Efficient charge/discharge Li-Ion battery

[AGRAIL]

Hi ,
I`m dealing with one schem of efficiently charge / discharge of lithium assembly (6P) and stuck with one problem.
In case charge / discharge Li-ion there are already many different devices , but main problem with them is heat - all discharge will go into heat.
This is not problem with one battery - but can cause an real problem in case scaling this job .
So idea is to use hybrid inverter - solar input as load and AC output and source for charging .
In case of high input voltage of inverter can be manage with serial connection of few assemblies and higher q-ty input BMS .
But main concern how to limit solar input current -  many of them use mppt and draw maximum power of the solar input .
This can shutdown BMS and stop discharge.
Maybe someone use more hybrid inverters and maybe some vendors have some SW limitation for solar input ?
Thanks in advance .

[Siwastaja]

This just word salad, completely non-understandable. Please try again, what is the battery and what it is used for?

[radiolistener]

So idea is to use hybrid inverter - solar input as load and AC output and source for charging .

This is a bad idea, because AC inverter has very poor efficiency, especially if you're converting AC=220V from 6S DC=6×3.7V=22.2V.

Much more efficient way is to use DC/DC from 22.2V to 5V, 12V or whatever you want. This is because conversion efficiency depends on voltage ratio. 22.2 / 5 = 4.4 which is better than 311 / 22.2 = 14 (220Vrms = 311Vpk). In addition, there will be additional losses because it also requires 50 Hz sine wave AC instead of DC.

For 6P it will be even worse, because voltage ratio will be 311 / 3.7 = 84, too much power will be lost due wire heating and bad conversion efficiency.

[AGRAIL]

Purpose - I need to charge / discharge 6S3P battery few times in a row and verify capacity of them . In case of one battery - it`s not a problem to use china battery testers 150W . But if this job must be scaled - all this energy lost as heat can be a problem.
So idea was somehow ( at least some part of energy ) transfer between batteries.
2nd idea was to connect 4 batteries in parallel with own BMS to AC inverter -> inverter connect to battery charger , which is connected to 4 parallel batteries with own BMS .
I was thinking about some DC-DC converter - but for such currents I can`t find widely available models . In my case it`s much easier to buy AC inverter and battery charger . When discharge cycle will finish - I`ll switch charger into grid for charging finish . After finish - I`ll reconnect batteries between the sides.

[Siwastaja]

Hacking it together from some off-the-shelf 12V-to-230V inverters and battery chargers is not going to work well and not going to be very efficient either.

In any case, sounds familiar. I have designed such battery tester, basically bidirectional DC/DC between two batteries. But it's quite a lot of design work involved. Professional solutions are thus very expensive. Just get the China tester and accept the heating.

[Rick Law]

Hi ,
I`m dealing with one schem of efficiently charge / discharge of lithium assembly (6P) and stuck with one problem.
In case charge / discharge Li-ion there are already many different devices , but main problem with them is heat - all discharge will go into heat.
This is not problem with one battery - but can cause an real problem in case scaling this job .
...

Purpose - I need to charge / discharge 6S3P battery few times in a row and verify capacity of them . In case of one battery - it`s not a problem to use china battery testers 150W . But if this job must be scaled - all this energy lost as heat can be a problem.
...

(Bold added to above quotes)

I think your choice of word "efficiently" may lead to some confusion.  Perhaps you mean rapidly?  And what do you mean by scaling this job - testing hundreds/thousands of packs per day?  Perhaps you mean repeatable easily since you stated that you need to do it a few times in a row?

When you are testing capacity of a battery, efficiency is not meaningful unless you are doing it commercially and the cost of recharging affects your bottom line.  As you already noted, the discharge energy are waste heat.  So, a design that could "waste it efficiently" is nonsensical -- unless recharge cost is an important part of the equation.

If you re-consider your problem, and if you conclude efficiently really means doing it rapidly, heat is not an issue at all.  Your discharge load can include a fan or fan(s) to move the heat else where.  With 6S3P, 6S will be about 24V, so have a few sets of 2x12V fans as part of your load.  Now some of those energy that was to go as heat will now be kinetic energy to get rid of the heat.

