Adding LifePo4 to BionX System

[chvidgov.bc.ca]

I have a newish PL350 Bionx system, with no battery. I would like to run it on my Ping battery, as a "discharge only" type setup. How can I disable the regen, and where do I connect my "Ping out" leads to? I got the bike for 100 dollars as they were fed up with Bionx and didn't want to buy a new battery (understandable), and I took off the system because it was a nice Devinci Vancouver bike to ride as a normal bike...

[GCinDC]

they were almost crying. One customer was able to blow up three battery pack circuit board in 2 week time. The same story, fully charge the battery, start the day by go down a steep hill with highest level of regen. They keep replacing the battery pack under warranty.

Hi Ken, Interesting story. But the battery replaced under warranty was NOT the original bionx battery though, was it? It was probably a lifepo4 with a bms, no? I wonder if the old (even dead) LiMn battery was there, if the board would have blown... It seems to me that if that is kept, it can absorb all the regen charge, esp if it's blocked by a bms.

I've been thinking about using lipo WITHOUT a bms - I use it now actually on my bike (see sig). My max regen current seems to be 8A, but the CA shows peak regen of 18A!

I'm very surprised if you get 60A regen off the bionx!

I wonder if I could test this with a multimeter. I should be able to run wires onto the + and - terminals on the board and ... oh, nevermind, I could just measure voltage, and that'll be the same as the diagnostic console... I was hoping I could switch it to measure amps too, but the multimeter can only measure max 10A, and it would have to be in line w/ circuit, not simply across poles... nevermind!

I have a CA but no idea how I would wire that...

[GCinDC]

I have a newish PL350 Bionx system, with no battery. I would like to run it on my Ping battery, as a "discharge only" type setup. How can I disable the regen, and where do I connect my "Ping out" leads to? I got the bike for 100 dollars as they were fed up with Bionx and didn't want to buy a new battery (understandable), and I took off the system because it was a nice Devinci Vancouver bike to ride as a normal bike...

No battery or a dead battery?
Battery case w/ board inside?

[chvidgov.bc.ca]

No battery/no battery case or board from battery case...

[GCinDC]

No battery/no battery case or board from battery case...

Considering all the electronics inside my battery case pictured above, if you want to ride the bike - as opposed to conducting a science experiment - I'd suggest you get yourself a new battery, controller, and throttle, and just use the wheel like doc. have you seen his post? It'd be a shame to discard the console, but this is uncharted territory! And being a closed system not easy to mod...

[itselectric]

they were almost crying. One customer was able to blow up three battery pack circuit board in 2 week time. The same story, fully charge the battery, start the day by go down a steep hill with highest level of regen. They keep replacing the battery pack under warranty.

Hi Ken, Interesting story. But the battery replaced under warranty was NOT the original bionx battery though, was it? It was probably a lifepo4 with a bms, no? I wonder if the old (even dead) LiMn battery was there, if the board would have blown... It seems to me that if that is kept, it can absorb all the regen charge, esp if it's blocked by a bms.

The story that I have mentioned, was not using Bionx battery, it was lithium magnese cell (yes similar type as Bionx). it has its own BMS. They re-programmed the BMS a few time, still could not manage the re-gen overcharge. Until Bionx step in, they created a new version of I2C console version 2.9 (you can't get this). This new version of console stop the re-gen at certain voltage or amp.

I'm very surprised if you get 60A regen off the bionx!

I could be wrong with the actual number, because I was not the one that witness it, but remember, bionx did not put any limited condition in the re-gen/recharge, the bionx battery will just take whatever current that through at it. To my surprised, these Bionx or Sony cell is an amazing product, that Bionx had figured it out the best way to use them, and for a long time ago. My friend Bob who help me on this project had reverse-engineering the most of circuit board, he kept tell me how good the bionx design was.

I have a CA but no idea how I would wire that...

You need create t-connector or pigtail bewteen the motor trailer wire and to the battery mounting bracket.

Ken

[ericslai]

Hi Ken,

Could you or have you plug a CycleAnlasyt in between your lipo battery and the Bionx PCB board. Are you seeing any negative amp coming from the PCB to lipo battery?

