Lithium-ion balancer questions for 18650 cells

[Trancistor]

Hi - I am new to this forum, and thank you for reading my post.

I am getting one of those cheap eBay Lion balancer boards (about $8) from China, about 4 cm x 6 cm.

I am trying to build a L-ion BMS charger that balances the cells and charges to the limit of 4.2 V with power supplied from an unregulated 25V DC power supply.

It is a 5S 10A balancer.  It also protects the cells.  I am balancing 18650 cells.  Questions:

1)  Can I use this to balance 3S or 4S?

2)  When balancing 5S, the maximum voltage needed is I suppose 5 * 4.2 = 21 V.  Will this charge the cells to 4.2 V each? What happens if the power supply delivers 25 V constant voltage.  Will the board attenuate that?

3)  What happens if the power supply can only deliver 19 V?  19/5 = 3.8 V.  Will all five cells balance exactly at 3.8 V or will they be random with a total of 19 V?  I understand that the cells will not be fully charged. 

4)  When you say 4S3P, is that 4 series of 3P or 3 parallels of 4S?

http://www.ebay.com/itm/5-S-12A-18-5V-Lithium-ion-PCM-BMS-Battery-Protection-Board-LiMn2O4-Balance-New-/122199648617?hash=item1c73aaa969:g:JJoAAOSwal5YEbUr

Thanks!

[Sceadwian]

You do know that you're asking questions about a device you didn't even link when you should be looking at the PDF for the chip on the device you purchased in question which you haven't shared with us.

There is zero information to provide a response based on...

[vealmike]

1)  Can I use this to balance 3S or 4S?
Absolutely. Yes. If you're prepared to rework / redesign it. If not, maybe.

2)  When balancing 5S, the maximum voltage needed is I suppose 5 * 4.2 = 21 V.
That's a poor assumption. The voltage you need will depend on the exact chemistry of the cell. Look in the cell datasheet.

2a) Will this charge the cells to 4.2 V each?
Yes. if you're prepared to rework / redesign it. If not, maybe. You said it was a balancer, not a charger.

2b)What happens if the power supply delivers 25 V constant voltage.  Will the board attenuate that?
I can state with some vague certainty that it either will, or will not burn with the light of a billion suns.

3)  What happens if the power supply can only deliver 19 V?  19/5 = 3.8 V.
In this eventuality, with only 19V on this exact PCB from China, I would expect Steve Hogarth to leave Marillion.

3a) Will all five cells balance exactly at 3.8 V or will they be random with a total of 19 V?  I understand that the cells will not be fully charged. 
Balancers balance. The trick is recognising that the pronoun is also a verb. To be fair, I think this is considered post-grad stuff these days.

4)  The board also protects the cells from under voltage, over current, and short circuit.  Does this mean that during discharge if the cells are at 4 V each for a total of 20 V, the output of the board will only show 19.4 volts or less?
Yes, Steve Hogarth's potential will have dropped by 0.3V, Steve Rothery, Ian Mosley and Fish each take 0.1V. The other instrumentalists are not entitled to a drop in potential, but may try to wangle something. This latter effect will be difficult to measure with an oscilloscope. Please use personal protection when probing Marillion.

4a) Would I need a heat sink @ 5A?  If there is no drop in output voltage, then how does the board break the circuit?
P=IV. Even in China, except I suppose they may use squiggly characters.

5)  When you say 4S3P, is that 4 series of 3P or 3 parallels of 4S?
Ah, a sensible question, deserves a sensible answer. It's ambiguous.

Hi - I am new to this forum, and thank you for reading my post.

Hi Trancistor. Welcome to the forum. If you'd like more accurate details, please post a link to the technical spec. or schematics of the board you have ordered. I only say this as, without more details to back up your questions some users might be tempted to poke a little fun out of you. Please ignore them if they do, I'm sure they mean no offence.


And on a very serious note:
Please do not use this board with unprotected cells. There is a real risk of fire*. You will not be able to extinguish a lithium fire.

*Actually it's more like a slow explosion.

[Trancistor]

You do know that you're asking questions about a device you didn't even link when you should be looking at the PDF for the chip on the device you purchased in question which you haven't shared with us.

There is zero information to provide a response based on...

These are basic questions that anyone that has worked with L-ion charging would know, and is common to all balancers AFAIK.  Like asking how does a Buck converter work.  I have updated the post with a link.

Thanks for your reply.

[Trancistor]

1)  Can I use this to balance 3S or 4S?

.......

Thanks for the (non-) answers.

5) I believe 4S3P is not the same as 3P4S

[vealmike]

You do know that you're asking questions about a device you didn't even link when you should be looking at the PDF for the chip on the device you purchased in question which you haven't shared with us.

There is zero information to provide a response based on...

These are basic questions that anyone that has worked with L-ion charging would know, and is common to all balancers AFAIK.  Like asking how does a Buck converter work.  I have updated the post with a link.

Thanks for your reply.

Sorry no, Sceadwain is correct. What you asked was tantamount to telling us you've bought a vehicle then asking what it's on and offroad performance might be. There are lots of cars, with varying features. May not even be a car for all we know.

Now we have a link, I can provide some more useful answers.
Looks like you have also changed the questions in your original post.

1)  Can I use this to balance 3S or 4S?
Hard to say. I don't think so. It looks like it is based on the Seiko S-8029B series chips (or similar) which daisy chain to provide balancing and protection for batteries.

2)  When balancing 5S, the maximum voltage needed is I suppose 5 * 4.2 = 21 V.  Will this charge the cells to 4.2 V each? What happens if the power supply delivers 25 V constant voltage.  Will the board attenuate that?
No. This is a balancer, not a charger. You need a separate Li-ion charger closely matched to your cell chemistry.
Have you checked that the protection offered by this balancer precisely matches the requirements of your cells? The odd tenths of a volt could result in an unsafe pack.
 

3)  What happens if the power supply can only deliver 19 V?  19/5 = 3.8 V.  Will all five cells balance exactly at 3.8 V or will they be random with a total of 19 V?  I understand that the cells will not be fully charged. 
Please refer to my original reply.

4)  When you say 4S3P, is that 4 series of 3P or 3 parallels of 4S?
It really is ambiguous. You're right that 4S3P is clearly different from 3P4S, but not everyone has the same idea of how this should be interpreted. You'd be foolish to make an assumption here, always check.
I'd say 4S 3P means 3 cells in parallel, in series with 3 cells in parallel, etc. I'd say that's the correct way to interpret this. I'd bet that a majority would agree. I'd also guarantee that some will disagree. 


Do you know how to parallel up unprotected cells? It's another recipe for fire if you don't. Do you know if your cells have protection?

I really wish that unprotected Li-ions were not available to the general public.

[Trancistor]

Do you know how to parallel up unprotected cells? It's another recipe for fire if you don't. Do you know if your cells have protection?

Thank you for your replies vealmike.

How to connect unprotected cells in parallel?  Ok, you soldier a wire to the can of the 18650 and label it minus or negative. You solder another to the top and lapel is plus or positive.  Then you connect the plus to negative and vice versa.  Well, lemme see, on the other hand, with further thought, maybe you connect plus to plus and minus to minus?  Hmmm... I like this better.  It really does not matter.  It is arbitrary.  Just do as you wish.  Experiment.  Be creative.  I like plus to minus better because it also gives you a higher voltage for free.  So you double the Ah and double the V and quadruple the Wh with just two cells!  Yes, I highly recommend that!

Come on vealmike, is that a trick question?  Just hardwire the positives together and negatives with the negatives.

 

[Trancistor]

No. This is a balancer, not a charger. You need a separate Li-ion charger closely matched to your cell chemistry.
Have you checked that the protection offered by this balancer precisely matches the requirements of your cells? The odd tenths of a volt could result in an unsafe pack.

Ahem, how many different types of 18650s do you think exists?  Go to eBay, and find me one 18650 that is not a LCO and is not 4.2 V at 100% SoC.  Even if there is such a thing as a LMN 18650, they will not tell you, and who cares if it wants 4.15V.  As far as I know, Panasonic, Samsung, and LG are all LCOs.  And no, why should I bother as a hobbyist with LFP?  Do you know of any hobbyist that employs LFP?

But you have not answered the question.  After allowing for the digression and making sure things match, what happens if a constant voltage charger of 25 V is hooked to the 5S balancer?  Would the balancer charge each cell to 5 V?  Of course not.  It will protect to 4.2 V.  The MOSFETs will regulate that.  That is my guess.  In other words, the power supply need not be CV, as long as it is a tad over 5 x 4.2 V, we get 100% SoC.  I think this is obvious and am not sure why everyone is having difficulty with this question?  Maybe there is more to it, and you can explain to me.

3)  What happens if the power supply can only deliver 19 V?  19/5 = 3.8 V.  Will all five cells balance exactly at 3.8 V or will they be random with a total of 19 V?  I understand that the cells will not be fully charged. 

My guess is that balancing will not work.  In other words, there is really no balancer.  Just a cell protector at 4.2V and by fiat, balancing happens only at 100% SoC due to protection.  But I could be wrong.

4)  When you say 4S3P, is that 4 series of 3P or 3 parallels of 4S?
It really is ambiguous.

AFAIK, xSyP has a definite interpretation which is different from yPxS. And I agree with your interpretation that it is in reverse construction order and the sequence is right-to-left.  It means "xS of yP".

Thanks again.

[vealmike]

No, it's not really a trick question.
Unless you want to kill your cells you must ensure that they are at exactly the same voltage and that they have relaxed before you parallel them up.

These cells have a very low series resistance. Even if you only have 50mV difference between cells when you connect them a very high current will flow. What should happen is that the PID will blow in one of the cells and kill the cell. If the PID fails (it is considered by the manufacturer to be a tertiary protection), you could get a fire.

I'm trying to help, I'm also trying to illustrate that there is a lot here you don't know. Much more than you seem to think. If you think I'm being harsh or flippant it's because I'm trying to get you to realise that you're playing with fire without an understanding of how fire works. 

Lithium-ion Cobalt cells can have a fully charged voltage between 4.15 and 4.3ish volts. Exceeding the fully charged voltage can cause gassing, popping of the PID, loss of capacity or fire.
The lowest recommended working voltage for Li-ion cobalt is usually 2.7V. It varies a bit. Going below 2.5 ish volts can cause gassing, popping of the PID, loss of capacity but not fire.
Depending on what the cell is designed for the manufacturer will have put a slightly different chemical soup inside. This will vary where the safe limits are. Take a look at the number of variants of the S8029 to see just how important this accuracy is.

Lithium Iron Phosphate  has a fully charged voltage of 3.6V.

Respectlfully, ebay isn't a good source of accurate information or accurately spec'd components.

A protected cell will have a cct. inside that prevents the user abusing that cell. The cct. should be exactly matched to the cell chemistry.

The problems start when uneducated people buy unprotected cells and abuse them. This isn't safe.
Please note that I said uneducated, not stupid. I think you're in this category. It's not an insult, you're here and asking the right questions, which is good! All of us started with no education. A lawyer would think I'm uneducated in the law for example.

OK, to parallel up your unprotected (?) cells. Charge them. Let them relax (couple of hours should do it.) Connect the -ve terminals of two cells together.
Connect the +ve terminals together with a 10 Ohm 1W resistor. Wait a few hours, time to wait will depend on the capacity of your cell. Remove the resistor and connect the +ve terminals together with wire.
Connect the -ve terminal of another fully charged, fully relaxed cell. Connect the +ve terminal through the resistor and wait.
Rinse and repeat.

This allows the voltage on the cells to equalise in a safe way. Current flows between the cells, but maximum current is limited by the resistor.
This must be done whilst the cells are fully charged because the discharge curve is too flat ant any other point. If you're not fully charged (or fully discharged) a small change in voltage can equate to a very large change in stored capacity.


May I suggest that you post a link to the datasheet for the cells you have purchased?

[vealmike]

But you have not answered the question.  After allowing for the digression and making sure things match, what happens if a constant voltage charger of 25 V is hooked to the 5S balancer?  Would the balancer charge each cell to 5 V?  Of course not.  It will protect to 4.2 V.  The MOSFETs will regulate that.  That is my guess.  In other words, the power supply need not be CV, as long as it is a tad over 5 x 4.2 V, we get 100% SoC.  I think this is obvious and am not sure why everyone is having difficulty with this question?  Maybe there is more to it, and you can explain to me.

A constant voltage charger is not suitable for a Li-ion of any type. You need a charger with two constant current phases followed by a constant voltage phase.
If you tried to charge this pack with a 25V constant voltage (assume max current of less than the 30A that your cell protection cct uses), I predict blown PIDs or fire.

If that board is based on the S8209B (I am guessing) then the mosfets do not regulate. They simply turn off to prevent further charge or discharge. They do not operate linearly to regulate.

[IanMacdonald]

Basically you should not rely on a protection (or balancing) board to cut off the charge when the limit voltage is reached. It's there as a safety device to do so if your charger regulation fails.

The charger should be current limited and voltage limited. Ideally it should also shut off once the current has dropped to a low level at the limiting voltage, since continuous trickle charging is not good for lithium cells.

[Trancistor]

Unless you want to kill your cells you must ensure that they are at exactly the same voltage and that they have relaxed before you parallel them up.

Well of course.  The difference can be as much as 0.3 V and it will not blow up or stress out.  The interconnections could easily amount to 0.1 ohms in a hobby application.  I guarantee you that.  First off, 18650s do not blow up in fireworks.  What you do is that when the batch of cells arrive, you measure them and then sort them according to static voltage.  Then you group the closest ones together and assure they are within let's say 0.2 V.  How difficult is this?  You make it sound like the end of the world is at hand.   

These cells have a very low series resistance. Even if you only have 50mV difference between cells when you connect them a very high current will flow. What should happen is that the PID will blow in one of the cells and kill the cell. If the PID fails (it is considered by the manufacturer to be a tertiary protection), you could get a fire.

I have accidentally shorted 18650s or connected them in the wrong way and have had red hot wires or cells that are cooking, and no PID has blown.  As far as I know, the PID will not blow unless the cell is shorted/abused or is charging at a high rate and for a long time and exceeding its upper voltage limit where the energy is converted into thermal at 100%. Believe me that is not my intention to short things or abuse things - hobbyists do not try to destroy things, but to build things.  Let's say we connect two 2.5 Ah cells in parallel with a difference of 0.2 V.  They will equalize at the average.  Thus the charge/discharge is 0.1 V.  Looking at a capacity/potential chart, you will see that at the flattest part of the curve, about 3.7 V, the (0.1 V diff) capacity is about 30% of the cell, or 0.75 Ah.  Usually 90% of the energy is converted into chemical energy.  Thus thermal energy produced is 0.0075 Wh.  If this is dissipated over 15 seconds, you have a power of 1.8 W.  Tell me which 18650 cannot withstand 1.8 W for 15 seconds?  It will get warm, but not enough to boil the thing and blow the PID.  The PID will blow upon high current over charging, but not when you connect two cells with a difference of 0.2 V.  In my example, I took the worst case of a 3.7 V static potential.

vealmike, I think we are coming from two different cultures.  My culture is a hobbyist who is designing something complex where just a small part of the requirements is a mobile power source of 500 Wh.  I believe you may be in a commercial setting designing products that are powered by Lithium-ion batteries.  Safety is paramount to you.  For me, I am not going to put my battery pack in my shirt pocket and then charge them at 10C with an unreliable power supply, and then go to sleep.  This may explain why my naive seat-of-the pants questions creates a reaction.  Our material settings are very different.  Our contexts are different.  My tolerance of risk is much higher because I can manage the risk, and I have other things to do than build a perfectionist BMS. I don't go by the book, because if I do that I get nowhere. While for you, it is imperative to go by the book, or you will lose your job.

I have not received the cells yet.  But I bought 50 Panasonic 3.3 Ah (I believe the NCR 'B' designation) from Alibaba.  I am sure these are rejects and I will be testing them.  For my project they will do if they can hold above 3 Ah.  They may have been rejected for safety issues.  Well, that is life for a hobbyist, and if we cannot take some risk and expect nanny state, we should all stay in bed so as to not be run over by cars.

[Trancistor]

Basically you should not rely on a protection (or balancing) board to cut off the charge when the limit voltage is reached. It's there as a safety device to do so if your charger regulation fails.

The charger should be current limited and voltage limited. Ideally it should also shut off once the current has dropped to a low level at the limiting voltage, since continuous trickle charging is not good for lithium cells.

The battery charger is simply a power supply.  There is no regulation (to fail).  I am trying to build a L-ion BMS/charger from a simple power supply.

Yes, the BMS should cut off the current if the voltage never reaches the limit of 4.2 V.  It should look at the rate of delta-V per delta-Ah.  If that drops below a threshold, power should be cut off.  The battery could be defective for example.  I believe this is what high-end BMS do.  But I do not think these cheap protection boards/balancers do this.  Do you know if microcontroller monitoring is required, or is there some easier way to accomplish what you indicated that would work with a protection board?

[vealmike]

Well, that is life for a hobbyist, and if we cannot take some risk and expect nanny state, we should all stay in bed so as to not be run over by cars.

You have me all wrong.  Yes I'm a professional electronics designer, but in my spare time, I'm a paraglider pilot, a white water kayaker & a caver (although I've not been for a while!) I drive a three wheeled car with no roof, doors or impact protection.
There is a difference between taking a risk blindly and understanding all the elements of risk then picking the safest path. But you know best.

Psst, you still need a Li-ion charger. For a four cell pack, consider a BQ24618.

[Trancistor]

Psst, you still need a Li-ion charger. For a four cell pack, consider a BQ24618.

Again, there is some kind of miscommunication going on.  I don't understand.  Are you saying the the supply voltage for my cheap 5S balancer must be exactly 21 V CV?  What if I supply 24 V from a cheap 60W power supply that is NOT CV.  Maybe with a 4R resistor in series?  Upon charging, the balancer input voltage can drop as low as 18 V or so.  The balancer will then reach 21 V at full charge, and then the MOSFET will cut off the supply?  Why do I need a CV charger like BQ24618, which will be an order of magnitude more complex?

[vealmike]

Because just connecting Li-ion cells to a voltage is generally considered a dumb ass thing to do. You need a charger. If you want to know why or you don't want to take my word for it, click HERE for more info.

There is no miscommunication, the message being sent very clearly.

[Trancistor]

Because just connecting Li-ion cells to a voltage is generally considered a dumb ass thing to do. You need a charger. If you want to know why or you don't want to take my word for it, click HERE for more info.

There is no miscommunication, the message being sent very clearly.

But the Li-ion is not connected directly to the supply!  You are mistaken.  There is a MOSFET in series, and the balancer/protection controls the MOSFET.  Look at the S8029B datasheet (thanks for that BTW).  Are you reading what I write?  I clearly say in my last post that the supply is connected through 4R to a balancer which will protect for overcharging.  And the balancer/protection will cut off at 21 V.  Have you misunderstood?

Again, maybe, this is not how it is done in the commercial world.  But I want a quick and dirty charger and can rely on the balancer to cut off the supply.  If the balancer/protection will fail, then why can't a charger fail?  What makes you think that the charger is inherently more reliable than a balancer/protection?

I think you have made certain assumptions about the circuit or are considering a different context than I am.  It would be nice to know what these are.   

[SvanGool]

Are you reading what I write?

 

@Transistor:  I think YOU should re-read this thread. I know you are going to think differently about it, but I do think that "vealmike" spend an awful lot of time on your questions and that you shouldn't double post in the "Renewable energy" section, he doesn't deserve that .

[Trancistor]

@Transistor:  I think YOU should re-read this thread. I know you are going to think differently about it, but I do think that "vealmike" spend an awful lot of time on your questions and that you shouldn't double post in the "Renewable energy" section, he doesn't deserve that .

I posted the question in the Renewable Energy section before vealmike had answered anything.  It has nothing to do with him.  There are other members frequenting that place who do not come here, and are answering the questions.  The more replies, the merrier.  You mean the parody he made out of my questions (see first reply) is "spending an awful lot of time"?

Fact is that everyone is using balancers and protection circuits for Lithium-ion.  But vealmike insists one has to build a dedicated charger.  Can you explain to me why?

[janekm]

(snip)

Again, maybe, this is not how it is done in the commercial world.  But I want a quick and dirty charger and can rely on the balancer to cut off the supply.  If the balancer/protection will fail, then why can't a charger fail?  What makes you think that the charger is inherently more reliable than a balancer/protection?

I think you have made certain assumptions about the circuit or are considering a different context than I am.  It would be nice to know what these are.   

The result of a failure to adequately protect a cell during charging is potentially extremely unpleasant, to say the least. That's why the standard practice is multiple layers of protection. The standard is thermal cut-out in the cell, protection PCB, and a charger designed to accurately cut off at the target voltage (CC/CV). With the number of fakes/low quality product in the channel, the situation is even more critical... If you get a fake cell, the thermal protection is likely lacking, the balancer may have poor soldering / wrong choice of protection IC, etc.

A charger that implements the correct charge algorithm is easy to get and inexpensive, so there is little justification in leaving it out. You can get laptop wall charger style chargers for all the common cell voltages, for example.

[vealmike]

Because just connecting Li-ion cells to a voltage is generally considered a dumb ass thing to do. You need a charger. If you want to know why or you don't want to take my word for it, click HERE for more info.

There is no miscommunication, the message being sent very clearly.

But the Li-ion is not connected directly to the supply!  You are mistaken.  There is a MOSFET in series, and the balancer/protection controls the MOSFET.  Look at the S8029B datasheet (thanks for that BTW).  Are you reading what I write?  I clearly say in my last post that the supply is connected through 4R to a balancer which will protect for overcharging.  And the balancer/protection will cut off at 21 V.  Have you misunderstood?


Again, maybe, this is not how it is done in the commercial world.  But I want a quick and dirty charger and can rely on the balancer to cut off the supply.  If the balancer/protection will fail, then why can't a charger fail?  What makes you think that the charger is inherently more reliable than a balancer/protection?

I think you have made certain assumptions about the circuit or are considering a different context than I am.  It would be nice to know what these are.   

What is the point of asking for advice if you do not listen when it is given?

The job of the protection circuit is to be a last ditch attempt to prevent cell abuse. Adding a protection circuit doesn't make it OK to abuse the cell.

I am not going to reproduce the contents of the internet here for you. I've given you a lmgtfy link, that's the most I'm going to do.
I will suggest that you google the following:
 Lithium, Properties of, position in the periodic table.
 Lithium fires, extinguising.
 How to safely charge a Li-ion.
 Li-ion protection.

You need a charger.


Oh, and by the way, IF that balancer / protection  board is based on a S8029, then it won't work with four cells. Given that you seem determined to abuse your cells, I'm not going to tell you why, or tell you how to reconfigure it.

[vealmike]

BTW, I did not get very useful answers.  Only two of my 5 questions were answered in a half-hearted manner.

No, all five of your questions were answered fully and with as much detail as can be accurately given based on the sparse information you've given.
I think you're confusing usefulness with your inability or unwillingness to comprehend what's being said.

I'm out.

[Trancistor]

No, all five of your questions were answered fully and with as much detail as can be accurately given based on the sparse information you've given.
I think you're confusing usefulness with your inability or unwillingness to comprehend what's being said.

I'm out.

Your answers are of this nature: 1- First try mockery, to establish and rub in your high-priest attitude. 2- Then say it cannot be done, to discourage alternative methods. 3- When proven mathematically or otherwise that it can be done, then say that is not safe and it will blow up, while refusing to provide the reasoning for that.  You seem to have a condescending attitude towards anyone who does not want to do things your way - i.e. the run-around way.

Your answer to #5 was the most amusing:  "4S3P can mean 3P4S"

So long.

[Trancistor]

The result of a failure to adequately protect a cell during charging is potentially extremely unpleasant, to say the least. That's why the standard practice is multiple layers of protection. The standard is thermal cut-out in the cell, protection PCB, and a charger designed to accurately cut off at the target voltage (CC/CV). With the number of fakes/low quality product in the channel, the situation is even more critical... If you get a fake cell, the thermal protection is likely lacking, the balancer may have poor soldering / wrong choice of protection IC, etc.

A charger that implements the correct charge algorithm is easy to get and inexpensive, so there is little justification in leaving it out. You can get laptop wall charger style chargers for all the common cell voltages, for example.

Well, the 5S balancer I am getting and linked above does all of this.  Has thermal cut-out, has overcharge (& undercharge, & overcurrent) protection, and has accurate cut-off.  So what is your point?

No, I am not getting 18650 cells with built-in protection.  I need the full 3.3 Ah (or more) per cell, for my project.  This is a serious project and I can't use "kiddie-cells".

And if I can get a fake balancer from China, why can I not get a fake charger from China?

The reason I wish to do my own BMS is that the 5S is a pilot project towards an eventual 13S and 96S pack.  (Not that I will be using a 5S balancer.)  Would you please refer me to a 96S charger/balancer?  Thanks!

[vealmike]

OK, one last post...

No. This is a balancer, not a charger. You need a separate Li-ion charger closely matched to your cell chemistry.

A constant voltage charger is not suitable for a Li-ion of any type.

Basically you should not rely on a protection (or balancing) board to cut off the charge when the limit voltage is reached. It's there as a safety device to do so if your charger regulation fails.

The charger should be current limited and voltage limited. Ideally it should also shut off once the current has dropped to a low level at the limiting voltage, since continuous trickle charging is not good for lithium cells.

Psst, you still need a Li-ion charger. For a four cell pack, consider a BQ24618.

Because just connecting Li-ion cells to a voltage is generally considered a dumb ass thing to do. You need a charger. If you want to know why or you don't want to take my word for it, click HERE for more info.

There is no miscommunication, the message being sent very clearly.

I think it is best to use a dedicated li-ion charger, be it a module or complete charger unit. In theory, if you had a 21v 1A supply, it should charge at constant 1A current until it reaches 21v, but I would not encourage it or rely on it and think that you would be taking a big risk in doing so.

The result of a failure to adequately protect a cell during charging is potentially extremely unpleasant, to say the least. That's why the standard practice is multiple layers of protection. The standard is thermal cut-out in the cell, protection PCB, and a charger designed to accurately cut off at the target voltage (CC/CV). With the number of fakes/low quality product in the channel, the situation is even more critical... If you get a fake cell, the thermal protection is likely lacking, the balancer may have poor soldering / wrong choice of protection IC, etc.

A charger that implements the correct charge algorithm is easy to get and inexpensive, so there is little justification in leaving it out. You can get laptop wall charger style chargers for all the common cell voltages, for example.

