Saturday, December 10, 2016

BionX 48V 8.8Ah SL 350 HT XL Pack Teardown

I'm kind of a weird person.  I get really, really excited when I get new battery packs in to pull apart.

And I've got a new pack to tear apart!  Today, I'm disassembling a shiny new-to-me 48V BionX 8.8Ah pack!  What's inside?  I have no idea, but I'm really excited to find out!

This is the SL 350 HT XL pack - 8.8Ah of fury.  This isn't the biggest pack BionX sells - they've got a 11.6Ah 48V pack on the market now, but this is a much more recent pack than most of the ones I see.

Sadly, this is a "warranty bricked" pack.  If BionX is replacing a pack under warranty, they just brick the newer packs remotely and ship out a replacement.  I get that lithium battery shipping is complicated, but I'm especially annoyed about this policy after pulling the pack apart - and you'll see why shortly.

So, read on as I dive into this reasonably modern ebike battery!

Labels and Warnings

The outside of the pack has the usual assortment of labels and warnings.  I've got a fancy "Warranty Void if Removed" sticker (special printed for BionX International Corporation), a label with the pack specs (48V, 8.8Ah, and 423Wh), and what appears to be a bilingual QC PASS sticker.  Good to know.

The back of the pack includes the usual warning label.  I ranted about RoHS previously when I took apart a 26V BionX battery pack - so if you want to hear why I think it's a huge load of crap, go read there.  Lead-free solder is objectively worse than the stuff with lead in it across a huge range of metrics, and there are so many exceptions to RoHS that it's basically worthless.  You can't use lead solder (except where you can), but you can have 30 lbs of lead acid battery in a car.

Anyway, I'm going to disassemble it.  I might use it with a non-BionX charger on it too.  I'm not going to incinerate it unless I really screw up, though.

And this label.  Also mocked mercilessly in the 26V pack teardown.  I've no idea why this is supposed to look like a trash can.

Well, that's one VOID WARRANTY.  I tend to do that to things that are outside warranty.  Voiding warranties has served me well in life.

Opening the Pack

Unlike the older generation BionX packs that I'm so familiar with, this pack only has three bolts along the bottom.  They're still 4mm hex heads, though - nothing new there.

With the pack unlocked and the bolts out, things start come apart, then rapidly stop moving.  I've played with enough BionX packs of various ages that I've got a good idea why - adhesive.  The newer the pack, the more goop holds it together.  However, with the exception of the Trek Valencia Ride+ pack and the Trek Transport+ pack (which are both held together with some nightmarishly sticky black foam), the adhesives are mostly silicone based or similar, and do a great job of keeping the battery in place while not being too awful to open.

So, more force, and, pop!  There's adhesive!  On top of a foam pad that's sticky on both sides.  I do sometimes wonder about the engineering that goes into these packs... beautiful electronics, foam pads, and, "Hey, we should put some extra adhesive stuff in.  Just kind of squirt it around, it'll work."

This pack layout is somewhat different from the older 36V packs I've torn apart.  The BMS is at the end, instead of under the battery.

And... oh, my!  Removable connectors!  The older packs are all soldered together.  That's not really a problem, but you can't easily pull stuff apart without a soldering iron.  Look!  I've removed the charge port power plug without a soldering iron!

The end of the BMS has what look an awful lot like more removable connectors...

With the 4 wire plug on the right removed and the DC input plug on the other end removed, the whole charge connector comes free and that entire case side comes clear of everything!

I can't explain how much nicer than the older soldered together packs this is.  Everything comes loose without a soldering iron!  You could replace a single bad part!  Oh, never mind, it's BionX.

BMS Details

This BMS is radically fancier than the older units.  First, it's got plugs for everything.  This excites me greatly.  Second, it has balance leads, so it can deal with battery chemistries that aren't self balancing spinel LiMn - which means more energy density with newer battery chemistries!

The big connector pair is for power.  The outside plug goes to the 7W2 connector on the outside of the case, and the inside connector goes to the battery.  I'm not sure if you can swap them around or not - I wouldn't, just in case.  They're tight, but come out if you wiggle them a bit.

At the other end is the glorious 20 pin balance cable (presumably doing more than just balancing - I assume the thermal sensor/sensors are connected here as well).  It, too, comes out easily.

With the main power connector from the battery and the balance connector removed, the 7W2 connector harness is the only thing still connected to the BMS.  One plug for power, one plug for communication.  Neat!

With those out of the way, the full BMS is exposed.

On the left side, the lowest connection is for the 7W2 output.  Above that, an identical connection goes to the battery pack.  The separate connection above the large black connector is the DC input jack, where the ~26V from the charger comes in and gets converted up (why not just use a 48V charger?  I have no idea, but I'd guess it relates to voltage limitations for exposed connectors or something).

