E-Bike XB-502 conversion project

[ElectricGod]

I have purchased 10 16,000mah 4S packs for the moped. They will be here today. I'll configure those packs into 20S by 2P. I'm currently looking at about 60 amps load max. The Multistar packs are 16,000mah so they will see something like 2C discharge rates at most which is well within their limited 10C discharge capacity. I want to get this thing on the road sooner than later so I will build up 18650 battery holders for it as time arises, but the LIPOs will get me riding much sooner. I'm going to use the Kelly 7230S controller I already have which is good up to 84 volts. Later as time and money permits I can plan for going to 120 volts or 30S which is my end goal for the moped. Right now I have parts that will serve well for the time being. I might as well use them.

 

[Ianhill]

With my experience 60amp is bang on the money for a constant current on these after that they get hot even still thats 5kw constants thats enough to get you shifting around the 60mph mark I can't remember the motor choice now for this one you got so much going on.

 

[ElectricGod]

With my experience 60amp is bang on the money for a constant current on these after that they get hot even still thats 5kw constants thats enough to get you shifting around the 60mph mark I can't remember the motor choice now for this one you got so much going on.

I'm running two of these in parallel so it's really 30 amps per pack. This is the moped and it's getting the 3220 Astro in it. I need to buy some 1/4" sheet steel to make a plate for mounting the motor to and then that will mount to the moped frame. I'm also still waiting on 219 chain. After that the moped will be rideable again at 48 volts. I'm uber curious to see what that 3220 does for the moped. The next thing will be to adapt it use the Kelly 7230 I have and then upgrade voltage to 84 volts or two sets of 20S LIPO made of these 16.000mah packs. Then much later, get a controller capable of 120 volts and run on 30S. By then I will probably not use LIPOs as I will have had the 18650 battery packs built by then.

 

[ElectricGod]

dead post

 

[ElectricGod]

Edit: In the pictures the halls are centered over the stator teeth. this is wrong. They should have been centered over the gaps between the stator teeth. I later redid the halls in the motor to fix this mistake.

I took the 3220 apart so I could epoxy the halls in place and took some pics too. I started to put epoxy on one of the halls and realized I should take pics. Anyway, sorry for messing it up before you could see all 3 halls clearly. Essentially what I did was count the teeth in the motor and then divided 360 by that number to find out how many degrees a single tooth used. Since I wanted 120 degrees between halls, I then divided 120 by the degrees a single tooth took to find out how many teeth should be between halls. In other words every 4th tooth. I arbitrarily picked a tooth that had easy access to it and then counted 4 more teeth and that's where I placed the next hall and then 4 more teeth after that for the next hall. I lightly super glued them in place on top of the teeth so that if I was wrong about placement, I could always break them loose easily to be placed elsewhere. Since it all works correctly, they are now epoxied in place. I've looked at other peoples hall placement and also at the motors I have. I have 3 motors with halls dead center in the middle of the teeth and a couple more with them between the teeth. Go figure. People who have added there own halls also have done the same. I'm not sure that either position matters or not, but the gaps between the teeth are pretty narrow and my halls wouldn't fit between them so I used the top of the teeth as a experiment and it worked. The motors I have, that have the halls in the middle of the teeth all have a machined slot in the face so the halls are right next to the magnets. I didn't do that and instead just put them on top of the teeth. It worked just fine and every test I can try to make sure they work has been positive. I've run the motor on two different controllers that use halls and it runs on them as well.

