Cheap LiFePO4 BMS?

[jetzi]

Isn't it worrying that the battery will sit above 90-95% for over 24 hours (because it is still charging to 100%?)

 

It seems like I need to check into a marina with electricity for a week. A weekend in a house isnt going to cut it.

 

Should I be looking to spend some big bucks on a high current 3.2V charger?

 

Using a single resistor to top balance while charging from solar was like trying to tune a radio with a sledge hammer. I just want to accurately get the cells top balanced and check the capacity once so I can start out with a clean slate...

12 minutes ago, WotEver said:

your psu appears to be a 15W device

It supplies 0-40V at max 5A. So theoretically it can put out 200W. But not at 3.4V (which it is currently showing at).

[nicknorman]

No I don’t think it is worrying that the batteries will be approaching fully charged for a day or two. That is a lot different from keeping the batteries at 100% for weeks, months or years. Yes there will perhaps be a minor loss of life but in the context of the long lifespan of Li batteries, insignificant. Better that than imbalanced batteries with one set of cells overcharging.

[WotEver]

2 minutes ago, ivan&alice said:

It supplies 0-40V at max 5A. So theoretically it can put out 200W.

Nope. It can supply 0V to 40V. It can supply variable current at up to 5A. It can only do so at a combination of those two parameters that results in no more than 15W. It quite clearly can’t supply 3.6V at 5A (18W) as your OP demonstrates

[Dr Bob]

I didnt bother putting them all in parallel. Just used the 3P 4S pack that Jeremy gave me and balanced by taking some out of the top cell and putting some into the bottom cell.

Just had them up to 100% this week and got to 3.50V on each cell +/- 20mV which is pretty good. Cell 1 was taking off so got to 3.60V 70mV ahead of the others. Well enough balanced for me. My cell 1 also hits the bottom knee first so obviously the 3 batteries making up this cell 1 have a lower capacity than the others.

Not a problem at all between 30% and 90%.

Edited August 15, 2020 by Dr Bob

[Dr Bob]

3 hours ago, nicknorman said:

Unfortunately there is still absolutely nothing wrong with the 450Ah of Trojans. I will have to find a need to repurpose them! I am certainly not going to do any sort of girly half and half system.

Nowt wrong with a hybrid system. Mine is working really well without major surgery to the existing system - which is necessary with warranties on a new boat.......or if you have an older one that you might sell.....or if you dont want to be without power for days while you set up the Li bank. Why not use all the capacity of the batteries you have?

[jetzi]

38 minutes ago, WotEver said:

Nope. It can supply 0V to 40V. It can supply variable current at up to 5A. It can only do so at a combination of those two parameters that results in no more than 15W. It quite clearly can’t supply 3.6V at 5A (18W) as your OP demonstrates

Ah. Thanks Wotever for this explanation. I didn't understand before why the system wasn't at 3.6V while under charge. I haven't seen a power rating on it. If I turn the current down to say 0.5A though, the voltage doesn't rise - in fact it dips down to 3.3V.

 

30 minutes ago, Dr Bob said:

Just had them up to 100% this week and got to 3.50V on each cell +/- 20mV which is pretty good.

 

Ah, so when you say you have taken them to 100% you mean "100% for my system" which is not 3.65V for every cell?

 

But yes I would be very happy with 3.52/3.48V on all cells. I wasn't able to achieve that level of precision charging the whole battery with solar and using a single resistor that I switched between the four cells. Perhaps I just need to be more patient.

 

I guess ideal thing to do would be to buy a more powerful lab / bench variable power supply.

Edited August 15, 2020 by ivan&alice

[Dr Bob]

31 minutes ago, ivan&alice said:

Ah. Thanks Wotever for this explanation. I didn't understand before why the system wasn't at 3.6V while under charge. I haven't seen a power rating on it. If I turn the current down to say 0.5A though, the voltage doesn't rise - in fact it dips down to 3.3V.

 

 

Ah, so when you say you have taken them to 100% you mean "100% for my system" which is not 3.65V for every cell?

 

But yes I would be very happy with 3.52/3.48V on all cells. I wasn't able to achieve that level of precision charging the whole battery with solar and using a single resistor that I switched between the four cells. Perhaps I just need to be more patient.

 

I guess ideal thing to do would be to buy a more powerful lab / bench variable power supply.

