10 year old lead acid 36V e-bike needing an upgrade!

Thanks for yor replies. There's no physical obstruction anywhere. I have the wheel off now and have connected it to the 12v supply. It spins with the axle in one orientation but when I turn the axle the power quits when I get to about 90 degrees from where I started. Tomorrow I will open up the motor as much as I can.
 
Motor opened up:
P1070528_small.jpg

Closer look:
P1070530small.jpg

In this orientation the motor may just spin with sparks visible between the top brush? and the spinning grooved disc (apologies for not knowing what the parts are called). When I rotate the whole thing around so that the red brush/magnet is in the top 12 oclock position the motor spins ok. But when in the 6 oclock position it won't spin.

It seems to me that the magnet attached to the red wire is not close enough to the spinning grooved disc. Because when it's at the 12 oclock position gravity is helping it get slightly closer to the disc.

There is a spring inside the casing for the little magnet.

Ok, bit more thinking and I reckon I may have it worked out. In the top of the pic there is a strand of wire, loose. Looks like this used to be soldered down securely near the little hole next to it. Think I just need to solder this in place. The equivalent strand on the lower half of the pic (where the black wire goes in) is soldered securely in place.

<edit>
Fiddled about a bit with the magnets. The distance from the spinning disc is obviously critical. I think I've now set this correctly but I'm not all that confident it will stay fixed. Not actually sure whether the magnets should be at the top or bottom of the hub motor...
</edit>
 
The springs may fail if they get to hot. Do they have spring tension ? Yes clean and resolder with electrical solder. Heat first then apply solder after it heated ( hot ). Don't know about brush placement.
 
It is indeed that braided wire causing your problem. Without it's direct connection from brush (carbon block, rather than magnet) to power wire, there is only the slight connection between brushholder box and brush to feed power thru, and it's not enough to work under load.

Resolder that wire to the hole it came out of (the empty one with all the melted solder around it) and the problem should be fixed. Most likely it got so hot inside the motor that it unsoldered itself, from what the solder around the hole looks like.
 
The spring tension was ok. But I was concerned that the 'red' brush wasn't close enough to the spinning disc so I cut up a couple of little square shaped bits of metal, opened up the brush holder and pushed them in forcing the spring down a millimeter or two, in an attempt to improve any loss in springiness.

I have resoldered the braided wire but I may re-do the solder, cleaning it up a bit first and heating, as suggested. I really don't want it breaking again. I guess I may not necessarily be able to prevent the solder from melting again. I am over-volting the motor after all, and the fault developed after about 20 minutes of mostly full throttle riding...
 
One thought, since you are using a higher voltage than original, is that it appears that brushholder can be used to advance or retard the timing on the motor, by slightly loosening those two screws (circled in green below), rotating the whole brushholder board, and tightening the screws.

The catch is you would need a way to measure current into the controller, preferably under load at various speeds, in order to find out if the advance is enough or too much (or if it is accidentally the wrong direction, retarding the timing instead). I don't know which direction it needs to go, clockwise or counterclockwise. Perhaps someone else that has done that sort of thing successfully could tell you. (or a search on the internet will probably find a page on how to advance the timing on brushed DC motors by rotating the brush ring or brush holder).

Anyway, the idea is that advancing it sufficiently could decrease the amount of current required at higher voltages, thus keeping the motor a tiny bit cooler (maybe a lot cooler over the short term).

You might also consider running a thermal sensor into the motor, via the entrance for the mtoor power wires. BBQ thermometers are common enough wya to do this; there are threads on ES about it. THen you'd know when to stop pushing the motor so hard, before it gets damaged again. ;)
 
Thanks for all your replies.

Update:

I removed those two screws pictured about with amberwolfs green rings around them. Then realised that one of the springs in the brush holders (where the red wire goes) had actually lost its springiness. This was obvious after I examined both springs. One of them sprang out to over twice its length but the other one barely sprang out at all. My 'fix' for this was to pull out the spring evenly to match the length of the other one. Hey presto: springiness returned. Ideally of course a replacement spring would be better. If there was a decent hardware store nearby this wouldn't be so hard to do.

So now not completely sure whether the failure was the unspringy spring or the unsoldered braided wire. My feeling is the spring was the fault because the motor didn't work reliably after I had soldered the braided wire but before I had fixed the spring.

I have ordered a couple of different temperature sensors from dealextreme - thanks for the tip. There really isn't much space inside the axle for more wire but maybe the sensor probe wires are thin enough. In the meantime I have reassembled everything and things seem fine (zero load testing). For the moment anyway.

It amused me to think of having yet another handlebar device to monitor. At the moment I ride with a small sat nav GPS device (to give me my speed and trip distance etc), power analyser (mostly to gauge battery capacity used). I have a sneaky feeling that someone will tell me that a cycle analyst would do all 3 of these... :)
 
Wow - just had a really useful PM sent to me with some helpful info. Probably the most helpful being that this bike is a 'Strong GT' or 'Strong GT-S210'. Did some searching around and there is plenty of info about this bike that is going to help me. Like this one:

http://endless-sphere.com/forums/viewtopic.php?f=6&t=34672

I've ended up a bit surprised that nobody else had made the connection to this bike earlier. Up until now I had thought that this was a pretty unique bike and I was on my own a bit as far as upgrading it was concerned.
 
