Repurposing a BionX 48V battery for generic ebike use

This is the opposite form of the way this issue has come up on ES in the past, which generally tries to use a generic battery to replace the BionX branded one. I've read the hundreds of great BionX related posts on ES, and am fairly certain that what I'm attempting to do can be accomplished.

I have a 48V (Canbus) BionX battery that's only about two years old and has had mostly prototyping and light experimental use. The cells are all balanced and in good condition. I have a full assortment of BionX components, including the G2 console and dealer diagnostic tool (BiB?) I no longer wish to use the BionX proprietary system, so have started converting the BionX motor into a generic DD BLDC hub motor. I followed the wisdom of Justin and the ES BionX gurus in undertaking this task, and am well underway with that process -- with a possibility of even reusing the BionX affixed strain gauge on the axle with an instrumentation amplifier developed for another bike setup.

At any rate, I don't want to throw out the BionX battery, nor do I want to sell/pack/ship it due to the volatile nature of this type of thing. I just want to use it up on one my other ebikes, especially on one with a new Phaserunner and Grin All-axle motor. My current issue is that when I "turn on" the battery using the Power button (on/off switch) on the G2 console/display, it lights up the display and shows the battery as being fully charged, BUT it displays the "POWR PROT" (note spelling) message. The result of this is that the power output ports on the battery remain dead. That error message, per the BionX documentation, suggests a bad connection between the console/battery and the (BionX) motor. The BionX motor (and its internal controller board) is never going to be connected, of course, so I want to override/ignore the error message, not fix it.

There are some suggestions out there that a temporary short between pin 3 on the D-sub battery connector and the A1 (ground) pin on the same connector will "turn the system on". Having some bad experiences on this system with poor choices related to "shorts", I wanted to see if there's a way of turning the battery's output ON and keeping it that way that doesn't require a lot of BionX magic, or risk release of the magic smoke. Changing connectors, taking apart the battery and doing rewiring, that I can do.

Get rid of any pcb boards attached to the battery, that will be the only way to use that battery.

everything i've read over the years here and elsewhere shows that some of them will "wake up" with a short between two pins (don't recall which ones), but then the main output turns off, leaving the brain of the bms "awake", and able to talk to the rest of a bionx system (if it were attached to one), but the battery is unable to power a motor.

almost certainly it requires a canbus command from the head unit to actually power on the battery.

since you have working systems there, if you have a computer system (or canbus sniffer) that can read the canbus data, you could just connect that system/sniffer to the canbus data wires anywhere in the system, and then capture that data. then you could program that data into an mcu (arduino, etc) as a string for it to send out to the battery whenever you "power on" the system (button on the mcu board).

you might have to pare down the data so it only has the necessary bytes to tell the battery to turn on. i don't know what that might be, or how that works. there are people here on es that have figured out canbus data for controllers like sevcon, etc., so maybe they can help you with it?


alternately, you could just remove the entire bionx bms system, and replace it with a nonproprietary type.

amberwolf said:

since you have working systems there, if you have a computer system (or canbus sniffer) that can read the canbus data, you could just connect that system/sniffer to the canbus data wires anywhere in the system, and then capture that data. then you could program that data into an mcu (arduino, etc) as a string for it to send out to the battery whenever you "power on" the system (button on the mcu board).

you might have to pare down the data so it only has the necessary bytes to tell the battery to turn on. i don't know what that might be, or how that works. there are people here on es that have figured out canbus data for controllers like sevcon, etc., so maybe they can help you with it?

alternately, you could just remove the entire bionx bms system, and replace it with a nonproprietary type.

Click to expand...

I just love this forum. The creative DIY mentality of you guys just can't be suppressed. Here I am, looking for a straightforward solution that leaves the proprietary BionX messiness behind, and markz just puts it out there. But, since this is ES, we just can't resist coming up with a totally clever, elegant solution that not only fixes the problem, but provides personal bragging rights on top of that. So, amberwolf, if I was 50 years younger and hadn't just decided to leave BionX coding behind, I'd gladly charge down that rabbit hole! Thanks,

As it is, I'll go with markz's advice, and your "alternate" advice, and just remove the BionX battery logic (BMS?) board. My only remaining concerns are that:
1) the proprietary BionX charger, which charges at 26V, will also need to be replaced (not a real problem for me)
2) whatever protection (balancing, fusing, and thermal cutoff) the battery control board offered will disappear.

