Controller Stutter Issue - Help Needed!

TheEBikeNinja

TheEBikeNinja

1 mW
Joined
Aug 23, 2015
Messages
15
Location
UK
Hey guys, so I've just bought this sensorless controller:

http://www.ebay.co.uk/itm/191549381815

And it runs fine no-load. Lovely.

When it has to push me on my bike however, it stutters under acceleration.
If I pedal up to speed first, then increase throttle to match, and stop pedaling, then it will keep pushing me.

So yeah, just no acceleration.
I've attached the data sheet for the controller.

MI2BqCkjefpVuY_56yJkTQ6jkdUPpvAt6eaRmfAW2i8=w1080-h1440-no


It's powering a bafang SWXH from 8S LiPo.

Any suggestions?
 

Attachments

  • JY01_V1_20141202(无霍尔).pdf
    961.5 KB · Views: 313
Not every sensorless controller will run every hub motor.
Geared motors, with their reduction ratios, are especially a problem.
To simplify the problem, the gears have the motor spinning so fast that the controller can't sync up.
It sounds like that is what is happening to you.
Do you have a bad Hall?
 
that controller is not even wired up properly. where is the negative lead P- from the BMS? why is that capacitor soldered onto the back side of that pcb?

why is it there? it has no function and looks like someone just soldered it on for the hell of it, but with no ground on the controller in the lower right corner that motor should not run at all.
 
The negative battery connection is disconnected for safety while taking the photo.

The capacitor is connected to the throttle input, so the throttle can ramp slowly, which works.

Yes, the hall sensors failed, so I removed them.

I agree that the synchronisation seems to be failing, but only under load.
 
That link didn't work, but I know the drill well. Buck, studder, die, studder, buck, etc.
I have tried to run three gear motors sensorless and had no joy on two.
The MXUS, with a pole count of 4 pairs(8)and a reduction ratio of 5 :1 worked fine, but
none of my Q100's or my Ezee geared motors will run sensorless.
The Q100's pole count is 8 pairs(16) and the reduction ratios start at 8.5 : 1 and go up from there.
The ratio on the Ezee is only 5 : 1 , but the pole count is really high, 16 pairs, if I recall.
I looked for that Bafang spec.s, but didn't see them without a advanced search.
I have tried various controllers, elifebike, bms B., infineon and Lyen with the same results.
I have come to the conclusion that different controller don't "read" the commutations better than others. It's pretty much a go/no go situation.
I say this in the hope that it might save you some money by not buying more controllers.
Maybe just order another motor and swap cores into your whl. ASM.
Good luck
 
What's strange is I tried powering it from an RC controller, and it works OK (at a much lower power level). It pushes me along at 5mph, and there's no stuttering or stalling at all.

So I don't think it's the pole count/commutation frequency causing issues.

It occurs equally for all throttle values, almost as if it loses sync with the motor when trying to accelerate.
It uses an LM339 comparator for EMF sensing, so maybe that's where the problem lies?

This link actually works:
https://youtu.be/AtIL7tndPN0
 
RC controllers are generally intended for very high RPM sensorless operation, so it's not surprising the RC controller works where the ebike one doesn't.

But if it doesn't work with *any* of those controllers, on any of your motors, there may be a more basic problem, though other than connectiosn I don't know what it might be.
 
he should just buy a decent size controller. it is only about 100W so he needs something larger. he may be getting overcurrent cutouts as he tries to make it do more than it is designed for.
 
The thing is, If I pedal up to speed, and then slowly increase the throttle to that speed, and stop pedalling, then it pushes me along at 20mph.

At that speed, it's using about 7-8A at 30V, so it doesn't seem to be a current issue. (It stutters just as easily at 5mph, drawing 2A)

It seems to be just that any large difference between the speed it's currently spinning at, and the throttle value, throws it out of sync.
I presume there's some sort of feedback loop between the throttle, the commutation, and the back-EMF sensing. So that seems to be what's failing.

Perhaps if I adjust the feedback threshold/gain, that might shed more light on the problem.
 
that controller looks like it was shorted out and destroyed since the previous picture was taken.

is there any way to know what the condition of the controller was when this behavior started?

how much is not known by us? i no longer have confidence we know anything related to the behavior described since the pictures are not representative.

why not remove that capacitor from the throttle input if that really is the throttle input which i doubt since there is no throttle wire attached.

actually, nothing really matters now with so much destroyed.
 
d8veh said:
Did you try what I suggested?
Click to expand...

