Low cost BMS

[mora]

Principle is right. Charger will only see pack voltage (in most cases) and switches between full current/reduced current based on voltage information. It doesn't know if some individual cells have different voltage than others. When pack voltage reaches certain level charger will start to reduce current on its own. BMS might tell charger to ease a bit if certain cell voltage is reached or even tell charger to cut off completely.

If main goal is to have all the cells at same voltage in the end of the charge one should perform top balance separately for each cell. Ideally none of the shunts (load across the cell) shouldn't get activated during charge process and charger should finish the complete charge on its own. Only if some cells go out of balance and try to reach unwanted voltage levels should shunt resistors activate. And this should happen during recuded current charge. Pack requires manual balancing if BMS trips charger before it starts reducing current.

Shunt resistors will get hot when they are in use. If you want to remove more charge from a cell you need more powerful resistor. Then it gets even more hot and requires a big heatsink. It is not desirable to keep them on for a whole charge. It is all wasted energy there. The less you need to use them the better.

[Astro]

I started thinking about using the cell top BMS to load the higher voltage cells at the start of the charging process after i read post by harlequin2. Harlequin2 mentioned that some people had balanced their cells by leaving them in reduced charge current state for a few days. That was probably an extreme example but it made me think, the low current period of cell balancing is burning up kwh's whilst possibly waiting for just one cell to balance.

Originally Posted by harlequin2

...
I balanced all mine with an individual cell charger made from an old 5 V/ 20 A computer power supply modified to put out 3.6 V instead of 5 V. Might take a while if you have a few hundred cells!
Also, weber and coulomb on the AEVA balanced theirs with only 1/2 A just by leaving the charger in balancing mode for a few days - they have about 220 cells.

1/2 Amp * 220 cells * 72hrs = ~28.5kwh

I would imagine it would depend on how out of balance the pack is as to whether holding the charger at a reduced current until all the cells caught up would use more power than reducing the charge current only to the highest voltage cells for the whole charge process.

The total charge time would be reduced and i get the feeling that it would actually use less power to load the high cells during high current charging. Tthe calculations to work it out hurt my head.
But say your high current charge period was 8 hours then loading the highest voltage cells during that period would at most consume 8hrs * cell count * cell voltage * BMS load current.
For the example quoted from harlequin2 and assuming the high current charge period is 8hrs then if the cell balance was achieved during high current charging. Then it would be an absolute maximum of 3.16kwh rather than 28.5kwh burned up by the cell top BMS modules. As not all 220 cells would be loaded by the BMS, only the highest voltage ones, it would actually be considerably less than that.

This is all purely guess work so it may be very different in practice.

Another thought i had was that as more and more people take up electric cars it is likely that the cars will be treated more and more like a regular appliances. The users will be able to drive the car but have no idea of the mechanicals. Eventually it wont just be enthusiasts using them.
At that time having a reduced current balancing period at the end of the charge cycle may cause some very short lived battery packs as the users just plug them in and then when they are almost fully charged they unplug and drive. Because what difference does 98% full to 100% full make? So cell balancing may never actually occur and the pack drift more and more out of balance. If the balancing was done during the high current period of charge then no matter how short the charging period was cut, the pack would end up more in balance than it was before the charge.

I am hoping to build some of harlequin2's cell top bms modules and then play around with the software and see how it works in practice.

[P-hack]

BMS AE-LMD17 REV A1

Seen a 24 cell capacitive balancer for less than $200. 60 amp discharge, Capacitive is a lot more efficient than resistive, it basically uses a capacitor between adjacent cells and mosfets to shuttle charge from higher potential to lower potential:
and it should be able to operate while the vehicle is driven or charging (or just sitting, but that might lead to excessive discharge over time, lotsa leaky caps). IIRC it is like %50 efficient, depending on the distribution of imbalance of course. If you could alternate weaker cells and stronger cells it would work very well
but the cells on the end still don't get refreshed at the same rate.

There should be an inductive version, that takes pack voltage to bring up the low cells too.

[Astro]

@P-hack That is an interesting circuit. It would do the job nicely and if another capacitor and switch set was added then the last cell could swap charge with the first cell so that all cells received the same refresh rate. As i am looking at it i see the number of wires required and can't help but wonder if some sort of combined power transfer/communications could be achieved so that cell top BMS modules could communicate to the BMS master module along those wires and the power transfer capacitors could be integrated into the BMS master controller. Actually seeing as the BMS master controller has a micro controller maybe there is a way for the charge to be moved in a more controlled manner. The BMS master module just having one capacitor that it switched across the cell with the highest state of charge and then delivered that charge to the cell with the lowest state of charge. The BMS master could keep track of each cells performance and warn if a cell was going bad (high self discharge). This would avoid the scenario of one bad cell slowly taking charge from all the other cells and discharging the pack.
Just need a couple of high current switching arrays (does such a thing exist?). It would also mean that the charge swapping would be stopped once there was no more imbalance.

