Dual "fast wind" mini-motor build

I put this in this section because, as far as I know, this is the first documented 2WD install using the "328" "fast wind" mini-motors and one of the initial test results is a little suprising to me. But it shouldn't have been as the Ebike CA's simulator had predicted it.
But first, the install, which consisted of two Ananda Q100 36V 328 motors. These kits come with a very small 15A KU63 controllers. Except for exchanging the main battery to controller connectors to 4 mm bullets, the kits were installed "as supplied' and not modified.
Both kits were very easy to install. The 9-speed DNP free wheel fit in the standard drop-out with a plain washer inside the chain stay on each side.
The rear disc. lined up without drama, but the rim was slightly off center. As supplied, it needed truing and a spoke adjustment, so I combined those two operations with a bit of "dishing"[a misnomer here, as I only moved it over a few milimeters].
The frt. kit was, if anything, even easier to install. The wheel was ok out of the box and I had already clearanced[filed]the drop outs to fit a MXUS geared motor.
Fitting the mini- motor's 12 mm axle does not require 'deepening" the drop-out to center the axle as does a 14 mm axle. Only a small amount of widening and contouring to match the half-round portion of the axle. With the correct size round file, a few passes is all that is needed.;


The only change to the platform is the substitution of the original 26" wheels for a pair of 24 inchers. This is a large framed bike and with my 29" inseam, I always felt as if I was riding a step ladder.

This morning, I wanted to get a work-out on the way to Macdonalds , so it was rear motor only for the three mile round trip. The first thing noticed is how quiet the Q100 36V is. At half power, a walker three feet away wouldn't know it running. It's also very smooth with no throttle twitchyness at all. Just grab a handful and shift up and down according to load and speed. Of course, noise and twitchyness are things one would associate with power and at 12S Lipo and 13 to 15 Amps, there isn't much. And with this controller, what you see is what you get. The CA reports no wild peaks when hitting the first steep part of a hill, MaxA never excedes 15.5A. The Lyen 23A controller coupled with the MXUS, by contrast will see peaks up to a 38A
The Q100 with the KU63 controller seems soft and friendly, almost feminine. The MXUS with it's 17A controller[peaks in the 20'sA]has a bit of attitude. It growls at you if you "dog" the throttle too much at too low a speed and in a race, it will walk away from the Q100 and it's controller anytime, anyplace.
All this makes for a wonderfully blissful pedaling experience with the Q100 until the first hill comes into view.
My test hill is in the 10 to 12 % range.
The MXUS will churn it's way up with what help I can provide at about 9 mph with the stock controller and 12 to 13 mph on the Lyen. It's not liking it and let's you know, but it digs in and does it. With the Q100 and Ku63, I wasn't a third of the way up when it was clear that without getting into a gear that wasn't readily available to me, we weren't going to make it, so at 5 mph and falling, I made a u-turn and headed back down.
Ok, so a 2.0 Kg, high speed motor with current limits of 15 Amps trying to haul a 250 lb. rider doesn't climb well, not exactly ground breaking news. But how would it climb on two motors?

2WD is a blast, it's hard to stay out of it.
The Hill can be taken at 12 mph pulling 1200 Watts, showing "never overheat" on the Ebike Ca sim.[only 600 Watts though each motor]. Truely a case of each motor helping the other to stay in a reasonably efficient range[72%] and losing too much to waste heat.
Where the dual drive system really shines is off the line. The throttle response is smooth and controlable. A skilled rider could do great "stupid Ebike tricks" with this set-up.
My morning route takes me across a couple athletic fields[before school]and up one grassy slope, wet with dew. And with the frt. mounted motor I would usually spin-out near the top. In 2WD it motors up with nary slip. Even with the "street" tires I'm running..
Top speed runs, flat, no wind or pedaling, produce 26 mph on 12S, 28 mph on 13S. So, it looks like 30 mph is doable on 14S[and I weight 250 lb.s!]. One motor speed remains the same on either 12S or 13S, 22 to 23 mph. I knew, since one motor was "current limited", that there would be a significent increase when switching into 2WD, but I did not expect gains of 4 mph.
I'm very happy with the results and the versatility of this build. One motor drive is perfect for assist and the second motor is great for "squirting" across busy intersections and the ocassional "killer" hill.
Using one small geared "fast wind" motor comes with limitations, but by coupling a pair together, these limits are overcome, actually turning them into strengths

I also think, that for Dollar value, this set-up matches up well with any single larger geared motor.
And at these power levels, there is no need to mess with frt. to rear power differential. There is no tendencey for the frt. to over-power the rear or visa versa.