 

[ElectricGod]

Things to do on the moped...
1. Find a 200 amp 24S BMS that I like.
2. Mount the 3220 motor
3. Mount the KLS7230 controller
4. Build the 24S LION battery holders
5. Add a power distribution BUSS
6. Convert all lighting to 12 volts via a 20 amp DC-DC converter
7. Make a mosfet switch that turns off power when the key is turned off

 

[joshuahandrich]

Things to do on the moped...
1. Find a 200 amp 24S BMS that I like.
2. Mount the 3220 motor
3. Mount the KLS7230 controller
4. Build the 24S LION battery holders
5. Add a power distribution BUSS
6. Convert all lighting to 12 volts via a 20 amp DC-DC converter
7. Make a mosfet switch that turns off power when the key is turned off

https://endless-sphere.com/forums/viewtopic.php?f=31&t=81211&hilit=BMS

I purchased one of these but it's been sitting on a shelf as the batteries that I was going to use it on from a Ford Focus electric arrived damaged. I never ended up using it as I ended up purchasing a pack with a BMS. It sounds like exactly what you're looking for and I'd love to see it put to use on a decent moped project. A big thing that drew me to it is the ability to charge to a lower voltage if you'd like and the ability to easily monitor the voltage of each set of parallel cells as I had a balancing channel on a bestek BMS fail and draw one string down to zero and never noticed which could have led to a catastrophic failure, so the feedback that it is working correctly seemed appealing. It would also satisfy number 7 on your to do list. I'd sell it to you for around what I paid for it including shipping and you could have it in just a couple days rather than wait for shipping from Russia. Let me know what you think. Awesome project you've got here. I'm working on a moped project now looking for around 10Kw power output once I replace the controller. Just recently replaced 100lbs of lead on my bike with a 25lb 2KwHr lithium pack. PM me if you're interested and make me an offer. I have 100% positive feedback on eBay: joshuahandrich91 and have been active in the sales forum here before. Can't wait to see how your project turns out.

 

[ElectricGod]

New developments...

I've been working on the battery box. I made one out of plywood. It's a little deeper than the original plastic one and much wider. Length was limited by the space I had so I couldn't make the battery box longer. It looks like I could go wider, but actually I'm up against the plastic side panels right now with the box at this size. Clearly I have more room for more batteries or slightly larger ones. With no holes in the bottom of the battery box, dirt and water wont get in either. Later sealing up the top will be interesting and making that be the new floor in the moped.

This is the space that the motor goes into.

 

[ElectricGod]

I've never welded up an 18650 pack...not that it's rocket science, but the battery box could be efficiently filled with many small cells rather than LIPO bricks that don't fill the space. That's later...however I just came across a nice deal on 100 panasonic PFs for $3.50 each so of course I bought them. They are 2900mah and 3C. Can anyone build me an ARC reactor please?!!! I'll need a lot more 18650 cells. These LIPOS were cheap and they are a quick way to get this thing powered up, but it was never my intention to stay with LIPO.

 

[ElectricGod]

This moped isn't my larger priority right now as I'm trying to get the Currie kick scooter going first. However, I pulled the bulbs out a couple of nights ago and then found LED bulbs in the same socket types. I'll have to modify the electrical system for the lights to work at 12 volts since right now it's all wired to run at battery voltage. That's a fairly minor detail since it's 2 wires that I need to disconnect from the battery feeds and connect instead to a DC-DC converter. Anyway, the LED bulbs use 1/2 the amps and are twice as bright. The head lights on the moped are pathetically dim bulbs. I found some LED bulbs that are 100 watts in the same socket size. That ought to do it!

 

[ElectricGod]

I am getting narrowed down on controllers for this ride. I think I have settled on a Sabvoton 96 volt 150 amp controller. It can really run at 106.6 volts or 26S and I can probably push that to 28S...so that's my plan. It uses IRFB135 mosfets which are good for 135 volts. According to Sabvoton support, the large filter capacitors are 160 volt. IE: 28S or 114.8 volts should be doable. Sabvoton says the controller will shut off if run above 120 volts so I should be good at 114.8 volts.

