PINNED - Lithium battery conversion/info

[shirley_hkg]

It is 4S2P measuring 185 * 185 * 430

Cell price £28.00

Size don't fit .

[Burgerman]

http://www.wheelchairdriver.com/these.pdf are what I really want x8

???? www.wheelchairdriver.com/these.pdf

[terry2]

Yup, the PL8v2 will always be recommended for charging Lithium batteries. It will take years for a plug'n'play solution to be developed.

Steve

I'm hoping someone invents a programmable BMS system using a Pi and maybe a ramps board.
My 3D printer with 5 motors\two hotends and extruders, a heated bed and many micro switches with 1 ramps board.
So I think it can be done.

[Burgerman]

Its already been done. But its the WRONG solution. Its fundamentally flawed, the wrong thing to do and at the wrong place. Unless your only interest is in being able to sell lithium batteries (cells) to a bunch of clueless DIYers in a way that means you don't get almost all returned as "faulty" through ignorance...

Its wrong in the same way as wearing ear plugs, instead of oil to your cars engine.

[terry2]

Its already been done. But its the WRONG solution. Its fundamentally the wrong thing to do and at the wrong place. Unless your only interest is in being able to sell lithium batteries (cells) to a bunch of clueless DIYers in a way that means you don't get almost all returned as "faulty" through ignorance...

https://teutanic.com/electronics/compon ... n-arduino/

[Burgerman]

Thats not a BMS as such, or rather it is but all it does is control charge and cell balance (badly) on a solar array..

[terry2]

Thats not a BMS as such, or rather it is but all it does is control charge and cell balance (badly) on a solar array..

(badly) so you used one then? They are using that to charge\balance LF4Pro battery packs. I'm doing some searching now.

The code looks easy. One could use Marlin to run which is again easy to program.

[Burgerman]

Dont have time so this is far too simple. But...

BMS has 2 jobs. We will split them up.

Cell use protection.
Cell charge protection. (More later... But to be effective it must be able to control charger output too)

Cell use protection.
It cuts power like a switch (disconnects) if you do any of these:
Over an arbitrary current limit.
Low voltage limit.
Over voltage on say regeneration.

The value of these things is debatable, We don't want any of them! Since it doesn't control them, but switches power OFF! It depends on cell type, actual load, etc. (for eg it may cut out as you try to climb a curb or turn on grass due to over current, or because the extra current causes a temporary voltage drop. So they are always wrong. If ANY of these things trip, then the cells are undersized for the system, and or the CONTROL SYSTEM on your chair is not correctly programmed. As its this that should be monitoring voltages and current and PROPORTIONALLY reducing current or load as needed proportionally, or telling the user the battery is low. You don't want to stop dead crossing a rail line or getting on a bus!

Likewise for the charging. That requires a better proportional control system, proportional balance, and a much higher balance current than any generic BMS gives. Not a digital on/off charger/balancer. In a properly implemented system it would not ever do anything. If it did, you already have other problems.

[terry2]

Dont have time so this is far too simple. But...

BMS has 2 jobs. We will split them up.

Cell use protection.
Cell charge protection. (More later... But to be effective it must be able to control charger output too)

Cell use protection.
It cuts power like a switch (disconnects) if you do any of these:
Over an arbitrary current limit.
Low voltage limit.
Over voltage on say regeneration.

The value of these things is debatable, We don't want any of them! Since it doesn't control them, but switches power OFF! It depends on cell type, actual load, etc. (for eg it may cut out as you try to climb a curb or turn on grass due to over current, or because the extra current causes a temporary voltage drop. So they are always wrong. If ANY of these things trip, then the cells are undersized for the system, and or the CONTROL SYSTEM on your chair is not correctly programmed. As its this that should be monitoring voltages and current and PROPORTIONALLY reducing current or load as needed proportionally, or telling the user the battery is low. You don't want to stop dead crossing a rail line or getting on a bus!

Likewise for the charging. That requires a better proportional control system, proportional balance, and a much higher balance current than any generic BMS gives. Not a digital on/off charger/balancer. In a properly implemented system it would not ever do anything. If it did, you already have other problems.