You have a high discharge rated battery, that may have avoided the problem. But it also depends how the BMS was programmed in the factor. Weather it will actually takes 40amp charge rate.

There are certain lithium battery/BMS will not accept reverse charge to battery. A good example is PING battery. You cannot regen to recharge the battery through the discharge connection. The could be avoided the problem I was having. I guess I would have put a diode to block the regen going into battery. It will turn into the a lot heat.

There are also another setting with bionx that can low the regen current, i can't remember the exact setting right now. This may also avoided the blow up by over charge.

If you only going a small down hill, it may have been avoided.

I have not tried to measure the regen current to either pack. It may be a good weekend project.
http://www.batteryspace.com/pcmwithequilibriumfunctionfor37vli-ionbatterypackat40alimitpcm-l10s30-062.aspx

[GCinDC]

I have not tried to measure the regen current to either pack. It may be a good weekend project.

ericslai, I keep thinking about this amazing solution you've come up with. How did you do it? Did you have help? Are you an EE? I searched and searched for BionX modification storys, but yours is one of the few successes I've ever heard of. I think the world would be very curious!

[WeWerk]

Hi ericslai :
how do you charge both batteries ?
Do you charge them separately with two chargers or both using original Bionx charger thru XLR ?
What is your charging time ?
THX for an answer.

[ericslai]

I forgot to test the regen in the last weekend but I will try this week.

ericslai, I keep thinking about this amazing solution you've come up with. How did you do it? Did you have help? Are you an EE? I searched and searched for BionX modification storys, but yours is one of the few successes I've ever heard of. I think the world would be very curious!

I am an ME but have studied EE and computing, 4 courses of MCSE. A few jobs before the current one, I designed and helped manufacturing rechargeable products such as camcorder and power drill battery packs so I know a few basic principles of rechargeable battery (NiMH and NiCd) pack constructions. Like most Bionx users, I was not happy with the capacity. The fuel gauge is not accurate enough so I opened the plastic case and peeked a bit. There are red and black thick gauge wires that feed to the PCBA. These are the output and input of the battery pack that I tapped to with the Lipo pack. I did the hacking myself.

how do you charge both batteries ?
Do you charge them separately with two chargers or both using original Bionx charger thru XLR ?
What is your charging time ?

I have to charge the system at work and at home (less than 19 miles each way). So I use the Hi-Power (from Bionx) charger at work. It charges through the XLR charging port of the Bionx pack. Normally it takes 3 hours to finish charging, the charger will shut down.
At home, I charge it with a generic 2A charger but I don't like the full charge voltage which was set too high so I use a timer to shut down the charger. This is always done at home so I set the charging time to be 6 hours. The end voltage is about 41.2-41.5V. I consider that is good enough. The Lipo pack came with a Tamaya type charging connector so I charge the whole system with that.

Since both packs' output + and - are tied together, they are paralleled as one big pack in terms of capacity. The difference is the charge input. Bionx's controller monitors the energy input so the fuel gauge will show full when charging is complete vs the un-monitored Lipo pack. After the system has been fully charged at home with the generic charger, the Bionx fuel gauge will show no juice left! This is not a problem since I know the system's voltage is at 41.2-41.5V by measuring each pack's voltage. I did not get 2 Bionx chargers because of cost.

I tied the two packs at the output side because the LiMn does not need to be balanced at all. Each cell (cell group) is almost identical in terms of charging and discharging characteristics. The Lipo pack has a built-in BMS to monitor charge and discharge. During discharging, the Bionx will treat the whole battery pack as a 20Ah pack. During charging, the whole pack will get a full charge from either charger.

Please DON'T mix NiMH or NiCd this way. These cells need to be monitored with the negative delta V method for safe charging. Their self discharge characteristics will ruin the parallel packs if they were tied together for too long. They can be paralleled for discharge BUT they MUST be charged separately with appropriate chargers and discharged through diodes.

[GCinDC]

Wow, awesome information. I've been studying it for an hour here....

