Why hasn’t cell based charging systems for li with a integrated bms exist

[Goboatingnow]

I was looking at my system design for my boats Li solution.



Why has the industry not considered isolated dc dc chargers to do cell level charging. On a 3P4s system this is just four dc dc isolated chargers each handling a series string , the cell monitoring could be added as well to the electronics.



This removes any balancing circuits and allows full current balancing in effect as each parallel bank is simply charged to the cutoff point.



It also means ALL dc sources then go through this charging system , so the “ BMS” by default controls all sources and the settings in the source make little difference , ie you just need you mppt , battery charger , alternator to be in a constant voltage power supply mode.



How come this approach doesn’t seem to be offered.

 

[John Frum]

How come this approach doesn’t seem to be offered.

Is this effectively a dc2dc charger that all the other charge sources will be connected to?
Pushing significant current at 3.2 volts nominal requires big wires.

Example for a 16s1p LFP battery made with the big blue 280AH prismatics.
The optimal charge current starts at .2c which is 56 amps.

You would need 16 chargers in parallel that could deliver 56 amps or better per charger.

So 56 amps, 10 feet round trip, 2.5 volts = ~2.7% voltage drop with 1/0 awg pure copper wire.
You would need 16 pairs of these wires.

 

[Goboatingnow]

Let’s say I want a 12v 300Ah bank

So I build a 3P4S coefiguraton with 100Ah cells. That’s 4 dc dc chargers.

Now let’s say I have a max of 50A from my charge sources. ( not unreasonable ) that’s spread amongst 4 serial strings of 3 parallel cells , that’s 16-17A per parallel set.

Firstly that’s a cheap DC DC unit

Secondly 17A doesn’t need big wires especially as the dc dc is beside the bank( the voltage is irelevant ) AWG 14 or 12 would be more then adaquate.

In fact I was thinking of 10A DC DC. as I’ve 200W of solar a 30 anp mains charger and an alternator giving out about 30A running cool.

 

[John Frum]

Let’s say I want a 12v 300Ah bank

So I build a 3P4S coefiguraton with 100Ah cells. That’s 4 dc dc chargers.

Now let’s say I have a max of 50A from my charge sources. ( not unreasonable ) that’s spread amongst 4 serial strings of 3 parallel cells , that’s 16-17A per parallel set.

Firstly that’s a cheap DC DC unit

Secondly 17A doesn’t need big wires especially as the dc dc is beside the bank( the voltage is irelevant ) AWG 14 or 12 would be more then adaquate.

In fact I was thinking of 10A DC DC. as I’ve 200W of solar a 30 anp mains charger and an alternator giving out about 30A running cool.

17 amps over 6 feet at 3.0 volts = ~2.2% voltage drop at 3 volts requires 8 awg.

Seems like a very narrow usage model.

I think what you want could be described as a scaled up hobby charger.
IMO it doesn't scale well.

 

[400bird]

Let’s say I want a 12v 300Ah bank

So I build a 3P4S coefiguraton with 100Ah cells. That’s 4 dc dc chargers.

Now let’s say I have a max of 50A from my charge sources. ( not unreasonable ) that’s spread amongst 4 serial strings of 3 parallel cells , that’s 16-17A per parallel set.

Firstly that’s a cheap DC DC unit

So, you'd want to charge each of your 12 cells individually? This sounds like quite the contraption you've envisioned. I'm not sure if you're picturing 4 or 12 DC DC chargers. But, you'd need a master/controller of some sort to monitor and adjust each DCDC live as it's charging. This sounds significantly more expensive than charging the battery as a whole, in series.

What is the issue you're having or trying to prevent?

Every charger, ever made, charges through the series cells because everywhere in a series circuit the current will be the same.

 

[Goboatingnow]

17 amps over 6 feet at 3.0 volts = ~2.2% voltage drop at 3 volts requires 8 awg.

Seems like a very narrow usage model.

