The AC Induction Motor Ebike Project

Aussiejester worked himself up into a rage... I really never did anything to cause it. (so don't blame me)

Safe is half right he didn't have anything to do with the 3 day ban, was random posts that were reported by Augidog where i "cussed" that got me banned...

I believe you might be confusing "rage" with hysterical laughter to Safe...


KiM
 
AussieJester was laughing because I thought (for some reason) that I was using a 100 amp controller by mistake. By the time I actually got out to the garage and took the controller apart to find out what I was running he was banned.

As it turned out I had set the 60 mph downhill speed with a 36 volt 40 amp controller that had no modifications whatsoever. I was running it with 48 volts though.

It's true that I own a 100 amp 48 volt controller and was afraid I got it mixed up. (that was an error... I wasn't using it)

The "final realization" on the rewinds was about brush spring pressure.

Man I wish I figured that one out a few rewinds ago. :sick:

Once you increase the spring pressure it reduces the brush heating and that keeps the commutator cooler. That's something no one figured out... so I just got lucky to have found a paper online about it. The stock spring pressure is too low for the extra power. Brush timing is another factor, but it's not as important as getting the spring pressure right.

So on the negative side, AussieJester, from your perspective you still have to explain how I was able to reach 60 mph (downhill, flat is from 45-50 mph) with only 40 amps. :cool: (the answer is kind of obvious... it's all about aerodynamics, aerodynamics, aerodynamics :whistle:)

But it's good to be skeptical... never just take someones word...

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

I went really conservative on this last rewind, so the speeds and power are down compared to some of the others.

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

Getting Back To The AC Induction Motor

I honestly can't see it being popular like Safe appears to think it will be ;-S The other alternatives
are lighter, smaller, powerful, NOT 240v and look alot less complicated to setup...

I don't normally read things at Endless Sphere, but since I was consulted on some of the ideas in the process of this project going together I have added a link to it to watch.

Frankly... I don't know... if the AC Induction motor can be the winner in the long run. There is a clear need for a low cost, low rpm, high performance motor and "in theory" the AC Induction motor has a lot of potential. I was kind of shocked at the 50 lbs that he's using for the motor. My motor will be about 10 lbs and is roughly of the same size as the Unite motors. I plan to really drive mine hard and do all the "hard work" of figuring it out rather than just buying stuff off the shelf.

It's good to see a first project working... but it's far from where I think it needs to go before people accept it as the correct solution.

The main goals are:

:D Eliminate the need for Geardowns.

:D Eliminate the need for Multispeed Gearing.

:D Adhere tightly to the 1000 watt / 750 watt Power Limits.

:D Remove the Inverter and Rewind the motor to Optimize for lower voltage.

:D Equal (or exceed) the best that any other technology can produce.

...it's a tough problem. :rolleyes:
 
Last edited:
AussieJester was laughing because I thought (for some reason) that I was using a 100 amp controller by mistake. By the time I actually got out to the garage and took the controller apart to find out what I was running he was banned.

No Safe not "Thought" KNOW pal...


Man I wish I figured that one out a few rewinds ago. :sick:

Once you increase the spring pressure it reduces the brush heating and that keeps the commutator cooler. That's something no one figured out... so I just got lucky to have found a paper online about. The stock spring pressure is too low for the extra power.

Chr!st you really take the cake mate, i told you that half way through your motor rewind thread around about the same time i told you about motor timing but as per usual you ignore everything said to you and carry on in your own little world. I even linked you to several articles that explained the realation of brush tension to motor performance :-S once again YOUR NOT doing ANYTHING that hasn't been done before stop claiming you are Safe nobodies buying it anymore.

So on the negative side, AussieJester, from your perspective you still have to explain how I was able to reach 60 mph (downhill, flat is from 45-50 mph) with only 40 amps. :cool:
<--need a "full of **** smiley face there not the cool smiley LMAO


HAHAHA you are a sucker for punishment my friend...No negative side to it Safe i actually had a ripper of a post ready to submit as part of my first post top of page but decided not to get back into it since you insist i shall...