I don't know how you would do it with solar as the input varies too much. I took mine to 100% i.e. Until the tail current decayed to 4% capacity - on my old boat charging at 30A from a Victron charger. That was around 14.1V total so just over 3.5V per cell. It really doesn't matter if it's 95% or 100%.....as long as it's in the knee to do the balancing.

Go into a marina to fix it.

Edited August 15, 2020 by Dr Bob

[peterboat]

21 minutes ago, ivan&alice said:

Ah. Thanks Wotever for this explanation. I didn't understand before why the system wasn't at 3.6V while under charge. I haven't seen a power rating on it. If I turn the current down to say 0.5A though, the voltage doesn't rise - in fact it dips down to 3.3V.

 

 

Ah, so when you say you have taken them to 100% you mean "100% for my system" which is not 3.65V for every cell?

 

But yes I would be very happy with 3.52/3.48V on all cells. I wasn't able to achieve that level of precision charging the whole battery with solar and using a single resistor that I switched between the four cells. Perhaps I just need to be more patient.

 

I guess ideal thing to do would be to buy a more powerful lab / bench variable power supply.

Bob I a have bought battery balance boards I will let you know how they work in the future 

[Dr Bob]

17 minutes ago, peterboat said:

Bob I a have bought battery balance boards I will let you know how they work in the future 

I bought one for £40 that said it worked at 0.1A to 10A. Problem was that on the 3.5V cells, the max rate was less than 1A and I was finding I needed to move 10Ahrs + from the bad cells so it was going to take too long. Also, the voltage had to be in the knee so the bank needed to be on charge so impossible to keep charging for 12 hours plus in the knee. Once you get an idea of how many Ahrs to move its far easier with a set of resistors or 5A into one cell from a lab power supply. 

I wouldn't like to do it without a decent charger though.

Edited August 15, 2020 by Dr Bob

[peterboat]

11 minutes ago, Dr Bob said:

I bought one for £40 that said it worked at 0.1A to 10A. Problem was that on the 3.5V cells, the max rate was less than 1A and I was finding I needed to move 10Ahrs + from the bad cells so it was going to take too long. Also, the voltage had to be in the knee so the bank needed to be on charge so impossible to keep charging for 12 hours plus in the knee. Once you get an idea of how many Ahrs to move its far easier with a set of resistors or 5A into one cell from a lab power supply. 

I wouldn't like to do it without a decent charger though.

Impossible for me to do on the car setup James has been running a bank much larger than my 225 battery setup and it has remained balanced unaided. These are new quality lithium polymer jobbies, I have decided to put the boards in as belt and buckles arrangement because of the 400 amp draw possible on the little car

[nicknorman]

12 hours ago, Dr Bob said:

Nowt wrong with a hybrid system. Mine is working really well without major surgery to the existing system - which is necessary with warranties on a new boat.......or if you have an older one that you might sell.....or if you dont want to be without power for days while you set up the Li bank. Why not use all the capacity of the batteries you have?

I knew that would wind you up! But anyway we don’t have room for 2 large sets of batteries.

[Dr Bob]

56 minutes ago, nicknorman said:

I knew that would wind you up! But anyway we don’t have room for 2 large sets of batteries.

This hybrid system is quite interesting now! On our old boat with flooded LA's (cheapo's), the Li bank did all the work until the voltage got down to 12.7V (so I guess around 30% SoC on the Li bank). We only had a 90A alternator so we limited the current when charging go 50% so not to overheat - which worked well with the Sterling AtoB.

This new boat is different. Our alternator is 240A and is working fine with 170A out into the LA's (480Ahr of AGMs) when putting the engine on with the AGMs at 60% SoC (ish). What I am finding though is now I have fitteded the Sterling BtoBs I am getting some unexpected results.

I have a 5KVA victron quattro combi which can charge at 200A - so the wiring to the AGM bank is 2*90mm cable. Not a lot of voltage drop there then. The Li's have about 5M of 50mm cable to where it joins the boat circuits.

When both banks are near full ie over 13.2V, 99% of the power comes out of the Li bank but if I put a big load on (150A from the nesspresso) the power is split coming out of both, then Amps go from the Li to AGM to replace the surface charge once the draw is off.

