Hi All, I bought a 2018 crosscity e bike with a dead battery, it had fallen below minimum voltage to charge, I opened up the battery and charged it by the battery + & -, now when I turn it on it either turns on straight away or the display just flashes and then turns off, if you turn it off via the display and then turn it back on it works fine, if you turn it off via the battery on/off and then try and turn it on again it just flashes one battery bar and then turns off, I have tried unplugging the BMS board and plugging it back in after a few seconds, I've tried electrical spray on the switch, I've short wired the switch to rule that out but still it will not switch on properly, any ideas ?
Carrera crosscity E power on issue
- Thread starter Clive Norwell
- Start date
I've tested voltage on the battery at 41.7v, when the display works it all seems fine on pedal assist, the BMS has on/off on it which is my first thought, do you think I have a mix of over/undercharged cell groups which is causing the issue ?
Just tested the cell groups as follows
1, 4.13
2, 4.13
3, 4.13
4, 4.13
5, 4.13
6, 4.13
7, 4.13
8, 4.13
9, 4.13
10, 4.14
The cells seem to be ok, the battery charged as I would expect but the on/off issue is baffling me.
1, 4.13
2, 4.13
3, 4.13
4, 4.13
5, 4.13
6, 4.13
7, 4.13
8, 4.13
9, 4.13
10, 4.14
The cells seem to be ok, the battery charged as I would expect but the on/off issue is baffling me.
Thanks for the reply guy's, the switch solder is fine, VFR400, it's the combination of turning the bike off by the battery switch that causes the issue, if I turn it off by the control panel then it seems fine and works as it should, if left then when I turn it on the panel flashes once and then turns off, very confusing.
When the panel flashes on once, it's because there is no battery connection. what you're seeing is the capacitor discharging that's across the battery wires. It behaves like a tiny battery that can only last for a split second.
You need to find out where the battery voltage is being interrupted. Normal causes are faulty key-switch, faulty key-switch solder joints (always look OK - re-solder them to make sure), bent battery terminals and/or stretched receivers, loose bullet connectors to the controller.
You need to find out where the battery voltage is being interrupted. Normal causes are faulty key-switch, faulty key-switch solder joints (always look OK - re-solder them to make sure), bent battery terminals and/or stretched receivers, loose bullet connectors to the controller.
I've now managed to sort it I think, left the BMS unplugged overnight and tried it this morning, all seems to be Ok now, have tried it around 20 times with no issues, Thanks for your help vfr400 & Nealh.
This is a left-field thought...
I watched a youtube video of a guy recelling an ebike battery some while ago. It wasn't a mainstream brand (like Bosch, Yamaha, Shimano etc), but nevertheless, quite a common one found in Europe. Like your battery, the voltage had fallen to a level too low to recharge, but once recelled (or voltage is restored as in your case) there was a common problem of the bike misbehaving. The purpose of his video was to show that (with his specific type of battery/BMS), a memory chip on the BMS logs cell state, and you needed to reset the chip. He did this by temporarily shorting a pin on the chip to earth. It worked.
Maybe leaving the board powerless for a while has done much the same in clearing this memory (if yours had such a chip)?
Glad it's working now though.
I watched a youtube video of a guy recelling an ebike battery some while ago. It wasn't a mainstream brand (like Bosch, Yamaha, Shimano etc), but nevertheless, quite a common one found in Europe. Like your battery, the voltage had fallen to a level too low to recharge, but once recelled (or voltage is restored as in your case) there was a common problem of the bike misbehaving. The purpose of his video was to show that (with his specific type of battery/BMS), a memory chip on the BMS logs cell state, and you needed to reset the chip. He did this by temporarily shorting a pin on the chip to earth. It worked.
Maybe leaving the board powerless for a while has done much the same in clearing this memory (if yours had such a chip)?
Glad it's working now though.
Have you got a link to the video? I've never seen or heard of that sort of thing on normal Chinese batteries, and I've dealt with hundreds. even the smart BMSs don't have a feature like that. I'd like to see what he had.
