what you think of the "new sound" that it makes. (it starts off as a whistle with my controller)

The 11x2=22 really does behave like a MotoGP motor... it's a screamer.

 

[safe]

Differences

The first most obvious difference I see between your Rewind AussieJester and mine is that I used a FOUR gap wind and you used a THREE. In other words, you left three grooves between the coils and I did four grooves. FOUR grooves is actually a full 90 degrees while THREE is a little less than 90 degrees. Your advantage is that you probably have a slightly lowered resistance since the wires are shorter, but the motor efficiency might be slightly less. The two forces will tend to cancel each other out somewhat... but this is a clear difference. Many motors use the THREE groove pattern and so it's okay to do this. (some use the THREE others use the FOUR)

The second "concern" is that I'm not sure if your second wind is aligned correctly with the first wind. It looks almost like you have shifted the second wind over one groove by accident. I tested this as a way to try to change the field (on purpose) and the result was a lot of wasted heat because the magnetic field was rising and falling too much because of the spiral shift in the pattern. This can be an illusion because when you wind the way you lay down the wires can produce some weird visual effects. It's just important that you are certain you go it the right way.

Question:

Did you do the entire wind with a single strand of wire?

...if so then you are fine. Otherwise, did you take care to be sure to get the starting position lined up right? If the starting position is off by one groove that's enough to screw up the results. A commutator pole needs to send it's wire to the same grooves on BOTH winding layers. In effect the two strands of wire are going to combine to behave much like a single wire, so they need to be going to the same places.

It's really hard with the photo to be able to tell.

I know it's a sickening feeling to think you might have made a mistake, but it's better to realize it now and fix it than to continue on with an epoxy layer only to have to take it apart again later.

If you go backwards through my tests you will find that I tried out a "Spiral Wound" pattern and the results were terrible... way too much heat for no reason. In the end I had to remove all the wire and start over.

 

[duivendyk]

In electric motors improvement of both output power and efficiencyi s possible if the no-load speed can be increased.This can be done by running it at a higher voltage. since the speed is proportional to input voltage,If the motor current and the torque which is proportional to it, are kept constant .The armature loss is also constant,but the input power has increased corresponding,and with it ,the output power (disregarding increase in core loss& mechanical losses).So if you take say a 24 V,20Amp motor runing at 3000rpm nl .with an 80 %efficiency and run it at 36V,20Amp ,4500 rpm nl you will achieve 87% efficiency.
In terms of output power close to a 60% improvement in power with no increase in loss.Of course the motor has to be able to run at this higher speed
This overstates the case somewhat depending on the increase in other losses.The basic point though is that for a certain torque (current) the losses are constant but the percentage loss is inversely proportional to speed (input voltage),if you ran this same motor at 12 V , at the same current instead of at 24 V the efficiency would be only be 60%
(1-2*20)=60 % at 6V only 40%.So crawling up a hill at low speed to"save" your motor is a bad idea.
One thing to remember is that in contrast to a gas engine which is torque limited an electric motor is NOT ,until saturation occurs the torque is proportional to the current and can be many times higher than the dissipation-limited value. The normal controllers control motor voltage, not current,therefore the input current is determined by current overload protection in the controller if any.Riders like this because they have lot's of get up&go and are clueless as to how inefficient all this is.
In an IC engine the max.torque is inherently limited and a transmission has to be used to multiply the torque as needed or wanted,also the efficiency is a lot less variable over the speed&torque range.
What elctric bikes really need is a multispeed rear hub or a NV hub to keep the motor running efficiently at close to max voltage instead of changing speed solely by voltage control.It's surprising to me that this does not appear to be done.Cost may be,or they like to sell batteries,these voltage control controllers are for the birds too,you need current (torque) control,like in a car or MC.