On the top, the non-power pins for the 7W2 connector plug in.

On the right, the upper connector is for the charge port (touch sensor and LEDs), and the bottom port is for the battery pack as well (presumably voltage and temperature).  The other port in the middle is unused - it might be some diagnostic port, though the pack normally interfaces to the world via CanBus over the 7W2 port.

Of note, the whole board has a conformal coating - you can see the somewhat uneven surface in the flash glare at the bottom.  This is good - conformal coating a protective layer that keeps water and dust away from board components, and I'm glad BionX is doing this.  The very oldest packs I've worked with don't have any sort of coating, but everything newer does.  This is better done than some of the boards, as the buzzer isn't covered - I suspect this is why some BionX packs have a loud beep and others of the same generation have a very quiet beep.  In any case, I'm happy to see the coating.

The bottom is more of the same.  I'd love to get chip numbers off most of this, but the coating makes it really, really hard to read the chips - so maybe at some point in the future when I have a good microscope.

Into the Battery

Continuing on, it's time to pull the actual battery apart.

The label is, as usual, quite helpful.  This is an official BionX battery pack now, not the BMZ packs they used to use.  Well, or it's a DLG pack.  I'll figure it out in a bit.

This pack is a 13S/4P (4 cells in parallel, 13 of those groups in series) pack, 8.8Ah (which means 2200mAh per cell), with a fully charged voltage of 54.6V (4.2V/cell).  Excellent.  And I agree with the charging temperature recommendations as well.

DLG and Port Examiner

So, who's this "DLG" on the pack?

It appears to be DLG Battery (Shanghai) Co., Ltd.

In the land of "Things I didn't know before" - you can just go look at US Customs data online for imports! lets you do all sorts of interesting searches on shippers and receivers.  You can, for instance, search for DLG Power Battery and discover that they supply BionX, Clean Republic (they sell the Hill Topper ebike kit), and K2 Energy Solutions.

One can search by an importer of BionX and discover that they import quite a bit from from Hana Microelectronics and Phase Motion Control.

I'm actually quite surprised how much one can find about relationships with Chinese companies online.  In any case, BionX certainly imports a lot from China.

Back to the Battery

Right.  The battery.  I should stop getting distracted by shiny business relationship searching and go back to the battery teardown that people are actually interested in.

The first thing I noticed about this pack is that the fuse is in the battery wiring - not on the PCB.  This is a really good thing.  Previously, the pack could dump more or less unlimited energy into the board before it hit the fuse (I've seen pictures of the results of this, and it's not pretty).  In the wiring is a better place for a fuse.  A 30A fuse on a 48V pack works out to around 1500W, so there's plenty of spare power, and the fuse seems to be suitably sized for the wiring.

Every generation of BionX pack seems to be better padded and protected, and this is no exception.  There's a layer of double sticky adhesive foam (the left part is slightly thinner where the pack tapers down), and there's a layer of goop on top of it.

Below that is the normal plastic shrink wrap of the packs.  You'll note that this pack is weirdly shaped, and not the usual rectangular brick of the older designs.

On one end, all the wires come out.  They're nicely protected by a layer of foam and some plastic board.

Balance Lead

The balance lead is the most obvious change from the older packs.  Now, the pack can monitor the voltages for everything, and (in theory) have better temperature monitoring as well.  You'll note that some of the wires are two colored - that's standard when you run out of colors.  It's been a while since I've dealt with that (I think the last thing I rewired with multi-color wiring was my old RX-7, for, uh, purposes), but it's nice that they're not reusing colors for different things.

What's really interesting is the blob of shrink wrapped things coming out of the balance connector.

They're yellow.  Hm...

And, once you pull the heat shrink off, you're left with this.

I'm not 100% certain, but it looks an awful lot like a PTC or polyfuse.  This is more or less a solid state self-resetting circuit breaker.  Put too much current through it, it blocks the current (mostly) until things cool down.  And it's protecting every single wire here - fancy!  This is a nice touch, though it being at the connector side reduces the protection it offers a bit.  Were it on the battery side, it would protect more wire against shorts.  Still, I haven't seen this on any other pack I've pulled apart.

Around the Edges

We must go deeper!  With the heat shrink removed from the battery, this is what one is left staring at.  A somewhat oddly shaped battery bank held together with plastic insulators on the top and what, to my slightly trained eye, appears to be multiple thermal probes in the pack!  On the bottom, one sees a rainbow of balance wires snaking around to the various points where they tie in.

Above, I mentioned that I expected 2200mAh cells.  The pack delivers!  The cells, at least in this pack, are Samsung ICR18650-22P cells.  The "22" suffix means a 2200mAh cell in Samsung's notation.

