BionX 48V battery not responding but cells are ok

[crustulu]

Hello,

I just bought a used 48V 8.8Ah faulty (cheap) BionX battery in order to replace the one of my Kalkhoff Image B24. The previous owner of the bike just threw the old battery because it was not working anymore ...

I recieved the battery knowing it was not working and assumed that i could replace the cells inside as I successfully did before on the 36V BionX CANBus battery of my wife's bike).

I just plugged it on my bike and obviously nothing appened, impossible to wake-up the system. I precise that i tested the bike with the 36V battery of my wife and everything worked like a charm. I've also tested the voltage on the output pins of the battery bus, everything is at 0V, i guessed that the pack was probably dead.

Unfortunately it doesn't seems to be that easy on those new batteries... After opening it (not without some sweet) i discovered that the cells where fine, the pack voltage was 52.5 V well balanced. I assumed that maybe the battery was somewhat over-charged and the BMS was protecting itself, so i discharged it a bit, down to 51.8 V, but same result, everything at 0V...

The issue obviously comes from the BMS, but it looks like new, not even a small trace of heat (see pictures below, click to enlarge):



My question is :
1 - Do you know how to "revive" the BMS ? (i am very pessimistic about that, sadly)
2 - Should i let down the BionX internal system and switch to an external controller instead ?

About question n°2, i saw here that it is possible to re-use the BionX motor with a third party controller like DISH did here : https://endless-sphere.com/forums/viewtopic.php?f=4&t=20176 but the exact reference of the controller is not clear and the thread is a bit old (2010).
So do you know external controllers that are compatible with BionX motor ?
With or without the hall sensor ?
Is it possible the re-use the strain gauge for torque sensor ?

As a first guess i was thinking of this controller : https://bmsbattery.com/ebike-kit/66...imulation-sine-wave-controller-ebike-kit.html
But i dont see if regeneration is possible with it, and i dont know if the "torque simulator" worth anything...

Thank you in advance for any advice.

Alexandre.

 

[tomjasz]

Try charging the battery using the power leads. I saw several batteries recover doing this today as a matter of fact. It reportedly "reset" the BMS, regardless of how or what, it worked. Go figure.

 

[crustulu]

Thank you tomjasz for the tip. When you say to try to charge the battery via the power leads you speak about A1 and A2 on the picture below ?

Because rigth now there is 0V on theses pins. How do you sugest to charge via these pins ? Apply +48V with and external power supply ? Connect directly the + and - of the battery pack on it ?

Cheers,

Alexandre

 

[amberwolf]

Do any of these help?
https://endless-sphere.com/forums/search.php?keywords=bionx+canbus&terms=all&author=&sc=1&sf=all&sk=t&sd=d&sr=posts&st=0&ch=300&t=0&submit=Search

 

[crustulu]

Thanks amberwolf for that, i was mostly already aware of all these threads. The only helpful thing that i found is that i have probably the same ploblem as MercedesMbBionx on this thread : https://endless-sphere.com/forums/v...&sid=7523f920aae1ec0b912f3a53ae88ca57#p768112

I also have a small spark when i plug back the power plug of the battery on the BMS but nothing happens after.

When i plug the external charger (a simple 26V power supply in fact) the led aroud the plug ligth briefly in red-orange-green then stop.

 

[crustulu]

Some news. I tried to load the BMS on the power bus (A1and A2) with a 36v power supply, nothing happened, sadly....

I'm in touch with m'y local BionX dealer and i will see if he can help me but i'm not very optimistic.

I think that this kind of situation is typical of the limits of a proprietary system like BionX. What i need is a simple BMS and what they sell is a complete battery sold 800$ while the gross cost of the parts does not exceed 200$... Not acceptable.

I will wait for the answer of my local BionX dealer but i feel a bit jailed with the BionX system... Perharps it is time to break free !

Cheers,

Alexandre.

 

[icerider]

When i plug the external charger (a simple 26V power supply in fact) the led aroud the plug ligth briefly in red-orange-green then stop.

This successfully loaded the BMS, which appears to be working fine.



What you are not getting is the battery to turn on when you try to run the system.

There is an old workshop manual at: http://www.r-m.de/wp-content/uploads/Werkstatthandbuch_BionX.pdf

It includes a test for "starting the battery"

Note that I have never tried this personally.