Thanks for the report, motomech. I have been intrigued by the idea of running dual "Cuties" for quite some time now. I am glad to see that you are doing it and are pleased with the results. I am not sure if this is technically correct, but my view is that each motor pulls half of the air resistance load as well as half of the gross weight up inclines, resulting in a speed boost over a single motor. My first build might consist of a dual setup with the same motors in 20" wheels on 16S LiFePO4, so I run the simulator using the 20" eZee motor and wheel at 52V (16S) / 15A / 50kg (110 lbs), the latter being ONE HALF of my estimated bike+rider weight. I like the way both the efficiency and power curves are relatively flat from 50% to 100% speed, and that the motor power is around 600W max with efficiency above 75% in the whole of that range (12 to 24 mph). The eZee motors are built to sustain that level of power (presumably not overheating even on a 25% grade in a dual 20" setup!). Of course the cheaper and lighter Q100s are not going to perform that well, but perhaps ATF (oil) or other creative cooling methods could boost their performance. Then there is the question of how well the gears (double planetary, IIRC) are to going hold up while running over their rated power. Perhaps in the future - after your inevitable mods - you will be able to give us a better idea of the power handling capability of these motors in your nifty setup.

WhatcomRider said:

Thanks for the report, motomech. I have been intrigued by the idea of running dual "Cuties" for quite some time now. I am glad to see that you are doing it and are pleased with the results. I am not sure if this is technically correct, but my view is that each motor pulls half of the air resistance load as well as half of the gross weight up inclines, resulting in a speed boost over a single motor. My first build might consist of a dual setup with the same motors in 20" wheels on 16S LiFePO4, so I run the simulator using the 20" eZee motor and wheel at 52V (16S) / 15A / 50kg (110 lbs), the latter being ONE HALF of my estimated bike+rider weight. I like the way both the efficiency and power curves are relatively flat from 50% to 100% speed, and that the motor power is around 600W max with efficiency above 75% in the whole of that range (12 to 24 mph). The eZee motors are built to sustain that level of power (presumably not overheating even on a 25% grade in a dual 20" setup!). Of course the cheaper and lighter Q100s are not going to perform that well, but perhaps ATF (oil) or other creative cooling methods could boost their performance. Then there is the question of how well the gears (double planetary, IIRC) are to going hold up while running over their rated power. Perhaps in the future - after your inevitable mods - you will be able to give us a better idea of the power handling capability of these motors in your nifty setup.

Click to expand...

Now that I have some time riding this set-up, I can go back and look at the parameters I was using in the sim. and play with the varibles and see what changes will rectify the actual results with the predicted results.

...but my view is that each motor pulls half of the air resistance load as well as half of the gross weight up inclines, resulting in a speed boost over a single motor.

Click to expand...

If I half my weight, the climbing results on the sim go way askew of the real results. Much closer to use the actual number in the sim. I believe that Justin, in his genius,had made allowances for wind and rolling resistance so I consider that pretty much a non-factor. Not much we can do about it anyway.
I'm seeing that the thing that makes the most difference is Current. If I reduce the value the 13 Amps, all the climbing performance results seem to fall into place. Also, the sim" Batt. Power reading of 660Watts is almost exactly what I see as a one motor Max on the C.A.[2 motor 1280Wmax].
Using,
52V
13A
330 lb.
24" wheel
MTB
I get,
1 motor 10% climb below 5 mph-true
2 motor 10% climb 12.5 mph-exactly
I believe the biggest gain in 2WD climb is because each motor is helping the other maintain a higher rpm and thus, in a high efficiency level.
Then, if I reduce the Current even further to 10.5A[everything else the same] and run the top speed numbers, I get these sim results,
one motor 21 mph-actual 20 mph
two motor 27.5 mph, actual 28 mph
I'm not savvy enough about this stuff to understand why, but I need to ck. the Wmax-2 motor, top speed next time I'm out.
so If we use the same "correction factors" with your proposed build.
52V,
13A,
20" wheel,
220 lb.s
MTB,
we get,
one motor 10% climb-9.5 mph
two motor, 10 % climb-19.5 mph!!!
Climbs like a Mountain Goat.
A further correction to 10.5 A, we get,
one motor top speed-21 to 22 mph,
two motor top speed-25 to 26 mph
Still, a very nice range indeed.