Once my Li's are depleted down to 12.85V, the power is coming from both banks at the same time (50/50). I guess some of this is due to the much bigger wiring to the AGMs but also I think the rested voltage when full of the AGMs is 0.1V higher than the cheapo's on the old boat. This means that I am actually using some of the capacity of the AGMs if I go that low, but that is fine as recharging the AGMs from 90% SoC or even 80% is fine.

The really interesting thing though is the charging. With the BtoBs the alternator puts out 14.4V to the BtoBs so nicely tops up the AGMS (maybe 30A going in after 5 mins). The BtoB takes about 110A and gives 90A to the  Li's which is fine for what I want - and even in the hot week we have just had, I havent seen the alternator go over 67°C on the top of the casing. Add on another 5-10A that the boat is taking and we are likely up around 140A from the alternator - it is well cooled. It did actually go over 67C once (last Wednesday on the hottest day, following a boat for 45mins who insisted on going 0.5 miles per hour past long string of permanent moorings - and it got up to 80°C but soon cooled with a few revs).

I did try it direct to the alternator ie without the BtoBs and really interestingly, the alternator only put out 90A and in this mode again temp was below 67°C. I am wondering if the 'low' power is a function of the wiring and the AGMs being full or nearly full are restraining the alternator output. What this says though is that I can just operate with the alternator charging the Li's direct and not worry about overcharging (ie the auto disconnect will isolate at my target voltage - or votage/amps via the SoC calc) or alternator temperature.

At the moment though I am using the BtoBs for my normal crusing and tuning to find the right setting to get them to move to float excactly where I want.

The BtoBs have a input from the BMS and can be forced to turn off from the BMS - ie doing what my autodisconnect switch does - which is interesting for someone putting in a cheap system. I prefer to have my auto disconnect separate though - if the BtoB goes wrong and doesnt go into float then will it also fail to disconnect.

At the moment then, when cruising I am getting 90A charge into the Li's for a full 2 hours  or so depending on how much power we have used overnight so am quite happy with the way its going. Maybe I didnt need to buy the BtoBs as it is also fine direct from the alternator?

[nicknorman]

18 minutes ago, Dr Bob said:

This hybrid system is quite interesting now! On our old boat with flooded LA's (cheapo's), the Li bank did all the work until the voltage got down to 12.7V (so I guess around 30% SoC on the Li bank). We only had a 90A alternator so we limited the current when charging go 50% so not to overheat - which worked well with the Sterling AtoB.

This new boat is different. Our alternator is 240A and is working fine with 170A out into the LA's (480Ahr of AGMs) when putting the engine on with the AGMs at 60% SoC (ish). What I am finding though is now I have fitteded the Sterling BtoBs I am getting some unexpected results.

I have a 5KVA victron quattro combi which can charge at 200A - so the wiring to the AGM bank is 2*90mm cable. Not a lot of voltage drop there then. The Li's have about 5M of 50mm cable to where it joins the boat circuits.

When both banks are near full ie over 13.2V, 99% of the power comes out of the Li bank but if I put a big load on (150A from the nesspresso) the power is split coming out of both, then Amps go from the Li to AGM to replace the surface charge once the draw is off.

Once my Li's are depleted down to 12.85V, the power is coming from both banks at the same time (50/50). I guess some of this is due to the much bigger wiring to the AGMs but also I think the rested voltage when full of the AGMs is 0.1V higher than the cheapo's on the old boat. This means that I am actually using some of the capacity of the AGMs if I go that low, but that is fine as recharging the AGMs from 90% SoC or even 80% is fine.

The really interesting thing though is the charging. With the BtoBs the alternator puts out 14.4V to the BtoBs so nicely tops up the AGMS (maybe 30A going in after 5 mins). The BtoB takes about 110A and gives 90A to the  Li's which is fine for what I want - and even in the hot week we have just had, I havent seen the alternator go over 67°C on the top of the casing. Add on another 5-10A that the boat is taking and we are likely up around 140A from the alternator - it is well cooled. It did actually go over 67C once (last Wednesday on the hottest day, following a boat for 45mins who insisted on going 0.5 miles per hour past long string of permanent moorings - and it got up to 80°C but soon cooled with a few revs).