[P-hack]

I did some more investigating and that link above actually talks about inductive balancing. The parts for the power section look pretty cheap ($2.50/cell?) but driving those "floating" mosfets could get interesting. But you could use a 328 for starters (also floating?) with 6 cells per cpu, plus i2c communication between 328s or something. I don't really know what I'm doing just trying to figure it out. They discuss the losses in capacitive balancing in that page somewhere. You can see from the schematic below that one can "chain" power sections for any sized pack in an inductive setup too.

schematic from: http://www.ti.com/lit/ds/symlink/bq78pl114.pdf (discontinued).
power section (8 amps?):
2x 3300pf $0.60
2x 22uf $0.10
2x 20k $0.10
1x 2k $0.10
1x FDC6327C $0.33
2x MA21D3800L $0.17
1x 4.7uh $0.23
Total ~$2.53 per cell for the balance power section.

[Astro]

@arber333
Not sure if anyone else is having the same issue but renaming that pdf to a rar doesn't want to work.

Originally Posted by arber333

...
File is named .pdf due to server restriction with .rar files.
Download the file, rename it to .rar and browse. You should get idea of purpose from the file names .
There are gerber files for pcb manufacture as well as .pcb files for designing your own board dimension...

[Astro]

Originally Posted by arber333

What seems to be the problem? I tried the link and it downloaded .pdf fine. Then i right click to the file and select rename. Then just replace *.pdf with *.rar. System asks you if you are sure....yap! Then browse trough newly formed .rar archive.
It worked for lots of people untill now.
A

Arber333, finally got the .rar to decompress. Had to use a WinXP virtual machine running RarZilla. The .rar file wouldn't play nice with any of the Linux based archivers i tried.
Had a look through the files and there is a lot of stuff in there.
Most of this stuff is beyond what i want to have to learn. I am more a software person. I was hoping that harlequin2 would have revisited the forums. I had hoped he would have continued developing his BMS into a product/kit. I am sure i would have had no trouble soldering up the board and programming the software but going from a schematic or a pcb file to the collection of files that the PCB manufacturers want is daunting. I could just see myself ordering $100 worth of pcbs to find them all messed up when they arrived.

I think i will just wait and see if harlequin2 returns and is willing to order some extra celltop/master pcbs and sell them to forum members. Or even better sell a kit of pcbs and components.

I just want to get a working hardware base that i can then get stuck into developing software for. harlequin2's setup was nice and simple. All the smart bits were in the software. This is why i think it would make an excellent starting point for me.

A thought i had as i was looking at your pcb was the idea of an infra-red link between modules. I think you (maybe someone else) mentioned it earlier in the thread. Setting up an infra-red receiver and transmitter on each module. If you placed the transmitter and the receiver side by side and left enough length on the devices legs so they could be bent over to face one side of the pcb or the other depending on the modules orientation. Then i think you would get good alignment with only few cm or so separating transmitter and receiver. You could even put a sleeve (tube) from transmitter to receiver to eliminate all crosstalk and outside interference. If you used a tube between modules you could even use visible light rather than infra-red which would make trouble shooting easier as you would be able to see the flashing led of the transmitter. The tube wouldn't have to be anything special, i am imagining something like a drinking straw but of a material that would cope with the heat it may experience. Get a few meters of tube and cut the lengths required to join the modules. Automotive vacuum line would probably do and is very cheap.

[4Springs]

Hello Everyone,
I have an electric car and have built the Low Cost BMS. Many thanks to harlequin2 for creating this and sharing it for everyone!

I've created a SourceForge page here: sourceforge.net/projects/low-cost-bms/

The page is intended to be a place where anyone can download all you need to get started with this project. I have uploaded all of my files and I'd encourage anyone else involved with this project to do the same - either join up and become a contributor or PM me and I can upload the files for you. I'm hoping that someone browsing the page will be able to select from several different versions, depending on their cell size, compiler etc.

[mora]

Hello 4Springs.

Did you include your version of master board with TCcharger control circuit in the package? I don't have Designspark, at least yet, so I couldn't open any of those files to take a look.

This got me interested. I could try and build it with for my 12V battery (4x 90Ah LFP) which currently doesn't have any BMS. Has been so for years already. And same for other 12V pack which I use as trolling motor battery.

[4Springs]

Originally Posted by mora

Did you include your version of master board with TCcharger control circuit in the package?

No, I didn't put the TC Charger control circuit on the Master board. The reason was just that I made the Master board before I knew what charger I was going to use. I ended up making a small extra board for the charger PWM control. The circuit I used was straight out of the TC Charger user manual.

Originally Posted by mora

This got me interested. I could try and build it with for my 12V battery (4x 90Ah LFP) which currently doesn't have any BMS. Has been so for years already. And same for other 12V pack which I use as trolling motor battery.

Yes, this BMS is cheap enough to use even on small batteries!
Having said that, it sounds like BMSs are less important on smaller batteries. The more cells you have in series the more likely they are to get out of balance. Can you measure the voltage on individual cells in your batteries? If so you can get an idea of how unbalanced they are now...