The eZee motors are built to sustain that level of power (presumably not overheating even on a 25% grade in a dual 20" setup!). Of course the cheaper and lighter Q100s are not going to perform that well,...

Click to expand...

Of course lighter mean less mass to absorb and transfer heat, but I don't think "cheapness" is a factor here. I've been inside these critters and they seem well built and engineered.

but perhaps ATF (oil) or other creative cooling methods could boost their performance. ...

Click to expand...

Tried it, the ATF poured out the right side bearing, all over the disc...fail

Then there is the question of how well the gears (double planetary, IIRC) are to going hold up while running over their rated power.

Click to expand...

My sense is, that on the street and no curb jumping, no gear problems. I used to ride the MXUS off road and with these low powered motor, it's not power that's going to strip the gears, it's load stress. And it's pretty easy to know when you are doing that. You can almost feel them getting ready to bend. And my understanding is, that the compound gears, with more surface area, have the potential to be more durable.

Perhaps in the future - after your inevitable mods - you will be able to give us a better idea of the power handling capability of these motors in your nifty setup

Click to expand...

At less than $80, plus shipping for a motor ASM., I don't fret too much over the Q100's reliability.

Interesting read. I keep meaning to get around to a dual motor bike but never seem to do it. When I feel perky, I end up going for a romp on the trail bike instead.

Been planning to try dual 2810 nine continents on my longtail cargo bike. That would mean 2000w to climb hills, but more likely no more than 600w to 800w would actually pull through each motor. Like your setup, time to overheat, infinite. 8) Yet able to haul a bike load of weight up that hill.

Keep the data coming, it's priceless.

A dual motor setup with the smaller geared motors is the only thing left for me to want to try once I get my mid drive hub motor finished. My only concern is how well these gears hold up with trail riding abuse. They are cheap enough to try, but I'm not convinced they can last since the gears look to be just plain nylon. For regular street riding I'm sure they are fine, but it's the trail riding capabilities that I'm interested in.

I like what you're doing here. Getting ideas... 8)

dogman said:

Interesting read. I keep meaning to get around to a dual motor bike but never seem to do it. When I feel perky, I end up going for a romp on the trail bike instead.

Been planning to try dual 2810 nine continents on my longtail cargo bike. That would mean 2000w to climb hills, but more likely no more than 600w to 800w would actually pull through each motor. Like your setup, time to overheat, infinite. 8) Yet able to haul a bike load of weight up that hill.

Keep the data coming, it's priceless.

Click to expand...

Thanks D.man, kudos mean a lot to me coming form you.
Not much to report, did a cross town run yesterday where I had to deal with traffic. To maintain the pace, I was switching back and forth between 1 and 2 motors. As one would expect, power usage was right in the middle of one vs. two, in this case, 27 Wh's.

double post

Hello Motomech,

So you did it...DUAL CUTE's...very cool 8)

The more I think about the advantages of two small motors compared to one larger motor the more I like the idea! Some day I will replicate your setup but for now I'm still working out the details on my latest build.

One question; what caps and FETs are in the KU-63 controllers?

-R

I know the question was directed to Motomech, but I suggest you take a look at this impressive web page where you can get photos, specs, schematics, and mod tips for the KU63 controller: http://www.avdweb.nl/solar-bike/electronics/ku63-motor-controller.html.

The 36V power rail caps are rated at 63V and the six 2SK4145 MOSFETS are rated at 60V / 84A / 10 mΩ MAX.

motomech said:

Not much to report, did a cross town run yesterday where I had to deal with traffic. To maintain the pace, I was switching back and forth between 1 and 2 motors. As one would expect, power useage was right in the middle of one vs. two, in this case, 27 Wh's.
...
A pair of MXi[plural?]would be stronger performing than a pair of Cutes, with the survival rate of the gears being inversely proportional to the power applied. But in the end I just went with another pair of Q100's, but 201 winds.
...
And taking a page out of your book, I'll have two pair of controllers, one pair stock for summer and a pair I can shunt mod for winter. Or 1200 to 1300 Watts-summer and 1800 to 1900 watts for winter.

Click to expand...

Very interesting build. I would recommend that you strongly consider getting a CA v3 on there with a limit switch so you can throttle the two motors back to the power of one. I find that the two motors run together at half power outperform a single motor at full power - it seems they are hand-holding each other up into a better efficiency point giving better usable power for the same watts. I was originally just switching between one and two motors as you are doing, but recently added the two-motors-as-one mode and it's a hit .

