My Torq has died

imellor

Pedelecer
Oct 25, 2006
67
4
Hi all

Apologies, this should probably be in the "Technical" section, but my account will not let me create new threads in this area for some reason. I have emailed the mods, but I have not received a response yet.

My Torq started to judder and I traced the fault to a broken power wire. This was near to the top of the front forks and due to the wire being twisted backwards and forwards when steering.

I have joined the cable together and checked the other power leads and the thin sense leads, these all look OK visually.

However when I turn the throttle the wheel attempts to move and then I receive an 8 LED flash error. Which according the Flecc's site is "Hall sensor in motor faulty"

If my memory serves me correctly, I did read on this forum, that this can sometimes be caused by a faulty throttle and the connections on my throttle are a little corroded. So I was wondering, it is possible to short out a couple of the pins on this connector, to simulate the throttle being at full power (Assuming it is a potentiometer).

Other than that any ideas will greatly appreciated.

Ian
 

flecc

Member
Oct 25, 2006
53,559
30,848
This is the right section Ian, the Technical section intended for more permanent items.

Those controller signal causes are only the main ones specified by the manufacturer, the 8 flash signal can also be due to a connection problem.

The throttle isn't a potentiometer, it uses a Hall effect semiconductor with a magnet passing it to give the variable signal from 0 to 5 Volts, and I think it's unlikely to be the cause. However, you won't do any damage by shorting across the connecting wires to get the full 5 Volt signal. I can't say which wire is which offhand and the colour codes vary anyway, so just experiment by finding the 5 Volt lead and jumping it to the other two in turn.

Since you've had the one lead failure, one of the eight leads from the motor to the controller could be the trouble, but the motor Hall sensors can't be ruled out as the cause.
.
 

Tiberius

Esteemed Pedelecer
Nov 9, 2007
919
1
Somerset
The throttle isn't a potentiometer, it uses a Hall effect semiconductor with a magnet passing it to give the variable signal from 0 to 5 Volts, and I think it's unlikely to be the cause. However, you won't do any damage by shorting across the connecting wires to get the full 5 Volt signal. I can't say which wire is which offhand and the colour codes vary anyway, so just experiment by finding the 5 Volt lead and jumping it to the other two in turn.
Hold on. I'd be very careful here. Are you thinking of doing this with the throttle still connected, Tony? I'm not sure this is completely risk free. Even with the throttle disconnected, the controller may not like the 5V being shorted to ground.

If you have a voltmeter, you can work out which lead is which - 0 V, 5 V or throttle output, this one should vary between 0 and 5, or 1 and 4 V as the throttle is turned.

But from what you describe, it sounds more like a Hall in the motor, or the wiring to the motor. You can check these, but its a bit fiddly - what test equipment have you got?

Nick
 

imellor

Pedelecer
Oct 25, 2006
67
4
Hold on. I'd be very careful here. Are you thinking of doing this with the throttle still connected, Tony? I'm not sure this is completely risk free. Even with the throttle disconnected, the controller may not like the 5V being shorted to ground.

If you have a voltmeter, you can work out which lead is which - 0 V, 5 V or throttle output, this one should vary between 0 and 5, or 1 and 4 V as the throttle is turned.

But from what you describe, it sounds more like a Hall in the motor, or the wiring to the motor. You can check these, but its a bit fiddly - what test equipment have you got?

Nick
I have a multimeter, the problem being how do you get to the wires to do an end to end continuity test? I'm guessing you can get into the controller and test to the joiner on the forks, but how about from the fork joint to the motor?

I have also pulled the sense leads at the fracture point and they didn't break under a fair bit of stress.

Ian
 

Tiberius

Esteemed Pedelecer
Nov 9, 2007
919
1
Somerset
Ian,

You can't easily get at the ends of the 8 wires from the controller to motor, but you can still check things. (I'm basing this on my Torq 1; newer models may be different.)

There is a joint between the motor cable and the controller cable on the forks, and its not exactly mil spec. You've already have trouble with the cable, but it sounds as if that was on the controller side of the joint. I wouldn't be surprised to find more troubles along that cable. When I opened mine up, I found corrosion along the wire and had to cut it back to half way down the cross tube.

You could open up the joint for a clean and inspection. While you're there you can check some things with the multimeter. Note that the connections inside the joint may not be blue-blue, yellow-yellow, etc so make a record before you disconnect anything.

The 3 thick wires are the motor phase. They switch between 0 and 36 V with a complex 3 phase waveform. You won't make sense of this with a multimeter. But with the power off, you can measure the resistance between the wires. This is the winding resistance of the motor and the reading should be low between any 2 of the 3 wires - don't know the exact value, but it will be Ohms rather than 10s of Ohms.