With some tape out of the way, the top of the pack appears to have four thermal probes going into the pack.  I haven't tested them, but if it's anything like the older packs, these should be bog standard 10K/25C thermistors.  I like the addition of multiple - the BMS can better detect hot spots that might indicate a problem early on.  The older packs just had one.

This looks like balance wires to me...

The small wires run to each voltage region of the pack, and allow the BMS to both monitor the per-cell-group voltage, and to balance the pack, charging or discharging each cell group independently to bring all the group voltages to the same level.  It's something one needs with newer battery chemistries.  The older LiMn cells BionX used would self balance (quite well, in fact), but that's not a normal trait for lithium batteries.

One lone balance wire on the other side...

For those curious, the pack voltage (in aggregate) is 46.9V as I received it after sitting for a while.  That's about 3.61V/cell if everything is balanced.  Sometimes things are, sometimes you find a horror show at -0.25V/cell.  I don't see any obvious corrosion, though...

The foam end cap comes off to reveal more of the wires and more tape.

The other end is also padded, with a few balance wires going in.  This pack is very well mechanically protected from shock and vibration!

Into the Core

It's time to remove the plastic insulation layer and see how this pack is laid out internally.

The plastic is slightly brittle, and comes off easily enough, revealing the pack guts!  The metal layer is the nickel strip that connects the cells together into banks, and then into a full battery pack.  The balance wires go into each of these voltage regions to balance the pack.

The layout is certainly interesting, and not what I'd expect.  It's sort of a "cram things in wherever they fit" layout, with plenty of custom cut nickel strip.

The bottom is more of the same, but with temperature probes run through as well.


There are at least 7 different shapes of nickel strip I can count between the top and bottom.  Someone likes their custom cut nickel strip.

Here's a closeup of the strip and the spot welds.  Each point is hand welded (there are three welds/6 points per end, but they're not consistent), and then apparently checked by someone with a marker.

And look at those weird triangle spacers!  I've never seen those before.

What's interesting is that the nickel strip, the welding (three hand-blasted spots), and the marker remind me very much of a different pack I tore down - the very first one!

This is the 2011 iZip Ultra battery pack.  Same style connectors, same three hand-done welds, and same marker marks.  Also, Samsung cells.  If it's not the same company doing the build, there's a lot of similarity!

Finally, there are those triangle spacers.  I've never seen them before, but here's what they are.

They're spacers for a tightly packed 18650 pack.  I have no idea if they serve any other purpose (thermally or electrically), but I guess they're easier than hot gluing things together.

Technical Notes

If you're just reading along for fun, you can skip this section.  I promise you won't miss anything.  If you're arms deep in one of these packs, you may find it useful.

Balance wires by voltage region, as well as the measured voltage on my pack.  I don't know why this pack had been marked as failed - the battery voltages look fine to me.

Negative: Large black
1: Brown, 3.60V
2: Purple/white, 3.60V
3: Orange, 3.60V
4: Yellow, 3.60V
5: Green, 3.60V
6: Blue, 3.60V
7: Purple, 3.60V
8: Grey, 3.61V
9: White, 3.61V
10: Grey/White: 3.61V
11: Red/White, 3.61V
12: Blue/White, 3.61V
Large red: Positive, 3.61V

Thermistor wiring notes:  The "region" refers to which balance wire is feeding that region, or which regions it's between.  See also the teardown pictures.

Orange/Black and Blue/Black: Thermistor Ground
Green/Yellow: Grey region thermistor
Green/Red: Blue region thermistor
Red/Blue: Yellow-purple/White region thermistor
Green/white: Blue/white-Purple/White region thermistor

Is this Rebuildable?

Yes - at least the battery pack part.  Just not with my current tooling.  I'd need a different spot welder with a handheld welding pen, at a minimum.  And I really don't know if the BMS will be happy after changing out the pack - I know it stores a good bit of state, and I don't know enough about it to have the slightest clue what it would do if it ended up with larger cells.  Or even just new cells of the same original capacity.

Do you have one you'd be willing to experiment with?  Let me know.  I'm interested in trying, but I cannot make any guesses about success at this point.  It would have to be a pack that hadn't been warranty-bricked.  I'll only charge you materials cost for the cells and a welder, and if it doesn't work after being rebuilt, I'll give you 50% back.  Ideally, you'd be somewhere around the Boise area.

Final Thoughts

I was really excited to see what made up the newer BionX packs.  Having seen it, I'm well and truly pissed off at BionX.  This is a beautiful, modular, repairable pack - that you can't repair, because they won't sell you the parts.  Or even sell the certified dealers the parts.  You can buy a new pack for $1000, though, if anything fails!  They might even offer you a loyalty discount!  Twenty whole percent off a brand new pack if one part of your pack fails.