A £20 Raspberry Pi can do 1400 MFLOPS. With the right programing it could do all that with ease plus more.
A 3D printer is 10 times more complicated then a powerchair. But a Ramps 1.4 does the job great.

[Burgerman]

You dont understand. If you put a full hi spec PC in there with 10000k worth of software development, it doesn't help! Its doing the wrong job in the wrong place. It doesn't matter if its a supercomputer! It is fundamentally INCORRECT to do this. And redundant in a properly configured system.

Would you fit a cars BRAKES or REV LIMITER, on to the fuel tank? Because thats exactly what A BMS does! Its job is to protect the BATTERY from MISUSE not to make the vehicle work properly. Its job is to protect the battery supplier from refunding money from ruined cells fitted to a system that it wasn't suitable for.

The correct thing to control the battery packs safety already exists. The controller that drives the chair does *everything* we need for that job correctly. Or at least it can with the correctly choosen cell capacity and C rate. And programming. And no BMS can fix this if you DON'T have this correct. All it can do is cut the power before it gets to the controller!

As for charging only theres hundreds of different ways. And thats called a charging system. Not a BMS. And it can be embedded, on board, or external as we do for a bunch of reasons.

Lithium is NOT any old universal lithium battery. Stuck into the wrong place usually with inadequate C rate or capacity, with a non current limiting controller as used on cheap chinese ebikes etc. With a BMS added on to stop the bad choices from killing it by cutting the power to save the battery. A lithium system is a combination of CORRECT choices, in a seamless system that covers all the requirements. Once these are met, the BMS if fitted will never operate or do anything ever, since its not then required at all, and its "protection duties" have already been taken care of in a better correct way. For eg, what do you suppose the chairs controller will do if a super complex expensive but unneeded BMS reduced power or cut it off on a ramp? More importantly why do you think its required? The chairs controller already has proportional and programmable current limiting, and low voltage limiting built in!

[steves1977uk]

https://endless-sphere.com/forums/viewt ... 14&t=63863 Not as simple as you think terry...

Steve

[Burgerman]

And absolutely not required! A pointless widget that wont do anything unless you have the wrong battery in the wrong place! And then what it does is stop you getting off the train tracks! Or into your van.

[terry2]

https://endless-sphere.com/forums/viewtopic.php?f=14&t=63863 Not as simple as you think terry...

Steve

They said that about 3D printers years ago. Yet here we are. A very complicated machine made easy. Well I built one from the ground up
I believe a small computer can take charge of charging and monitoring the system. And using Arduino is great as it's very easy to program.

In other words it will use the same kind of chargers\balancers but the computer will be in charge. The Arduino forums have some good posts.
A monitor has already been dun for lithium.

[Burgerman]

I believe a small computer can take charge of charging and monitoring the system. And using Arduino is great as it's very easy to program.

But not required. Theres many ways to do a charge system. Thats easy.
Its all the rest of the stuff a BMS does that we don't want or need.

In other words it will use the same kind of chargers\balancers but the computer will be in charge. The Arduino forums have some good posts.
A monitor has already been dun for lithium.

The PL8 is exactly that ALREADY. With thousands of users, and massive flexibility. It is the computer in charge, and can do a better job than 100s of other also ran chargers as it is. And its very configurable. You are just reinventing the wheel. Its electronics could be copied or used on board. As they are in for e.g. many full sized cars/bikes, or inside professional solar systems. But you don't want that.

Unless you like extra on board fragility (on board chargers are notoriously unreliable for a bunch of reasons) its better from a safety and a ease of use/connectivity, and for saving of space and weight on a powerchair to use an off board charger. Which is the way powerchairs have already gone for those reasons. On board chargers are less and less popular. And lo and behold thats what the PL8 already is.

Tell me again why we want a BMS?
What exactly will it do for us?

A monitor on a cell by cell basis, for low voltage end of charge is the only part of the equation that the stock controller you already have on a chair doesn't do. A manufacturer can add that easily, and add a few software settings to reduce power or tell the user the battery needs charging (battery meter) very easily. Or we can just watch a monitor for 2 dollars!

[LROBBINS]

The other things we lack are:

(1) a reasonably compact, single box, containing both a power supply and the PL8 (or equivalent). Having two boxes on the kitchen counter with mains cable, power supply to charger cable, and charger to chair cable would definitely not be appreciated by others in my household, and if the PS to PL8 wires could be detached (or worse yet shorted) that would certainly happen.