Amberwolf suggests using a "nonproprietary" BMS, and I'm aware of multiple other threads on ES that are trying to find the 'ultimate BMS'.

Do you guys have any additinal suggestions in this regard, such as "going commando" (no BMS of any kind), or going with a cheap chinese BMS replacement (I think battery is 13S, but would confirm), or some other system?

if the cells were new, well-matched, good quality, and used well within "the middle" of their specifications, going bmsless would probably be fine.


but since they're not new, and i have no idea what they're quality is, or how well-matched they were to start with (or how they are now), it's probably safer to run a bms to limit things. if you know what their specs are, you can at least determine if your usage will be well within their limits, or if you might be pushing them to some degree. the harder you push them vs their limits, the more you might want a bms.

i don't use a bms on my main packs on the trike, because theyre large, well-matched, good qualtiy, cells (even though they're old they stay well-balanced), and i don't run them down very far (though i do charge them all the way), nor do i push them very hard currentwise. but there is one on an em3ev pack on my brother's trike (so he can just use it without worry about cutoff, charging, etc.

there's always an "ultimate bms" goal...but what that goal is depends on your needs, pack size, etc. for a big ev pack, an ultimate bms is probably modular, with nodes within the pack at the cells, to minimize wiring that carries widely different voltages that could short and cause a fire. for a small pack, a centralized bms wired to cells directly works fine, and greatly simplifies wiring. etc.



there's anumber of threads about bms's, including one about a bluetooth type you can monitor/program with your phone (i don't know any more than that, though).

amberwolf said:

if the cells were new, well-matched, good quality, and used well within "the middle" of their specifications, going bmsless would probably be fine.

Click to expand...

This BionX 48V pack uses NCR18650PF cells, which apparently have a decent reputation for maintaining balance according to a number of German pedelec forum members who are familiar with BionX batteries. While not "new", they've been lightly used ("middle" of specs, as you say) on a prototype project here, so I think I might just try "bmsless" (did you just coin a word for this?). I'll monitor them carefully, and I've always been curious about doing this anyway.

amberwolf said:

there's anumber of threads about bms's, including one about a bluetooth type you can monitor/program with your phone (i don't know any more than that, though).

Click to expand...

I've been monitoring that bluetooth BMS thread for quite a while now, and while it seems like a great idea, some of the posters on that thread give the impression that those systems aren't quite ready for prime time yet. I just don't feel like jumping into another project right now that still has software issues. The day that Grin endorses one and sells it, I'll pull the trigger.

Thanks for the great guidance!

My point was get rid of the BionX "p.o.s." bms and go either or, BMS'less or aftermarket BMS. Either or, ying or yang.



Bestech has quality BMS' but their Minimum Order Quantity (MOQ) is 2.
https://bestechpower.com/

I see EVFITTING aka Greentime has BMS'
https://evfittinggreentime.aliexpress.com/store/group/Battery-Management-System-BMS/313864_259134332.html?spm=a2g0o.store_home.pcShopHead_11248317.1_10_0

BMS Battery - https://bmsbattery.com/66-bmspcm

markz said:

My point was get rid of the BionX "p.o.s." bms and go either or, BMS'less or aftermarket BMS. Either or, ying or yang.

Click to expand...

Thanks markz! I think I'll try the BMS'less method on this pack and just carefully monitor the cells. This whole BionX venture has been an experiment anyway, so I might as well continue on that path.

markz said:

Bestech has quality BMS' but their Minimum Order Quantity (MOQ) is 2.
https://bestechpower.com/

Click to expand...

I followed some of your links, and apparently Bestech also sells their products via a "distributor", which will sell qty:1 of any of the Bestech models, for instance:

https://www.litechpower.com/product-detail/HCX-D140LI12S15A_05.html

I'd be interested in thoughts about the utility of the Bluetooth (or any wireless communication protocol) feature re BMS's. Is this truly useful or just a gimmick?