Then I have motor options...
I have another big block inrunner coming. It will have the end plates opened up so I can run it much harder with a fan. My existing big block runs at 150-160F on a hot day and at about 4000 watts. There are reports that people are doing 5000 watts on this motor without it over heating. I'm skeptical, but who knows until I try it myself. Then...I have the 12090 18Kw outrunner. I'm getting an adapter made that mounts in place of the prop adapter that has a 3/4" OD to direct mount a 219 driver sprocket. 18Kw is too much for 219, but I'll see how well it holds up. With 3/4" OD on the shaft, I can easily get other sprockets that will fit. Anyway, 18Kw on a moped would be crazy fast. Then there is option 3. I have two AstroFlight 3220 inrunners. It's such a tiny motor with crazy power. I'm not sure what I'm saving these motors for, but it's an option too.

 

[Ianhill]

219 chain might work if the gearing is not to steep and not a lot of weight but that depends on the ride and motors final rpm to how steep a gearing it needs, The end turns of the coils are the hottest part in the motor when pushed hard that's normally the point of failure where the wire has poor heat exchange not being in contact with the laminations just floating in air sometimes blowing air in the motor can move the heat into the lamination stack from one side and out the other, some motors use epoxy or encapsulation on the windings to give a better thermal path than air for the end turns to cool back into the laminations combined with forced air with a blade shaft driven from the motor It will defiantly be more robust to peak bursts.
Check out skin affect on a conductor at higher frequency's with large pole counts combined with high erpm, When erpm equals over 400hz then more than a single strand of wireis needed, the skin affect will take place and heat will gather on the conductors surface where the electrons are moving on the surface of the wire rather than through it with the higher switching frequency's, So a single strand may be lowering the motors effective output before overheating if the wrong conductor is selected.
If the erpm is high and the stator has lots of slots like a outrunner it needs thinner strands low slots and low erpm then a solid conductor is used.

 

[ElectricGod]

219 chain might work if the gearing is not to steep and not a lot of weight but that depends on the ride and motors final rpm to how steep a gearing it needs, The end turns of the coils are the hottest part in the motor when pushed hard that's normally the point of failure where the wire has poor heat exchange not being in contact with the laminations just floating in air sometimes blowing air in the motor can move the heat into the lamination stack from one side and out the other, some motors use epoxy or encapsulation on the windings to give a better thermal path than air for the end turns to cool back into the laminations combined with forced air with a blade shaft driven from the motor It will defiantly be more robust to peak bursts.
Check out skin affect on a conductor at higher frequency's with large pole counts combined with high erpm, When erpm equals over 400hz then more than a single strand of wireis needed, the skin affect will take place and heat will gather on the conductors surface where the electrons are moving on the surface of the wire rather than through it with the higher switching frequency's, So a single strand may be lowering the motors effective output before overheating if the wrong conductor is selected.
If the erpm is high and the stator has lots of slots like a outrunner it needs thinner strands low slots and low erpm then a solid conductor is used.

Thanks for the info. I've run across a few threads on winding this motor and similar 80mm motors. They claim to get more wattage out of them with larger strands than with many smaller strands. I have the 14 awg wire and I can always wind it for lower Kv.

 

[Ianhill]

No problem what's a forum for if not sharing, My plan is to rewire the boma to half the kV so i can up the running voltage to 28s range. Because its packed like a pot noodle at the moment i wont have to use a winding thats exactly half the size but double the length, I can use slightly bigger than half of the existing gauge and still fit it all in and give me higher power than i push through it now ,so it will run more like 120v×30amp for 3.6kw peaks instead of 60v×60 amp for the same effect, I've been running 3.6kw at 16s now with very good results heat wise it is more robust to heating at higher volts than it was with the stock controller and the lead acids on 1600w crawling along.
I've got my next build all in my head at the moment the main thing I need to sort is 28s BMS and charging I will be using 28s of 10ah lipo for a lightweight pack that just over 1kwh and fairly punchy, I never use more than 600wh at the moment so a small lack with high volts is plenty, main thing is i need to keep my charging solution small and mobile.