BionX claims they care about "environmental sustainability" - their about page says, "BionX is on a mission to revolutionize the way people move. Our award-winning electric drive systems are answering a growing worldwide demand for alternative transportation solutions that enable environmental sustainability and a higher quality of life."

Having to throw away an entire pack because one modular component failed and you can't replace it is a really weird way to implement "environmental sustainability."  And having your customers spending $1000 on a new pack because something failed inside it is a really weird way to implement "higher quality of life."  Higher corporate profits?  Sure.

If they intend to treat each pack as a single unit to be thrown away when worn out, why bother with modular connectors?  They're a complete waste of money if the pack will never be repaired.  Solder is perfectly good, and the older packs are built like that.

If you do put in nice modular connectors such that you can literally replace each component as a unit, and go to the distance of protecting the balance leads from shorts - why refuse to sell components?  I get not selling them to end users, but a bike shop with a tiny bit of training should be able to replace these cells, or the board, or a bad charge connector.  I know that a 52 cell pack can be had for less than $1000, because I do this in small volumes, and I charge a lot less.

I suspect the 11.6Ah version of this pack is simply the same thing with 2900mAh cells (2.9 x 4 = 11.6).  So, in theory, one could swap the packs around.  I have no idea what the BMS would do about this, though.

It just frustrates me that BionX users are left out in the cold if something fails.  It's not environmentally friendly to trash an entire battery system because one part failed, it's utterly consumer hostile to build a repairable system and then refuse to let it be repaired, and it's a waste of some very, very nice engineering.

But that's how BionX has operated from day 1.  Their first I2C packs are utterly abandoned, and while the cells can be replaced, if your BMS dies (which happens), you're up a creek.  There are literally no replacements available short of salvaging other systems - and good luck finding those.

Buy accordingly, I guess.  They still make great systems to ride, but you're more or less at their mercy when anything fails, and as you can't just buy a replacement part, you have to buy a whole new replacement battery pack.

If you do have a working-but-weak 48V pack you're willing to let me experiment on, I'm interested!  Get in touch with me with the contact form on the right.


  1. How does BionX remotely brick a pack? There's no communication coming into or leaving the pack...right? Also, if all the voltages were good, then what is the failure point? Thanks for another cool post!

    1. It's not done over the air - it's done at an authorized BionX dealer. The dealer plugs the pack into the computer, a BionX tech remotely queries the pack, and if they decide it's going to be replaced, they issue commands that brick it, and tell the dealer to recycle it locally.

      I don't know why this particular pack is dead, though I know a number of the 48V packs have had problems with output transistors failing.

  2. Physically what happens to a bricked pack? Is it just deeply discharged?

  3. As far as I can tell, the controller is disabled. That's about it.

  4. Have you explored getting a CANbus toggle and software that can connect to a Bionx system so you can view those controller parameters? I saw a link on Endles Sphere by a German poster.

    1. I do have the Canbus interface hardware - I just haven't had time to get it all set up yet.

  5. Russel
    That is a most interesting insight into the Bionx world of 48V powerhouses! You are enlighten us Bionx enthusiasts about what goes on under the plastic covers. Bionx has done a pretty good job keeping our paws out of their systems to discourage battery rebuilds as you have been doing. It's a wonder that they have not cemented the plastic battery housing halves together so you destroy it on disassembly. It's also remarkable that this housing has not change significantly since 2003, except for the contents.

    I have been commuting for more than 10 years with the assistance of Bionx and have sampled its many flavours. The very early batteries were of course (heavy) NiMh and I had one sent back to the factory for a rebuild. With the move to Lithium and the horror stories around exploding battery packs we are all now in a straight jacket. Bionx did not and could not "brick" the battery packsk in the early days. That was also when they still built their systems with the Philips I2C communications standard and before the BBI interface that is now part of the the Bosch automotive CAN bus standard. I remember being disappointed at the time when Bionx abandoned I2C in favour of CAN bus as Bionx decided to no longer support the older system and I had to fork out big bucks for a new kit. I think in retrospect it was the right move to adopt CANbus and I have been impressed with the latest 48V offerings and in particular the DV500 series.

    Speaking as a user, I like that Bionx has put more of the battery charging regulation in the battery housing in that the charger is no longer this huge fan driven brick that is a pain to take along should one have range anxiety. Your teardown shows me that this arrangement results in a weight saving overall when you factor in the charger weight. I think more than that, integrating the charging electronics in the battery pack has allowed Bionx to capitalize on its advantage of powerful recouperation when in regen while going downhill. With the higher torque of the DV500 for example, recouperation can be quite significant, and perhaps without subjecting the lithium battery to unwelcome votages/currents. I have ridden my DV500 to the 3000 ft elevation on our local (Vancouver) mountain from almost sea level and discovered that the console was again showing full battery charge before I reached sea level again. With this advantage and having 555 Wh in the new battery pack my range anxiety is a thing of the past.