Do you get increased speed and/or torque with the additional pack? Or is it just increased range?

Since both packs' output + and - are tied together, they are paralleled as one big pack in terms of capacity. The difference is the charge input. Bionx's controller monitors the energy input so the fuel gauge will show full when charging is complete vs the un-monitored Lipo pack.

Any idea how it monitors the energy input? The actual Amps/Ah or the voltage?
Also, when you reconnect the lipo pack to the bionx, won't there be current flow between the packs to balance out any difference in voltage?

If the charger limits amps, my plan is hosed -- I've been planning to parallel a 7s 5p (11Ah) lipo pack without a BMS by connecting it to the BionX + & - leads AND charge it ONLY with the BionX charger.

My logic to think I could is:
- Once the packs are paralleled, they'll balance each other
- Since they're paralleled, the voltage will be the same for 'each' pack
- The lipo won't be isolated, so it will accept charge (& regen) current
- The charger will shut off when the pack reaches the voltage, so it'll just take twice as long to charge a pack that's twice as big

Think it will work? Any suggestions?

I'm not crazy about pumping the lipos up to 4.2, even though I will balance them via the balance taps.
I was going to do 8s but thought that'd cause me more problems...
And I was originally going to test a 6.6Ah 8s lipo pack, but decided it'd be better to duplicate capacity as much as possible, so ordered 5 3s 2200mah lipo packs to connect in series with 5 4s 2200mah packs ($175)

Oh yeah, so did you cut the big black & red wires going to the PCBA and splice in a connecting wire to the lipo leads? Or did actually solder the lipo connecting wires to the same spot on the board?
And, on my image on page 1, the yellow charge wire going from board to the pack, does it just go to the - side of the pack, where the black must be soldered?

[ericslai]

I just did a simple test to test the regen. Yes, the current flows both directions between the 2 packs. The connection between the Bionx pack and the Lipo pack is not Anderson so I cannot hook up the CA to record the current on the road. I just ran it on a stand (the system is mounted on a recumbent trike).

Do you get increased speed and/or torque with the additional pack? Or is it just increased range?

My goal was range only so I have achieved that. The 36v motor is limited to 25mph max no load (350rpm).

Any idea how it monitors the energy input? The actual Amps/Ah or the voltage?
Also, when you reconnect the lipo pack to the bionx, won't there be current flow between the packs to balance out any difference in voltage?

I don't know how it monitors the energy input. My wild guess is the same as the CA with some software. If the voltages are very close to each other like .2 to .3V, the current will not be too much before the voltages become identical. When I connected the 2 packs together, I made sure that the voltage different was less than .1V.

If the charger limits amps, my plan is hosed -- I've been planning to parallel a 7s 5p (11Ah) lipo pack without a BMS by connecting it to the BionX + & - leads AND charge it ONLY with the BionX charger.

You mean the High Power charger from Bionx? My 36V charger says 2Amp only. My generic charger says 2Amp but in reality, it only consumes ~50W from the mains so it took much longer to charge the whole system. The High Power (Bionx) charger has 3 wires. Sometimes it won't start charging when I just plugged in the XLR charging connector. I had to unplug and plug in again to enable the charger. The yellow wire on the charging port must be looking for either normal temp or just connection to enable the charger.

Your logic makes sense. My 36V charger stops at 41.6V. The charging time did not change from only Bionx pack to parallel pack because my route did not change. The consumption did not change. I seldom run off course and the recumbent trike helps on the aero differences.

Maybe I have missed that. Your is a 24V LiMN or NiMH? Can you verify the cell type and model number that they are actually LiMN? I would not mix different chemistries to be on the safe side.

On the secondary side of the board, the Bionx pack's + and - wires were protruding out long enough. I just soldered a wire harness to them and run the harness out of the plastic case. I had to cut a slot at the seam to let the wires come out. Then I can plug and unplug the two packs easily. The connectors are not Anderson as I said earlier, they are 20Amp Molex free samples.