(2) once programmed, a "no switch" and at most a "1 button press" way to start the charge. Even with a simple lead-battery charger that has an on-off switch, I manage to once in a while forget to turn it on, and Rachi's personal aide never remembers to turn it on.

With large enough high-C batteries, I really don't think we'd have any need of a monitor, and if I wanted cell-by-cell monitoring I would indeed use an Arduino to do that so I could feed the result either into 1 pin of the Roboteq, or, better yet, into the CAN bus system so that the low cell warning would be shown on the display as well as trigger an inhibit (probably at a limp-home speed well before reaching a level that's a danger to the cell).

[steves1977uk]

Lenny, the PL8 can be unplugged from a PSU. It can remember the last preset used so all that would be needed is a way to press the start button when the PSU is connected to start charging if it was mounted as an on-board charger.

Steve

[Burgerman]

There are many that have built a PL8 or other hobby charger into a Pelicase. Complete with power supply etc. So it really is one box. Press go, wait for a cell count check, press go again. And then go to bed.



Heres a few dual ones https://www.google.co.uk/search?q=pl8v2 ... KY&imgrc=_

The problem is that while a typical lead charger is just 225 watts, the fact that a lithium battery can store 2 to 3x the energy, and can be charged faster, means that we need a much more powerful setup. So an all in one unit is too big for on board. We can charge at 5x the rate, 40A instead of just 8 for eg. So your typical lead charger would need to be physically 5x the size and weight. So its not easy to put that sized system on board, or make it portable.

The PL8 is 1344 watts. Not 225w like your lead charger. So something like the above, with just 1 charger, and a power supply in the same box would need to be in your kitchen. This one has 2 chargers, and could do my 45V 13S battery for EG. Its 2688 charging watts! 10x that of a typical lead charger. And yes 1 button start.

[Gnomatic]

After my mishap, I completely tore down the pack, and removed the two cells that were compromised by scraping against bolts when I was doing a test fit a couple weeks ago. I also took out a third cell that seemed it may have taken a bit of abuse as well. Huge thanks to expresso for sending me a few replacement cells and orange blocks so I could bet the rebuild going much sooner than otherwise.

I individually tested a bunch of cells for capacity ect. Everything seemed fine. Reassembling the pack went much, much easier and smoother than when my friend and I initially put it together. This is all new territory for me, and there's no teacher like experience. Once assembled, I charged the pack, did a complete discharge to 2.9v, and then fully recharged it.

A little bit of bouncing around at the top of the first graph, maybe a little less pronounced than before. My charge cables are long so I'm chalking it up to that.

Ahrs returned seems fine I think.

So all this put me back two weeks, but all in all, I think the pack is better assembled than before. It really helps having some experience putting one of these together. I get my chair base back from the shop I took it to next week. All the bolts sticking above or inside the battery area have been ground off and welded in place so the inside of the battery area is snooth. I'll be lining the entire inside of the battery area with drawer liner, and also am wrapping the pack in rubber sheeting. At that point, I think I'll be ready to install the pack in the chair! (hopefully no sparks this time )

[expresso]

this is all true - BUT - trust me - if the user is not involved most of the times and you want to just leave it to YOUR AID to do it - even a simple on and off dosnt get done - you know how many times - i watch my worker plug the lead chair in and then never flip the switch to turn it on - !!!!

its plugged in a power strip with on off switches - all they have to do is plug in the XLR cable in the joystick and flip a switch - and yet they miss that last step many times if i am dont watch to make sure its charging - and i know many others with home aids - that are so bad - its comic - - not everyone can get it done -

i control my setup - my girl just plugs the Sub D together - and even that part - i have to put a Orange sticker on both ends so they dont try to connect it wrong and break a pin - have to stress not to bend while pushing it together - etc, or wiggle it to remove it - i watch those steps - to be sure -

would be a real pain if i have to redo a cable - what i do now is use a pen stick it between the two to wiggle it off disconnect them very very easy this way - no force or stress - Connection them also start slow line them up and ease them together with some force - no wiggle or bending force -

i was worried also with this PL 8 - PSU - cables etc, - but it worked out fine - again its because i have to be involved and do it - NO way you can leave a home Aide to do it themselves - will end up bad at some point -

this is why a good plug it in and flip a switch charger is a great thing for ALOT of users -

[expresso]

After my mishap, I completely tore down the pack, and removed the two cells that were compromised by scraping against bolts when I was doing a test fit a couple weeks ago. I also took out a third cell that seemed it may have taken a bit of abuse as well. Huge thanks to expresso for sending me a few replacement cells and orange blocks so I could bet the rebuild going much sooner than otherwise.