  6. Getting back to your teardown! I also like that Bionx is now using more connectors rather than wiring. You were able to strip down the circuit board quite nicely! As you astutely pointed out, these battery packs with internal balancing of cells appear to be designed to accept a larger variety of battery chemistries. Given that Bionx is designing its battery packs to be disassembled more easily (?), would this suggest that they will be more willing to allow Syonyk Enterprises (or a dealer?) to rebuild their battery packs since shipping a case of individual cells does not incur the steep regulator expenses. I hope so! As you also uncovered, these battery packs are coming directly out of China rather than Germany. I'm sure Bionx has a healthy markup on the batteries but their expertise and advantage must be at the motor and controller side. Perhaps in the future battery packs will last much longer, be cheaper to buy and easier to rebuild as with computers of the past.

    I have read that long term, Bosch is targeting their CANbus eDrive systems to have the capability to accept batteries from any supplier. Perhaps this is what is driving the decisions in the current Bionx battery designs. I have discovered that the current Bionx-CANbus systems allow a 48V battery to drive say a 24V system. This apparently will even work in reverse, (which I hope to fully test) where the 24V LiM battery operates a 48V DV500 system. Bionx is not particularly open about this as but Bosch has already captured a huge part of the ebike market in a very short time, Bionx has to go with the flow. Bosch must have learned from Bionx since Bionx adopted CANbus some 7 years ago for ebikes. As users we know that the battery cost and life affect our purchasing decisions and ownership costs. Perhaps going forward we will see more options available to us so "bricking" a battery intentionally will be a thing of the past?

    Thanks Russell for your explorations and insights into the world of batteries and Bionx. Maybe someone will ship you an expired Bosch ebike battery to tear down?

  7. Hi Syonyk,
    Impressive tear down, as always. it is annoying that companies boast of sustainability while doing business like this. Worst offender is Apple, IMO, who started the trend of consumer devices with non-easily-consumer replaceable batteries on phones and laptops. If it needs specialty screwdrivers, it will put off some consumers, so cause devices which are repairable to be trashed or recycled instead of reused, which is far more sustainable. I'd forbid the manufacture by law of devices like this except for very few exceptions.
    Sorry about the rant. Anyway, about the trash can icon, have you really not seen trash bins like this?
    That's the most common residential sort here and in quite a few European countries -- the pair of wheels is to allow them to be wheeled from their niche near the building to the road, along which the garbage compactor truck is slowly moving.
    ~~~ wavelet

    1. Apple isn't as bad as BionX. You can get replacement parts easily, and the tools are an annoyance, at worst. Your local phone repair shop can replace the battery/screen/etc, and the other bits don't fail terribly often without some sort of actual abuse.

      As far as the garbage can, yes, I'm familiar. I just don't think the icon has much in terms of resemblance.

  8. I'm wondering if one will be able buy circuit boards for the battery system for the Bosch systems like one can for other automotive products. It seems to me that when you buy an ebike or ebike kit the life expectancy of the motor is much longer than the battery with the battery last maybe 5 years if cared for. Given the cost of replacement battery packs from OEM's it would be good for the ebike business if people like Russel were given the opportunity to become certified to rebuild them. That would also help keep motors out of the recycle heap. Most times a used bike appears for sale, the deal falls apart because one has to factor in a new battery. As we see more ebikes on the market this could all improve. I certainly hope so.

  9. Hi Syonyk,
    I have a similar Bionx battery pack, that has a cell balance issue:the BMS goes in error (blinking green) while charging, in the first 30 seconds. Measuring the voltages on the balancing connector I have the lowest at 2.6V and the highest at 4.2V. What would you recommend? Maybe a discharge/depleting of the pack would level out the cell voltages, and then I could try to recharge it? But in this case what is the lowest voltage I should discharge?

    1. I wouldn't attempt to discharge the pack - you'll drive those low cells deep into the damaged region (arguably, 2.6V is already well into that region). If you do anything, you should try to charge the low banks back up to 4.2V, then cycle it and see what the voltages are.

    2. Following your advice, I recharged the lower value banks to 3.5V (charged with 4V for about an hour), but looks like it's still not enough as the highest bank is 4.2V. What is the maximum allowed voltage difference for cell balancing?

    3. I don't know what the BionX controller will tolerate, but an hour probably wasn't enough to charge that bank. Let it run longer, and try to get it up to 4.2V.