 

[ElectricGod]

No problem what's a forum for if not sharing, My plan is to rewire the boma to half the kV so i can up the running voltage to 28s range. Because its packed like a pot noodle at the moment i wont have to use a winding thats exactly half the size but double the length, I can use slightly bigger than half of the existing gauge and still fit it all in and give me higher power than i push through it now ,so it will run more like 120v×30amp for 3.6kw peaks instead of 60v×60 amp for the same effect, I've been running 3.6kw at 16s now with very good results heat wise it is more robust to heating at higher volts than it was with the stock controller and the lead acids on 1600w crawling along.
I've got my next build all in my head at the moment the main thing I need to sort is 28s BMS and charging I will be using 28s of 10ah lipo for a lightweight pack that just over 1kwh and fairly punchy, I never use more than 600wh at the moment so a small lack with high volts is plenty, main thing is i need to keep my charging solution small and mobile.

Wire is predictable stuff. A wire the thickness of a hair or the thickness of a tree trunk can equally well carry a million volts...OK since the tree trunk wire has lots more surface area, it can carry it better...but neither wire will burn out with 1 milion volts potential on them. Now add a tiny bit of current to that hair thin wire and it melts in two, but the tree trunk wire is still good at 10,000 amps. Wire doesn't heat up and burn out from voltage, it's the amps. IE: Run higher voltage with less current while maintaining the same wattage and the motor wont heat up as much while still producing the same amount of torque. On that subject, why rewind the motor for a lower KV? Why not just change your gearing? I assume you want to maintain pretty much the same speed you are right running now, but want the motor to run more efficiently right? A slightly different gearing ratio will be much easier to do than a motor rewind. I would use it as is until I burned it out.

Of course if a rewind is something you really want to do, well by all means do it. And yeah the BOMA motors are not well wound. I haven't bothered to rewind my burnt out motor because of the hassle factor. It's not exactly the easiest motor to rewind and pressing out the stator from the can is a pain, but can be done. I guess you could rewind it in the can. It looks like 24 awg strands in the motor from the factory. What awg wire do you intend to use?

Regarding a 28S BMS...they do exist, but there is another option too. Run 2 BMS in series. High voltage BMS are much more expensive than their lower voltage siblings because you need high voltage mosfets. The support electronics for each channel are minuscule in cost compared to the mosfets. IE: Get a couple of 16S or 20S BMS that can handle the amperage you want and run them in series. Each BMS divides the total voltage in half so even though you might be running two 20S BMS in series, each BMS at 28S is only seeing 14S so no one gets unhappy and lets out it's magic smoke. This is probably how I'm going to run 28S on my moped (two 20S BMS in series), if I don't find a reasonably priced BMS that can do the amps and volts I want in a single unit.

Whatever you find in a BMS, please post it here.

 

[Ianhill]