I think what you want could be described as a scaled up hobby charger.
IMO it doesn't scale well.

Dc dc chargers can be paralleled

It also removes the need for start mains chargers , external alternator regulators etc.

No need to worry about 2% or 5% voltage drop, easy to adjust the dc dc to have battery voltage sensing

Anyway for 12AWG 6 feet total run 12v

Voltage drop: 0.21
Voltage drop percentage: 1.71%
Voltage at the end: 11.79

Thd resulting 0.21V is easily compensated for by voltage sensing.

 

[Goboatingnow]

So, you'd want to charge each of your 12 cells individually? This sounds like quite the contraption you've envisioned. I'm not sure if you're picturing 4 or 12 DC DC chargers. But, you'd need a master/controller of some sort to monitor and adjust each DCDC live as it's charging. This sounds significantly more expensive than charging the battery as a whole, in series.

What is the issue you're having or trying to prevent?

Every charger, ever made, charges through the series cells because everywhere in a series circuit the current will be the same.

No all parallel cells are charged from one dc dc charger , so maximum 4 chargers for a 12 v system.

There’s no need for any master controller. Each dc dc unit is a simple CCCV unit , charging to a fixed set point voltage with the addition of tail current monitoring to ensure accurate charging. ( ie each unit senses complete battery discharge ( load ) current )

What I’m doing here is

No balancing circuit needed active or passive. Each set of parallel cells is charged to the correct cell voltage and then charging stops. Hence the series string is automatically balanced to the same voltage

No smart mains chargers are needed , cheap AC to DC smps power supplies are all that’s needed , similarly no external alternator regulator is needed stock alternator is fine. Again mppt controllers simply need to power supplies not 3 stage chargers.

Low current dc dc <= 10 amps are as cheap as chips ( $4-10 ) , all that’s needed is a monitor circuit added to effect charge current sensing and temperature sensing. The monitor board can start or stop the dc dc module and does cell temp , lvc , HVC monitoring.

Conceptually much simpler circuit the sone of the complex stuff presented here.

 

[Goboatingnow]

here's a conceptual block diagram showing two chargers of the total of 4 for a 12V system

advantages

Cost , BMS/cell monitoring functionally included
No balancing circuits needed, cells automatically balance as part of charging process
simple charge source controllers needed, charge regulation control is purely CV , no multi stage or cutoff logic needed
networking features could be easily added , CAN etc
possible to do good SOH calculations as exact charge records can be maintained
Dc DC can be paralleled for higher charge current configs

 

[John Frum]

Dc dc chargers can be paralleled

It also removes the need for start mains chargers , external alternator regulators etc.

No need to worry about 2% or 5% voltage drop, easy to adjust the dc dc to have battery voltage sensing

Anyway for 12AWG 6 feet total run 12v

Voltage drop: 0.21
Voltage drop percentage: 1.71%
Voltage at the end: 11.79

Thd resulting 0.21V is easily compensated for by voltage sensing.

But its not 12 volts.
The charge voltage will be between 2.5 and 3.65 volts dc for LFP cells.

 

[Ampster]

Unless the question in the title to the thread is costed out we won't know the economics? Presumable some one some where in the world has raised the same question and decided either the boating market is too small or other alternatives are more cost effective.
I have a 12 volt pack for my tiller and a 24 volt pack for my chipper shredder and once the cells got balanced I don't need the expense or complexity or cell level balancing. For RC helicopter hobbyists it make sense because the packs are 2S to 4S and very small amperage because those packs are small and light for helicopters.

 

[Goboatingnow]

But its not 12 volts.
The charge voltage will be between 2.5 and 3.65 volts dc for LFP cells.

sorry yes you are right , moving to AWG10 , , which remains a manageable size, results in 5% which can easily be compensated by terminal voltage sensing

 

[Goboatingnow]

Unless the question in the title to the thread is costed out we won't know the economics? Presumable some one some where in the world has raised the same question and decided either the boating market is too small or other alternatives are more cost effective.
I have a 12 volt pack for my tiller and a 24 volt pack for my chipper shredder and once the cells got balanced I don't need the expense or complexity or cell level balancing. For RC helicopter hobbyists it make sense because the packs are 2S to 4S and very small amperage because those packs are small and light for helicopters.