      However, pay close attention to the temperature while charging - a bad cell in that parallel group could be responsible for the voltage drop, and dumping more energy into a bad cell is quite unwise.

      There's no valid reason for one bank to be that far off - something has to be wrong inside that bank that exceeds the balancing current's ability to correct. I don't have any real speculation beyond a bad cell, but if one bank in a pack I owned was at 2.5V when the rest of the pack was at 4.2V, I'd probably replace that whole bank. I'm normally unhappy if I see more than a few hundredths of a volt difference between banks, and the older BionX LiMN cells, with no balancing gear whatesoever, will reliably maintain a balance to within 0.01V.

  10. I have an old IC2 Bionx and the battery is still OKish. But after reading your post I've tried to find a solution for a 'bricked' pack by looking into some German forums (it seems Bionx is quite used/hacked there), but I couldn't find any. I don't speak German so I had to use a known translator in my search, which quite often gave incoherent results. But from the little I gathered and understood, there's maybe a company in Germany that can (apparently) fix this, it's called Liofit. Maybe I'm wrong. Cheers. PS: I know you're aware of what 'Crustulu' (Endless sphere) did to get his CANbus system to work with a dead BMS. I wonder if his solution would work in bricked pack.

    1. I'm working on some options with regards to failed BMS boards, but nothing fully developed yet.

  11. Recently exchanged 48v, 11.6 BionX battery under warranty. Although OHM/BionX (OHM makes bikes with BionX systems) claims these "smart" batteries go into deep sleep mode when stored, and can be shelved for 18 months (check the charge every 6? months) my battery was behaving badly six months after purchase. I learned the battery purchased in Feb of 2017 was manufactured in 2015 July. Some experts suggest recently manufactured batteries are preferred. I had no idea where this 18 month old battery was stored or if it was stored with a full charge as recommended.

    I could not find anyone to rebuild the "suitcase" battery (case with handle) that came with the 2013 OHM bike. Canbus. Some experts maintained BionX goes to extremes to prevent rebuilding. Your tear down reveals why my new battery has a hump/bump vs tear drop design that went back to BionX. I am referring to that portion of the battery assembly that sticks out. The hump design case is supposedly newer model.

    OHM arranged the exchange with a BionX dealer. Dealer upgraded the bike firmware during the swap and told me they are sending the returned unit back to Bionx, which impressed me as strange since I was told the dealer was going to disable the old battery, giving me the impression it was going to be trashed. Is BionX interested in why I was experiencing decreased performance/longevity?

    Equally perplexing is the charging suggestions for this new beastly battery, which impress me as contradictory. They suggest recharging after every use while also mentioning charging it after depleted 50 percent (or greater) is advisable. Hmmm. There are 8 bands to reveal remaining charge and each band can achieve 12 to 18 miles depending on power demands by rider. It is referred to as an 80 mile battery. I typically recharge with 3 bands remaining after 60 to 75 miles. My point? It is very difficult for the average rider to deplete this battery below 50 percent on a single ride. I will probably continue to charge after 3 to 4 rides, when the battery is below 50 percent, or should I adopt the charge-after-every-ride method.

    Regarding the original suitcase battery that was set aside about 1 year ago. I've been hearing an eerie noise for the past few months and was unable to find the source. It sounded like a smoke detector battery running low but that was not the problem. Turns out the original suitcase battery, which had been boxed up to ship to some battery rebuilding service, who offered to buy it but never followed through, was making the noise !! The battery was emitting low charge signal approximately 1 year after it was boxed up for a shipment to rebuilder that never happened. I should toss the old battery and charger but hoarding habits prevent me, despite the fact I swore to smash the annoying noise making device with a sledge hammer when it was discovered. Never imagined that geriatric, 35v? battery maintained enough juice to produce noise one year after it was set aside. It would squeal once or twice every 24 hours. Just enough to be a haunting annoyance but not enough to track down the source.

    1. Your battery charging method is fine, and probably will lead to longer pack life than charging after every ride. Lithium is most stressed when fully charged and fully empty.

      I have lost an Annoy-a-tron in my dryer before and taken many hours to find it...

  12. Thanks for confirming battery charge procedure. This $1,200 battery was the same amount paid for 1968 Chevy Malibu (exc condition with 60,000 miles) in 1972. Absolutely bizarre.

    One battery expert was surprised I only charged the battery that was replaced 30 times in approx 6 months, implying it should be charged more frequently. Others suggested either method is OK: after every ride OR below 50 percent, but instinct told me too frequent charging is detrimental.

