Help!! pos short circuit on electric bike

[roaringfire]

I have a Pedego Umbrella 16" bike and through a pretty random thing the wire going into the wheel got ripped through and the wires exposed and I think there might have been some kind of short circuit as once I'd soldered the wired back together the geared motor is not turning. When I switch on the ignition the lights come on on the throttle. Now I'm wondering whether the controller might have been damaged or whether the small circuit board in the wheel might have been affected? Trying to figure which to try to replace? Would two wires connecting that have come out of the controller still be able to damage the controller itself? I'm very much a novice.

It's very frustrating and any help would be gratefully received.

Thanks

 

[RallySTX]

A bit more info might help out here. Pics of the damage and bike in general wouldn't hurt either. What exactly happened to the bike? Was it turned on at the time. I'm no expert, but I will do what I can too help.
Pedego site.
http://www.pedegoelectricbikes.com/index.php?option=com_content&view=article&id=35

Brian L.

 

[dnmun]

brushed or brushless, do you have a voltmeter? pictures.

 

[dogman dan]

Cut wires at the hub can blow controllers. So you could be out of luck this time.

 

[roaringfire]

Many thanks guys for the quick responses.

Basically I was transporting he bike in parts from the US to the UK and got caught in the blizzards a couple of weeks ago so instead of neatly carrying the bag out of a car I had to walk up a long snowy hill with the bag on a trolley and as my bad luck would have it the bit where the cable goes into the powered wheel ended up touching the snowy ground and scraped right through the cable. Very random and unfortunate. Anyway I managed to solder it all together and it worked but then someone carelessly moved it and the insulation between the cables were loosened and when it started next it suddenly cut out an from then on no power to the wheel. I've re-soldered it all back together.

The bike has a geared rear hub motor that I assume is brushless (though am a bit of a novice), it has a 36v battery and 250w controller (spec is supposed to be 350w and think previous user swapped controller for some reason).

To me it can be one of three things. Either I've blow the controller (if so can this be fixed with replacement part and a soldering iron?) or something has blown in the wheel motor, or in the process somehow one of the wires going through the wheel has split internally under the strain of being fixed. I've gotten inside the wheel and all the cables going into the motor all seem fine and there is no signs of anything that may have blown.

I don't have a voltmeter, if I did would I use it to check the cable going into the wheel to see if current is reaching it? And if there is would I be right in assuming then that the controller would not be the issue?

I'd imagine replacing the controller would be the easiest to fix if that's the problem? Also could give me a chance to put back in a 350w controller which is the original spec. If so can I use any 36v 350w controller with the right cables to connect as there are some that are not too expensive or are they very manufacturer specific?

Many thank for taking the time to help. (btw too dark in the shed too take pics tonight)

 

[RallySTX]

The link I gave you will provide a contact us email link, that you can use to ask for technical data if available. I looked all over the site for data on your folder but found nothing. You can buy a volt ohm meter, or DMM, digital multi meter, at most hardware, auto parts, or electronic supply places. They can be had for less than ten bucks. You don't need a very expensive unit to tell you the basic values of the things you need to check. Sometimes the controller still sends voltage to the wheel even when it's toast. You need to narrow down the specific model of controller and motor, as well as the type. Then the guys here can offer their expertise on what to do. I have learned that if the motor is a hall sensor equiped unit, you must spin the wheel and check the wires for pulses, to determin the state of the sensors. There is much to learn here. Sorry you had to learn the hard way about the care that must be taken to prevent damage to the wires attached to the motor. You may yet get lucky. Could just be a fuse somewhere. Make sure the battery is charged, and all the connections are good and dry. Best of luck.
Brian L.

 

[John in CR]

If you're lucky it's just a wiring issue in/near the axle, and no matter what that harness needs to be redone to avoid continuing problems. Pop the motor cover off and get pics first. It's a common issue and even a broken controller can be repaired if you're electrically competent. Motors typically aren't easily messed up, though a short could burn a hall sensor if it's a brushless motor.

 

[roaringfire]

Thanks, here are a series of pics. If I can pull this off I'll be so chuffed. Where do I find fuses within this chain? Thanks guys

I've just temp insulated the parts of the cable soldered back together so please don't be horrified over how bad it looks

http://picasaweb.google.com/roaringfirefilms/BIKE?feat=directlink

 

[roaringfire]

Well I've gotten a volt meter and established that the 36 V is going in to the controller fine. There are three main wires coming out of the controller yellow, blue and green that I assume carry the current though could be wrong (and five lessers wires) and when I connect the volt meter in any combination between the wires and turn on the throttle no current shows, so sounds like the controller to me. Any thoughts? And in the controller photo is there any component I can replace that could fix this?