The 5 thin wires are for the Hall sensors. There's 0 V (probably black) 5 V supply (possibly red or brown) and the 3 signals back (blue, green, yellow). With the power on and the throttle closed, check the voltages. If you turn the wheel backwards slowly the signal wires will switch between 0 V and 5 V. You have to go slowly; I think its 86 cycles per turn of the wheel on the Bafang motor, so its not easy to see on a digital meter. Its a bit fiddly to do this without accidentally shorting the wires, so you may want an assistant and some bits of insulating tape. That will tell you if the Hall sensors in the motor are working.

I replaced the joint with connectors. In fact I used the same connector system on several bikes and controllers. Now I can test things simply by parking them close together and connecting the controller of one bike to the motor of another.

HTH

Nick
 
Last edited:

torrent99

Esteemed Pedelecer
Nov 14, 2008
395
36
Highgate, London
Ian,

You can't easily get at the ends of the 8 wires from the controller to motor, but you can still check things. (I'm basing this on my Torq 1; newer models may be different.)

There is a joint between the motor cable and the controller cable on the forks, and its not exactly mil spec. You've already have trouble with the cable, but it sounds as if that was on the controller side of the joint. I wouldn't be surprised to find more troubles along that cable. When I opened mine up, I found corrosion along the wire and had to cut it back to half way down the cross tube.

You could open up the joint for a clean and inspection. While you're there you can check some things with the multimeter. Note that the connections inside the joint may not be blue-blue, yellow-yellow, etc so make a record before you disconnect anything.

The 3 thick wires are the motor phase. They switch between 0 and 36 V with a complex 3 phase waveform. You won't make sense of this with a multimeter. But with the power off, you can measure the resistance between the wires. This is the winding resistance of the motor and the reading should be low between any 2 of the 3 wires - don't know the exact value, but it will be Ohms rather than 10s of Ohms.

The 5 thin wires are for the Hall sensors. There's 0 V (probably black) 5 V supply (possibly red or brown) and the 3 signals back (blue, green, yellow). With the power on and the throttle closed, check the voltages. If you turn the wheel backwards slowly the signal wires will switch between 0 V and 5 V. You have to go slowly; I think its 86 cycles per turn of the wheel on the Bafang motor, so its not easy to see on a digital meter. Its a bit fiddly to do this without accidentally shorting the wires, so you may want an assistant and some bits of insulating tape. That will tell you if the Hall sensors in the motor are working.

I replaced the joint with connectors. In fact I used the same connector system on several bikes and controllers. Now I can test things simply by parking them close together and connecting the controller of one bike to the motor of another.

HTH

Nick

I'm not a Torq owner.... but didn't you say that a power wire to the motor had broken due to flexing upon steering. Do the hall sensor wires also run the same way?

If so could one of those wires have suffered a similar fate (maybe intermittently)?

Nick's procedure will help you tell...
 

Sector

Pedelecer
Mar 5, 2007
102
0
Leicestershire Le8
My faulty controller

My Quando developed a fault after it had been left unusued for a few months while I favoured the Sprint. The Quando motor ran, but it felt like a car running on three cylinders when starting from stationary. Above about 5 mph it ran normally.

I suspected damp, so went to great lengths to dry everthing out. No improvement.

I opened up the motor and had a look inside. All seemed fine.

After that I removed lots of shrink wrap and examined all the wiring connections for corrosion. I re-made the joints. No change.

I opened the controller enclosure and blew a hair drier in there to eliminate any moisture. Still no change.

Then I drastically simplified the wiring on the bike. I chopped it back to the bare essentials for throttle, controller and motor. I removed the voltage indicator and lighting wiring, and just brought a single new three core cable direct from the controller to a (new) twistgrip throttle. No change.

I checked the firing of the Hall effect sensors and they all did their stuff as I turned the wheel backwards.

I got rid of the 5-way Hall sensor connector altogether and soldered the wires. No change.

At this stage I thought that the controller was probably faulty, but I was reluctant to spend £125 to get a new one when I couldn't be sure that it would clear the fault. I compromised and ordered a much cheaper controller, advertised on American Ebay and sourced in China.

When it arrived from Hong Kong I lashed up some temporary wiring using connector blocks and the bike ran perfectly right away. I've done a proper wiring job now, although I've still kept the wiring as simple as possible because I prefer it that way. I'm now using the Quando every day. I was beginning to think of it as a write-off. Now it's back to its old nippy and powerful self and is as much fun to ride as it ever was.
 