I individually tested a bunch of cells for capacity ect. Everything seemed fine. Reassembling the pack went much, much easier and smoother than when my friend and I initially put it together. This is all new territory for me, and there's no teacher like experience. Once assembled, I charged the pack, did a complete discharge to 2.9v, and then fully recharged it.

A little bit of bouncing around at the top of the first graph, maybe a little less pronounced than before. My charge cables are long so I'm chalking it up to that.

Full Recharge Volts.JPG

Ahrs returned seems fine I think.

Full Recharge Ahr.JPG

So all this put me back two weeks, but all in all, I think the pack is better assembled than before. It really helps having some experience putting one of these together. I get my chair base back from the shop I took it to next week. All the bolts sticking above or inside the battery area have been ground off and welded in place so the inside of the battery area is snooth. I'll be lining the entire inside of the battery area with drawer liner, and also am wrapping the pack in rubber sheeting. At that point, I think I'll be ready to install the pack in the chair! (hopefully no sparks this time )

NIce - You Saved the day - thats great - got back in shape in 2 weeks time - i bet the graphs will get even better with time and using it daily etc, -

[Burgerman]

Well theres hundreds of ways to do stuff. But as long as you are using a powerchair controller, and have access to low level programming tools you can actually set ut the battery gauge to tell you when to stop. And the chair to actually stop at a safe voltage. At least with the Dynamic software. Which is extremely useful. It means that the entire job of the dumb BMS can be left to the chairs controller with no risk of fire, error, increased impedance, cutting out on hill/curb/ramp/while crossing a junction etc. Leaving just the charging. And theres many ways to accomplish that.

As for the posts above, this shows why you need to be able to see graphs, monitor cell voltages while charging, change termination points, balance start points, etc on the fly to keep in control and not leave it to an on board simplistic system.

[Gnomatic]

NIce - You Saved the day - thats great - got back in shape in 2 weeks time - i bet the graphs will get even better with time and using it daily etc, -

It would have been much longer than two weeks if you hadn't sent me some spare cells and and orange blocks! Thanks again.

You know, one of the things this episode illuminated for me, is that starting with an ADD ON pack like you did is a very good option if you don't have much experience with this stuff. Once you have that, you already have all the tools and experience etc to do a full conversion later down the line. As it stands for me, my conversion is behind schedule from where I though I'd be when I started, primarily because of my own inexperience. But after having to tear it all down and start all over, I feel I have enough experience to carry forward now!

[expresso]

yes i agree with you Genomatic - doing my ADD ON first - gave me the little bit of experience - i redid the first ADD ON - 4 times for one reason or another - and you did great jumping right into a full pack - considering - you only lost 2 weeks
not bad - since its not even warm daily yet - but soon it will be and you be done by then

[LROBBINS]

I definitely would not want an on-board charger. Even if large, however, I would want the PSU and Charger to be permanently hooked together and housed together. I really don't care if it's not compact, though under 6 kg would be better for travel. I agree that getting as much information as possible while setting things up is an extremely good idea, but once it has all been programmed, I would want something that at most needs 1 button push and no switches. The ideal would be a setup that needs neither switches nor button presses, could be left permanently attached to the mains, and would start the charge safely once plugged into the chair. I forget to push the single toggle switch we have now once in a while, and Rachi's aide forgets near 100% of the time -- if she moves the chair to the charger and plugs it in in the first place.

There are major other things, however, that she does very well indeed. Yesterday, a national holiday here, for example, she and Rachi spent the day at the sea at a beach at least 2 hours from here. Left at 10:30 AM, returned at 11:30 PM and Rachi had a blast. Had she been with me on an excursion like that we would have both been bored out of our minds.