I did try it direct to the alternator ie without the BtoBs and really interestingly, the alternator only put out 90A and in this mode again temp was below 67°C. I am wondering if the 'low' power is a function of the wiring and the AGMs being full or nearly full are restraining the alternator output. What this says though is that I can just operate with the alternator charging the Li's direct and not worry about overcharging (ie the auto disconnect will isolate at my target voltage - or votage/amps via the SoC calc) or alternator temperature.

At the moment though I am using the BtoBs for my normal crusing and tuning to find the right setting to get them to move to float excactly where I want.

The BtoBs have a input from the BMS and can be forced to turn off from the BMS - ie doing what my autodisconnect switch does - which is interesting for someone putting in a cheap system. I prefer to have my auto disconnect separate though - if the BtoB goes wrong and doesnt go into float then will it also fail to disconnect.

At the moment then, when cruising I am getting 90A charge into the Li's for a full 2 hours  or so depending on how much power we have used overnight so am quite happy with the way its going. Maybe I didnt need to buy the BtoBs as it is also fine direct from the alternator?

Yes it sounds as though the 5m of 50mm wiring has sufficient resistance to limit the current somewhat, and also of course the alternator’s soft regulation will only allow it to produce max output with its voltage pulled down to perhaps 0.5v or more below its regulated voltage. It doesn’t take much wiring and switch/isolator resistance to drop a few 10ths of a volt and thus reduce alternator output current significantly.

 

All of which is fine for use during a day’s cruising, but not so good if you want to belt a couple of hundred Ah in when moored for a few days.

[Dre]

On 15/08/2020 at 22:34, ivan&alice said:

Ah. Thanks Wotever for this explanation. I didn't understand before why the system wasn't at 3.6V while under charge. I haven't seen a power rating on it. If I turn the current down to say 0.5A though, the voltage doesn't rise - in fact it dips down to 3.3V.

 

 

Ah, so when you say you have taken them to 100% you mean "100% for my system" which is not 3.65V for every cell?

 

But yes I would be very happy with 3.52/3.48V on all cells. I wasn't able to achieve that level of precision charging the whole battery with solar and using a single resistor that I switched between the four cells. Perhaps I just need to be more patient.

 

I guess ideal thing to do would be to buy a more powerful lab / bench variable power supply.

Hi, how did the balancing go? 

To add to what's been said... If the cells are still a long way to being charged, say below 3.3/3.2v you can turn the voltage on your power supply all the way up until you reach the max current that it can supply thus using the full capacity of your power supply and cutting down charging time. Your supply will be at the Constant Current stage and voltage is irrelevant (always monitor the voltage at the battery terminals). When you start reaching full capacity you will see that the voltage (at the battery) will rise more quickly and you now need to turn the supply down to 3.6v as it will soon be entering the Constant Voltage phase, where the current will start diminishing (think of filling bottles... You start at full flow and then slow the flow down to the top). You will soon reach 3.6v or just under at the terminals and the current supplied will go down to zero. 

 

It pays to reach 3.6v and then leave all cells in parallel for a few hours (or days) and then re top balance to 3.6v one last time. 

 

I initially didn't balance my cells properly as I didn't have a power supply and while they were all exactly the same voltage at up to 3.3v they started showing up to 0.3 to 0.5v difference at full charge. 

 

After top balancing they haven't gone over 0.01v difference..... Which is very satisfying

 

 

 

 

PS What BMS do you use? 

 

 

 

[jetzi]

Hi Dre, that does sound very satisfying! Wish I could say my cells were as balanced as that. 

 

Unfortunately as a CCer I don't generally have any access to mains power. I didn't have enough time at my relatives' house before I had to get the hire van back. I got the 16 cells in parallel up to 3.323V before giving up. The power supply I had was limited to only 5A at 3.3ish V - turning up the voltage made no difference so as @WotEver said it was already the most that the power supply could put out.

 

I wired up to solar and took a bit of charge out of two of the highest cells with my resistor. In a test, I took the battery to 14.2V before stopping - at this battery voltage, two of the cells were showing 3.6V and two were showing 3.5V. So they are not too far out of balance. I have set my absorption & float values at 13.8V on my MPPT so individual cells should never get anywhere near fully charged - at 13.8V they are all around 3.45V.

 

I'll keep an eye on them, though, until I can balance them properly. The only danger behind not balancing them is if they are so far out of balance, and the charge source is set to a voltage high enough, that a cell goes overvoltage before the charge source shuts down. Based on this test I could probably push it a bit higher than 13.8V but I really have no need to do so, with my MPPT set to 13.8V I'm going to make do with unbalanced cells for now. 