If you probe (carefully) it, you will find that the yellow is not connected to the negative of the pack I think. It has been a while. It must be connected to a thermistor or something like that. The charger will not be enabled if the yellow wire was not connected to the charger.

[Canis Lupus]

Question, touched upon in the previous post by ericslai:

If I have a BionX battery which charges to 41-42 volts and I plug into the BionX's battery charging slot a LifePO4 or Lith Mag battery which charges to 46-47 volts, am I inviting problems?

Looking at Ping's 36 volt batteries, they appear to charge to something around the 46 volt mark.

[itselectric]

I made a XLR charge port connector for one of the our Bionx customer. She had a dahon bike. Here are the picture of the XLR charge port connector:

XLR-chargerPort.jpg (93.2 KiB) Viewed 1310 times

Here is LiFePO4 battery connected to the charge port.

bionx-connected-to-lifepo4.jpg (103.76 KiB) Viewed 1310 times

Here is her communication to me:

Hi Ken!

Well, I finally drained my BionX battery after 54 miles of riding on level 1 & 2 at 10 to 15 mph on flat and gentle rolling hills.

No bars are left on the control panel battery indicator. So I get off the bike and hook up the connector to the LiFePO4 battery and turn it on. Then I get on the bike and pedal a bit and on comes the power again. You are a genius!!! It works. And as you can see by the control panel and computer mileage photos 1 & 2, after 17+ miles, the battery indicator shows 4 1/2 bars. Photos 3 & 4 show the hook up to the LiFePO4 battery and I just put it on the bench so you can see, I put it into a pannier which works fine. Then when I took the LIFePO4 battery off and turned on the control panel, it showed that it recharged my original BionX battery to 4 1/2 bars. So you can unhook and ride without LiFePO4 battery and rehook it when you run out again, or just leave it hooked up and keep going.

I will let you know in a week or so what mileage I get from the LiFePO4 battery.

What else is nice about the BionX/LiFePO4 extended battery is that it keeps you at a constant speed when you are going on flat and uphills. Usually your rpm drops when you go uphill or into wind but BionX/LiFePO4 keeps you steady giving you the boost and bionic feel and that's why I like it so much.

So thanks for the connector, you should patent it and make serious $$$$!!!!

So, not only she was able double her range without any modification or opening battery, she was able get better uphill because of double amp. She currently had a PL350 20" and 38.4v 10Ah LiFePO4 battery pack. This is simplest way to get more range and doing it very safety, no modification, no opening the battery, no overcharge regen problem, warranty should be the same. In addition, she also had another eZee kit, she can use the LiFeP04 battery on both system.

Ken

[itselectric]

Question, touched upon in the previous post by ericslai:

If I have a BionX battery which charges to 41-42 volts and I plug into the BionX's battery charging slot a LifePO4 or Lith Mag battery which charges to 46-47 volts, am I inviting problems?

Looking at Ping's 36 volt batteries, they appear to charge to something around the 46 volt mark.

Canis:
Do you know how many cell are your ping battery. LiFePO4 cell when it is fully charge, it 3.65v each cell. Normally, a so call 36v (actually it is 38.4v) battery is make up of 12 cell. so 12 x 3.65 = 43.8v, if you have 13cell, then 13 x 3.65 = 47.45v. A fully charged so call 36v battery, should not be higher then 44v, unless it is 13cell, which is very odd. Check you charger, what does it say the output voltage?

Ken

[Canis Lupus]

Ken, that's a really elegant add-on pack for the BionX you've made. Well done! That's the type of set-up my OP envisaged, with all the benefits you've stated. I'm really impressed and agree with your client's comments about marketing the setup. I think you will get plenty of demand!!


Please don't let BionX dissuade or pressure you not to offer it more generally.

Question, touched upon in the previous post by ericslai:

If I have a BionX battery which charges to 41-42 volts and I plug into the BionX's battery charging slot a LifePO4 or Lith Mag battery which charges to 46-47 volts, am I inviting problems?

Looking at Ping's 36 volt batteries, they appear to charge to something around the 46 volt mark.