Recall the falsified graphs you upped on ES before you were banned the FIRST time Safe? If you deny it i'll find them tomorrow and post them here for all too see they are in either #1 or #3 build thread... You compared motor efficiencies of 4 types of motors, ring any bells buddy? You listed rc brushless with gear setups, x5 hub motor and the Unite Brushed motor and the Cyclone IIRC?...You concluded that the RC outrunner was hugely more efficient than the brushed Unite...With that in mind Safe explain to us all how he can do what NOBODY else is able to SH1T a 2 wheeled rc outrunner recumbents use ~1400watts to achieve a constant 40mph (yes Matts piece of perfection Safe) yet YOU CLAIM 60mph with 1200watt on a hugely heavier less aerodynamic bicycle with a massively less efficient Motor and drivetrain... doesn't take Einstein to figure out your FULL OF **** Safe...been caught out before you will continue to be caught out... To put it bluntly Safe your clueless to how many amps you are using you have no idea whatsoever nothing you report is accurate you have NEVER used ANY form of real time onboard monitoring..EVER!.....I dont need to explain a thing pal you need to explain why YOU are the ONLY one on this earth that can hit 60mph using 1200watts.


But it's good to be skeptical... never just take someones word...

Very true with in this instance, Safe lost his creditability many many moons ago. I continue to get the occasional chuckle from your posts they are so ridiculous you have to laugh. I give up though pal...say what you like i'm sure some people will buy it...im done with this arguement same old same old...

Now carry on with your AC induction motor Safe or it will be another 8 pages of dribble before you even buy a motor ...pft...its going to be at least 8 pages either way isn't it Safe...not sure i can hang in for this one sorry.

KiM..
 
Last edited by a moderator:
I even linked you to several articles that explained the realation of brush tension to motor performance :-S

I remember you talking about brush timing and I immediately attempted to modify the bike to adapt to that.

Could you go back and find where you posted those links?

So your motor has higher spring tension than stock?... that explains why it's holding up well.

If you want to believe my video was faked (57 mph downhill) then that's up to you... all that can be said for a fact is that my controller was stock 36V 40A.

Enough said.

(for those that might be reading this thread from the start you might just skip a page as AussieJester and I chat for a while)

Jump to the Rewind Thread (AussieJester)
 
Last edited:
I remember you talking about brush timing and I immediately attempted to modify the bike to adapt to that.

Could you go back and find where you posted those links?

So your motor has higher spring tension than stock?... that explains why it's holding up well.

If you want to believe my video was faked (57 mph downhill) then that's up to you... all that can be said for a fact is that my controller was stock 36V 40A.

Enough said.

(for those that might be reading this thread from the start you might just skip a page as AussieJester and I chat for a while)

OH i believe your bike is well capable of 57mph not questioning that...what i am saying is its definitely not using 1200watts to do it Safe unless the "hill" was Mt Everest..Do what you said you were going to do in the motor rewind thread get a WattsUp meter to VARIFY your results. Better yet Get an Egale Tree they give you a printout of the motors performance IN GRAPH form already!!!

And No my motor has stock spring tension i never said i changed it i linked you to several sites that explained the relation of spring tension and brush hardness to motor performance AND commutator wear, same links that explained the motor timing IIRC. Can go over it again though no probz there you want to find the original you sift through 60 od pages of dribble im not about too-->

Harder Brush You will get more power. You will have also less runtime. The motor may get hot and the commutator will show faster signs of wear.

Softer Brush You will get less power but more runtime. The motor won't get as hot as with the hard brush and the commutator will show slower signs of wear.

Heavier Springs More power at low speed. Slightly lower top speed because of the friction of the harder springs.

Softer Springs More power at high speed and less punchy. Slightly higher top speed because of the less friction of the springs.

Heavier Spring on the positive side You will have slightly more punch and the commutator and brushes will show slower signs of wear

http://www.rctouring.eu/electric_motor_set_up.htm

2am here im off to bed...

KiM...
 
First... AussieJester go to the Rewind thread for chatting... (not here, thanks)

Pulse Width Modulation

I like to think that we are all like students in a classroom with no teachers. We seek the truth (just like a group study session) and yet we have to find it on our own most of the time.

To the studies...

I was having some difficulty imagining how one thinks about the mathematics of generating the Pulse Width Modulation for Six Phase power. If I'm going to do this with a generic PIC (which looks like the only way to do it) I will need first to be able to conceptualize what to do, then be able to do the math for it. The MOSFETS are pretty easy to figure out, so that's not the hard part, but the control logic is where the effort ends up going. I'm going to do just simple VF (voltage / frequency) control which is considered the most simple technique. (things get radically more complex for vector control) I will add a current sensor so that the battery delivers precisely the 1000 watts of power allowed by the racing group I'm focused on. (the purpose of it all)

This image helps a lot. The thing to focus on is that the PWM pulse is the intersection of the lines and the actual curve. If you are in the VF (voltage frequency) area of the frequency where you do not want to allow excessive current then the intersection would be LOWER down at a place that corresponded to a lower voltage. In the image the curve is shaped at a little over half of the possible voltage of one.

attachment.php


...the moment I saw this image I was able to instantly understand how the two would relate and the math is going to be easier to deal with now. Sometimes a picture really is worth a thousand words... or a page of math equations.