I get what your saying with the tree trunk analogy and with DC that is always the case but on inverter driven ac motor skin effect comes into play and the current moves to the bark of the tree using less of its area so by having many small conductors you give yourself more bark surface area within the same size conductor so it runs cooler by using many fine strands, astro, alien and even the Chinese use the technique they can't all be wrong.
I guess i want to do a rewind so i can put all this info I've been stocking up on to test see if what i believe is going on in there actual is.
The volts are the top speed the amps are what gets you there so If u run my stock boma on 117 volts it will want to spin at 16000rpm and self destruct but let's imagine it can take the centrifugal forces etc, the more amps I add the more torque it will produce and the closer to its top speed it will get but because its top speed is so high it will never have the torque to get there so it will produce a lot of waste heat getting up to revs where its fairly efficient or more likely burn up before it gets there, plus it would need like 12-1 ratio massive rear sprocket.
If I rewind it to half the kV from 105 for the stock 1600watt to 52kv for the rewind it will spin to a more reasonable rpm range and stay in one piece with out the magic smoke and magnets flying off the rotor hopefully, I can add a bit more beef with the added copper mass around 4kw peaks at 28s spinning around 7krpm just a bit more torque so I could gear it to 50mph or so.
I won't go to fine on the strands with the boma the erpm won't be that high with only 9 stator teeth its running around 400hz at flat out so I'll brake down the existing winding size by half then double its length work out how many strand they used and make it slightly finer plus add a strand or two for luck as it will fit easy that way it will spin up to stock rpm and handle a tad more wattage through put with the extra strand or two ive added, so it runs 28s 35-40amps it will be light work for the battery and the cables to the motor will be overkill with 12Awg it will be a nice efficient build run on sinewave, if I pull it off.
I'll keep u updated with bms and the dc charger side of things for my new build i was going to do a mountain bike but with a 26inch wheel the motor sits to far from the wheel if placed where the crank once was, A bmx size wheel would be the limit for cutting the crank out and put the boma in place without the chain being massive, I can fit 219 sprockets to a trials bike hub that laces in a bmx wheel too but what ever I choose there will be a thread and I will keep u updated along the way, its always a budget build with me but the next one will be very different to whats come from me before.

 

[ElectricGod]

Ian,

Did you see this BMS from Bestech? 32S-96S at 200 amps.

http://www.bestechpower.com/27Sto32Spcmbmspcbforli-ionli-polymerbatterypack/BMS-D170V1.html

Or this from Chargery. 24S at 300 amps

http://www.chargery.com/BMS24.asp


BTW...this style BMS tend to be lower quality and don't balance well. They are much less expensive for a reason.
http://www.bestechpower.com/74v20spcmbmspcbforli-ionli-polymerbatterypack/PCB-D166.html

 

[Ianhill]

The bestech is a monster I like the fact of all the systems talking together with canbus then one display to calibrate it all that would be a car level of build 96s or maybe a superbike
The chagery looks sweet and quite flexible for smaller packs.
At the moment I'm tempted to set me sights a little lower and go the 20s route like yourself a 20s 80amp BMS and 84v 2amp charger can be had for peanuts it makes sense cost wise for the 20s route everything is a lot cheaper and my charging solution would be fairly compact but up to 5 hours to charge that's a step backwards by at least an hour for myself I would like to charge at 5amp with 10ah cells what ever cell count for a two hour full charge ideally.
It seems graphene prismatics are coming to the market in higher numbers 10ah, 20ah and 30ah nmc cells the future looks bright for a high cell count converted classic car beating stock performance by a long margin.
There is a UK car that runs the same subframes front and rear so it can easily be twin engined with a bit of trickery the makers MG made an electric hybrid version back in 2003 The MGTF 200hpd with a 28kw/40ps kick that give it 4wd and over 200bhp Output and clocked 0-60 under 6 seconds with what looks to be lead acid at 72v 400amps it would make a perfect platform for a diy electric many upgrade parts and go like the clappers when tuned but that was before the brand turned to a shameful Chinese existence unbacked by our government and a scandal called the Phoenix five with 30 million of missing money.

 

[ElectricGod]

The bestech is a monster I like the fact of all the systems talking together with canbus then one display to calibrate it all that would be a car level of build 96s or maybe a superbike
The chagery looks sweet and quite flexible for smaller packs.
At the moment I'm tempted to set me sights a little lower and go the 20s route like yourself a 20s 80amp BMS and 84v 2amp charger can be had for peanuts it makes sense cost wise for the 20s route everything is a lot cheaper and my charging solution would be fairly compact but up to 5 hours to charge that's a step backwards by at least an hour for myself I would like to charge at 5amp with 10ah cells what ever cell count for a two hour full charge ideally.
It seems graphene prismatics are coming to the market in higher numbers 10ah, 20ah and 30ah nmc cells the future looks bright for a high cell count converted classic car beating stock performance by a long margin.
There is a UK car that runs the same subframes front and rear so it can easily be twin engined with a bit of trickery the makers MG made an electric hybrid version back in 2003 The MGTF 200hpd with a 28kw/40ps kick that give it 4wd and over 200bhp Output and clocked 0-60 under 6 seconds with what looks to be lead acid at 72v 400amps it would make a perfect platform for a diy electric many upgrade parts and go like the clappers when tuned but that was before the brand turned to a shameful Chinese existence unbacked by our government and a scandal called the Phoenix five with 30 million of missing money.