Canis:
Do you know how many cell are your ping battery. LiFePO4 cell when it is fully charge, it 3.65v each cell. Normally, a so call 36v (actually it is 38.4v) battery is make up of 12 cell. so 12 x 3.65 = 43.8v, if you have 13cell, then 13 x 3.65 = 47.45v. A fully charged so call 36v battery, should not be higher then 44v, unless it is 13cell, which is very odd. Check you charger, what does it say the output voltage?

Ken

Ok, the only charger I have atm is the BionX one. It says 37volt, but I remember putting the voltmeter on the battery once and it said something like 41.9 volts. I believe BionX sets the charger to cut off slightly lower than capacity to prolong overall life of the battery.

I will check with ping and the other battery makers.

What is the max voltage of the add-on battery pack you have used? Obviously, the way it is being used, ie waiting for the BionX battery to use up most of its charge first, then plugging it in, there is an initial voltage disparity between the batteries, which must be around 5-6 volts at least, maybe more.

[Canis Lupus]

Actually, as long you as you have the LiMn in place to receive regen current, the lifepo4 could be isolated (not connected to the LiMn) to provide power only. Regen would just recharge the LiMn. Discharge only provided by lifepo4. Make sense?

Yep, thanks for your post. It was very helpful.

[GCinDC]

Looking at Ping's 36 volt batteries, they appear to charge to something around the 46 volt mark.

Do you know how many cell are your ping battery. LiFePO4 cell when it is fully charge, it 3.65v each cell. Normally, a so call 36v (actually it is 38.4v) battery is make up of 12 cell. so 12 x 3.65 = 43.8v, if you have 13cell, then 13 x 3.65 = 47.45v. A fully charged so call 36v battery, should not be higher then 44v, unless it is 13cell, which is very odd. Check you charger, what does it say the output voltage?

Ken

Hi Ken, your solution is brilliant. Nice work! Is it a 12 cell LiFePo4?

Correct me if I'm wrong, but I don't think Canis has a Ping yet. He's probably looking at: http://www.pingbattery.com/servlet/the- ... PO4/Detail

where it says:

Specifications:
Suitable Wattage of Motor: up to 400 Watt, 300 Watt suggested
Applications: E-Bike, Electric Bike, E-Scooter, Electric Scooter
Voltage: 36 Volts
Capacity: 10 Amp Hours
Dimension: 150x105x150 mm / 5.9x4.1x5.9 inches
Weight: 3.70 kg / 8.2 lbs
Charging Voltage: 45-46 Volts
Charging Current: <5 Amps
Rated Discharging Amperage: 10 Amps
Max Continuous Discharging Amperage: 20 Amps
Maximum Discharging Current: 40 Amps
Discharging Cut-off Protection: 25-30 Amps
Lifecycle of the whole pack: >85% capacity after 1000 cycles. Lifecycle of single cell: >85% capacity after 1500 cycles, >70% capacity after 3000 cycles. (<1C discharge rate and <1C charge rate)

[Canis Lupus]

Yes, that's right, I haven't taken the plunge yet, and the specs you have quoted were the ones I was referring to, in particular what you have highlighted in red. Thanks again, GCinDC.

[GCinDC]

Here's the approach I'm taking, from here:

I have no idea how it will work, but it's more similar to ericslai's.

- Made 11Ah 7S LiPo pack (7 x 4.2V = 29.4V) (~$235 for the cells)

Still to do:
- Parallel each 4S & 3S balance taps.
- Balance charge the 7S pack on my RC iCharger
- Splice BionX external connector to BionX battery main power leads to controller (drill hole in seam, a la ericslai)
- Hook it up and test! (and pray)
- I should probably get a watt meter to put in line to watch.

Since there will be no diodes or BMS on the lipo pack, it will receive regen current and be paralleled w/ BionX pack. My hope is that the BionX charger will charge the lipo pack as well, and that the battery will not be close to drained to compensate for the shortening of the lipo lifespan by charging it up to 4.2V.

I think I will also put a switch on the connector so they don't sit together while resting.

[GCinDC]

PS. This is funny.