    The dealer who swapped the battery did a firmware? update that changed the function of dashboard controls (not appreciated) and could be responsible for another quirk. The battery case is now super-sensitive and activates the automatic voltage light ring around the charging port, aka touchport, too frequently. This battery has GREEN, ORANGE and RED light ring to provide voltage status. GREEN = 75% and higher I believe. The 48v battery that was replaced correctly activated only when finger was placed inside the receptacle indentation. The new battery activates if the top of case is touched 6 inches from the charging receptacle. I recall reading this feature should not be over-used; more than two or three times in 60 seconds. WHY ON EARTH would they make this feature as adjustable, and why would default setting be high sensitivity. It isn't a major concern since the battery is covered. It was also mounted upside down, with 3 corners of the case contacting the bike frame benefiting from foam mounts to minimize shock and vibration.

    RESPONSE FROM COMPANY...........As for the touchport, the sensitivity is adjustable through a dealer meaning you can make it less sensitive or more sensitive. Right now it might be fairly high, it requires you to bring the bike to a dealer though.

    1. You can find my thoughts on BionX's business practices and actual sustainability elsewhere on this blog.

      I'm not particularly familiar with the 48V packs, though. They're still for sale, and I try to avoid stepping on BionX's toes. They have to know I'm out here doing this stuff if they have any interest, and they haven't contacted me to tell me to knock it off. I figure that by staying out of the way of their current pack sales, I'm not directly competing with their systems.

      And, sadly, on the 48V packs, most of the failures I know of aren't actually cell failures. I've got a few 48V BMS boards with fried output transistors and the like.

  13. I've covered the entire blog several times. Although it is beyond the scope of my technical prowess there are tidbits which can provide a novice useful insight, such as the complexities involved regarding BMS and the issue regarding transistors. If this battery behaves badly I may benefit from the insight.

    This BionX system came installed on an OHM bike. Both companies located in Canada. They generally receive favorable reviews. In 2015 June I happened to search C-List and found a desperate dealer intent on liquidating OHM and BionX stock. He had 2013 OHM Urban XU700 with 36v, 9.6AH battery with handle known as suitcase battery. 350 watt motor. $3,500 MSRP with 37 miles. He let it go for $700 cash and told me a new battery would cost about 500. I got one year of use (800 miles) before the original battery was fading. Several top notch battery rebuilders told me they could not refurb it. It undoubtedly has the same guts as the 36/9.6 teardrop design you dissected. You can have the battery and charger. I only need a shipping label if you have Fed Ex or UPS account.

    1. Are you interested in a replacement battery for it? I know at least some of the Ohm systems use the same brick, and people have successfully installed my brick in Ohm systems.

  14. There could be a demand for rebuilt suitcase battery. eBay battery rebuilder in Ohio has sold one.

    You can include the charger. The Sport XS700 is virtually identical to XU700 Urban.

    --- Rebuild-service-for-OHM-Sport-XS700-and-XS750-Battery. $475-----

    OHM CEO told me "50 percent of older bike owners upgrade to more potent battery". A costly outlay for battery, charger and mount. $100 installation for those lacking necessary skills.

  15. No need for a rebuilt suitcase battery after installing 48v, 11.6AH. Your offer would have had greater appeal before ordering new battery in Jan 2017. Do you concur with those who have a suggest getting a newly manufactured battery (ideally less than four to 5 months old) while implying lithium batteries begin to decline from day of manufacture.

  16. I'm just reading old BionX stuff now that they are in reorganization and users are stuck with what they have.

    Is there a way to separate the battery from the Canbus controller? To make it a supplier of volts/amps to use on other motors and systems?

    I have a 48 volt battery. Yet, the charger is a 26 volt charger. How do they do that? DC to DC converter? I'm really old, and from the early 1970s, the DC to DC converters we built back then were made from ferrite ceramic cores to handle the amp range power. Now?

  17. Hi Russel,
    I have a HT 250 XL Pack here that has lost current over the winter. The BMS wouldn't charge it therefore I'm looking into recharging it through the BMS Connectors to make sure no single pack gets overcharged. The wiring to the pack seems to be slightly different and also the board is the SMC6.2B instead of SMC6.3B.
    Would you have specs for the 6.2B BMS connector pins to make sure i charge the right port?

    1. No, sorry. I don't have any technical details on BionX packs beyond what I've posted on my blog.

  18. Great post! Especially for me when I have problems with the exactly the same battery. I had a hard time opening it and reading about the adhesive helped a lot. Thanks!

    Now to my problem. Last week my battery just died on me from being at good health. I connected to the charger after a long ride as I normally do and the day after it was blinking red. I then reinserted the charger the led just flashed and nothing more. Since it is 3.5 years old (no warranty) I decided to open it up and measure the cells, they were scary low (24V) and I charged them back to 42V with 100mA, hopefully saving them. What I also found was that a mosfet (Q307) on the BMS (SMC6.3B) had burnt. Further investigation I also found Q104 and D102 burnt.