It is from a Bionx workshop manual in German published by Reise and Muller. Google and I agree on the translation of this small section.

Note that this only works with a "charged battery". For testing the circuit card of a 48-volt system, I could visualize a 13s1p string of 18650s with sense lines and the right connector to attach to the BMS. Attach the battery leads, the BMS sense lines, and the thermistor wires (thermistor salvaged from bad battery of any style -- bet money they always used the same thermistor.) to the BMS board and try the test below.

It is a test to see if battery will come on. I do not know how much built-in-test there is in the startup routine, but at least some apparently.

The battery will turn off when the paperclip is removed (or worst case 15 minutes later -- timeout)

To see if the battery can be switched on: connect pins A1 and 3 in the battery connector with a paperclip (see picture below)

An I2C battery should beep 4 times
A Canbus battery should beep 3 times

 

[crustulu]

Thank you iceride for the tip !

Tried this morning to short circuit pins A1 and 3 and unfortunately nothing happened ... :? No bip, no voltage on the other pins

 

[crustulu]

Tonight I will try to bypass the BMS and see if the console and the controller can speak to each other without the BMS. In order to do that I will directly connect the battery leads to the A1 and A2 connectors and I will feed the Pin 4 with + 5V and hope that everything will wake-up !

 

[crustulu]

It works !

I feed the red wire with +5.5V, the brown one with +12V and the system woke up (except the BMS). After that I connected the + and - leads of the battery to A1 and A2, one spin on the wheel and the motor started !

The only strange thing is that I need to let the original BionX BMS connected in parallel of the +5.5V and +12V feeds, if not it doesn't works.

Obviously the battery gauge doesn't works and if I go into diagnostic mode I get an error 26 (problem of communication with the battery).

The next step is to add 2 DC-DC converters, a proper BMS, a wattmeter and a charger. That is to say all the functions of the dead BionX PCB.

What we can learn from this, is that it is possible to bypass the battery PCB of a CANBus BionX system. The console and the controller can talk to each others without it.

Good night and thanks to all the Endless Sphere community, without you I wouldn't succeeded.

Alexandre

 

[Syonyk]

Ensure the system remains on for a period of time before you spend too much effort on it.

I believe that without regular signals from the BMS, the unit turns off every 5 minutes.

 

[crustulu]

Thank you Syonyk for the advices.

I am well aware of that and that is why i plan to use an Arduino + CanBus Shield in order to re-establish proper communications between the console and the battery. I am still waiting some hardware to continue my project (it migth have juste arrived today!). Thanks to the analysis of the CanBus dumps from Dimitrios Simitas i managed to isolate the ID of some crucial CanBus blocks in order to reproduce them with the Arduino. I can reproduce the battery SOC, voltage and periodic wake-up signal for now.

I will post the wire diagram and the source code for the Arduino when ready.

 

[Syonyk]

I'm very interested. Being able to build a "legitimate" replacement pack for CanBus BionX systems is of great interest to me.

 

[rowbiker]

I am also interested in acquiring more information that can be used to get more use out of BionX systems, especially those that have aged out of the warranty period. I downloaded the German workshop PDF file above -- thanks for the link!! -- and it is a goldmine of information, as some of the above posts already have demonstrated. Although I live in Minnesota and am now a US citizen, I remember a fair amount of my native German language and volunteer to translate any of the parts of the PDF manual upon request by an ES forum member. l have a bit of experience with the newer Canbus system, having done a custom bike conversion using the 350 series motor, but have not disassembled anything (except the throttle) yet.

 

[crustulu]

Some updates from my side.

My 48v CanBus system with dead BMS works like a charm. I use it everyday.

For now i just removed the BionX BMS, installed a standard 13s Chinese BMS with balancing function, added 2 DC/DC converter one for +5V and one for +12V. I also added a opto-magnetic 30A relay operated by the +5V and a simple rocker switch on the outside of the battery. When the switch is off, the DC/DC converters are disconnected and the relay is in idle position witch means that the + of the battery is connected to the charge plug. When the switch is on the DC/DC converters are on, the relay is activated and the + of the battery is connected to the socket. Do not plug a battery in the bike with tension on the socket, it will produce a huge spark.

The CanH and CanL wires are shorted with a 240 Ohm resistor for the moment.