What is becoming pressing now as Autumn draws in is that I get my alternator charging set up. First step here is going to be to test it out while watching the current output and the temperature of the alternator. From there, I'll need to find a way to turn off the alternator without turning off the engine (interrupting the field current). Then, it's a case of automating that with an arduino. I'm planning on getting started with this project this weekend, so I'll let you all know how it goes soon.

 

Regarding the BMS, there are several components. The MPPT is the first line of defence to prevent overcharge. I have a ISDT BG-8S which I use to display the cell voltages and sound a low voltage alarm. I have a BMV-712 that sounds per-battery voltage alarms but doesn't drive any relays. Then I have a GWL cell performance monitor that drives 3 190A tyco relays on high (charging), low (load) and emergency (both - on negative) voltages per cell. My aim is to never have the high voltage relay trip, at least until the alternator and lead acid battery are connected, because that will potentially damage the MPPT. So I'm still being very conservative and keeping a close eye on everything. I'm far from satisfied with this setup - in retrospect I actually think it would be better to build the BMS from scratch with an arduino or similar. 

[peterboat]

1 minute ago, ivan&alice said:

Hi Dre, that does sound very satisfying! Wish I could say my cells were as balanced as that. 

 

Unfortunately as a CCer I don't generally have any access to mains power. I didn't have enough time at my relatives' house before I had to get the hire van back. I got the 16 cells in parallel up to 3.323V before giving up. The power supply I had was limited to only 5A at 3.3ish V - turning up the voltage made no difference so as @WotEver said it was already the most that the power supply could put out.

 

I wired up to solar and took a bit of charge out of two of the highest cells with my resistor. In a test, I took the battery to 14.2V before stopping - at this battery voltage, two of the cells were showing 3.6V and two were showing 3.5V. So they are not too far out of balance. I have set my absorption & float values at 13.8V on my MPPT so individual cells should never get anywhere near fully charged - at 13.8V they are all around 3.45V.

 

I'll keep an eye on them, though, until I can balance them properly. The only danger behind not balancing them is if they are so far out of balance, and the charge source is set to a voltage high enough, that a cell goes overvoltage before the charge source shuts down. Based on this test I could probably push it a bit higher than 13.8V but I really have no need to do so, with my MPPT set to 13.8V I'm going to make do with unbalanced cells for now. 

What is becoming pressing now as Autumn draws in is that I get my alternator charging set up. First step here is going to be to test it out while watching the current output and the temperature of the alternator. From there, I'll need to find a way to turn off the alternator without turning off the engine (interrupting the field current). Then, it's a case of automating that with an arduino. I'm planning on getting started with this project this weekend, so I'll let you all know how it goes soon.

 

Regarding the BMS, there are several components. The MPPT is the first line of defence to prevent overcharge. I have a ISDT BG-8S which I use to display the cell voltages and sound a low voltage alarm. I have a BMV-712 that sounds per-battery voltage alarms but doesn't drive any relays. Then I have a GWL cell performance monitor that drives 3 190A tyco relays on high (charging), low (load) and emergency (both - on negative) voltages per cell. My aim is to never have the high voltage relay trip, at least until the alternator and lead acid battery are connected, because that will potentially damage the MPPT. So I'm still being very conservative and keeping a close eye on everything. I'm far from satisfied with this setup - in retrospect I actually think it would be better to build the BMS from scratch with an arduino or similar. 

I have just finished putting my banks together, didn't balance cells just fitted battery balance boards next day all cells balanced! 

125 for 3 15s boards worth every penny 

[jetzi]

1 minute ago, peterboat said:

I have just finished putting my banks together, didn't balance cells just fitted battery balance boards next day all cells balanced! 

125 for 3 15s boards worth every penny 

But using a balance board, you'll only end up mid-balanced, right? My understanding is that this basically doesn't matter at all. At normal operating voltages all my cells are balanced to well within a hundredth of a volt.

 

We're talking about top balancing - getting all the cells to be near-100% SoC; 3.6V or even 3.65V - in order to maximise capacity while minimising chances of overcharge in normal use. Also in order to capacity test the cells, get a good baseline for your SoC monitoring, etc.

