Where to start - broken Claude Butler Glide 1 and Carrera Subway E

[benuk]

I bought two non working bikes, both sold without working battery. Additionally I bought batteries, two that appear to work which I plan to use on the Glide, and a broken one matching the broken one that came with the Carrera (subway e (v1)).

The glide has a 250w/24v Bafeng read hub (which is apparently perfectly happy at 36v). The batteries I bought are 36v, from a similar rear rack mounting system. I am not sure if the batteries have any communication channels but they show full charge on the test button and I measure 40v on the terminals. There are only two wire from the battery connection to the controller so no handshaking to deal with. With a battery connected to the bike I see all power level LEDS on the bikes 'displayer' are all lit. However I get no output from the motor if I turn the pedals or twist throttle, or any combination of the two. There is a disc on the crank with holes in it that rotates with the pedals and is within 3mm of some kind of (hall?) sensor. I think I see some of the holes have something in them - magnets? dirt? I don't know what to check yet but I am assuming the control is the likely fault (they go more often than the motors I imagine?).

The carrera is a suntour system and had I done my research I would not have bought it at any price. The battery that came with it showed as dead according to the test button and would not respond to the charger. The second battery I bough showed four out of five LEDS with the fifth one flashing. It also did not respond to the charger. I opened both batteries up and found the first to have 17 volts before the BMS, and the second had 25v. I charged both directly until 4v. Both retained that charge overnight and I reconnected the BMS but I do not get a voltage on the output side of the BMSs and neither respond to the charger. Both however show 4 out of 5 LEDS now when the test button is pressed (4 with 5 flashing). I'm unclear what is normal behaviour, these batteries have three wire comms and I guess the controller might have to tell the BMS to switch on the output. When attached to the bike, I get nothing on the display and no response to pressing the power button on the display - which I assume should switch things on.

At this stage I am stuck. Usually when I try to fix things I would attempt to simplify and reduce, eliminating possible problems until I isolate the culprit. On the glide I have disconnected the brake sensors -they are not the problem. Since one bike is bafeng and the other is suntour, I can't swap things around, and with no known working parts to substitute, I'm unsure what to do next. Ironically, I think I have working batteries and dead bikes - the opposite of what I though I was working with ;-)

So, any advice?

On the bafeng, what tests can I do with a multimeter to try to establish if the control or the motor is the likely problem?

Anyone know these early carrare subways and whether I should get output at the battery terminals or how I should be able to switch them on?
I guess I can't just supply power from the glide batteries because the control requires some handshake?

Can these suntour HESC motors be made to work with generic controller and display, and if so which? And if I did that, would I also need to change the BMS in the batteries so that they were not waiting for handshake to enable output?

 

[Annieolnie]

Boy, have you given yourself enough variables to run in circles 'til the next millennium. The Glide is not a fancy bike. It shouldn't have a closed system. I don't think it is looking for any kind of handshake. On the other hand it may dislike you for asking it to run on 40 volts when it is a 24 volt system. Do you have two 12 volt batteries you could put in series and see how it responds to that? If you go over to the third page of the thread titled "Corroded Molex Connectors" you will see some information on testing a controller and a throttle with 2 household batteries (1.5volts) put in series. You need to take a good look at your controller anyway and your pedal assist. You will have to pull your cranks and take off the magnetic ring to clean it. You can also just start going over everything and testing for continuity. Where is the controller on that bike? Is it in the battery base? That type loves to cause trouble. Lastly, where is u.? Google doesn't recognize that as a country.

 

[benuk]

Country should be u.k, although I've escaped the cold for now which means no ready supply of unwanted broken Halfords Carrera throwaway ebikes for parts.

The glide certainly should work on 36v - it was a common user upgrade back in the day, according to a number of threads Ive read. The controller was indeed on the battery base, or in fact hanging out when I got it. It's also rusty and once opened it smells the way power components should not smell. So I'm fairly certain the controller is dead.

 

[Annieolnie]

Well, there is your beginning.

 

[benuk]

Cheap controller and display arrived in post. Instructions not great and I wonder what way to test motor without wiring up pedal assist sensor or throttle?

 

[benuk]

Starting to suspect a dead motor on the glide as the controller has intelligent motor phase function which should mean the motor spins regardless of external throttle or PAS sensors when connected, but I get nothing. The throttle sensor wires seem easy enough with identical colouring, but the PAS sensor wires are different with no obvious way to confirm whether wired the same. But given the motor does not spin, not much point exploring that further.

So now I am wondering whether this control can be used to test the suntour HESC motor on the Carrera.

...

 

[benuk]

The suntour has proprietary control. It has an extra unused connector, no additional external wires not connected. The controller box cover had screws missing and was placed wrong way around so obviously someone has been in here before.

The suntour motor has LOAD of additional wires in addition to the phase wires, I assume for inbuilt hall sensors or similar.

Is it possible to use the controller I have to test the suntour?

 

[Annieolnie]

Cyclechat.net has a thread for the Carrera. I no little about them. ebikes.ca/learn/connectors shows you many, many connectors, but that one in your middle picture is unknown to me. The Carrera Subway supposedly has a cadence and torque sensor, so the controller would have to be one designed for that, for it to work. If you want to test the motor and i.d. those pinouts, electricbike.com has some instructions on how to do that. If you like the frames on one of these bikes, I guess you could start over and create a new build, but otherwise, in my opinion, this is all a frustrating dead-end.