Last edited:

imellor

Pedelecer
Oct 25, 2006
67
4
Ian,

You can't easily get at the ends of the 8 wires from the controller to motor, but you can still check things. (I'm basing this on my Torq 1; newer models may be different.)

There is a joint between the motor cable and the controller cable on the forks, and its not exactly mil spec. You've already have trouble with the cable, but it sounds as if that was on the controller side of the joint. I wouldn't be surprised to find more troubles along that cable. When I opened mine up, I found corrosion along the wire and had to cut it back to half way down the cross tube.

You could open up the joint for a clean and inspection. While you're there you can check some things with the multimeter. Note that the connections inside the joint may not be blue-blue, yellow-yellow, etc so make a record before you disconnect anything.

The 3 thick wires are the motor phase. They switch between 0 and 36 V with a complex 3 phase waveform. You won't make sense of this with a multimeter. But with the power off, you can measure the resistance between the wires. This is the winding resistance of the motor and the reading should be low between any 2 of the 3 wires - don't know the exact value, but it will be Ohms rather than 10s of Ohms.

The 5 thin wires are for the Hall sensors. There's 0 V (probably black) 5 V supply (possibly red or brown) and the 3 signals back (blue, green, yellow). With the power on and the throttle closed, check the voltages. If you turn the wheel backwards slowly the signal wires will switch between 0 V and 5 V. You have to go slowly; I think its 86 cycles per turn of the wheel on the Bafang motor, so its not easy to see on a digital meter. Its a bit fiddly to do this without accidentally shorting the wires, so you may want an assistant and some bits of insulating tape. That will tell you if the Hall sensors in the motor are working.

I replaced the joint with connectors. In fact I used the same connector system on several bikes and controllers. Now I can test things simply by parking them close together and connecting the controller of one bike to the motor of another.

HTH

Nick
Nick

Thanks some excellent tips there, I'll give them a try and report back. Although I probably won't get a chance until the weekend.
 

Andrew harvey

Pedelecer
Jun 13, 2008
188
0
Wyre Forest
www.smiths-cycles.com
Juddering motors are normally a sign of faulty hall effect signals, most commonly the faults tend to be in the cables were they move ( handlebars ect.) or twist as they exit the motors.
To test hall effect sensors , disconnect them at the controller.
Find the positive and negative wires, these are normally red and black.

If you then attach a small battery pack it doesn't need to be 5 volts anything above about 3 volts seems to work, to these two wires you can then measure the voltage between the battery and each of the return wires. get a third hand to turn the wheel slowly and the voltage reading should vary from 0 volts to about your battery voltage.

If the voltage on all the return legs varies equally then your sensors and the cables should be ok.
Double check, turn the wheel so you get a return voltage then gently pull the wire straight, but don't pull so hard as to stretch it. Wire cores can break inside the sleeving when repeatedly twisted, in such a way as to make a contact only when in a specific position.

If you find that one or more return wires does not give a signal look to the obvious bending place.
If the wires are broken outside of the motor just replace them.

Is the Ezee controller £125, I've just found a supplier who will ship me 8Fun motors with controllers for less, if I buy enoght.

If the you need a controller I have some spares that should do, but you might have play with the phase wires to get them to run the right way, and the Cycle Analyst will need adapting if you have one.
 

Sector

Pedelecer
Mar 5, 2007
102
0
Leicestershire Le8
8Fun

I'm interested in getting another 8fun motor and controller. Flushed with success at fixing my Quando I now fancy converting one of my pushbikes to electric. Please let me know what the kit price would be. Which dealer is it? Are they UK based or overseas? If they are overseas then customs might be levied.

I have heard that the Bafang factory accepts orders direct for motor/controller kits, but they probably want to sell in batches rather than one off.
 

Andrew harvey

Pedelecer
Jun 13, 2008
188
0
Wyre Forest
www.smiths-cycles.com
I tried contacting the Bafang company about 5 years ago and never got a single response to my e-mails. They were I have been told only interested in orders of hundreds of units.
I eventually found Ethinkar who I dealt with for several years, buying motors in lots of 5. There website disapeared about 2 years ago and they no longer respond to emails.
I have used the Cycleman kits mentioned on another current thread and had no complaints, but they seem to have gone the way of others before them. http://www.pedelecs.co.uk/forum/electric-bicycles/3673-cycleman-mark-rao-warning.html
As as a result of the Cycleman / Mark Roa thread I will not be publishing the name of my source until after I receive the goods.
Buying anything from so far away you have to be prepared to lose your money, the main reason I only buy in small quantities.
If and when they arrive I will send you details. I'm ordering a couple of motors so I should have a spare, but I've asked for them to be sent airmail which should add about £100/150 pound to the cost. at about $90 with controls their not the cheapest but it gives a taxable cost of about £180, with import duty and VAT added say £220 plus delivery, the postage cost is not taxable. So the cost should be about £150 each to my door, which I think is reasonable, considering what you can pay.
 