 

Without an external charging source, top balancing isn't physically possible unless some cells were over 100% charged to begin with. So I'm not sure what the point of fitting such a board would be.

Please correct me if I'm wrong?

[peterboat]

22 minutes ago, ivan&alice said:

But using a balance board, you'll only end up mid-balanced, right? My understanding is that this basically doesn't matter at all. At normal operating voltages all my cells are balanced to well within a hundredth of a volt.

 

We're talking about top balancing - getting all the cells to be near-100% SoC; 3.6V or even 3.65V - in order to maximise capacity while minimising chances of overcharge in normal use. Also in order to capacity test the cells, get a good baseline for your SoC monitoring, etc.

 

Without an external charging source, top balancing isn't physically possible unless some cells were over 100% charged to begin with. So I'm not sure what the point of fitting such a board would be.

Please correct me if I'm wrong?

You are wrong  it balance boards not by resistors but by active balance  my cells were out of the box I didn't check them just made them into 5P 15S  x 3 banks and connected boards checked a couple of the cells and they were different volts next morning all the same volts.  I will only change to 60 volts maximum is 63 so they will always balance over a period of time just like my valence batteries.  I don't want 100% as I know the damage it causes 80% is best and gives the longest battery life, all those EV makers are not wrong. 

[jetzi]

1 minute ago, peterboat said:

You are wrong  it balance boards not by resistors but by active balance  my cells were out of the box I didn't check them just made them into 5P 15S  x 3 banks and connected boards checked a couple of the cells and they were different volts next morning all the same volts.  I will only change to 60 volts maximum is 63 so they will always balance over a period of time just like my valence batteries.  I don't want 100% as I know the damage it causes 80% is best and gives the longest battery life, all those EV makers are not wrong. 

So yes - you are talking about mid balancing. The active balancer uses power from within the battery, not from an external charge source. What is the purpose of mid balancing your cells?

 

Top balancing does not imply charging to 100% as a general rule. It just means charging once to 100% so that you can know where 80% is (and then discharging them to prevent any damage from holding them at 100%). Top balancing does not mean breaking the 80/20 rule. 

 

If you have never taken the cells to 100% how do you know where 80% charged is?

[peterboat]

3 minutes ago, ivan&alice said:

So yes - you are talking about mid balancing. The active balancer uses power from within the battery, not from an external charge source. What is the purpose of mid balancing your cells?

 

Top balancing does not imply charging to 100% as a general rule. It just means charging once to 100% so that you can know where 80% is (and then discharging them to prevent any damage from holding them at 100%). Top balancing does not mean breaking the 80/20 rule. 

 

If you have never taken the cells to 100% how do you know where 80% charged is?

By working it out of course that and the internet.  Plenty on there about safe voltages,  I don't know what you are trying to do to be honest and I have been using LifePo4s for 3 years with no issues and no knackered batteries.

The 3 banks I have built for the car should work fine with the boards and the charger cutting out at 60 volts.  That's what James has done and his bank is staying balanced,  we never go to 100% as their is no need 

[Dr Bob]

1 hour ago, ivan&alice said:

Unfortunately as a CCer I don't generally have any access to mains power. I didn't have enough time at my relatives' house before I had to get the hire van back. I got the 16 cells in parallel up to 3.323V before giving up. The power supply I had was limited to only 5A at 3.3ish V - turning up the voltage made no difference so as @WotEver said it was already the most that the power supply could put out.

I was late to this comment but it is interesting that the Marine know how site suggests a different approach. Take a +ve and -ve from a charged up 12V lead acid and connect across the +ve?-ve of the 3.6V 'super' cell of the 12 cells all in parrallel and disconect when the super cell gets to 3.6V. Leave 24 hours and then repeat. This is the way you can balance if you are not on shore power but it is a right faff to dismantle the bank to get them all in parallel. I didnt bother.

 

1 hour ago, ivan&alice said:

wired up to solar and took a bit of charge out of two of the highest cells with my resistor. In a test, I took the battery to 14.2V before stopping - at this battery voltage, two of the cells were showing 3.6V and two were showing 3.5V. So they are not too far out of balance. I have set my absorption & float values at 13.8V on my MPPT so individual cells should never get anywhere near fully charged - at 13.8V they are all around 3.45V.

