Idiots guide to rewinding Revolt 120 pro

How it drives the bicycle forward, what the gearing is, how many volts you run, what wh/km you see.
Just general info, seems like a nice commuter and its a shame the motor doesn't work.

I did use my revolt for maybe 1500km, but changed due to the the bad efficiency after that.
I still liked the motor, but I like the hubs better for the particular application.
Hoping your rewind turns out great as well, and that the motor changes to the better.
 
Gearing is from motor to the normal freewhel on the rear hub. I milled the sprocket to fit shimano freehub splines.
It uses moped/420 chain, derailleur had to be adapted to fit this width with longer bolts and new/wider idler wheels.
Gearing 8/53
72V
I think top speed is something like 50km/h at full battery.

The ride is actually really nice, it's like an american car with soft suspension and automatic gearbox (since only one gear) :D
Wh/km ? don't know, didn't measure. I am sure it is really poor with the revolt motor..
Battery pack is 2x100 Samsung 25R 18650 cells with BMS

Once the motor is OK then it is a super successful build!
 
larsb said:
I REALLY hope the wind turns out OK. Otherwise i think i'll have to buy another motor - or stop the ebiking. This is taking too much energy. I'd love to work with a quality motor instead of this piece of sh*t.
Click to expand...

i can understand your feelings, but why should it not work?

from myself i know that it can be really time consuming if you going to make someting good out of crap, but you safe a lot of money (a comparable motor probably costs 5-10 times as much), and on the other hand your motor is unique 8)

thats why i like DIY projects -> you put your brain and skillfullness into something for having something that is superior to products everyone can buy.
 
Exciting. You are on the final lap now, closing in on the last stretch. I really look forward to see this custom revolt come to life. It will be very cool to see how the new outside hall sensors works compared to the stock ones. Plenty extra of copper and better timing should mean a totally new experience and motor should handle much more current without turning it all to heat.

I must say I envy your stamina, you just keep pushing forward. If I would have taken on such a task I am pretty sure after a few completed turns the motor would have met the garage wall, or ended up in the scrap pile. Lets hope Revolt is paying attention to this thread and that they are being a good ol boy and are taking notes. Maybe we will see improvements on their side as well in the future.
 
The rewinding is pretty and that square magnet wire is sooo sexy :p.

How glad am I that this isn't my project? Extremely.
 
kelly KLS controller has 12v hall signals, that's the reason the encoder did not work. I've searched for a way to shift 5v signals from encoder to 12v and found this board:
http://m.ebay.com/itm/251604416861?_mwBanner=1

The connections to the board and dc/dc converter are done, quite sensitive work.
image.jpeg
I'll do the connectors during easter holiday since i'm on business trip next week. Stay tuned!
 
I hooked up everything to the RLS encoder and IT WORKS LIKE A DREAM!!!!

Bike pulls like crazy, wheel lifting acceleration, all the vibrations are gone :D I am so psyched, i even got this error message when posting:
error msg.JPG
I can actually say that the motor would be OK for my use already with the poor copper fill. So how it will be with 2,5 times the copper.. we shall see.

Safe conclusion from this is that poor sensor function is root cause of the stalling, vibrations and heating issues i've had. Probably a combination of the hall placement deep in the slots with interference from coil fields and/or the hall wiring getting interference from coils.
Only remaining for the sensor is to waterproof it. I'll have someone 3D-print a cover and then i'll fill it with silicon potting compound
encoder cover.JPG
 
That's Awesome! The hard work has been repaid. Nice
 
I hooked up everything to the RLS encoder and IT WORKS LIKE A DREAM!!!!

Bike pulls like crazy, wheel lifting acceleration, all the vibrations are gone :D
Click to expand...


This is done on a stock motor? Where only the halls are swapped for the RLS encoder?
Wow. Any chance you could do test of the motors performance?
 
Hold your horses there macribs, i must correct myself:
Before the encoder i ran 20% on the Kelly, 60A max and got stuttering all the time starting from 35A.
now i changed the phaseA to 66% or 200A and it works. That is the main reason for the power and happiness :D
I did a little riding and motor didn't get more than fingervarm from acceleration and some quite steep but short hills (100m elevation)

I notice that the vibrations are gone as well and comparing with the old screaming sound the motor is running smooth now.
Until i have some waterproofing done i cannot ride more.

unfortunately i haven't got any dyno and i don't like to give people subjective data. What i can say is that if motor had been running like this from the beginning then i wouldn't have started the rewind. That doesn't mean that motor is OK compared to the Revolt specs, only that my power needs are fulfilled as motor is now.
 
I'm very pleased to hear that you found the solution for the problem. :) This is another proof of how important it is to have good positioning signals.
On more expensive motors i have often seen little capacitors and pull up (or down?) resistors on the circuit borad for the halls, and shilded wires against noise coming from the windings and phase wires. maybe the addition of that parts would be helpful too for that motor.

Let us know how that encoder works, and please share pics of the attachment.
 
Attachment of encoder on the left side motor support. It's hard to get the overview from the pic..
I use the crank arms to support the motor (no pedalling), encoder sits on the bracket and can be adjusted to get phase shift by rotating it. I will try that next time i have access to a scope.
image.jpeg
 
Larsb, your motor woes sound similar to my 160 short so am keen to try your encoder route. Can you advise the version of the rmc22 you are running as there are a heap of options when you go to order (eg. I assume 7 pole pairs?). Thanks heaps as I thought I had hit a brick wall and have been watching this thread with much interest.
 
Encoder part number is RMC22UG12BAA10 for the one i bought. If the rv160 is 7 pole pairs then this encoder will work also for you.

The encoder is quite dependant on installation tolerances (distance from magnet to encoder body specified to 0.3mm, also concentricity) and it is a lot more expensive than placing outside hall sensors. the big advantage is that the signals are perfect between them, no offset introduced by positioning tolerances, no stray fields will affect it

Outside hall sensors will be my preferred route to go for the rewinded motor. More on that to come later in the thread if you can wait. or by all means: claim a warranty replacement by Revolt.
My motors are so bad they should be replaced under warranty if i hadn't already started to improve them
 
Took it to work today. It runs really well now, I think i'll try to set the kelly att 100% (300A) just to see how it holds up for heat. Shouldn't be possible considering the original poor winding but as it's road driving and only bursts to full speed it might work.

I will connect a 2 speed switch to the Kelly so i can run slower to work and full blast when i am on leisurely business :D
 
You must log in or register to reply here.

Similar threads

U
BLDC Motor Rewind
Replies
1
Views
485
SafeDiscDancing
SafeDiscDancing
R
    • Like
Really good article comparing Bafang vs. Tongsheng vs. CYC
Replies
12
Views
10,409
999zip999
999zip999
mxlemming
Building a motor
2 3 4 5
Replies
106
Views
7,435
mxlemming
mxlemming
sn0wchyld
Same debate again, motor windings!
2 3 4
Replies
96
Views
9,613
Hummina Shadeeba
H
M
DIY Spot welder power- limiting factor?
Replies
9
Views
5,291
SniperGaulois
SniperGaulois
Back
Top