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

Just to bring things back up to speed...

The reasons for using Six Phase is that smaller motors tend to have more difficulty being efficient than larger ones. Six Phase power has been shown to increase efficiency while lowering eddy currents and temperature and this will go a long way to counteract any negatives that exist because of lower voltage.

:unsure: The chain reaction of thinking goes like:

Lower voltage demands lower resistance and higher currents.

Rewinding to accomodate that means lower inductance.

Lower inductance means more eddy currents and less efficiency.

Six Phase power smooths out the cycle and will compensate for the lower inductance motor.

...if this is going to be a motor that works well I need to give it a fighting chance to win. Apparently with Six Phase power you can drive the motors much harder without as much heat and that's important if you are going to use lightweight motors as your foundation.
 

Attachments

  • Pulse Width Modulation.jpg
    Pulse Width Modulation.jpg
    108.3 KB · Views: 668
Last edited:
Real World Data

It's nice to get some actual dyno data about AC Induction motors and the control methods that might be used.

If you use the "naive" standard sensorless Voltage vs Frequency control technique the actual dyno data is not very good:

attachment.php


...you can see that at the lower frequencies that the torque is much lower than it could be. What is pretty much ideal is to use Sensorless Field Oriented Control where the software makes a mathematical model that is abstract and then monitors simple things like voltage and current and then applies that to how it reacts. The results are near perfect:

attachment.php


...the difference show up at the lower frequencies and it all comes down to an increase in current for the Voltage vs Frequency technique that does not deliver on the power you hope for:

attachment.php


Conclusion?

The most obvious conclusion is that the more complex Sensorless Field Oriented Control is significantly better than standard "naive" Voltage vs Frequency control.

But I purposely held one set of dyno data for last...

If you can modify the "naive" Voltage vs Frequency control with just the smallest change and add a current sensor to it, then you can adjust the frequency at low rpms so as to get a better result. This last set of dyno data shows this "compromise" solution:

attachment.php


...so there is "perfection" and "close enough" (if you add the current sensor) and that's good to know. The advantage of the "perfect" solution is so small that one can choose the "close enough" solution without worry. Since I need to limit power input to 1000 watts anyway (since that's the central focus of the the motor in the first place) the need for a current sensor is already in place.

Even the most "naive" Voltage vs Frequency control works fine at higher frequencies.... so if you couple the pedal power to the motor and allow the pedal power to "peak" in the low rpm area of the motor's powerband then you can accept the "naive" solution as okay.

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

There is hardware and there is software... with AC Induction motors the software is complex, but the hardware is not.

Computer chips these days can achieve fast enough processing speeds that you can simply offload the complexity to the software and that means the hardware can remain simple.

The "bottom line" on all this is that top speed should not be effected by the choice of control method. Acceleration off of a slow start will be improved with the more advanced technique.

0-10 mph (can be improved with a better control method)

10+ mph (does not matter)

...however, on a hill one might drop down to 10 mph in the worst cases, so one can't so easily dismiss the value of the better control method.
 

Attachments

  • Voltage Frequency Control (Standard).jpg
    Voltage Frequency Control (Standard).jpg
    63 KB · Views: 645
  • Sensorless Field Oriented Control.jpg
    Sensorless Field Oriented Control.jpg
    50 KB · Views: 674
  • Torque Amp Comparison.jpg
    Torque Amp Comparison.jpg
    50.6 KB · Views: 614
  • Voltage Frequency Control (Torque Limited).jpg
    Voltage Frequency Control (Torque Limited).jpg
    55.1 KB · Views: 634
  • PWMDrives01.pdf
    202.2 KB · Views: 284
Last edited:
Mind In The Motor

Inductance... it acts like a dampened spring and resists the natural flow of electical current. More inductance means a slower changing magnetic field. Less inductance means faster changing fields and that translates into more eddy currents. Given that electricity is much faster than the motor rpm as a general rule you want to have high inductance matched with high voltage so that you can create smooth gradually changing magnetic fields and this increases efficiency.

For an ebike we need things to be lightweight so what would be acceptable in weight for a 1 hp industrial motor can be many times too heavy for anything useful for ebikes. In a recent AC Induction motor project a 2 hp industrial motor was used and it weighs 50 lbs. That's simply too much for an ebike.