 

[benuk]

I tested the befang motor. it only has three wires, the phases, and i measure about .5 ohms between each which suggests it is fine eletrically. There are no hall sensors in these old motors. I suspect that the controller I bought, a brainpower, might not work with hall-less motors, or the controller itself is a dud, as from what I have read I should expect the motor to do something when I have the intelligent learning wire connected on the controller.

I hate problems with too many variables!

 

[benuk]

As for the Carrera, I'm thinking to try to brianpower controller on it but all the wires to the motor from the original controller will need to be cut in order to do so as they have no connectors (they don't build these halford things with repairs in mind). The new controller will not fit where the original if housed so it will need to be stuck in a box/bag somewhere, and the original batteries (of which I have two) can not be used without bypassing the BMS for output because suntour use handshaking between battery and controller and there is no output from the BMS if it doesn't see it's own proprietary controller. I'll also need to buy and fit hall effect PAS sensor. Hopefully I can avoid much of this work just testing the new controller with the motor (using another battery and relying on the learning wire to show whether the more works) but I can't avoid having to cut all the wires at the original controller.

 

[benuk]

PROGRESS on the Carrera!! I bite the bullet and fitted a hall pedal speed sensor (in the process I found the torque sensor was f**ked), and then cut the wires between the suntour to the original controller. I then rigged up the brianpower controller. Unlike the befang which merely twitched for a second, the suntour motor span up instantly with the learning wire connected. Changing the assist speed setting on the display changed the speed of for the motor, and when I spun the pedals the speed of the motor changed. However, when I disconnected the learning wire and instructed in the 'manual' for the controller, the motor stopped and no longer responded to the pedals. I dont have anything else connected, brakes sensors or throttle. I also did not connect the motor phase/hall sensor wires, only the three phase power cables. Might it be that I have to connect the other five wires even though I got the motor spinning without them?

 

[Annieolnie]

The learning wire takes less voltage to function. You have such a hodge podge there I cannot tell what is going on. I'll post what AI said regarding the controller and hall sensors.

 

[benuk]

I am really close now. My previous issue with the motor spinning with the learning wire connected but not when disconnected turned out to be very simply, I had not actually connected the PAS sensor at all! Pedalling resulted in changes of speed from the motor, well, because was increasing the speed ;-) So, having discovered I'd not yet connected the PAS sensor, I did that today, and yes, when I disconnect the learning cable and pedal I do hear the motor getting power, but it cuts out after a second or two. You continue to hear it trying to come on, but it never stays.

I suspect the problem may relate to the non trivial differences in wiring between the BrainPower controllers and the suntour HESC motor. I have three hall effect sensors wired up, and a red wire I assume to be +5v for the hall sensors, but although my controller has a black ground cable for the hall sensors, the suntour cable does not. Additonally, the Suntour has a white speed whire, but despite it showing on the instructions for the controller, the controller does not (I suspect because this is the version for the simple LED display that doesn't show speed).

So, I'd really appreciate hearing from somebody who have converted a suntour based carrera to generic controllers.

 

[benuk]

I managed to go for a cycle. This thing pulls like a train up hill, very impressed. I unplugged the hallsensors and retrained the controller without them. They could not have been working since I did not connect negative to the sensors. It's kinda of working, but not well. Seems kind of random when it decides to provide assistance, very sensitive to pedalling speed. And when it does deliver, it might stay on long enough to get me up a hill, or it might cut out. Happy to have proved the motor works anyhow. Should have another controller arrive tomorrow, will see what happens.

 

[benuk]

Fixed it! Fixed both bikes in fact. The problem with the Carrera was the new PAS sensor, and I solve it by removing 6 of the 12 magnets. I only figured this out by swapping the throttle from the Glide to the Carrera and in doing so I found it worked fine with the throttle, the problem was the PAS.

I did a 10k hill trail loop and it all worked perfectly.

I'd pretty much given up on the Glide, guessing either I needed to sensorless controller or there was something wrong with the motor. However, while refitting everything on the Carrera once I knew it worked, I noted that when I had not fully connected one of the phase wires, the results were almost identical to the problem i experienced on the Glide. Armed with this knowledge and another Brainpower controller (with throttle), I retained to the glide yet again. I'd already checked the motor and phase cables with a multimeter but I took another look at the connector, pouring some vinegar into it and repeatedly plugging and unplugging it. I then rigged up the controller, twisted the throttle and woosh! it worked!

I then examined the PAS sensor fitted, popped out the five magnets and pushed in the six I'd removed from the new PAS sensor I fitted on the Carrera, then plugged it in and spun the peddles and what do you know, it worked!!

So, I still need to do a bunch of stuff on the glide - a battery mount, somewhere for the controller, replace a seized brake cable, but I know it works now and it should fly as the controller is 500w.

Which brings me finally to batteries. Anyone who knows the carrera suntour system may wonder how I resolved the issue of the batteries refusing to work without handshake comms with a suntour controller. I'll post a separate thread about this as I know others have faced this dilemma.