As much as a pain in the back-side this is it's kind of a fun challenge figuring it out (thanks for the help)

 

[Jeremy Harris]

Ok, what you have is a fairly common brushless three phase hub motor. First some basics. The three fat wires that are yellow, green and blue are the three phase power wires from the controller to the motor. When the motor is running these will have what looks like alternating current on each, so you can't measure anything with a voltmeter by way of solving your problem.

The five thinner wires are sensor connections. Three of them (again coloured yellow, green and blue) are the outputs from Hall effect position sensors inside the motor. The red and black wires are the power to the sensors, usually around 5V (red is + black is -). The purpose of these sensors is to tell the controller the exact position of the rotor in the motor, so that it can keep the three phase AC in the phase wires in sync with the motor rotation. You can do some simple diagnostics on these sensor wires.

First, check that you have around +5V or so (it maybe anything up to maybe 8 or 9V, but 5V is typical) between the red and black wires at the motor. If you have, then that means that the controller is at least partially alive. Next, put the meter between the black wire and each of the three sensor outputs, the thin yellow, green and blue wires. If you power up and gently turn the motor you should see the voltage on each of these wires go high then low, then high again as the sensors switch. If all three sensors are switching, then the good news is that the motor looks as if it might be OK.

Next, check the motor with the controller disconnected, to see if the three phase wires (the thick yellow, green and blue wires) are OK. At the controller end of the motor cable, with your meter set to the resistance range, check between each of the wires in turn (yellow to green, green to blue, blue to yellow). Each should show a very low resistance if they are OK. If any show a very high resistance then the chances are you still have a broken wire somewhere. Next, to make sure that the phase wires aren't shorting to the motor case inside the axle, connect the meter to
each phase wire in turn and the case of the motor. You should see a very high resistance. Virtually zero resistance indicates a wire that's shorting.

Finally, with the power off and the battery and motor disconnected, and with the meter on the resistance range, check between each phase wire on the controller, as you did for the motor. The resistance should be high, if it's very low then you have a shorted FET in the controller.

Hopefully this will narrow down the possible fault. I gather you're in the UK, whereabouts? We have a few UK forum members, one of us might be near enough to help out.

Jeremy

 

[roaringfire]

Hi Jeremy

This is extraordinarily kind of you to right such a great step by step response. Most importantly you do so with great clarity as it sounds easy to follow. I look forward to trying it out in the morning when I can see what I'm doing (no light in the shed). So does the three phase AC you describe mean that the controller converts the DC to AC on the way to the motor?

I'm in Edinburgh till Monday then up to Aberdeen for a few weeks. Visiting family here for a few weeks. Spend most of my time overseas these days, mostly Los Angeles.

 

[Jeremy Harris]

Hi Jeremy

This is extraordinarily kind of you to right such a great step by step response. Most importantly you do so with great clarity as it sounds easy to follow. I look forward to trying it out in the morning when I can see what I'm doing (no light in the shed). So does the three phase AC you describe mean that the controller converts the DC to AC on the way to the motor?

I'm in Edinburgh till Monday then up to Aberdeen for a few weeks. Visiting family here for a few weeks. Spend most of my time overseas these days, mostly Los Angeles.

No problem, glad to try and help. It's a pity you're so far North, I'm not aware of any forum members in your neck of the woods.

You're right about the controller, it does pretty much exactly as you say, turn the DC from the battery into AC at a frequency that matches the motors rpm and at a voltage that is set by the throttle position. Increasing the throttle increases the voltage to the motor, which makes it spin faster. As the motor spins faster the Hall sensors inside it give feedback to the controller to increase the frequency of the three phase AC feed to it. The amount of current the motor draws is also limited by the controller, it senses the battery current and will reduce the voltage to the motor if the current rises above a set limit.

Those tests I've suggested will just get you to a point where we have enough information to be able to narrow down the possible causes.

Jeremy

 

[roaringfire]

Through the process of elimination. Small black and red wires in motor registering about 9.5V so look fine

(yellow to green, green to blue, blue to yellow) to motor. all okay

"put the meter between the black wire and each of the three sensor outputs, the thin yellow, green and blue wires." the yellow wire goes lower and higher as I turn the motor but with the other two it seems to stay around the same. Though the motor doesn't spin that freely (not sure if the cold in the garage makes it not so smooth) should it spin easily or is there a lot of friction when it's not operational? Also it's hard to get a good connection as the solder joins for the wires to the board are so closely connected. The readings are 8.0 for the yellow and blue wires and about 9.9 from the green.