There's an inexpensive BMS I have bought several of. I found them on ebay for $65. They are 20S and good for 50 amps. Very compact at approximately 3.5x3.5x3/8". They have status LEDs for each channel and they are not the dumb variety of BMS made from all discrete components. You can't communicate with the BMS, but I bet since the charge controllers are made by Microchips, that there is some form of onboard communications in those chips for battery status. I'll probably continue to buy more of them in the future and run two in series when I want to go over 20S. I'm using two of them in my blue scooter. The LIPO packs in the main battery bay are on one of them and the LION cells in the back are on the other one. I took one apart a while ago and the mosfets are n-channel T290 or BV4V1, but I can't find any datasheets for the part numbers. They are not surface mounted so it would be easy to replace them with irf4110 mosfets for more current. In the second picture, you can see that there are another 4 channels that are not populated with components so there is a way to add parts to the board for 24 channels like is common on a lot of BMS. I can't tell from my picture what the values are for the smaller parts, but that's a Microchips lithium controller chip. It probably would be fairly easy to upgrade this BMS to 24S. Those two black objects to the right of the mosfets are current shunts. I had an over voltage event happen some time ago due to regen and the BMS shut off to protect itself. Obviously the voltage sensing circuits work. I'm thinking maybe the shunts would need to be upgraded along with the mosfets to get this BMS to handle more current. Maybe go with active current sensors and then I could actually measure current.

 

[Ianhill]

So the BMS you use is 20s max and its possible to overlap the BMS so they both maintain some of the same cells for a portion of the pack ?.
I get you can series the bms but is this how you would achieve 28s with 2 20s bms overlapping them so both bms would see 20 cells on each BMS but the second BMS is overlapping its first 12 cells for 28s total and no magic pixies getting out.

 

[ElectricGod]

So the BMS you use is 20s max and its possible to overlap the BMS so they both maintain some of the same cells for a portion of the pack ?.
I get you can series the bms but is this how you would achieve 28s with 2 20s bms overlapping them so both bms would see 20 cells on each BMS but the second BMS is overlapping its first 12 cells for 28s total and no magic pixies getting out.

No...two different things. These BMS have unpopulated solder pads for an additional 4 cells per BMS. IE: that would make these into 24S each or 98 volts.

Series running two or more BMS is a bad idea. I was partly wrong. It sounded great in my head, but then I drew this thing out and realized I had made a HUGE mistake. It is possible to series string multiple BMS, but also a really bad idea. The reason why is Load- on the BMS and Batt+ are not the same thing...in fact they are exactly 180 degrees opposite of each other. In order to string two BMS together in series, that means the two 16S packs are also in series. The positive end of the first pack connects to the negative end of the second pack. To series wire two BMS means Batt- on the negative most BMS is the final Batt- for the whole 32S battery pack. It's Load- needs to connect to Batt- on the second 16S pack where the second BMS also connects its BATT-. So far so good...sort of. Lets not forget that the Batt+ on the negative most 16S pack needs to connect to Batt- on the second 16S pack to make a 32S pack. Well that means that Load- on the negative most BMS is also connected to it's own Batt+. This would create a dead short across the negative most pack via it's own BMS. So much for protecting the battery pack! I assume the BMS would shut off, but still...completely unworkable.