Also, the throttle ramping on the v3 is a real game-changer for gear motors - you can go WOT off the line and it just eases away smooth as can be w/o shredding clutches and gears.

A CA Vaux limit switch would also let you get away without the two seasonal sets of controllers - just build up one pair with 'winter' capacity and let the CA limit them back for summer mode. I appreciate you are going for a simple setup, but the new CA v3 is a huge improvement over the previous model and well worth the minor complication. Also, by using it to source the power for your throttle, you can get around your current issue of needing a particular controller active to juice the throttle.

Anyhow - just a thought :wink:

Russell said:

Hello Motomech,

So you did it...DUAL CUTE's...very cool 8)

The more I think about the advantages of two small motors compared to one larger motor the more I like the idea! Some day I will replicate your setup but for now I'm still working out the details on my latest build.

One question; what caps and FETs are in the KU-63 controllers?

-R

Click to expand...

So I guess you saw Cassino's web site on KU63's. He seems rather obsessed about that little controller

I've been happy so far with the controllers I've purchased from BMSbattery, they're inexpensive and feature-packed, but I've never used the little KU63. The two KU93's I have (a 48V and a 36V) both use 63V caps and 9 HY1707P mosfets (70V/80A, Rds on = 0.06 mohm).

-R

teklektik said:

motomech said:

Not much to report, did a cross town run yesterday where I had to deal with traffic. To maintain the pace, I was switching back and forth between 1 and 2 motors. As one would expect, power useage was right in the middle of one vs. two, in this case, 27 Wh's.
...
A pair of MXi[plural?]would be stronger performing than a pair of Cutes, with the survival rate of the gears being inversely proportional to the power applied. But in the end I just went with another pair of Q100's, but 201 winds.
...
And taking a page out of your book, I'll have two pair of controllers, one pair stock for summer and a pair I can shunt mod for winter. Or 1200 to 1300 Watts-summer and 1800 to 1900 watts for winter.

Click to expand...

Very interesting build. I would recommend that you strongly consider getting a CA v3 on there with a limit switch so you can throttle the two motors back to the power of one. I find that the two motors run together at half power outperform a single motor at full power - it seems they are hand-holding each other up into a better efficiency point giving better usable power for the same watts. I was originally just switching between one and two motors as you are doing, but recently added the two-motors-as-one mode and it's a hit .

Also, the throttle ramping on the v3 is a real game-changer for gear motors - you can go WOT off the line and it just eases away smooth as can be w/o shredding clutches and gears.

A CA Vaux limit switch would also let you get away without the two seasonal sets of controllers - just build up one pair with 'winter' capacity and let the CA limit them back for summer mode. I appreciate you are going for a simple setup, but the new CA v3 is a huge improvement over the previous model and well worth the minor complication. Also, by using it to source the power for your throttle, you can get around your current issue of needing a particular controller active to juice the throttle.

Anyhow - just a thought :wink:

Click to expand...

Honored Sir that you would drop in. I have drooled over and studied your spectacular 2WD Yuba build on more than a few occasions. It's quality of craftsmanship certainly something to aspire too.
I need to start reading about C.A. V3, but frankly this years Ebike budget is dwindling. The Lipo Gods have not been kind to me as of late and I have had to buy more of the spiteful stuff recently.
At any rate, by the time I finish reading the 30 page[and growing]thread on the V3 and understand it's functions, it will be next year and I can up-grade. It's funny, even though I don't use the functions of my Sm. Screen CA, I've become rather attached to it. Destined to be a classic.

Jeeze, what's up with these dual posts?

Ok, more changes, I just can't leave anything alone.
Well, there was something that was bothering me about this build, namely running two controllers off a single throttle signal
We talked about the 'interaction effect" of running two like systems in conjunction and I think I was experiencing one of an undesired nature.
Namely, the throttle feel. Running both systems at the same time produced a harsh feeling twitchiness that I didn't care for. Just not smooth and I know it's not the throttle itself because I have used the Half-twist before w/ a single motor
Perhaps any "harshness" in the controller's operation is acerbated when they are summed.
To be continued...