It was a while ago, but I'll see if I can find it again and post if I do. It was a rack battery IIRC. Many smart BMS's (Bosch, Phylion etc) have chips that store number of charges, state of health etc for dealer reading... maybe this logged the fact that the battery was below voltage and un-chargeable. It needed to recalibrate. Bosch totally craps out if you disconnect power, doesn't it? Kalkhoff needs to fully drain and recharge fully for theirs to "calibrate". This guy just shorted the chip briefly to reset whatever it had remembered - it was a hack. The OP did recharge/recover his pack through unconventional means - any memory/charge cycle chip wouldn't know that, as it wouldn't by fitting new cells?
The Kalkhoff batteries and some other similar ones have to be run through a complete charge cycle to calibrate the meter - nothing to do with the main functions of the BMS.
Some laptop batteries have BMSs that switch off permanently if they detect certain conditions, including the maximum number of charges. They're reset by bridging the positive cell pack to the positive output wire, but I've never seen or heard of that on an Ebike BMS. Normally, to reset an intelligent BMS, all you have to do is disconnect the multi-pin connector if they don't get a separate positive from the cell pack. If they have the positive wire, you have to unsolder it as well. That's only for BMSs that have a CPU. A lot of cheap and older BMSs don't have a CPU, so they have no intelligence to do anything like that.
@vfr400: Had a quick look, can't find the video... but it's there somewhere.
Pretty sure it was a Tranzx BL03 rack battery - common on German Winora and Hercules bikes, big manufacturer selling only in mainland Europe.
Pretty sure the youtube video was linked to a thread from one of the other forums.
Here's a discussion "TranzX Battery 36V BL03 defective.... Cell swap", pedelecforum.de page 7/8/9. Similar story - of which there are a few - abridged:
User colourful cake: The old BMS has been taken over so that the display display works again. Installed is now 10s5p [new 18650 cells from old pouch totally flat]... 250w engine and 36V...
Now I only have the problem that he shows with a full battery (41.1V) that he has only one bar of the battery on the display...
So maybe a BMS reset could help?... Which of the 5 pins in the charging plug do you have to bridge for how long?
User Tommmi: Connect pin 2 and 5 on the port for 2sec... when the wrong ones have been bridged BMS is also gone.
User colourful cake: I now bridged contact 2 and 5 again with another cable and then it went!... Definitely he now shows the full battery!
User Garret: [theorising] ...then it has to work via the I2C bus... picture is an OZ 890 ( i2C ), where probably reset on ground lies.
IIRC in the video the guy just shorted two points internally to reset the BMS, but I guess it likely amounts to the same thing?
Pretty sure it was a Tranzx BL03 rack battery - common on German Winora and Hercules bikes, big manufacturer selling only in mainland Europe.
Pretty sure the youtube video was linked to a thread from one of the other forums.
Here's a discussion "TranzX Battery 36V BL03 defective.... Cell swap", pedelecforum.de page 7/8/9. Similar story - of which there are a few - abridged:
User colourful cake: The old BMS has been taken over so that the display display works again. Installed is now 10s5p [new 18650 cells from old pouch totally flat]... 250w engine and 36V...
Now I only have the problem that he shows with a full battery (41.1V) that he has only one bar of the battery on the display...
So maybe a BMS reset could help?... Which of the 5 pins in the charging plug do you have to bridge for how long?
User Tommmi: Connect pin 2 and 5 on the port for 2sec... when the wrong ones have been bridged BMS is also gone.
User colourful cake: I now bridged contact 2 and 5 again with another cable and then it went!... Definitely he now shows the full battery!
User Garret: [theorising] ...then it has to work via the I2C bus... picture is an OZ 890 ( i2C ), where probably reset on ground lies.
IIRC in the video the guy just shorted two points internally to reset the BMS, but I guess it likely amounts to the same thing?
TransX is similar to Bosch, Kalkhoff, etc, that use comms between the controller and BMS. The procedure was probably to reboot the comms system. You don't have that on most Chineses batteries.