Dc DC convertors are cheap and ICs for them are widely available , look at the low end Vicron isolated units, The additional cell / DC DC monitor circuitry is adding little expense ( < $10 in components ) as there is no heavy current handling needed

certainly i would see a BOM costing under $25 in volume, resulting in a typical 6x pricing ,markup to retail ie $150 or $600 in total for 12 V , street pricing probably around $400 after discounters

The system doesn't suit " drop in 12V Lithium cause these are 4S3P solutions , its designed to suit Cell level batteries arranged in the better parallel first , series second configuration.

Ive not seen these type of concept discussed anywhere, The closest is RC land, where cell level charging is sometimes used , largely because of the high currents used in RC ( ie high C ) , this method avoids the need to do high current balancing or to slow charging to allow a low current passive or active balancer to work

 

[Ampster]

Dc DC convertors are cheap and ICs for them are widely available , ...........pricing probably around $400 after discounters
.......
Ive not seen these type of concept discussed anywhere, ......

I think a $400 price tag is probably the reason this concept is not often discussed.

 

[Goboatingnow]

I think a $400 price tag is probably the reason this concept is not often discussed.

have you looked at a full 12V 300Ah all cell monitor BMS lately and that's only part of the install as you need a wakespeed 500, etc etc etc . a BMS from Victron is 2x that as well !!!

 

[John Frum]

Charging the big blue cells that we know and love at below .2c cause the SEI layer to sequester lithium that is no longer available to the chemistry.
Just curious, what BMS you use on your 3p4s batteries?

 

[BradCagle]

Conceptionally it's pretty sweet, and something I've thought of myself.

Practicality is not so great, expensive isolated dc/dc converters for each cell bank, and they would probably need to be mounted directly at each bank to avoid giant mess of #2 wire everywhere.

Also you would need separate charge, and discharge ports.

 

[BradCagle]

Dc DC convertors are cheap

I have not found fully isolated DC/DC converters that are cheap.

 

[Goboatingnow]

I have not found fully isolated DC/DC converters that are cheap.

From a design perspective they are , there’s a few more key components then non isolated and the transformer is more expensive ( and bigger) but that’s about it.

 

[Goboatingnow]

Conceptionally it's pretty sweet, and something I've thought of myself.

Practicality is not so great, expensive isolated dc/dc converters for each cell bank, and they would probably need to be mounted directly at each bank to avoid giant mess of #2 wire everywhere.

Also you would need separate charge, and discharge ports.

As I said , 10AWG would be sufficient coupled with voltage sensing , the system is not suitable for drop in 12 v batteries and in a cell by cell arrangement the necessary parallel connected cell terminals are readily available so nothing further is needed . Thd cell chargers are also fulfilling the bms function so would be situated normally where the bms goes , ie close by.

 

[A.Justice]

I think @John Frum hits the crux of the issue in his first post. You would need multiple, very large, 3.6v chargers to make that happen.

Lets say I had a 100ah, 4s battery. I could buy a single, 50a charger, and charge that pack at 14.4v x 50a = 720 watts.

Let's do it your way with 4 chargers and try to get the same charge rate. 3.6v x 50a = 180 watts. With 4 cells, 200a of total charge from 4 separate chargers, to get less than 1/4 the the same charge rate above. To get the same 720w charge rate, I would need four separate 3.6v chargers, all cranking out 200a 50 amps each, 200a total) That's some brutal wiring, not to mention the cost of four 200a, chargers.

I see how the idea would work conceptually, but I think the application would end up being wildly cost inefficient.

-Edited the some early morning mis-steps.