Higher inductance is achieved with more turns of wire. But more turns of wire also means higher resistance of that wire because it becomes longer. You can increase the number of turns and also increase the wire thickness and then keep the resistance the same, but now it's going to be using more copper and that means it's harder to fit all of that into the grooves.

Three Phase power is the most common. Six Phase is very uncommon to the point of being essentially never done. But for the needs of a small AC Induction motor (that isn't going to be able to have high inductance if it's going to work well at low voltage) the Six Phase power approach might be a way to lower the eddy currents by design, thus increasing the efficiency.

Is there any other way?

...I just don't see another way. There are no small lightweight AC Induction motors that when stock will work. (to my knowledge) The VFD (variable frequency drives) that can be bought on ebay require high voltage because they tend to be built around the household 120 VAC or 240 VAC power source, in order to use them you need to step up the voltage with an inverter and each of those processes lose at least 5% per step making the overall efficiency not so great.

Either you have to give up on the idea and fall back onto permanent magnets or you have to push forward and do the homework and complete an optimized small AC Induction motor from the ground up.

This is not going to be easy.... :beta1:

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

Chorus Motors

You can't find these on ebay and probably need to special order directly from them because they are so specialized (and expensive) but these people really do seem to have a solid understanding of how to pull off high performance from an AC Induction motor. For the most part the industrial world has not bothered with electric motors for a century because the stuff out there already is good enough to run a fan or a converyer belt while sitting on the ground. For EV's the need for performance is reviving the technology and taking it to new levels.

They have a REALLY COOL tool that simulates the motors !!!

Executable:
http://www.chorusmotors.gi/technology/chorus_4.exe

Java Based Browser Version:
http://www.chorusmotors.gi/technology/simulator2.shtml

...hmmmm, I wonder if I can get them interested in building an ebike motor? :geek:

Try it at Three Phase, then switch to Six Phase... this clearly shows how much smoother things get as the phase count increases.
 

Attachments

  • Chorus Simulation.jpg
    Chorus Simulation.jpg
    239.9 KB · Views: 301
Last edited:
Email Response From Chorus

Safe writes:

Chorus Motors,

Has your company ever considered creating a small lightweight
AC Induction motor/controller using Chorus technology built
specifically for the ebike market?

Permanent magnets have a narrow usable powerband, but the
AC Induction motor is being used in the more advanced electric
motorcycles (like the races at the Isle of Mann) and electric
cars. Your technology is superior to all the older designs.

In America the law limits ebikes to precisely 750 watts (1 hp) as
output from the motor and so what is needed is a high performance
light weight ebike motor that delivers a flat powerband from the
bottom to the top speed. It needs to operate at low voltage (48
volts at most) because you can't build machines for young kids that
are high voltage.

I'm not a company, just a hobbiest, so I have no business to do
with you directly, but if you could at least pass this idea on and
get people to consider the idea I would be grateful.

Thanks


Chorus responds:

We have thought about this for a while. I guess we will think some more.

Thanks!!!!


...which is very non-committal, but at least shows they are aware of the potential.

Maybe we will be talking about tricked out Chorus Motors in the coming years? :cool:

(I hope so)
 
Drilled Out Stator

While I can dream of one day using some high performance AC Induction motor specifically designed for an ebike... that's not likely to happen any time soon.

So this is the progress on my motor so far...

Using just a regular drill I managed to expand the grooves on the Single Phase motor I started with so that Six Phases of wires can be wound into it. (I expanded the grooves making them symmetrical... or as good as I could anyway)

The machining is very crude, the shape of the stator is wrong (it should be round for a polyphase motor) but it's what I have to work with for now. You do what you can with what you've got. The metal is steel laminations and it drills about the same as most steels you would be used to dealing with. It would have been really nice if I owned a drill press (with an attachment to hold the part) so as to be able to use a router to have expanded these grooves. They turned out "okay" for this project I guess. (this is more "proof of concept" that a lightweight AC Induction motor is possible than anything else) We already know you can use industrial motors (too heavy) so this one weighs less than 10 lbs and should deliver 1 hp when I'm done.

Before rewinding I plan to cover the stator with a thin layer of fiberglass epoxy resin much like the way I did my other motor rewinds. The epoxy creates an insulation barrior so that the magnet wires do not short out.
 

Attachments

  • Drilled Stator 001.jpg
    Drilled Stator 001.jpg
    141 KB · Views: 287
Last edited:
Back
Top