The Winora/TranzX BL03 battery referred to pre-dates the whole TranzX CANbus era by some margin (2017/2018) although it may have used some earlier smarts/comms method. And Halfords say of their Crosscity "As standard on all Carrera E-bikes, the Crosscity comes with an integrated smart control system that protects the battery...", so the OP's battery may not be a totally dumb/basic BMS Chinese cheapy; the inference is there's some smarts in there (all the other Carrera's do).
I dare say you're right though (you usually are).
I dare say you're right though (you usually are).
Hi Clive, can I please ask using which points on the battery did you charge it? Perhaps you could post a picture, I have two faulty batteries now for this bike, and I think both are undervoltage cells, I want to try to charge them but not able to work out the groupings and how to exclude the bms when charging. Thanks.
First, you open the battery, find the BMS, Unplug the multipin connector. Using the "B-" pad on the BMS as ground, you then check the voltage on each pin on the multipin connector. If all 10 voltages are equal and in the range 3.0 to 4.2 volts, the cell-pack is OK. Clive then left that Multipin connector off all night, and when he reconnected it, everything came back to normal operation. If one or more of the voltages is below 3.0v, you'll have to charge it/them individually to bring it the same as the others.
Somebody else solved it by temporarily (short time) attaching the charger to the "B-" point and the battery positive, which seemed to reset the BMS.
Here are the symptoms I had with a Carrera Crosscity: it was working correctly last year, then it was left over the winter and now on the first attempt to use it this spring it did not work. The bike display did not come on and the charger LED went immediately from RED to GREEN when connecting the battery.
With the battery removed and the switch turned on, there was no voltage present on the +/- terminals of the battery connector. I checked that there was voltage on the charger's plug, so I knew that there was something wrong with the battery.
It goes without saying, if you are going to open up the battery pack then you need to be very careful, it is energy dense, and you don't want to short anything out! I disassembled the pack and measured about 40V across the main connection points to the end cells, so I knew that it was unlikely that there was a problem with them.
I then suspected the BMS and so I carefully removed that from the pack to inspect. In doing so, I discovered that 4 of the 0603 SMT inductors (arrowed in the photo) were measuring open circuit. They should be less than 1 ohm. They didn't look electrically damaged, they looked like the metal end-caps were corroded, so it could have been a manufacturing problem since there were no signs of water ingress.
I replaced the faulty inductors with some new ones and that fixed the problem with the pack, the battery charged, and the bike worked. The exact inductance of the inductors is not that important, they are probably only there for EMC purposes. You could just remove any faulty inductors and replace them with a thin piece of wire, just be careful that you don't short anything else out!
I hope that is helpful to someone else.
With the battery removed and the switch turned on, there was no voltage present on the +/- terminals of the battery connector. I checked that there was voltage on the charger's plug, so I knew that there was something wrong with the battery.
It goes without saying, if you are going to open up the battery pack then you need to be very careful, it is energy dense, and you don't want to short anything out! I disassembled the pack and measured about 40V across the main connection points to the end cells, so I knew that it was unlikely that there was a problem with them.
I then suspected the BMS and so I carefully removed that from the pack to inspect. In doing so, I discovered that 4 of the 0603 SMT inductors (arrowed in the photo) were measuring open circuit. They should be less than 1 ohm. They didn't look electrically damaged, they looked like the metal end-caps were corroded, so it could have been a manufacturing problem since there were no signs of water ingress.
I replaced the faulty inductors with some new ones and that fixed the problem with the pack, the battery charged, and the bike worked. The exact inductance of the inductors is not that important, they are probably only there for EMC purposes. You could just remove any faulty inductors and replace them with a thin piece of wire, just be careful that you don't short anything else out!
I hope that is helpful to someone else.
Related Articles
-
MTF Enterprises announces acquisition of EMU Electric Bikes- Started by: Pedelecs
-
Wisper 806T folding bike wins Which? ‘Best Buy’- Started by: Pedelecs
-
Sustrans calls for protected cycle lanes- Started by: Pedelecs
-
Amazon launch their first UK e-cargo micromobility hub- Started by: Pedelecs