 

[Goboatingnow]

I think @John Frum hits the crux of the issue in his first post. You would need multiple, very large, 3.6v chargers to make that happen.

Lets say I had a 100ah, 4s battery. I could buy a single, 50a charger, and charge that pack at 14.4v x 50a = 720 watts.

Let's do it your way with 4 chargers and try to get the same charge rate. 3.6v x 50a = 180 watts. With 4 cells, 200a of charge from 4 separate chargers, to get less than 1/4 the charge rate above. To get the same 720w charge rate, I would need four separate 3.6v chargers, all cranking out 200a. That's some brutal wiring, not to mention the cost of four 200a, chargers.

I see how the idea would work conceptually, but I think the application would end up being wildly cost inefficient.

Sorry 50a is 50a ie as you say 720 watts

Over 4 parallel cells that’s still 720W , ie 180W per parallel bank. ( or cell )
That’s at a nominal 4v @ 45a . That’s no more wiring then your 14v chsrger

Not to mention it’s not a good ideas to do 0.5C charging

And remember the TCO costs. NoBMS costs , no specialised alternator regulator no specialise mains chargers etc.

 

[A.Justice]

Sorry 50a is 50a ie as you say 720 watts

Over 4 parallel cells that’s still 720W , ie 180W per parallel bank. ( or cell )
That’s at a nominal 4v @ 45a . That’s no more wiring then your 14v chsrger

Not to mention it’s not a good ideas to do 0.5C charging

And remember the TCO costs. NoBMS costs , no specialised alternator regulator no specialise mains chargers etc.

You have an error in your ohm's law conversions or general principal.

50 amps @ 14.4v = 720w
50 amps @ 3.6v = 180w
720w @ 3.6v = 200a

The method you described gets even less efficient with higher voltage packs.

50 amps @ 24v = 1200
50a @ 48v = 2400w

To achieve the equivalent of a 200 amp 3.6v charging on a 48v system, it would use 3.75a @ 48v = 180w. That's the difference between using a solder trace and a welding cable.

 

[Goboatingnow]

You have an error in your ohm's law conversions or general principal.

50 amps @ 14.4v = 720w
50 amps @ 3.6v = 180w
720w @ 3.6v = 200a

The method you described gets even less efficient with higher voltage packs.

50 amps @ 24v = 1200
50a @ 48v = 2400w

To achieve the equivalent of a 200 amp 3.6v charging on a 48v system, it would use 3.75a @ 48v = 180w. That's the difference between using a solder trace and a welding cable.

sorry , if you charge the whole battery with 720w of power and that is made up of 4 cells, each cell gets 720w/4 of power !!!!, nothing to do with ohms law, its the first law of thermodynamics ! """""

each cell gets fed 180W !! , now apply ohms law so at 3.6V thats 50A, remember the current does not travel through every cell, in series as its an isolated dc dc convertor, electrically its no different to dismantling the battery and charging each cell separately

higher voltage packs, have more cells in series , so all that means is duplicating the same cell based chargers , ie one for each parallel set of cells , so in a 24V battery thats 8 cell chargers etc of the same capacity ) , ie 24v charger 700W , means each cell charger is now only doing 700/8, or 25A per cell charger , a 50A cell charger would actually support 1400W charging power in a 24V system ( cause there are more cell hence more cell chargers )

___________________________

***** think it this way , 4 LFP cells at 3.6V nominal in series gives you 14.4V , lets say you push 50A at that voltage , thats 720W from your charger
hence each cell sees the 50A ( kirchoffs laws) at a voltage of 3.6v or 180W
to replicate this with a cell charger is 180/3,6 , ie 50A, conversation of energy applied, certainly 200A is "right out" with deference to Monty Python

 