To anybody reading my nonsense about series connecting two BMS...it's a really bad idea...don't do this!!! LOL. My mistake was I was thinking of the BMS as both the positive and negative connection to the battery pack it protects and it's not. A BMS is only the negative connection of a battery pack and then the balance cable is all the intermediate cell connections. There's just no way to connect together two battery packs and 2 BMS without a really high power diode between the two 16S packs to prevent current flow from Batt- to Batt+ of the negative most 16S pack. You would need a diode that can handle 2X the current that the pack is capable of delivering just to be safe. It's doable, but at a minimum you are going to see a .7 volt drop across the giant silicon diode that is going to need a heat sink. Essentially...never mind, not really doable or practical...just buy a 32S BMS.

A BMS serves three roles.
1. Overload protection
2. Cell under voltage shut-off
3. balance charger

!. This is easy, current sensors attached to high voltage mosfets that turn off the mosfets when X amps is reached. I could build this in a couple of hours.
2. This is pretty easy too, but I would need an op-amp for each and every channel of the pack that compares cell voltage to a threshold voltage and then sends a shut-off signal to the mosfets when the threshold is crossed. One channel would be a couple hours of work, 20 channels would be 20X the effort.
3. You can buy balance charger boards. It used to be that RC chargers didn't balance charge, you needed a separate balance board connected to the balance connector on the pack to balance charge. Now days, this is built into RC chargers. So...with sufficient numbers of balance boards (number 3) and a signal off those boards for under voltage and a separate current sensor/mosfet switch (number 1), you could build a BMS any size you wanted as long as your mosfets could handle the total voltage.

 

[Ianhill]

http://www.bestechpower.com/27Sto32Spcmbmspcbforli-ionli-polymerbatterypack/PCM-D170.html

This one looks promising.

 

[ElectricGod]

http://www.bestechpower.com/27Sto32Spcmbmspcbforli-ionli-polymerbatterypack/PCM-D170.html

This one looks promising.

I was looking at that one too and meant to post its URL earlier, but forgot. It looks like the BMS to get of what I've found so far.

There's a Chinese company...AyaaTech or something like that. They make a variety of BMS solutions. I bought 2 of their more advanced BMS 8 months ago. At the time I wanted 20S. Anyway, they support, bluetooth, LCD, PC access, data logging, etc, but they were so badly implemented that it never worked reliably. Then I noticed that the BMS didn't see all 20 channels and would vary how many channels it saw from 12-16 based on whether a sparrow vs a robin flew over my house and then if the price of tea in China was appropriately valued, sometimes they would see all 20 cells. LOL! Anyway, at $200 each, I expect them to work pretty damned flawlessly. Oh yeah...and the PC software was uber buggy. Everything about them was half done and poorly implemented. Even the on board LCD would not work reliably. I eventually got a refund and they never sent me return shipping so I still have them. I wouldn't trust that BMS to manage to trip over it's own feet, never mind keep tabs on an expensive battery pack. I waited for 6 months for them to get things worked out. They kept promising a new PC software version that worked and BMS's that worked and LCDs that worked, but only ever delivered one thing. The way they displayed cell status on the LCD was really clunky and inefficient use of the screen space. They made that one change and sent me pictures of it, but I never got hardware that incorporated them. Eventually...and I thought 6 months was gracious plenty patient...I insisted on a refund. If you ever see this BMS anywhere and think "Oh man that is a sweet BMS! It does everything I could want.", run away and buy something else.

 

[ElectricGod]

This is uber cheap and will do 4S. They could be strung together in whatever length you wanted for 2billionS if you wanted. It's just unfortunate that it's the discrete variety which don't balance particularly well.

http://www.ebay.com/itm/4S-4-2v-li-ion-balancer-board-lithium-balncing-full-charge-battery-balance-board-/331949476978?hash=item4d49bb0c72:g:U5UAAOSwgZ1XvXSl

 

[ElectricGod]

Besttech has a lot of balance boards.

http://www.bestechpower.com/balanceboard/