Dealing w/ one throttle-two controller harshness;
Since my backround is more mechanical than electrical, I choose to try a simple mechanical solution first, two throttles. It looks like this;
View attachment 1
I added the BMS Battery kit's thumb throttle up close to the half twist, retaining the half twist for the rear and using the thumb for the frt.. The motion using both is very natural because it's the same motion;

motomech, I enjoy your posts so I opened this thread for the first time. I'm a noob and haven't read much about 2WD bikes yet, but I had a thought (dangerous!) after reading through this. I wonder what would the performance be like if you went back to a 26" tire on the front? How would the power balance be with both motors running at the same time? The rear motor would give you more torque off the line and push you up to speed, and then maybe the front would pull you up to a higher speed and help fight the wind resistance as you go faster. I've seen other bikes on here where people use a smaller tire in back, but I'm curious how that would work with 2WD and how it would affect the efficiency.

anarce said:

motomech, I enjoy your posts so I opened this thread for the first time. I'm a noob and haven't read much about 2WD bikes yet, but I had a thought (dangerous!) after reading through this. I wonder what would the performance be like if you went back to a 26" tire on the front? How would the power balance be with both motors running at the same time? The rear motor would give you more torque off the line and push you up to speed, and then maybe the front would pull you up to a higher speed and help fight the wind resistance as you go faster. I've seen other bikes on here where people use a smaller tire in back, but I'm curious how that would work with 2WD and how it would affect the efficiency.

Click to expand...

Yes, If I was doing it over, I probably would have gone with a 26"-frt./24"-rear, although I think the fast wind in the 26" wheel would be too tall, even with some help from the rear.
When I selected the original components, I had 2 things in mind. Using 2 of the same, simplest kits out there. And changing from 26" wheels to 24" wheels.
The wheel change was prompted by the height of the MTB, it's a 20.5 inch frame and with 5" of rear wheel travel, the seat is way up there[I'm 5'11", but with a short 29" inseam]. It also has a very high top rail, so that even off the seat, I can't really flat-foot it. But what I found, to my suprise, is that it's too low with the 24" wheels, I can't pedal though a corner without the pedals hitting the pavement. part of that problem is due to the soft rear suspension. It's an air shock and as they are want to do, it looses a bit of pressure over time and after a week or so it's back down to 95 psi form where I set it at 125 psi. I have this little hand pump, but it's lame. I recently aquired one of these and now it's easy to keep the shock where I want it;
EDIT: That inflator is impossible to regulate. This is much better;
http://www.amazon.com/gp/product/B002AGI1QW/ref=oh_aui_detailpage_o01_s00?ie=UTF8&psc=1

So to recap, I didn't really like the throttle feel of two identical Q100 kits hooked together by a "one wire" throttle. I don't think I imagined it either as it was the first thing I noticed when I first rode it..
The easiest way I can think of to describe what was going on is to use an analogy comparing an electric motor system to a sound system. We all know the sound reproducing systems do not have a perfectly linear freq. response[not that our ears would like that anyhow]and due to introduced artifacts, there are peaks and valleys and usually,, they are considered distortion. Usually[but not always], the less expensive the system, the more distortion. In this case, one low-powered system doesn't produce anything unfavorable to be noticable, by running the two Q100 systems in sync., the result was to amplify any "distortion" that resulted from the interaction of the motors and controllers.
I don't want to make a big deal out of this. as anything to get the two systems out of lock-step would probably make the issue disappear. It could be running a "fast wind" and a slow wind" motors together would do it, or mixing wheel sizes, or isolating the throttle by using two or simply running one throttle signal though the Cycle Analyst.
But I've moved on and probably won't take the time to re-install the 24" Q100 frt wheel to try out any of these "solutions.
I went back to the MXUS 26" frt. because that is what I had on hand. And while the lowly BMS Battery 15A controller worked fine with the MXUS,

+1 subscribed.

I have one of these controllers, 24v 350W - found the info very handy.

double post

Hey there,
you mentioned that you are running 14s, but elsewhere people say not to take the KU63 over 50 volts. Am I reading right?

Cheers.

Yeah, I did run 14S for a while, it turned the Cute into a tiger[cub] with a top speed around 26 mph.
It's been really hot here[in the high 90's F] and today during a routine inspection I found this;



That's the insulation on the kit supplied phase wire connectors that are starting to melt.
I will be replacing the stock connectors with Andersons or 4 mm bullets, but I figure the wires themselves would be right behind in terms of melting.
No more 14S, although I don't know if this happened on 12S, 13S or 14S.
This is with the stock controller that never really pulls more than 15.5 Amps.

Motor never really gets hot, but the controller and wires sure do.

EDIT-the melted bits was caused by running the fast wind motor at slow speeds, it's this combo. that really heats up the controller.