[A.Justice]

sorry , if you charge the whole battery with 720w of power and that is made up of 4 cells, each cell gets 720w/4 of power !!!!, nothing to do with ohms law, its the first law of thermodynamics ! """""

each cell gets fed 180W !! , now apply ohms law so at 3.6V thats 50A, remember the current does not travel through every cell, in series as its an isolated dc dc convertor, electrically its no different to dismantling the battery and charging each cell separately

higher voltage packs, have more cells in series , so all that means is duplicating the same cell based chargers , ie one for each parallel set of cells , so in a 24V battery thats 8 cell chargers etc of the same capacity ) , ie 24v charger 700W , means each cell charger is now only doing 700/8, or 25A per cell charger , a 50A cell charger would actually support 1400W charging power in a 24V system ( cause there are more cell hence more cell chargers )

___________________________

***** think it this way , 4 LFP cells at 3.6V nominal in series gives you 14.4V , lets say you push 50A at that voltage , thats 720W from your charger
hence each cell sees the 50A ( kirchoffs laws) at a voltage of 3.6v or 180W
to replicate this with a cell charger is 180/3,6 , ie 50A, conversation of energy applied, certainly 200A is "right out" with deference to Monty Python

Correct me if I'm wrong, but I guess your saying that you would charge the individual cells at a lower individual (same overall) charge rate to keep the amperage of each charger and charging circuit lower, but I feel like that's just adding a bunch of devices for no reason. I was under the impression that the goal was to increase the charge rate and efficiency by using more chargers, and my first, incorrect (now edited), response was based off of that. Instead of a hundred amp charger, putting out 25 amps per cell in series, you're proposing four separate 25 amp chargers. I thought you were proposing for separate 100 amp chargers. I stand corrected on the wiring size issue.

I don't think anybody's saying that you can't charge a battery like that, but I'm saying that buying four separate chargers, and four separate sets wiring, is much more difficult and expensive than using a single charger, with a single set of wiring. It's just adding an unnecessary amount of complexity to a problem that can be solved with a single charger and supporting electronics.

I wonder what the cost difference would be for a 16a 48v charger, vs 16 separate 1 amp, 48v chargers. Installing a single unit would be much easier than installing 16.

 

[Goboatingnow]

Correct me if I'm wrong, but I guess your saying that you would charge the individual cells at a lower individual (same overall) charge rate to keep the amperage of each charger and charging circuit lower, but I feel like that's just adding a bunch of devices for no reason. I was under the impression that the goal was to increase the charge rate and efficiency by using more chargers, and my first, incorrect (now edited), response was based off of that. Instead of a hundred amp charger, putting out 25 amps per cell in series, you're proposing four separate 25 amp chargers. I thought you were proposing for separate 100 amp chargers. I stand corrected on the wiring size issue.

I don't think anybody's saying that you can't charge a battery like that, but I'm saying that buying four separate chargers, and four separate sets wiring, is much more difficult and expensive than using a single charger, with a single set of wiring. It's just adding an unnecessary amount of complexity to a problem that can be solved with a single charger and supporting electronics.

I wonder what the cost difference would be for a 16a 48v charger, vs 16 separate 1 amp, 48v chargers. Installing a single unit would be much easier than installing 16.

I’m not focused on charging per se.
I’m specifically looking at Li solutions , these solutions contains BMS system which are essentially cell level systems.
this solutions integrates the bms , without significant additional hardware since voltage and current monitoring is already integrated and the only real addition is cell temp monitoring (arguably temp monitoring should be included in chargers anyway ) equally bms have to add cell balancing hardware and the common passive balancing system throws away energy and balancing delays charge completion anyway.

hence I would argue , that cell level charging incorporating cell level monitor , which as a by product removes any need to cell balance is or should be a viable option and for typical 1/10 C charging common on boats should be cost effective when a BMS cost is removed

as an added benefit no change to stock alternators regulators is needed and smart mains chargers or multi stage solar regulators are NOT needed again lowering the total cost of ownership of a good lithium system

no cell level charge / monitor , no bms , no other specialised charge controllers needed