Hey monkeychops, at least you know your brushes don't need replacing anytime soon. The spring and the solder on the braided brush wire must have got too hot. Any way you could put some more metal at that point to carry heat away? Or to vent air thru?
 
My guess is that after the brush wire solder came loose, the spring began to carry current (which it is not supposed to do) and got so hot it lost it's tension. Typically heat in a motor isnt' enough to undo springiness wihtout seeing other major failures as well, often including smoking windings. But passing current thru a spring could do it, if it's thin enough (like brush springs) and current is high enough, plus all the other heat buildup already happening.
 
A couple of lipos arrived. 2 x 4000 mAh Turnigy 6S, 20C. Wired them up in series and they work!

Am going to buy a couple more to get me to 8Ah capacity. That should be enough for a while.

Battery weight will drop from 16.2Kg (SLAs) to about 2Kg, with basically the same usable capacity. And less voltage sag too. Speaking of which, went out on a ride yesterday with my SLA pack (lipos not yet 'broken in'). At the start of the ride the load voltage was about 48V (on flat, and wide open throtle). By the end of the ride, 6Ah and 14 miles later it was down to 40V under roughly the same load - resulting in a significant drop in top speed (about 20 mph down to 16 mph).

Anyway, here they are, looking quite lonely in my spare battery box:
new_lipos.jpg

There is theoretically space for 4 of these packs in each of the 3 compartments you see here (giving 24Ah capacity).

Oh, and btw, I mentioned a few posts back I had bought some lipos on back order thru HK. These are not them! I ended up cancelling the order after waiting about 3 weeks (as dogman predicted!). I believe they still don't have them in stock. Weirdly, on cancelling I didn't get an immediate Paypal refund as one might expect. I was basically given a credit note. Then when I ordered the cheaper lipos I *still* had to pay a small amount towards them (i.e. for some reason I couldn't use all my credit note). So then I had to chase them to get them to refund the remaining money. Quite a hassle.

Yes, never order anything from HK on back order!
 
Plenty of room there for some fun.

It's hard to see in the photos but it looks like those wires that exit the battery box are far smaller in diameter than the lipo ones. They may get pretty hot. Are they soldered or crimped? Bad connections will get hot. Don't fry your family jewels.

Sometimes a second wire in parallel will be a good thing. You want as little volt drop between battery and controller as possible...

You are getting there! Full points for not putting it in the too hard basket on any occasion!
 
Well spotted. Yes, the serial test harness from HK uses 14 awg wire and the battery +/- wires are 10 or 12 I think. The 14 awg wires are slightly thicker than the stock wires that were there before. I won't be pulling more than about 22A, so I'm not too worried. Although I appreciate the thicker the wire the better.

At the point where the wires exit the box, there are ring connectors (crimped). Again this is as per the stock setup (I appreciate the doesn't necessarily make it right). Worth adding some solder to these connections to improve the conductivity do you think?

I've invested too much time and effort to give up now!
Credit to my good chum CB as well for all his help - hello CB!
 
22 amps thru anything crimped is possibly risky. If you can get quick access then ride hard and quickly check the temp of those, and any other components. Heat will dissipate quickly, but its a measure.

Even better to check the voltage drop between battery and controller. Or measure the resistance of the crimps...
 
Hi Sam. Thanks for your suggestions.
I just measured the 2 batteries. Both showing 24.9V on one voltmeter and both 25.0V on another. When I plug them in to the battery box and check my handlebar mounted power analyser it shows 50.19V. So no voltage drop that I can see. Or is this only a valid test under load? Please excuse my ignorance.
Resistance I think I can measure with my multimeter, I will give that a try and report back.
I have a couple of little temperature probes which I can wire up. I could tape 2 of them to the crimp connections and go for a ride.
It's so cold at the moment in the UK, not fun cycling weather.
 
Voltage check sounds pretty good to a mug like me. Perhaps someone else will know?

The main thing is that you have a got a good bike to ride now I guess. Well done!
 
On a test ride yesterday I noticed voltage under load measured on the handlebar power analyser was about 0.5 - 0.7V less than that measured at the battery box.
Measuring resistance between the 2 points (battery box and handlebar power analyser) showed about 1.2 ohms.
The thermometers didn't show any temperature rise. But when I got back from the ride, I noticed the ring connectors where the wires exit the battery box were warm. I wouldn't say they were hot but noticably warm. The batteries the same.
I guess this is all ok?
 
You must log in or register to reply here.

Similar threads

T
Indentify hub motor
Replies
3
Views
443
99t4
9
MojoSlim
  • Question
Old Ping Battery - Is my BMS Busted?
Replies
5
Views
453
amberwolf
amberwolf
sgtkahuna117
I think my scooter lead acid batteries died, I want to go with Lithium
2
Replies
37
Views
4,939
6700scooter
6
R
Evercross Ev65m
Replies
3
Views
1,309
Radek
R
I
  • Question
Project bike from e-scooter and vanmoof parts
Replies
4
Views
1,244
ikawabunga
I
Back
Top