Hi Ken! Well, I finally drained my BionX battery after 54 miles of riding on level 1 & 2 at 10 to 15 mph on flat and gentle rolling hills.

54 miles w/ the BionX alone, dragging an additional LiFePo4 pack... She's a hard core cyclist! Level 1 is like the wind at your back.

How much further did she go!?

Also, I was very happy to hear about the extra power she had going up hills.

[itselectric]

Here's the approach I'm taking, from here:

I have no idea how it will work, but it's more similar to ericslai's.

- Made 11Ah 7S LiPo pack (7 x 4.2V = 29.4V) (~$235 for the cells)

Still to do:
- Parallel each 4S & 3S balance taps.
- Balance charge the 7S pack on my RC iCharger
- Splice BionX external connector to BionX battery main power leads to controller (drill hole in seam, a la ericslai)
- Hook it up and test! (and pray)
- I should probably get a watt meter to put in line to watch.

Since there will be no diodes or BMS on the lipo pack, it will receive regen current and be paralleled w/ BionX pack. My hope is that the BionX charger will charge the lipo pack as well, and that the battery will not be close to drained to compensate for the shortening of the lipo lifespan by charging it up to 4.2V.

I think I will also put a switch on the connector so they don't sit together while resting.

I really DO NOT recommend to do parallel lipo with Bionx pack without the BMS, and have the regen turn on. At least put a cycleAnlyst to test and review the result first. May be i am just being too caution what I have seen so far, how someone is losing big money on this setup. The worst thing could happen is your lipo blowup in your behind instead of the circuit board. With 10 of them, you are playing with fire.

You may be able get away with PL250, may be the PL250 doesn't had as much regen current like a PL350, may be you have not encounter a big steep hill like others who blow up 4 batteries in 2 weeks. May be your lipo able take high current recharge. But just one day under certain condition, the overcharge kick in, you don't know what could happen. Are u riding this setup or your wife riding this? I would not put my love one on this.

As fast as I know the worst possible condition of regen overcharge problem that I am talking about is. The battery was being dischage 1 volt or half volt off, if the battery is fully charged, this may no happen. But if it's 1 or 2 volt drop off the voltage, going down on very steep hill, with maximum regen, it had no condition because it has not reach the 42volt limit yet, as a result, it will regen at the maximum current at 40 - 60amp. In some case, you may never hit this extreme condition. Now, do you want to put your behind on a experimental risk?

Ken

[GCinDC]

Thanks for your honest feedback! The truth is that my wife and I haven't been getting along well, lately...

Just kidding! Of course, I would not want such a thing. That is why I was thinking about a switch - maybe a thumb press momentary type - and that she should use it only towards the end of the trip when then Bionx is weak. I will be doing many tests before she rides it, but i was tired of theories! If the voltage jumps up a lot, I diode would solve the problem, wouldn't it?

Have you seen liveforphsyics tests on lipos? He subjects them to incredible punishment before they catch fire. Also, I have a much bigger motor (9C) on a 72V 45A controller and the max regen I do is 10A... Everyone on the forum thought 60A regen would stop them so fast they would be flying out of a canon.

Thanks!

[ericslai]

Hi Canis,

Question, touched upon in the previous post by ericslai:

If I have a BionX battery which charges to 41-42 volts and I plug into the BionX's battery charging slot a LifePO4 or Lith Mag battery which charges to 46-47 volts, am I inviting problems?

Looking at Ping's 36 volt batteries, they appear to charge to something around the 46 volt mark.

I would not try to put such high voltage (46-47V) on the charging port of the Bionx. As I remember, the Bionx manual recommended to switch off the console before hooking up the charger. There are ecaps on board that are 50V rated (I think).

A fully charged 12S LiFePO4 is about 43V only. Ping's charger was set higher so the cells can be balanced at higher than 3.65V, something like 3.7-3.8V to boost capacity and balancing speed. The pack voltage will drop back to the nominal voltage in a few hours once the charger has been unplugged.

[Canis Lupus]

Thanks for the info although I must confess to being not too much the wiser. Sorry for my thickness.