So, my brother wheelied my bike.......

B

benkels

100 mW
Joined
Aug 31, 2011
Messages
40
Location
Devon, UK
pretty hard into the ground catching the rear mech and snapping off doctor bass' torque arms. I should say at this point that I couldn't get the epoxy type that he'd suggested so used some other supposedly special metal formula epoxy. Otherwise I'm sure they would have survived the crash.

So I designed some new torque arms and fitted them and while the bike was out of action I routed the controller on/off switch wires back in through the controller housing and out with all the other wires so I could mount it up near the bars.
I just went to ride it and noticed a little more rolling resistance while pushing it along. Once I got on it it wouldn't put any power through to the motor although the CA fires up. I couldn't see any damage to the hall wires although they must have got a pretty good yank when the motor/wheel broke free of the frame. When I pedal it there feels like it's alternating between higher and lower mechanical resistance. Does this sound familiar to anyone? Any ideas of where to start my fault finding process would be great as I haven't got a clue where to begin. There's also a very quiet click that comes from the controller when I twist the throttle, it may have always done this of course?

It's an HS3540 with halls and an Infineon 72v fitted by the way.

oh, and the CA says it's drawing 6 to 8 watts.

My brother's an idiot.
 
what special metal formula epoxy?

his torque arms should still work again, but sounds like you have a dead motor now.

check continuity and then check to see that the hall sensors are toggling when you rotate the wheel.
 
it's some stuff called Araldite metal that I got at the local hardware store here in England. When you say check that the halls toggle, what do you mean?
 
Also, do you reckon it's a motor type fault and not the controller? Or at least start at the motor and work back to the controller?
 
Doesn't that motor get hot at 72v? What is the amp rating on the controller as well? 40?
 
maybe read some threads on how the motors short out the hall sensors when the phase wires are cut and short to the hall sensor wires. to get up to speed on what happens when you do a twistoff.

the hall sensors turn on (and off) when the magnets pass by them, and when on, they drain current from the controller down the sense wire to the ground in the hall sensor. when the current flows out of the controller, it goes through the resistors in the controller so that on the outside of those resistors there is a voltage drop that the controller microprocessor detects in order to know what phase to switch on and off. sounds like your hall sensors are bad now.

with the controller on and with the 5 pin connector attached to the wheel, put the pins of the voltmeter between ground and each of the hall sensor wires while turning the wheel. the voltages should toggle on and off. obviously put the black wire of the voltmeter on the black or ground of the hall sensors in the 5 pin plug. you can force the probes into the back side of the plug to measure, it must be connected to the wheel!!
 
That's great, thanks for that. I've just been out there and had a proper good look at it and it looks like at least one of the wires has damage to the outer sheath, very slight though. My next problem is removing the side cover. Been trying to work my way round it with screwdrivers and slowly prizing it up but no dice. I've removed the screws from both sides in case that makes any difference. Theres no way to get the bearing out of the middle first either, I'll do a search on side cover removal.

Is it likely to have caused serious damage to the motor or controller?

thanks again.

and trackman417, it does get a bit warm but to be honest I'd only done about 20 miles on it before it got crashed and all fairly short rides, and it is a 40 amp controller, yes.
 
use a razor blade and hammer it in between the cover and magnet ring. all the way around then switch a flat screwdriver. then fix those wires!! sounds like a shorted phase ground. ht hs35 are shit from the factory unless its "newer"
 
benkels said:
it's some stuff called Araldite metal that I got at the local hardware store here in England.
Click to expand...
From what I can find there is one called Araldite Rapid Steel; if that is what you got it seems to be basically like JBWeld, which does not have any real resistance to shear, and wouldn't work for this application very well. But I have never used Araldite of any kind, so I don't know for sure. You might have to ask the manufacturer.
 
Take pics of how it broke.

To choose an upgrade adhesive, you need to know how the previous adhesive failed. For example, if it's still bonded to the surface of both sides, and the adhesive body itself tore, you need to seek different properties than if the adhesive came unstuck from one side or the other, but the adhesive itself did not tear. Or sometimes the failure is an obvious surface failure, like the glue ripping paint off etc.

As a rule of thumb for most situations, the best possible thing you can do for any adhesive application to be stronger is to properly prep the surfaces. Even the strongest adhesive in the world can't stick any harder than some loose sticking crap paint is bonded to the metal, if the adhesive is only bonding to the paint.

That said, I personally drill where appropriate and use fasteners rather than adhesives. However, I do believe adhesives can do a great job if done right.
 