Crucially in the yellow green and blue coming out of the controller when connecting the combinations of them the resistance drops from 1 to about .019 so I guess from what you're saying there is a short in this part of the controller. Since it looks like a shorted FET is this something I can solder out and replace? If so what should I be looking for on the controller board? Looking at photo's online and my pics of the controller board and don't spot anything on the board that quite looks the same?

Many thanks again

Steve

 

[Jeremy Harris]

That's useful information, thanks.

It looks as if two of the Hall sensors (the green and the blue ones) aren't working properly, either because of a wiring fault of possibly because the broken wires managed to blow two of the sensors in the motor. Also, it very much looks as if at least one, most probably two, of the FETs in the controller may have blown. That low resistance of around 0.019 would seem to indicate that to be the case.

It seems likely that you actually have two faults then, one in the motor (the two Hall sensors that don't seem to be working) and one in the controller (blown FETs plus possibly one or two other parts).

The controller can be repaired, but if you aren't that familiar with electronic repair then you might find it a bit difficult. The FETs are six of the thing screwed to the alloy bar at the bottom of your photo (the seventh will probably be a voltage regulator for the controller). You need to find the blown ones (a meter set to measure resistance between the pins will show which are blown, this thread has some useful tips: http://endless-sphere.com/forums/viewtopic.php?f=16&t=19645).

Getting hold of new FETs means looking at the part number printed on the existing ones and then going to a semiconductor suppliers website, like RS Components, Farnell or Digikey, and buying some replacements. My guess is that you'll be looking at a cost of maybe £5 to £10 inc postage for a couple. If you have a problem identifying them, post the numbers you can read on them and I'll have a go for you. Replacing them is a matter of unsoldering the old ones (often quite difficult, the best way is to cut the leads off the old ones, remove the FET and then unsoldering the remaining stubs of lead one by one). Once you have all the excess solder off the board (use either a solder sucker of some solder wick to get rid of the excess) and the holes clear where the new FETs need to go, then it's just a matter of screwing them up to the heatsink bar (to make sure they are aligned) and then soldering them in place. Make sure the insulating washer and pad are OK and that a small smear of heatsink paste is applied to the face of the new FETs.

Replacing the Hall sensors is a harder job, as they are quite well buried inside the motor. If you use the search function of this forum and search on replacing Hall sensors then you will find some threads where others have done this. You can buy Hall sensors from the same suppliers that you get the FETs from, they are usually Honeywell SS41 or SS411, either will work OK. They are usually around one or two pounds each. Replacing them is fiddly, but not hard. It does often need some fine soldering work though.

Jeremy

 

[roaringfire]

Great, thanks.

I can get to the Hall Sensors okay, just they're quite firm in there so a bit of digging out to do. There's nothing that I can see written on the FET's I'll scan the websites and see if I can find anything that looks like a match. Might just buy a new controller as there seems to be one on ebay with the same connectors and spec that I need. The one that blew was a replacement I think put in by the person I bought the bike from as its a 250W controller whereas the spec of the bike is 350W so I might just buy one at 350 and get a bit more bang for my buck on the bike.

I'll do some online research and give you a shout if I hit any brick walls.

You've really been a star with this, thank you so very much.

 

[Jeremy Harris]

Sounds like the Hall sensor replacement should be reasonably straightforward on your motor, which is good news. A useful tip is to take some close up photos of the orientation of the old Hall sensors before you remove them, as they are often fitted with the centre one flipped relative the the outer ones. The new ones need to be fitted in exactly the same orientation as the ones you take out.

A new controller is a sound move and probably an awful lot less hassle than repairing the one you have. To be honest, given the price of the things they aren't worth spending the time on repairing, unless you're doing it for fun or can't wait for a new one to arrive in the post.

Good luck with your repairs!

Jeremy

 

[roaringfire]

Many thanks again

 

[roaringfire]

Hi Jeremy

A quick question. I've finally gotten back on to fixing my e-bike as I was waiting for the controller I'd ordered. I replaced the hall sensors easily enough and plugged it all back in to test it but there seems to be no power going into the throttle on the handle-bars as it is not lighting up and the multi-meter shows no current going into the cable that the guide indicates is for the throttle. (the cables are slightly different and there are a some additional ones to the ones I need). When tested the only one showing the 36V current is the cable marked as the Alarm System power. My bike doesn't have an alarm system so wasn't going to use these wires but is it required for these wires to be connected to something to make a circuit for power then to be sent through the throttle cable?

I hope all is well with you and thanks again for all your help.

Steve