The console and the motor can communicate without the battery original BMS, the only things that lacks are the battery gauge, the regeneration and the odometer that acts very strangely. No automatic shut-down after 10 minutes like it was said above.

On the CanBus Hack side, for now i just listened to the communications of the system on my wife's bike witch is fully functional. I think that i have discovered the codes for battery voltage, state of charge, regen on/off. The problem is that the Arduino doesn't fit inside the battery rigth now, i have to reprocesses the overall wiring for that. I will continue that later i think.

Cheers,

Alexandre.

 

[izzzzzz6]

Interesting discoveries here! I believe i have the same BMS. recently i was lucky and bought a Smart Bike for €400 from the pound. There was no charger so i just used a power supply but had to disassemble the battery to connect the 26V (while i wait for what is hopefully the correct dc input cable). I noticed there appear to be 4 small SMD shunts around the power terminals to the BMS 2 on each side of the board. Does anyone know if a shunt modification is possible or do i risk messing with the charging limiting circuit?
Did bypassing the original BMS remove any of the speed restrictions?
I would like to over-volt my System, perhaps by adding a 10ah single cell or a 2s sting of cells in series with the cable that takes power to the motor, (I live in the mountains and sometimes assist level4 doesn't cut it and i have no access to mountain mode). Does anyone know if this would be effective in delivering more power to the motor? Could it damage or interfere with the rest of the system? I'm also looking for a simple hack to de-restrict the top speed as i'm locked out of the code menu with the Smarts rev62 on the G2 console. I saw some threads with some info on flashing the system with a "Crumb128-CAN V5.0 AVR CAN Module" but i am a bit lost as to how to actually program the console. What software do i use and how do i find it/execute the new commands? Has anyone seen more detailed instructions on using the Crumb128-CAN V5.0?
How is the work going with the arduino project? What will the advantages be over using the stock BMS? Was it just too difficult or expensive to buy a bionx BMS?
Do you think you can get the re-gen function woking again? Thanks for the info so far.

 

[icerider]

On the CanBus Hack side, for now i just listened to the communications of the system on my wife's bike witch is fully functional. I think that i have discovered the codes for battery voltage, state of charge, regen on/off. The problem is that the Arduino doesn't fit inside the battery rigth now, i have to reprocesses the overall wiring for that. I will continue that later i think.

Cool, please keep us posted.

 

[izzzzzz6]

Does anyone know where we can find detailed instructions on how to flash the bionx controller? I have been reading about using the Crumb128 5.0 from chip45.com in Germany. Seems simple enough but would be nice to have some step by step instructions or a video.

 

[Syonyk]

No idea. If you find anything, I'm interested as well.

I know there's a shop tool that connects the system and lets you change things around, but I think you have to be a licensed BionX dealer to get that.

 

[crustulu]

About the speed limitation, it is still active. I looked inside the source code of the German program that is supposed to unlock it with a Crumb128-CAN V5.0 AVR CAN Module. What I saw is that the limitation is build in the motor controller. To unlock it you have to inject a specific unlock serial that i am unable to find in the source code. My opinion is that that serial is compiled in ones of the dlls given with it and since i'm not really interested in unlocking the speed limit i didn't pushed further.

@izzzzzz6 adind 2 more cells to the battery seems very hazardous, you will over volt the controller which might be dangerous. You will also struggle to balance the whole thing which may results in over-charging / over-discharging a cell element and destroying the pack.

About the 26v power supply, i don't know what you want to do with it but since my BionX BMS is dead, I connected it to a CC-CV DC-DC power boost that i tuned to 2A/54.6V volts. I dont know how your BMS is but i wasn't able to recover the charging function of mine.

About buying spare BMS from BionX, my local reseller refused and wanted to sold me a new battery at 750€ !

About the BionX USB connect, I think that it is a bad idea, because it is expensive and you are not sure that you will get the rigth software if you are not an official bionX reseller. In addition, from what I see here and there, the 48V BionX BMS are crap, they always end up non-functioning. You are better off them with a system that you know and can repair.

Cheers,

Alexandre.

 

[icerider]

Help !!!

My wonderful 2011 Trek Valencia+ (with Bionx SL350 Canbus) has gone crazy. After 2 years and ~2000 miles, it has decided to malfunction.