    Fixing the board is probably hard/impossible without schematics and stuff so I was wondering if you have any idea were to buy a replacement board or dead battery back from were I can take a board?

    1. I wish I had better news. You've suffered the standard 48V failure, where the BMS dies, and takes the pack with it. Since there are no schematics, and BionX doesn't sell replacement boards, you're sort of up a creek. Any "bad pack" you can find of the 48V variety almost certainly has a bad BMS board (I've got half a dozen of them in my storage unit for research purposes), and BionX isn't currently in business, so I'd buy a replacement pack quickly, if you can still find one.

      Good luck.

      This is the sort of reason I rail against proprietary systems so much on my blog...

    2. Continuing this story I have almost repaired my SMC6.3B BM but still no cigar... Further investigation I found out that the 5V supply had died and was currently 12V, ooops. The 5V is generated with a TPS62110 DC/DC (U103). I managed to replace this using hot-air reflow station, it was really hard because of the thick copper. I also replaced the mosfet Q307 (IRF6668) and Q104 (guessed PNP transistor) and D102 (guessed schottky). After doing this my system is working again except for the charging part... So I did further checking and found that Q309 mosfet also was burnt (lower part of the H-bridge). Together with that the two 4.7ohm resistors connecting the mosfets to the IR2011S driver. And finally I think I found the failing part, the IR2011S driving Q307 & Q309. I have just ordered a new one so next week when I can test it will be exciting!

    3. When they fail, they fail... wow. That's an awful lot of blown parts.

    4. Exchanged the IR2011S and it is charging again! This must be what failed causing the H-bridge to short. Don't know why the 5V failed though. Anyway it now seams to be working however I have some suspicion something more is broke, time will tell. Also while charging it gets pretty hot. Around 75 deg Celsius around the big inductor. Wonder if this is normal...?

  19. I just had to read about your teardown of the 48 volt battery again. I have one that is the same specs as the article except that it is a rack-mount which simplfies construction a bit. I bought this battery as part of a system in October of 2014 and I had good service for three years and over 10,000 miles. Last Fall I noticed that the range was decreasing. And this year the system on the trike was acting funny. Sometimes I had support, I ride pedal assist, and other times no support. It seemed like I had to spin the pedals very fast with a moderate effort to get any assist. Another problem I had was that the system would momentarily shut off (and back on) a few times at the beginning of the ride. I noticed it because there was no assist and I had to start assist again. The daily distance would go back to zero.

    I spent time checking all the connections using contact cleaner and silicon grease trying to make them good.

    Then the other day I did an experiment. I was on my way back home so decided to use the throttle. Well, the power came on and shut off. On and off, up to almost 20 mph and then coasting below 10 mph before the power came on again. Jiggled wires from the throttle to the control, G2 control console, to no effect.

    I have a wimpy 36 volt BionX battery, from a PL 350 HT RR M, 37 V / 6.4 Ah / 237 Wh, and I put it on the trike today. The system was fine, worked like almost new. I ran on the throttle only around a few blocks in the neighborhood. I had to put a charge back into that battery, because I hadn't used it for a while.

    Could the behavior of the 48 volt pack be explained by having the PTCs making and breaking the circuits on the remaining good strings of cells? The cells that are now trying to pump out 350 watts?

    1. I doubt the PTCs are involved. It's probably, like almost all 48V pack failures, a BMS board on the way out.

  20. Hi!Can be rebuilt this pack from new cells,if the BMS board is good?Thanks

  21. Hi there,

    i have a weird problem with my S350DL. Its circa mid-2016 bought brand new on my hybrid bike. 48V 6.6Ah version with 600+ recharge cycles on it.

    At this point the motor cuts out when i get a bit of load on teh system e.g. minor hill, seems worse when carry more stuff in my pannier. I am on assist level 2 or 3, getting up to speed and the motor/assist will cut out, with speed going down to 0 km/h but power remaining on the console. I can adjust assist levels but cannot never get the assist to kick back in. I have ot power off the unit and power back on. When on flats on a more difficult gear, generally does not cut out. When using Throttle only , it can cut out more than not.

    The store tried a different battery and seemingly the problem goes awaybut they have not put it through my exact ride scenario. I have checked all cables, connections, cleaned them all, etc. What do you think?

    1. It's a 48V BionX system. Replace it with something you can work on and reason about.


Comments on older posts are moderated due to spam issues. If you don't see your comment immediately, and you weren't just spamming me with some irrelevant comment and a link to whatever site you're trying to SEO, your comment should show up relatively soon. If you're trying to use my blog for your SEO purposes, your comments will never show up, so don't waste your time.