Don't forget to mark both covers and the same place on the magnet backing ring, so you can get the covers back on in the original position to avoid rubbing problems afterward. I like to use a center punch and make one mark on each on one side, and 2 marks on the other to know left and right. I do it when motors first arrive, so I don't forget down the road. Center punch dents work well, because they're visible even after a fresh paint job, and not unattractive like a scratch would be.

John

PS, I hope this epoxy thing works out, because if it's ok for torque arms then we should be able to glue entire bikes together.
 
Pictures are good.

Typically, adhesives really need to be properly cured. It can be dry and hardened, but that does NOT mean its strong.
 
Buy an electric motor tester from ebay- they are a huge help. It'll tell you straight away if any wiring is bad...

http://www.ebay.com.au/itm/260927223352?ssPageName=STRK:MEWNX:IT&_trksid=p3984.m1439.l2649

Certainly sounds like a phase short- see if there's continuity between a phase and the axle. If there is, that needs to be fixed.

Spin the wheel and note any difference in resistance if you connect one phase to another with a piece of wire. The controller should be unplugged for this! What you should get is an equal increase in resistance to spinning when any of the three phases are connected together. If there is no difference to when no phase wires are connected together then that suggests a short somewhere. The HS35 hubs do have quite a documented history of phase wires shorting to the axle.

Good luck!
 
Thanks for all the advice. I've checked the phase wires with a continuity test between the axle and each wire and there's no bleep so it all looks good there. I've ordered a motor / controller tester from eBay as suggested too but it'll take a while from China. I thought it might come in handy in the future anyway.

I wondered if I could remove the side plate by attaching the claw of the puller to the brake disc there by leaving the motor still in its housing and just removing the side cover that I need to get to? All the examples I've found seem to involve pushing the whole thing out and then trying to get at stuff.

Has anyone tried it using the brake disc?
 
I removed the side of my HS3540 by using a three legged puller against the centre of the axle and as close to the centre of the disc as I could get. I did not want to bend the disc!
 
Don't even think about pulling on that disc!

Take the disc off then do whatever is needed to get the side cover off. On my MAC it was as simple as undoing the ?6 screws then gently pulling the cover off- it didn't even need prying.

Did you do the check where you connect the phase wires against each other and look for the change in resistance to turning the wheel?

As has been suggested, check the halls if you can before you pull it all apart. You might have a broken hall wire just outside the axle which means you could probably fix it without taking the hub apart...
 
Right, finally got a minute to do some of the testing suggested by your good selves. I got a short length of wire and shorted between the three phase wires L1 to L2, L1 to L3 etc and all the combinations gave the same bumpy type drag/mechanical resistance when pushing the wheel, it felt the same as when pushing the wheel with all the phases connected to the controller. So is that good or not?

I tested for continuity between the axle and each phase and no beeping so I know that's good.

I then connected the controller, switched it on, placed the black probe of my meter into the black wire of the halls and then looked at the voltage between that black wire and the other four wires (I had the motor connected to the controller too). between the blue to black, yellow to black and green to black the voltage went up when I spun the wheel and then gradually dropped back to zero when the wheel stopped. But between the red to black it seemed to be going all over the place not really settling any where although it did react to the wheel being spun.

So is that the culprit? What says the hive mind?
 
I'm putting the probe into the back of the connector as you suggested to get those readings. Sorry I thought that was clear. I haven't taken the side covers of yet and am hoping I can avoid doing so.
 
if it is a dead hall sensor, the signal does not toggle when you slowly rotate the wheel. you do not have to connect the phase wires to the hub to test it but the 5 pin connector from the controller must be connected and the controller on in order to test them. you had better hope it is only a short in the phase wires, replacing the hall sensors is no fun so you have to know which one is dead first or opening the hub is a waste of time.

the wheel will not rotate once you take the covers off.
 
Okay dnmun, I've done that and I'm not getting any sort of read out from my meter apart from a bit of fluctuation from the probes not getting great contact up inside the back of the connector. So I'm guessing that might mean that the halls have had it?

Thanks for all your help.
 
You must log in or register to reply here.

Similar threads

jetpackjbd
    • Like
1000 Miles In - JoyKie 350W 36V Mountain Bike Review (Risky Cheap Chinese Bike Purchase)
Replies
10
Views
587
2old
2
S
Hello
Replies
3
Views
1,084
amberwolf
amberwolf
E-HP
Searching for a spare/upgrade controller
2 3
Replies
58
Views
6,283
999zip999
999zip999
D
52V battery now 20V - dead?
Replies
12
Views
1,084
amberwolf
amberwolf
T
Phaserunner V1 issues - Maybe you can help?
Replies
0
Views
329
thunderstorm80
T
Back
Top