Last edited:

imellor

Pedelecer
Oct 25, 2006
67
4
Thank you all for your help, I have spent the day tinkering with my Torq. I believe a have a faulty hall effect sensor.

I took the controller apart and measured the voltages on the sense leads, I was getting +5v pulses on all but the yellow lead, as I tuned the wheel backwards as suggested.

I then removed some of the insulation below the fork connector (motor side) and checked continuity of the yellow wire from that point to the controller, which was OK.
Just to be sure, I also looked for the +5v pulses, there whilst turning the wheel, but these did not show on the meter.

I also removed the front wheel and hub cover and cut in to the insulation of the yellow wire within the motor, I then tested continuity OK, to the fork connector.

Finally I put the wheel back on and connected a battery to the fork connector and took some measurements, I couldn't get any positive voltages, but I could get -5v pulses on the green and blue leads as I spun the wheel and a permanent -5v on the yellow.

So my conclusion is that my "yellow" sensor is stuck at -5v.

So it looks like I need a new motor, unless any one can suggest otherwise!!!

Ian
 

Tiberius

Esteemed Pedelecer
Nov 9, 2007
919
1
Somerset
Ian,

It does sound like that Hall sensor has gone.

It may be possible to replace the sensor inside the motor. The sensor itself is a standard part - made by Siemens or Honeywell and costing a quid or two. I don't think its buried that deep inside the motor.

I haven't tried to change one, though. Maybe there is someone on here who has.

Nick
 

flecc

Member
Oct 25, 2006
53,559
30,848
I haven't changed one but remember a superficial impression that they may not be buried deeply.

The way it's laced up and varnished is a bit offputting, but on the basis that it's dud anyway, there's nothing to be lost in trying.
.
 

imellor

Pedelecer
Oct 25, 2006
67
4
Tiberius / Flecc

Thanks, I'll have a go at dismantling the motor and report back!!!!

Ian
Flecc

From your picture, it looks like you have previously removed the armature. If I need to extract the armature from it's casing, is there anything I should be aware off? Flying springs or gears etc?

Ian
 
Last edited:

imellor

Pedelecer
Oct 25, 2006
67
4
All

Thanks for your help so far, I have now taken the armature out of the hub.
That's as far as I can dismantle the motor for the time being, I have cut some the lacing away and the yellow wire looks intact and not shorting anywhere.

I have also tried to tease the sensor out, but have ended up snapping the wires off. However this has confirmed that the sensor is faulty, because the voltage has gone from the yellow wire!!!! ie it wasn't shorting anywhere.

If you look at the outer ring, there are many places you could insert a hall sensor. From the factory, the sensors are laid out as follows:

Starting from a position I will call "1"

1 - Yellow
2 - Empty
3 - Empty
4 - Green
5 - Empty
6 - Empty
7 - Blue
8 - Empty
etc

I propose to buy a new sensor and insert in position 10 as follows:

1 - Empty
2 - Empty
3 - Empty
4 - Green
5 - Empty
6 - Empty
7 - Blue
8 - Empty
9 - Empty
10 - Yellow
11 - Empty
etc

If I then cross over the wiring like so:

Bike Motor
Yellow Green
Green Blue
Blue Yellow

I should have effectively the same set up, but further around the ring, do you think this will work?

Also, can anyone advise an appropriate sensor, I was looking at:Rapid Electronics - Electronic Components

Ian
 

flecc

Member
Oct 25, 2006
53,559
30,848
I can't be sure on that position change without knowing all the spacing relationships Ian, but instinct tells me you should avoid it and use the simplest replacement option.

I feel sure it should be possible to chip out the existing dud sensor head and just replace that. Looking at this photo from Endless Sphere you may be able to insert a screwdriver just below a bar magnet and gently lever out the sensor. If not, cutting into it from side to side should fragment it enough for removal. A further option with care is to gently drill into the sensor at an angle from just below a bar magnet above, either to break it up or to insert a screw to give you something to get hold of to wiggle it free. If you break a little of the paxolin surround removing the sensor, you'll be able to build it back up again with epoxy resin when you cement in the new sensor.

I'd be happy to try that Hall sensor you found as it seems to match what's needed.
.
 
Last edited: