Cheap LiFePO4 BMS?

[peterboat]

13 minutes ago, Dr Bob said:

If your cell 3 reached 3.9V then you dont have the right settings for you BMS.

When any cell reaches 3.5V and you are charging at 30-40A (and certainly over), you need to be watching cell voltages by the minute.

 Take power out of cell 3. Maybe aim to get 10Ahrs out first then charge the whole bank again and watch the cell voltages. Repeat until cell 3 is not the highest cell.

To get 10Ahr out, you will need something that draws 10A at 3.3V. Not easy. I didnt think you could do it with a 12V element when I tried to do the sums. I built my own resistor circuit. My highest cell was 3.6V when the others were just getting to 3.4V and I had to take 20Ahrs out of the high cell to get it to balance. It took me around 4-5 hours.

As you charge, record the cell voltages every 5mins then every 1 min as they get to 100%.

Once you have finished your balancing, I would set the bulk voltage to 13.8 or 13.9V to keep safe.

The balancing circuit in the BG8S will be useless. It will do a max of 100ma - and only if cell voltage is high. You are looking at taking out 10A. You dont want to keep your cells at those high voltages for days on end.

Forget the yttrium in the cells. I have exactly the same cells as you and 13.9V is my limit. Never go over it.

Getting the settings right on my truck mppt controller was a pain! It charges to 56.2 volts dividend by 4 gives 14.05 it's set at 13.9 which as you can see it doesn't do! If I set it at 13.8 it charges to 13.7!!! Don't ask me why, anyway in a years running the 14.05 per battery hasn't been an issue batteries balanced and flashing green so alls well. I am using a cheap Powmr controller on the truck which might be a mistake but it seems to work 

[jetzi]

46 minutes ago, Dr Bob said:

If your cell 3 reached 3.9V then you dont have the right settings for you BMS.

The BMS will only shut down the charge at a cell voltage of 4.0V, because otherwise it is going to damage my MPPT. I'm relying on the MPPT to stop charging, but of course that's based upon the voltage of the whole battery, not individual cells.

 

48 minutes ago, Dr Bob said:

Take power out of cell 3. Maybe aim to get 10Ahrs out first then charge the whole bank again and watch the cell voltages. Repeat until cell 3 is not the highest cell.

I should point out that the cell was only in the 3.8x volts under charge. When I removed the charge, it fairly quickly subsided to 3.418V. 

 

This makes it quite a lot harder to understand. When we're talking about all these voltages, are we talking about the voltage while under charge or when the charge source is removed from it?

 

 

I have been using the power today and I've disconnected the solar for now. My voltages are currently sitting as follows:

 

Cell 1: 3.314V

Cell 2: 3.324V

Cell 3: 3.322V

Cell 4: 3.314V

 

 

 

[nicknorman]

1 hour ago, ivan&alice said:

The BMS will only shut down the charge at a cell voltage of 4.0V, because otherwise it is going to damage my MPPT. I'm relying on the MPPT to stop charging, but of course that's based upon the voltage of the whole battery, not individual cells.

 

I should point out that the cell was only in the 3.8x volts under charge. When I removed the charge, it fairly quickly subsided to 3.418V. 

 

This makes it quite a lot harder to understand. When we're talking about all these voltages, are we talking about the voltage while under charge or when the charge source is removed from it?

 

 

I have been using the power today and I've disconnected the solar for now. My voltages are currently sitting as follows:

 

Cell 1: 3.314V

Cell 2: 3.324V

Cell 3: 3.322V

Cell 4: 3.314V

 

 

 

When charging and considering an upper limit we are talking about the voltages under charge. When considering the minimum voltage we are considering voltage under discharge.

 

It is the nature of the “knee” that if you get the voltage up to say 3.9v on charge, when you remove the charge it will fairly quickly subside even if you don’t discharge it. I’m not sure what the Yttrium does to the start of the knee so can’t comment on ho far up it 3.9v is, but it sounds as though it’s some way up it!

 

One thing I’ve found when tinkering with the little AA cells is that it’s hard to know what is going on during a bit of discharge, in terms of cell balance. Well, unless they are very well matched which I shouldn’t think your second hand ones are. IMO you really need to observe which cell starts to climb the knee first during charge and then take some charge out of it using Dr Bob’s big resistor. Having the cell voltages matched at mid-charge isn’t really the issue.

 

My plan is to have the alternator or charger hold the voltage at a value that gets the highest cell just onto the knee then switch in the big resistors onto the highest cell(s) until they are top balanced. Hopefully it doesn’t have to be done often!

[peterboat]

I will be building a bank for my electric car, I am definitely putting in a good BMS to balance the batteries under charge and discharge plus sticking to the 80% SOC.

It's a 48 volt nominal system based on 15 cells in parallel then in series 15 times should be about 17.5kwhs, I expect a range of 120 miles on the Aixam 

[jetzi]

18 minutes ago, nicknorman said:

It is the nature of the “knee” that if you get the voltage up to say 3.9v on charge, when you remove the charge it will fairly quickly subside even if you don’t discharge it.

Thank you this is super helpful.

So in the case of that cell at 3.9V, which was probably 95% SoC at that point, when I removed the charge (around lunchtime) it was in a bit of a "danger zone" there - although I did discharge all the cells right away so it should be OK now I think (BMV is showing 25Ah consumed out of the battery so Cell 3 should be no more than 90% charged now)?

 

Or do you think I should take a bunch of energy out of the whole battery to be safe?

 

22 minutes ago, nicknorman said:

you really need to observe which cell starts to climb the knee first during charge and then take some charge out of it using Dr Bob’s big resistor.

Alright so plan of action - tonight I will disassemble my battery and heat up a bucket of water from just Cell 3 (I'll maybe run my 600W 12V heating element for 15mins or so). Tomorrow I will put the bank on charge by solar, and set an over voltage alarm for 3.5V. I'll then monitor the voltages every 5 minutes. When I see the voltage of one cell climb to say 3.6 or 3.7, I'll disassemble the battery again and try to take some charge out of that cell. Rinse and repeat until the batteries are all kind of balanced at around 3.5V or so?

 

OR do I really need less of a mallet and more of a scalpel - should I try to construct a big resistor to use instead of my heating element?

If so can you maybe give me some pointers on what's required @Dr Bob preeeetty please?

[peterboat]

14 minutes ago, ivan&alice said:

Thank you this is super helpful.

So in the case of that cell at 3.9V, which was probably 95% SoC at that point, when I removed the charge (around lunchtime) it was in a bit of a "danger zone" there - although I did discharge all the cells right away so it should be OK now I think (BMV is showing 25Ah consumed out of the battery so Cell 3 should be no more than 90% charged now)?

 

Or do you think I should take a bunch of energy out of the whole battery to be safe?

 

Alright so plan of action - tonight I will disassemble my battery and heat up a bucket of water from just Cell 3 (I'll maybe run my 600W 12V heating element for 15mins or so). Tomorrow I will put the bank on charge by solar, and set an over voltage alarm for 3.5V. I'll then monitor the voltages every 5 minutes. When I see the voltage of one cell climb to say 3.6 or 3.7, I'll disassemble the battery again and try to take some charge out of that cell. Rinse and repeat until the batteries are all kind of balanced at around 3.5V or so?

 

OR do I really need less of a mallet and more of a scalpel - should I try to construct a big resistor to use instead of my heating element?

If so can you maybe give me some pointers on what's required @Dr Bob preeeetty please?

Just stick to 13.9 absorb and 13.4 on float see what happens 

[nicknorman]

13 minutes ago, ivan&alice said:

Thank you this is super helpful.

So in the case of that cell at 3.9V, which was probably 95% SoC at that point, when I removed the charge (around lunchtime) it was in a bit of a "danger zone" there - although I did discharge all the cells right away so it should be OK now I think (BMV is showing 25Ah consumed out of the battery so Cell 3 should be no more than 90% charged now)?

 

Or do you think I should take a bunch of energy out of the whole battery to be safe?

 

I would have thought that if the cell was at 3.9v it was probably above 95% but I guess it depends on the charge current at the time. There is no particular issue with having the batteries near 100% SoC in the short term, it’s just not good for them to be kept at 100% for days and weeks. So no, I can’t see any need to discharge the whole bank especially as you’ve already taken out 25Ah.
 

I think at this time of the year when nights (and hence energy required for lighting etc) are short and solar is plentiful there is no point in trying to cram the batteries full. Once you have top balanced, that is. If you are operating between say 100% and 50%, that is sub optimal in the long term and you would be better to operate in the 80% to 30% range. In winter when nights are long and sunny days are rare, it would make more sense to fully charge the batteries on an opportune basis (ie when It’s a sunny day, or you are cruising).

[Dr Bob]

2 hours ago, ivan&alice said:

 

OR do I really need less of a mallet and more of a scalpel - should I try to construct a big resistor to use instead of my heating element?

If so can you maybe give me some pointers on what's required @Dr Bob preeeetty please?

At 3.9V, that cell was 100% charged and you MUST stop charging. As Nick says, it is voltage under charge you are interested in at the top end not rested voltage.

 

I dont think your 12V heating element will work. You need to get something that will take 10A or so out at 3.3V. Use Ohms law to work out that resistance of the circuit needed. If you have an element that heats at 12V, then it isnt going to heat at 3.3V. Check it on ohms law.

My 'resistor array' is in storage for the next few weeks. We took all our lithium gear off our old boat when we sold it and all the kit is in storage. We are currently on our new boat on a delivery trip back to our 'base' (for the next 2 weeks) using LA's to make sure everything is working so we have a working 'guarantee' (new boat). ....so I cant remember what the resistor values are and cant show you a photo.

I have had a quick look at the laptop and think I was using a 0.3Ohm/60W resistor so 3* 1 Ohm 20W resistors in parallel, mounted on a heat sink - it gets hot.  It may have been 6 * 2Ohm 10W resistors in parallel.

I am sure Nick can put you straight on resistor values.

[nicknorman]

52 minutes ago, Dr Bob said:

At 3.9V, that cell was 100% charged and you MUST stop charging. As Nick says, it is voltage under charge you are interested in at the top end not rested voltage.

 

I dont think your 12V heating element will work. You need to get something that will take 10A or so out at 3.3V. Use Ohms law to work out that resistance of the circuit needed. If you have an element that heats at 12V, then it isnt going to heat at 3.3V. Check it on ohms law.

My 'resistor array' is in storage for the next few weeks. We took all our lithium gear off our old boat when we sold it and all the kit is in storage. We are currently on our new boat on a delivery trip back to our 'base' (for the next 2 weeks) using LA's to make sure everything is working so we have a working 'guarantee' (new boat). ....so I cant remember what the resistor values are and cant show you a photo.

I have had a quick look at the laptop and think I was using a 0.3Ohm/60W resistor so 3* 1 Ohm 20W resistors in parallel, mounted on a heat sink - it gets hot.  It may have been 6 * 2Ohm 10W resistors in parallel.

I am sure Nick can put you straight on resistor values.

Yes if you want 10A from 3.3v you need 0.33 ohms. And the power dissipation will be v x I, ie 33 watts.
Of course you probably want it to work at max charging voltage ie 3.7 volts so 0.33 ohms would give you about 11A and 41w

 

something like this would do : https://uk.rs-online.com/web/p/panel-mount-fixed-resistors/0160887/

but note that the 50w rating is only when the resistor is mounted on a large heatsink using thermal paste to make a good thermal connection. And at maximum power dissipation even with a heatsink it can reach 200C which is definitely ouch territory!

 

 

Edited July 8, 2020 by nicknorman

[Dr Bob]

14 minutes ago, nicknorman said:

Yes if you want 10A from 3.3v you need 0.33 ohms. And the power dissipation will be v x I, ie 33 watts.
Of course you probably want it to work at max charging voltage ie 3.7 volts so 0.33 ohms would give you about 11A and 41w

 

something like this would do : https://uk.rs-online.com/web/p/panel-mount-fixed-resistors/0160887/

but note that the 50w rating is only when the resistor is mounted on a large heatsink using thermal paste to make a good thermal connection. And at maximum power dissipation even with a heatsink it can reach 200C which is definitely ouch territory!

 

 

Thanks Nick for doing the sums!

At that price you could buy 4 and series parallel them so they don't get too hot. If you could get a circuit that would do 20A then that would make balancing far quicker but I couldn't find resistor values of Ohms and watts that worked. 

[jetzi]

1 hour ago, Dr Bob said:

At 3.9V, that cell was 100% charged and you MUST stop charging. As Nick says, it is voltage under charge you are interested in at the top end not rested voltage.

I certainly did, I had a mild panic. I was keeping an eye on it and it suddenly jumped up to 3.8 something. I turned off the solar input as soon as I saw it.

 

The problem is with unbalanced cells the MPPT can't know when to stop. Because even if one cell is at 3.9 if the others are only at 3.3 the average is only 3.5. So I think I need to balance them as a matter of urgency before I can rely on the MPPT shutting itself off. Actually even then I shouldn't be relying on the MPPT shutting itself off. I think I am going to need to go with Peter's idea and have the BMS cut the feed from the PV as a backup. It will just have to drive two relays when the alternator is set up. Or, perhaps I can wire it such that the solar also uses the LA dump load, so the MPPT is never without a battery connection. But I think switching from Li to LA might confuse it too...

 

21 minutes ago, nicknorman said:

if you want 10A from 3.3v you need 0.33 ohms. And the power dissipation will be v x I, ie 33 watts.
Of course you probably want it to work at max charging voltage ie 3.7 volts so 0.33 ohms would give you about 11A and 41w

Is there any particular reason I want 10A specifically?

 

I don't understand what the sum is that I'm trying to do with these resistor values - surely I just want any resistance and then keep checking up on the voltage?

 

Lastly when you talk about charging voltage - that implies that I will be doing this balancing with the battery fully assembled - I thought I'd need to isolate the cells - is it possible for me to just short out one cell with a resistor to bring the charge of just that cell down?

 

[nicknorman]

Well 20A would give you twice the power, obviously, ie 74w at 3.7v. The resistance would be 0.185 ohm. But to be simple about it, why not just use two of the resistors I mentioned in parallel? You’d need a big heatsink though!

 

Since 3.3v is a popular supply voltage for modern electronics, there are plenty of 3.3v cooling fans cheaply available so it might be a good idea to construct a force-cooled heat sink to mount the resistors on.

[nicknorman]

7 minutes ago, ivan&alice said:

 

The problem is with unbalanced cells the MPPT can't know when to stop. Because even if one cell is at 3.9 if the others are only at 3.3 the average is only 3.5. So I think I need to balance them as a matter of urgency before I can rely on the MPPT shutting itself off. Actually even then I shouldn't be relying on the MPPT shutting itself off. I think I am going to need to go with Peter's idea and have the BMS cut the feed from the PV as a backup. It will just have to drive two relays when the alternator is set up. Or, perhaps I can wire it such that the solar also uses the LA dump load, so the MPPT is never without a battery connection. But I think switching from Li to LA might confuse it too...

 

Is there any particular reason I want 10A specifically?

 

I don't understand what the sum is that I'm trying to do with these resistor values - surely I just want any resistance and then keep checking up on the voltage?

 

Lastly when you talk about charging voltage - that implies that I will be doing this balancing with the battery fully assembled - I thought I'd need to isolate the cells - is it possible for me to just short out one cell with a resistor to bring the charge of just that cell down?

 

Your first para - this is exactly the problem and why you need individual cell monitoring and the ability to stop charging or at least reduce the charging voltage to prevent any one cell from being over charged. The alternative, as promoted by Peterboat, is to leave a large margin and never charge the batteries near to 100%.

 

2nd/3rd para. No there is nothing magical about 10A it is just a convenient figure to make it easy to work out how many AH you have extracted. But if you have no idea what current you are taking out you risk either overheating the resistor or having to spend decades discharging the cell!
If you want to top balance the cells during charging then it would be advisable to be able to dump more current into the resistor than the charging device is providing, otherwise the cell voltage will continue to rise.

 

Final para. Provided the discharging device is isolated, as a resistor would be, you simply put it across the cell with the battery assembled. No need to disassemble. The cell discharges through the resistor without any current flowing through the other cells.

Of course this is passive balancing where the excess charge is wasted. If you want to do active balancing where the excess charge from one cell is used to charge other cells, that is a lot more complicated to achieve. And not necessary IMO.

[peterboat]

3 minutes ago, nicknorman said:

Your first para - this is exactly the problem and why you need individual cell monitoring and the ability to stop charging or at least reduce the charging voltage to prevent any one cell from being over charged. The alternative, as promoted by Peterboat, is to leave a large margin and never charge the batteries near to 100%.

 

2nd/3rd para. No there is nothing magical about 10A it is just a convenient figure to make it easy to work out how many AH you have extracted. But if you have no idea what current you are taking out you risk either overheating the resistor or having to spend decades discharging the cell!
If you want to top balance the cells during charging then it would be advisable to be able to dump more current into the resistor than the charging device is providing, otherwise the cell voltage will continue to rise.

 

Final para. Provided the discharging device is isolated, as a resistor would be, you simply put it across the cell with the battery assembled. No need to disassemble. The cell discharges through the resistor without any current flowing through the other cells.

Of course this is passive balancing where the excess charge is wasted. If you want to do active balancing where the excess charge from one cell is used to charge other cells, that is a lot more complicated to achieve. And not necessary IMO.

The 80% is promoted and used by the majority of EV makers it gives the batteries a longer life and keeps them balanced. I understand where you are coming from but for me its easier to build a bigger battery bank 

[jetzi]

Brilliant. Thank you.

 

9 minutes ago, nicknorman said:

this is exactly the problem and why you need individual cell monitoring and the ability to stop charging or at least reduce the charging voltage to prevent any one cell from being over charged

I can't go back through the last 50 pages now to double check, but why hasn't it come up? Surely this is the case with everyone charging via solar - there has to be something to turn off the MPPT other than the battery voltage, or is it just a case of making sure the batteries stay top balanced? It just seems at odds with the 3, 4 or 5 layers of protection to just leave the MPPT to its own devices on this.

 

9 minutes ago, nicknorman said:

there is nothing magical about 10A it is just a convenient figure to make it easy to work out how many AH you have extracted.

Sorry that I have missed this point. Why do you need to know how many Ah you have extracted? Surely the only goal is to get the voltages synchronised - so you discharge the highest cell until they balance (using any resistance you have), then charge the bank, discharge the highest cell, then charge the bank, etc, until all cells are as close to 100% as you are willing to go (say 3.8 V)?

9 minutes ago, nicknorman said:

If you want to top balance the cells during charging then it would be advisable to be able to dump more current into the resistor than the charging device is providing, otherwise the cell voltage will continue to rise.

Understood. So that means a seriously beefy resistor (or I could shade 3 of my 4 panels).

Edited July 8, 2020 by ivan&alice

[Richard10002]

FWIW: My understanding is that 3.65V is “full” / 100% SOC. This corresponds to a 4 cell voltage of 14.6V.

 

My Tracer MPPT has a control panel, (MT50 I think), which allows me to set the Absorption Voltage to whatever I like, (within limits). If I set it to 13.9V it won’t charge to higher than that voltage. Allow a bit for voltage drop and Bob’s your Uncle.

1 hour ago, ivan&alice said:

So that means a seriously beefy resistor (or I could shade 3 of my 4 panels).

I’m pretty sure that Will Prowse uses hefty resistors to deplete cells, and shows how he does this in one/some of his videos. Pretty sure it would be prior to October last year - maybe a long time before.

[Dr Bob]

8 hours ago, ivan&alice said:

 

 

I can't go back through the last 50 pages now to double check, but why hasn't it come up? Surely this is the case with everyone charging via solar - there has to be something to turn off the MPPT other than the battery voltage, or is it just a case of making sure the batteries stay top balanced? It just seems at odds with the 3, 4 or 5 layers of protection to just leave the MPPT to its own devices on this.

 

 

I isolated mine if the voltage got above 13.9V so that isolated the solar as well, but my solar turned itself down around 13.8V so never had a problem. You have to be pretty far off in cell balancing it 13.9V sends an individual cell to 3.9V. Do as much balancing as you can then just turn all your voltages down to 13.8V and run like that until you want another crack at balancing. This is what Peter an I do. Try 13.7V as the cut off - you will still be up near 80% - until you get a bit more experience.

 

Nick's spot on with the 10A stuff. If you have a circuit that takes 10A then if you record your data of cell voltages vs time, when you come to rebalance in 6 months time, you know exactly how long to attach the circuit. Nick is also right, you dont have to disconnect all the cells to deplete one. Just wire across the terminals of that cell.

[WotEver]

9 hours ago, ivan&alice said:

so you discharge the highest cell until they balance (using any resistance you have), then charge the bank, discharge the highest cell, then charge the bank, etc

No, you don’t need to stop charging while discharging the high cell. As Nick wrote, so long as the current through the resistor is higher than the charge source then you can pull that one cell down while the others continue to rise. When you’ve done this once then limiting future charges to 80% should allow the cells to stay pretty well balanced. 
 

If the charge current is higher than your discharge resistor gives then yes, you need to stop charging while pulling that cell down. 
 

 

Edited July 9, 2020 by WotEver
Typo

[nicknorman]

9 hours ago, ivan&alice said:

Brilliant. Thank you.

 

I can't go back through the last 50 pages now to double check, but why hasn't it come up? Surely this is the case with everyone charging via solar - there has to be something to turn off the MPPT other than the battery voltage, or is it just a case of making sure the batteries stay top balanced? It just seems at odds with the 3, 4 or 5 layers of protection to just leave the MPPT to its own devices on this.

 

Sorry that I have missed this point. Why do you need to know how many Ah you have extracted? Surely the only goal is to get the voltages synchronised - so you discharge the highest cell until they balance (using any resistance you have), then charge the bank, discharge the highest cell, then charge the bank, etc, until all cells are as close to 100% as you are willing to go (say 3.8 V)?

Understood. So that means a seriously beefy resistor (or I could shade 3 of my 4 panels).

2nd para - well, it’s been discussed extensively before but which thread, no idea! This is why I am developing my own BMS system that can control the Combi charging voltage and do automatic balancing, and why it will be linked to my alternator controller. The lower tech alternative is simply to get the cells balanced and then keep the charge voltage low enough to avoid one cell running up the knee (with some safety margin), which is what everyone else is doing.

 

3rd para yes it doesn’t matter that much, it is just to get an idea of how much charge you’ve taken out for future reference. Not critical at all, but then not hard to implement.

 

Ideally you want 3 resistor systems, you switch them onto the 3 highest cells and turn them off (probably at different times) when those cells reach the voltage of the lowest cell. You can of course do it with 1 resistor but it will take a lot longer and be more faffy as you will have to bring 3 cells down to the level of the lowest one, one at a time.

[nicknorman]

9 hours ago, peterboat said:

The 80% is promoted and used by the majority of EV makers it gives the batteries a longer life and keeps them balanced. I understand where you are coming from but for me its easier to build a bigger battery bank 

A bigger bank is one solution, obviously. But Li batteries are expensive and long lasting so it may not make sense to spend a lot of additional money on a large bank that is going to massively outlive the owner! For us as leisure users it seems unlikely that Li batteries bought now and used fairly hard would be at risk of approaching their useful lives before I become unable to boat any more (due to being dead, etc).

[peterboat]

43 minutes ago, nicknorman said:

A bigger bank is one solution, obviously. But Li batteries are expensive and long lasting so it may not make sense to spend a lot of additional money on a large bank that is going to massively outlive the owner! For us as leisure users it seems unlikely that Li batteries bought now and used fairly hard would be at risk of approaching their useful lives before I become unable to boat any more (due to being dead, etc).

I agree Nick, but nearly all batteries on boats are secondhand so unless its like my valence batteries you have no idea of how many cycles they have completed.  So it seems sensible to continue the makers recommend 80% to make sure they do outline us.i have in the recent past destroyed some LifePo4s cells it was deliberately done and didn't take much to achieve. I am building a 12v battery for my electric car and wanted to see if I could get away without a BMS the answer is no, so I will be buying a couple shortly for the job

[jetzi]

10 hours ago, Richard10002 said:

FWIW: My understanding is that 3.65V is “full” / 100% SOC. This corresponds to a 4 cell voltage of 14.6V.

Yeah, but only if the cells are perfectly balanced.

 

 

10 hours ago, Richard10002 said:

My Tracer MPPT has a control panel, (MT50 I think), which allows me to set the Absorption Voltage to whatever I like, (within limits). If I set it to 13.9V it won’t charge to higher than that voltage.

 

3 hours ago, Dr Bob said:

my solar turned itself down around 13.8V so never had a problem. You have to be pretty far off in cell balancing it 13.9V sends an individual cell to 3.9V. Do as much balancing as you can then just turn all your voltages down to 13.8V and run like that until you want another crack at balancing. This is what Peter an I do. Try 13.7V as the cut off

Alright thank you for this. I have changed the Absorption voltage to 13.7V for now (Float is still 13.5). 

 

This means if one cell is overcharged more than 3.7V, the others will have to be unbalanced such that they are less than < 3.33V. Since my cells are right now 3.332, 3.339, 3.347, 3.330 this can't be the case, so I can stop furtively checking my BG-8S every 60 seconds... (maybe just every 10mins now)

 

 

4 hours ago, Dr Bob said:

I isolated mine if the voltage got above 13.9V so that isolated the solar as well

Ah but you have a LA battery in parallel right, so you can safely disconnect the solar from the Lithium battery without risking damage to your MPPT (assuming that the MPPT is OK with having it's battery changed under load). 

 

3 hours ago, nicknorman said:

The lower tech alternative is simply to get the cells balanced and then keep the charge voltage low enough to avoid one cell running up the knee (with some safety margin), which is what everyone else is doing.

Fair enough, it's a detail I must have overlooked. I'm going to see if I can find a bistable relay that will support 100A / 250V and have my BMS shut down the feed from the PV at the same time. I still intend to only charge to 80% but I don't want to rely on my batteries being balanced - I feel like we're putting in so many layers of protection that leaving this as a weak link is not sensible. 

 

3 hours ago, nicknorman said:

Ideally you want 3 resistor systems, you switch them onto the 3 highest cells and turn them off (probably at different times) when those cells reach the voltage of the lowest cell. You can of course do it with 1 resistor but it will take a lot longer and be more faffy as you will have to bring 3 cells down to the level of the lowest one, one at a time.

Perhaps - ideally - I would have 4 resistor systems that I permanently mount over each of my cells, with switches that I can flip whenever I want to reduce the charge of one cell. 

 

Being that I'm continuously cruising getting deliveries isn't that easy, I generally have to order to ebay click and collect locations or amazon lockers - so I will see if I can find these from one of these merchants. 

[nicknorman]

I use RS a lot. They do have some trade counters where you can collect from, depends where you are.

[WotEver]

1 hour ago, nicknorman said:

I use RS a lot. They do have some trade counters where you can collect from, depends where you are.

Ditto but...

1 hour ago, ivan&alice said:

Perhaps - ideally - I would have 4 resistor systems that I permanently mount over each of my cells, with switches that I can flip whenever I want to reduce the charge of one cell. 

Ensure that the switches are rated for 10A DC or higher.

[nicknorman]

7 hours ago, peterboat said:

I agree Nick, but nearly all batteries on boats are secondhand so unless its like my valence batteries you have no idea of how many cycles they have completed.  So it seems sensible to continue the makers recommend 80% to make sure they do outline us.i have in the recent past destroyed some LifePo4s cells it was deliberately done and didn't take much to achieve. I am building a 12v battery for my electric car and wanted to see if I could get away without a BMS the answer is no, so I will be buying a couple shortly for the job

This brings up a point that I don’t think has really been discussed. Which source of LiFePO4 cells (without BMS) for a boater? By which I think I mean Jeremy (as per quite a few on here) or new, direct from China (as per Dre). I can see the cost of the new Chinese cells but have no idea what people have paid for Jeremy’s second hand cells.