[MrAl]

Hi ,
I`m dealing with one schem of efficiently charge / discharge of lithium assembly (6P) and stuck with one problem.
In case charge / discharge Li-ion there are already many different devices , but main problem with them is heat - all discharge will go into heat.
This is not problem with one battery - but can cause an real problem in case scaling this job .
So idea is to use hybrid inverter - solar input as load and AC output and source for charging .
In case of high input voltage of inverter can be manage with serial connection of few assemblies and higher q-ty input BMS .
But main concern how to limit solar input current -  many of them use mppt and draw maximum power of the solar input .
This can shutdown BMS and stop discharge.
Maybe someone use more hybrid inverters and maybe some vendors have some SW limitation for solar input ?
Thanks in advance .

Hello,

Let me see if I understand what you are saying here in brief.

First, you want to discharge Li-ion cells, and because there will be more than one there will be high current and high power wasted.
Second, you want to charge Li-ion cells using a solar panel.

In the case of discharging the Li-ion cells, you do not want to use a resistive load because that will cause too much heat, so you want to use that energy to power something else with maybe an inverter.

In the case of charging the Li-ion cells, you want to limit the current because it's too much for the cells to handle.

Is all of this correct?   If not, indicate what is not correct and what is correct.
I have to ask you questions because your post is not clear on what you want to do exactly.

[AGRAIL]

First, you want to discharge Li-ion cells, and because there will be more than one there will be high current and high power wasted.

- correct.

Second, you want to charge Li-ion cells using a solar panel.

- no , idea was to use solar input as load for battery discharge ( to not use resistive load ).

In the case of discharging the Li-ion cells, you do not want to use a resistive load because that will cause too much heat, so you want to use that energy to power something else with maybe an inverter.

- correct.

In the case of charging the Li-ion cells, you want to limit the current because it's too much for the cells to handle

- most probably , this I`ll limit with supply itself as I can have connection to BMS .

[AGRAIL]

I think your choice of word "efficiently" may lead to some confusion.  Perhaps you mean rapidly?  And what do you mean by scaling this job - testing hundreds/thousands of packs per day?  Perhaps you mean repeatable easily since you stated that you need to do it a few times in a row?

- more efficient way that use resistive load and convert all energy into heat. Few times in a row - it`s requirements per each battery to verify their capacity ( 3 charge and 3 discharge cycles ). About scaling - yes, this can be up to hundreds per day indoor .

When you are testing capacity of a battery, efficiency is not meaningful unless you are doing it commercially and the cost of recharging affects your bottom line.  As you already noted, the discharge energy are waste heat.  So, a design that could "waste it efficiently" is nonsensical -- unless recharge cost is an important part of the equation.

Heat is much more critical then bill for electricity at the moment ( but if I can manage both problems - why not ).

Your discharge load can include a fan or fan(s) to move the heat else where

Sounds easy "elsewhere" , when you don`t know price for such system installation , which can transfer heat / air outdoor in this volumes ( for information - ventilation or cooling cost more that stations for this batteries ).

[Siwastaja]

So what size of batteries? In kWh? I.e., tell us what you are actually doing. Battery testing scales quite far using dissipative discharge. Of course at some point thermals and cost of electricity start to hurt but without numbers, I'm not convinced you are anywhere near that point.

In any case, you can also get regenerative Chinese RC hobby pack testers in the price range of a few hundred $. The name escapes me but I used to have one with 12V-24V storage battery connection and 10s output (so up to 42 or so V). But these also come with limitations. You need to charge the storage battery to compensate for the energy loss and if you charge it too far, then the tester has to resort to dissipative discharge anyway.

[AGRAIL]

Size is 6S3P 250Wh, but maybe will be even more .

[Rick Law]

...
About scaling - yes, this can be up to hundreds per day indoor .
...

Doing it merely few times in a row as you initially stated, the cost of implementation vs energy cost saving probably doesn't work.  But if you are doing it hundreds of times per day, the math would likely work.

Then again, may be one can draw enough benefit in doing it merely for learning or fun.  That is a value beyond quantification.

[MrAl]

First, you want to discharge Li-ion cells, and because there will be more than one there will be high current and high power wasted.