Unfortunately, I don't have any other large capacitors that are rated for that voltage. I shall try that soon, when I get one, but there's already 300uF on the supply, which is a Turnigy Multistar Pack, so I'm not sure if that will help, especially if current limiting is disabled.


dnmun said:
that controller looks like it was shorted out and destroyed since the previous picture was taken.

is there any way to know what the condition of the controller was when this behavior started?

how much is not known by us? i no longer have confidence we know anything related to the behavior described since the pictures are not representative.

why not remove that capacitor from the throttle input if that really is the throttle input which i doubt since there is no throttle wire attached.

actually, nothing really matters now with so much destroyed.
Click to expand...

I have provided multiple photos, a link to where I bought it (where there are more photos), the data sheet for the controller (which includes a full, labeled circuit diagram), and a video of it in 'operation' (or lack thereof).

I really don't know why you seem to doubt my credibility, or what I have shown. The controller behaved like this out of the box. All I have done is soldered some MOSFET traces thicker, which has not changed the behaviour, and which I did before taking any photos or asking my original question.

Nothing is shorted, or damaged via heat/current/voltage, and my video clearly shows how I have connected the throttle and motor correctly.
 
that controller is damaged. it is obvious when you look at the picture of the controller in it's present condition and not the picture you put up originally.

below that capacitor you soldered to the edge of the pcb is a series of surface mount capacitors and resistors that are blown up.

the short seems to go all the way to the ground on pins #2 and 3 on that 16pin IC below those blown up parts. it looks like that IC is damaged too since there appears to be melting of the plastic package right above the ground on the package.

imo they are the high side charge pump circuit and the base resistor for the hi side gate drive transistor. that would totally account for the loss of the phase wire current.
 
Those components are not blown up. They are a few low voltage power smoothing capacitors, and a few potential dividers to set over-current and startup torque for the JY01 control chip.

However, that bike has just been stolen (combination locked through frame to massive lamppost - someone must have cut it).

Fortunately, none of my batteries, controller, or motor were on it at the time, so I will continue to get those working properly for when I eventually get a new bike.

I am removing the comparator and drive chip, and building the RC controller into the existing MOSFET drive circuitry. I've flashed it with BLHeli, and disabled low rpm/high torque protection. It should work.

The bike was stolen while I was in a shop, buying a 63V 2200uF capacitor. So I guess I'm fine for capacitance.

:(
 
motomech said:
Not every sensorless controller will run every hub motor.
Geared motors, with their reduction ratios, are especially a problem.
To simplify the problem, the gears have the motor spinning so fast that the controller can't sync up.
It sounds like that is what is happening to you.
Do you have a bad Hall?
Click to expand...

motomech said:
Not every sensorless controller will run every hub motor.
Geared motors, with their reduction ratios, are especially a problem.
To simplify the problem, the gears have the motor spinning so fast that the controller can't sync up.
It sounds like that is what is happening to you.
Do you have a bad Hall?
Click to expand...
I have exactly the same problem with my geared hub, without load it runs well but under load it starts to stutter at higher speeds, it's a KT controller, could it have something to do with the erpm? Since it is a very fast winding, which controller could be suitable for it?

Thanks
 
For reference:
endless-sphere.com

geared Motor problem

hello guys, i have a problem with my newly arrived mxus xf19 geared motor, the motor runs unloaded without any problems but when i drive, it starts to jerk a lot at high speeds, as if it were misfiring, and at full throttle it cuts off briefly (only the performance, display stays on) It's...
endless-sphere.com endless-sphere.com
in relation to the above post.
 
You must log in or register to reply here.

Similar threads

N
Help: Motor Suggestion
Replies
2
Views
814
larsb
L
P
2014 S-Pedelec Lishui controller replacement
Replies
4
Views
871
PeterT12
P
T
*USED FOR SALE* Grin Technologies 52V eBike DIY Build
Replies
2
Views
1,389
tomceg123
T
A
we are having an issue getting our Kelly controller to work.
Replies
2
Views
785
alexis.persinger
A
C
Handbike Rebuild - Controller help needed please
Replies
2
Views
397
skeetab5780
skeetab5780
Back
Top