I found it in the garage charging but it now has front and rear light on and they will not turn off. The G2 controller does not respond to inputs -- e.g. the assist level does not change when the + button is pushed, and the mode button does not switch the display miles, ODO, elapsed time, avg speed, etc.

I disconnected the battery and is is at 42.8 volts (nominal is 3.7 volts per cell is 40.7 volts) so the battery is 70+% SOC.

When I disconnected the battery, the rear light remained on for 20+ minutes but the battery did not beep (did not time-out)

When I reconnected the battery, the G2 console was briefly responsive and allowed me to send lights on and lights off (4 second push on the power switch) but neither command was obeyed, the lights remained on the whole time. Then the console remained unresponsive after that.

I just plugged the charger back in to support the battery and left it for the moment. -- if I can get the assist level turned on, the bike will be rideable and I will be perfectly happy since at 70% SOC I can do my morning rides and then just put the bike back on the charger. The lights consume about 6W so the charger can certainly keep up.

Otherwise, CRUSTULU, can you tell me which wires I have to energize (AT THE BATTERY CONNECTOR) to get the system running WITHOUT the battery. I would be fine riding if the assist feature worked. I would happily give up all of the battery functionality to have a rideable bike powered by LIPO.

UPDATE 30 April 2016:
OK, got it fixed....

When I initially found the bike, the battery was at 42 volts, the front and rear light (6-8W total) were on and the display was awake, but un-responsive. At that point I disconnected the 11s 5 Ah RC LIPO parallel boost battery and checked balance -- all OK.

Over the next 36 hours, the Bionx battery discharged below the ~24V MINIMUM charging voltage and the charger declined to charge the battery. I verified the low voltage by reading the voltage at the tap for connecting the LIPO parallel boost pack. At this point, I removed the Bionx battery (with taillight still running) from the rack mount and set it on the bench.

I used an external CV/CC charger (at LOW current) to bring the main battery above the Bionx charger minimum voltage requirement and then fully charged the battery using the Bionx charger. At some point, the battery electronics woke up and timed out, turning off the current flow to the taillight. Not too sure how that worked and I am reminded that "problems that go away by themselves ..." I do remember reading somewhere that if you are running with lights on and your battery falls below the point at which the system turns off, the lights remain on (regardless of battery state). I think this was cited as a safety feature to let you get home safely and I would certainly agree that having a dead system on the road in the dark without lights would be SCARY.

So, with an apparently normal charge on the Bionx system battery, I slowly equalized the voltage of the RC LIPO parallel boost battery and reconnected it to the Bionx battery.

I now have two fully normal daylight ride and charge cycles on the system and it appears to be fine. I have not tried the lights-on and lights-off commands, but my auxiliary lighting is wildly brighter than the Bionx lights so the loss of the Bionx lighting does not bother me. Perhaps at some point I will try the lights, but for the moment I think I will just not use them.

During this "crisis", I did contact Trek about a replacement battery and discovered that the 41V (11s) battery is no longer available and the recommended replacement is the standard 48V 6.6Ah Bionx battery. Price quoted was ~$900 US. At that price, I would probably opt to buy a used standard Bionx PL350 48V system and move the battery onto the Valencia+ while putting the motor and display into my spares box.

 

[meeki]

Hello all,

Did you made any progress on emulating the battery gauge signals with a arduino. I also have a Gen2 battery with a broken (read: killed by Bionx because they said the cells where broken) Bms.

Very interested in your project and progress.

Greets

 

[Syonyk]

I haven't spent any time on it yet, but the project is getting bumped up my priority list based on the number of people who have emailed me about damaged BMSs. Including one that appeared to short the battery through the BMS, somehow... it's pretty spectacular.

 

[micheleu1]

Hello, I was reading your thread on the forum about the BionX not responding.. Well, I have got the same problem.. but I'm far from being an expert of this kind of things..
Would you be so kind to write me how you managed to revive the battery, step by step, and mostly, what kind of equipment I will need to do that, so that I can look for what I need..
I hope it is not going to be too much trouble, and I hope I will hear from you..

Thank you!

 

[djolex]

when charging the battery, the bionx 48v has reached a short circuit in the cable and since then the battery has been dead but the cells are good, the voltage is 52v, the battery has been removed from the bike.
I opened the battery and checked the 20A fuse, which is correct and both electronic boards are new, as can be seen on the pictures. Can somebody help me. thanks in advance.