- correct.

Second, you want to charge Li-ion cells using a solar panel.

- no , idea was to use solar input as load for battery discharge ( to not use resistive load ).

In the case of discharging the Li-ion cells, you do not want to use a resistive load because that will cause too much heat, so you want to use that energy to power something else with maybe an inverter.

- correct.

In the case of charging the Li-ion cells, you want to limit the current because it's too much for the cells to handle

- most probably , this I`ll limit with supply itself as I can have connection to BMS .

Hello,

Not sure what you mean when you say:
 "idea was to use solar input as load for battery discharge ( to not use resistive load )".

I assume this is a solar panel?
Why would you want to use a solar panel as a load?  Do you mean connect the solar panel to the batteries directly as a load? There would still be energy lost as heat in the solar panel.

I am not sure there has been much research on this but here are a few notes...
First, you would have to have enough voltage to push enough current through the panel.  The panel looks like a lot of diodes in series so you'd have to have a high enough voltage to allow all those diodes to conduct.  For example, for a typical 12v nominal panel you might need as much as 21 volts to get enough current to flow through it.  It would act as a load, but the load current could not exceed the max current of the array.
Second, you might be taking a chance on this not hurting the solar panel.


I can guess that you want to use the panel because it's already available and it has a larger surface area so it may not get too hot.  Resistors have a relatively small surface area so they get very hot with a lot of power, even though they may be rated for that power level.
This could work I would think, but you would need a boost converter to boost the cell voltage up to the right voltage for the panel if the panel was a typical 12v panel.  The current through the panel would have to be limited that's for sure.
Perhaps you can test this idea with a lower cost panel first.  To understand what will happen you can look deeper into what happens when you use rectifier diodes in series as a load.  It can be done but you should look at the current/voltage characteristics doing it this way first.

Does the above sound like something you were thinking about for using the panel as a load?

[Rick Law]

I think there is a lot here that would be the learn-as-you-go kinds.

- "Capacity" alone doesn't mean a whole lot.  Capacity varies a lot depending on current drawn.

- Batteries degrades as you use them (or not).  Just sitting there the battery is degrading slowly.  A charge and discharge is a cycle and each cycle degrades the battery a bit.  A "few hundred cycles" is the life of the battery.  If you do test the same battery pack a few hundred times, that battery pack is close to totally done.  Google search for "lithium battery cycle life" returns this: "... For example, a lithium-ion cell charged to 4.20V/cell typically delivers 300–500 cycles. If charged to only 4.10V/cell, the life can be prolonged to 600–1,000 cycles..."

- You will need to pump in measurably more power than it actually stores.  Discharging to charge another battery is degrading two batteries at the same time.  Besides, efficiency is never 100%.  You cannot use a full 6s3p pack to fully charge another 6s3p pack of the same nominal capacity.

- At the risk of over-simplification:  The longer the battery stays at high voltage (> nominal), the more it hurts the battery.  The higher the demand (current), the more it hurts the battery.  Lithium battery is like a typical cat, very finicky.  You need to treat it well and it will return the favor; mistreat it, it will scratch you.

I've seen at least one advertisement saying this EV's battery can go a million miles.  I wonder, as they are nearing the million miles mark, how many miles can the car go with a full-recharge?  Is yards or feet a better measuring unit by then?

Edited for typos and grammar.

[AGRAIL]

Hi
Most probably you miss my point - solar input of hybrid invertor used as a load for 6S3P battery packs discharge cycle ( forget about the solar panels ) .

I`m also not talking about life cycles , battery degradation or similar.

Batteries travels quite some time to vendor , which is welding this packs (6S3P) and he would like to test it 3 times with charge / discharge cycles.
It`s not a problem with 20 packs per day , but this can have a problem if he would like to do it for all batteries in production .
So one of the ideas was to use hybrid inverter with solar input as a load - even with all losses on inverter, it`s better then simply use resistive load in case of heat dissipation (1 pack can dissipate up to 700W as heat , so  with 100 packs tested we are talking about 70kW heat  - but it`s only 100 packs ...).
So this for me is a bigger problem that electrical bills.