Well, the 5S balancer I am getting and linked above does all of this.  Has thermal cut-out, has overcharge (& undercharge, & overcurrent) protection, and has accurate cut-off.  So what is your point?

I think his point is that you need a charger.

No, I am not getting 18650 cells with built-in protection. 

 

This is a serious project and I can't use "kiddie-cells".

  Priceless!

Your answer to #5 was the most amusing:  "4S3P can mean 3P4S"

That is not what I said. I said that it is sufficiently ambiguous that someone will use the alternative meaning. You should always check what is meant, never underestimate the ingenuity of dolts.

Your answers are of this nature: 1- First try mockery, blah blah

You are correct, whilst I'm happy to help people learn, I do have a low tolerance for window licking lard gargling hamster tamers with delusions of attaining the dizzying heights of mere inadequacy.

Mods, can we close this before I resort to personal insult please? This is clearly going nowhere.

[Trancistor]

Well, sorry to disappoint vealmike - but I had a 7S board and I hooked it up to seven LCO 18650s that were not even matching.  The cutoff voltage is 7x4.2 or 29.4 V.  I fed it 32 V with a 12 resistor in series.  As expected, the board balanced the cells, even though they were far from matching, and cut off the current to zero at exactly 29.8 V, for an average of 4.25 V.  The current at cutoff was 80 mA, which is about right.  Just as expected, there were no overcharging.  It has now relaxed to 29.4 V and the cut-off is still in effect.

And you said it cannot be done.  And if it is done, too dangerous.  Then can you explain why are people selling balancer/protection boards all over the place, if they should not be used to charge Li-ion cells?

[janekm]

Well, sorry to disappoint vealmike - but I had a 7S board and I hooked it up to seven LCO 18650s that were not even matching.  The cutoff voltage is 7x4.2 or 29.4 V.  I fed it 32 V with a 12 resistor in series.  As expected, the board balanced the cells, even though they were far from matching, and cut off the current to zero at exactly 29.8 V, for an average of 4.25 V.  The current at cutoff was 80 mA, which is about right.  Just as expected, there were no overcharging.  It has now relaxed to 29.4 V and the cut-off is still in effect.

And you said it cannot be done.  And if it is done, too dangerous.  Then can you explain why are people selling balancer/protection boards all over the place, if they should not be used to charge Li-ion cells?

It's simple... no sane person will give advice in a public forum that it is OK to divert from industry standard safety norms when it comes to charging lithium cells. The risk factors are well known and the protections are standard and frankly cheap. Even with all the protections a certain risk remains (ask Samsung about that one...).

The purpose of the balancing/protection boards you see being sold is to help build battery packs for notebooks, power tools, small electric vehicles etc. These are always going to be used together with a charger that regulates the charge current and voltage (implements the CC/CV algorithm at a minimum, typically thermal monitoring as well). They are intended to protect the battery pack against malfunction of the charge circuit, and against over-discharge.

Nobody is doubting that a functioning battery protection PCB will stop the charge when the float voltage has been reached. It's supposed to, but it's only intended to have to do so if your charger failed.

My point is that the battery, the protection PCB, and the charger could each be fake or faulty. The likelihood that all three are is less.

Yes you can easily find 13S chargers, 96S not so much.

[Trancistor]

These are always going to be used together with a charger that regulates the charge current and voltage (implements the CC/CV algorithm at a minimum, typically thermal monitoring as well). They are intended to protect the battery pack against malfunction of the charge circuit, and against over-discharge.

Thanks for explaining this so well. What you are saying is that the 5S protection board is a SECONDARY line of defense, i.e. a backup safety, to that of the CV charger - when charging.  That does make sense.

(I believe, the other fellow was saying don't use this board, just use a CCCV charger.)

[Trancistor]

Well, I managed to get the 7S balancer-protection board to actually charge, discharge, protect and balance only five cells - namely act as a 5S.  So this answers my first question positively.  A 7S board has been reconfigured to balance and protect a 5S configuration.