[Gnomatic]

I'm glad I have the PL8. It's so flexible. And if you miss torquing a bolt on your pack, or a cell dies an untimely death, it shows you. You can see in real time if something needs addressed.

Still, a BMS + dumb charger would be useful. It would be a lot more convenient, especially when traveling etc. One end of the charger plugs into the wall, and on charge cable end, have the BMS built into the charge cable, which has a Dsub on the end.

That way you can still use the PL8 and PC to charge most the time like say at home. But if traveling with an aide for a couple weeks, they can just plug the chair in at night like before. One end onto the wall, the other into the chair. No buttons to press or anything else. Very similar to using a mobility charger.

Then when you get back home, go back to using the PL8.

A BMS can't be THAT bad for the pack if used sparingly like the above scenario, can it? I mean, pretty much every EV or balance board etc. is sold w/ one. I know its not ideal. But it would be real nice to have that convenience from time to time, especially in certain situations

[Burgerman]

Most of what a BMS "protection" does isn't wanted. If you only need it for charging, thats a small part of its function. Various charge controllers are available for that.

And you could do that. Is it bad? Well that all depends! On a perfectly built and regularly checked (by PL8? ) pack it wouldn't do that much harm. But if for eg a cell was far out of balance it would rapidly get worse or hold every cell at a high CV voltage for 8, 16, or days till its weedy balance circuit fixed it. During this time it switches charge power on/off repeatedly bouncing the high cells up and down over and over... And of course you wouldn't know anything was wrong.

But remember that your dumb charger and BMS will be no smaller or simpler than a PL8 unless its also only the same power as your typical mobility charger. And that wont do! So I dont see any real gain.

[Gnomatic]

If you only need it for charging, thats a small part of its function. Various charge controllers are available for that.

And you could do that. Is it bad? Well that all depends! On a perfectly built and regularly checked (by PL8? ) pack it wouldn't do that much harm. But if for eg a cell was far out of balance it would rapidly get worse or hold every cell at a high CV voltage for 8, 16, or days till its weedy balance circuit fixed it. During this time it switches charge power on/off repeatedly bouncing the high cells up and down over and over... And of course you wouldn't know anything was wrong.

But remember that your dumb charger and BMS will be no smaller or simpler than a PL8 unless its also only the same power as your typical mobility charger. And that wont do! So I dont see any real gain.

The real gain I see is convenience, which is important in certain circumstances.

Yes, charging only. No other use for BMS for me. I had not heard of charge controllers before, so that's something new to me. I'll see if I can find anything.

I know there are "smart" BMS's out there that allow you to set the CV voltage etc and monitor with a phone app.

[Burgerman]

Since you will need something the same sort of power, which means a 40A charger, I doubt its going to be any smaller or lighter though. And both a PL8 or a dumb charger need to be connected (one cable both cases) regardless. So what part is more convenient? Remember it can be configured for 1 button start.


If you are happy with a low power charger, like the 8A type you are used to, then an 8S all in one plug into the wall hobby charger could be used. But 8A on a 135Ah pack like espressos, will take 16.8 hours plus an hour at CV/Balance.
Or 180Ah in the case of Nandol, over 24 hours. Or even my 120Ah pack in my BM2.5L chair, 15 hours plus another to balance and finish.

Remember we have to put back a huge amount fairly fast. These are not lead.

So I am not sure how a dumb charger IS more convenient? The only thing you are doing is moving the cell balancing from the charger where it can be properly controlled, to the chair where it cannot throttle the charger. And that part is very small.

[Gnomatic]

If you are happy with a low power charger, like the 8A type you are used to, then an 8S all in one plug into the wall hobby charger could be used.

Have a link to any?

I an 8A charger is far less than ideal. But it could still return 64Ah to the pack in 8hrs. My pack is 105Ah. So even a puny 8A charger could return 60% of my pack's capacity on an 8hr overnight.. I wouldn't mind having something like that in the toolkit.

[Burgerman]

I had no specific one in mind. But you need 8S and more importantly correct voltages, decent balance current, accuracy, and a flexible termination point that you can choose. And no capacity or time limitations, etc. Quite how you find all that out without buying, testing, measuring exactly what it does beats me. It all depends on its firmware programmer etc.