OK, well done. They are balanced enough. No need to do any more other than check it in 6 months time. Go back to charging to 80%.

 

1 hour ago, ivan&alice said:

What is becoming pressing now as Autumn draws in is that I get my alternator charging set up. First step here is going to be to test it out while watching the current output and the temperature of the alternator. From there, I'll need to find a way to turn off the alternator without turning off the engine (interrupting the field current). Then, it's a case of automating that with an arduino. I'm planning on getting started with this project this weekend, so I'll let you all know how it goes soon.

Why not just put a 110Ahr Lead acid in parallel to accomodate the voltage surge on disconnection or buy one of the Sterling devices that allows a disconnection.

 

 

2 hours ago, ivan&alice said:

 I'm far from satisfied with this setup - in retrospect I actually think it would be better to build the BMS from scratch with an arduino or similar. 

Why not use the relay on your BMV to terminate at a voltage just in case the alternator goes too far. You may need a circuit to activate the tyco latching relay. My system seems pretty idiot proof with a BMV relay and a cell monitor cut off.

 

1 hour ago, ivan&alice said:

Without an external charging source, top balancing isn't physically possible unless some cells were over 100% charged to begin with. So I'm not sure what the point of fitting such a board would be.

You need a charging source to top balance. You then need to be in the voltage knee to balance which means not far of fully charged. Perhaps the clever balancers are working out how much to take out of cell 1 to put into cell 2 and then doing this once the charge has stopped and the voltage gone back to normal but what is sure that if a balance circuit is only transferring low currents (ie 1A ) then you would have to be charging in the knee for days if badly out of balance.

 

Ivan, I think you are now fine if you have a spread of 3.5 - 3.6V. No need to go over 3.45V for another 6 months.

 

Before you embark on any project to control your alternator, try it first. It might work without any intervention. I dont know how much you run your engine but you may find over winter if not moving, you just run your engine and hour in the morning and an hour in the afternoon to put back in 180Ahrs (or so) so the alternator will not overheat or overcharge ie that is exactly what I am finding with our new boat.

[jetzi]

59 minutes ago, Dr Bob said:

I was late to this comment but it is interesting that the Marine know how site suggests a different approach. Take a +ve and -ve from a charged up 12V lead acid and connect across the +ve?-ve of the 3.6V 'super' cell of the 12 cells all in parrallel and disconect when the super cell gets to 3.6V. Leave 24 hours and then repeat. This is the way you can balance if you are not on shore power but it is a right faff to dismantle the bank to get them all in parallel. I didnt bother.

This is clever!

 

But potentially still dangerous since the lead acid could take the cells above 3.6V? So would need to monitor, and it becomes a little bit not so easy again (the nice thing about the lab power supplies is that they stop charging when they reach the set voltate).

 

Or the opposite - and this ruins the LA? Since the LiFePO bank is 640Ah @ 12V, won't the 110Ah LA deplete first? "so what" about the naff old LA, but you'd still be unsuccessful in the balance...

 

Sometimes these circuits make my brain hurt a little. If I were to connect a 12V battery in parallel with a 3.2V LiFePO4 cell, then the whole circuit would dip down to 3.2V right. So you'd need to disconnect the LA to check what voltage it is at. 
 

Either way this is a clever way to do this without shore power. Thanks for the tip.

 

1 hour ago, Dr Bob said:

OK, well done. They are balanced enough. No need to do any more other than check it in 6 months time. Go back to charging to 80%.

Sure, but I demand the satisfaction that @Dre experiences when his cells are only 0.3-0.5V different at the top and only 0.01V in the mids!!!

And well, I don't know where 80% is exactly, because I can't set my BMV to 100%, because I haven't taken the battery to 100%. But, I can stick with charging to 13.8V per the battery / 3.45V per each cell. That's what I'm going to do for now. 

 

 

 

1 hour ago, Dr Bob said:

Why not just put a 110Ahr Lead acid in parallel to accomodate the voltage surge on disconnection or buy one of the Sterling devices that allows a disconnection.

The sterling devices we covered a while back in this thread, I was really keen on the idea but a) my alternator is a bit piddly and b) it's rather inefficient given the fact that I'd be using a much lower current than the alternator is even capable of.

 

Moreover, there is a second part to alternator charging other than preventing overcharge (and safely disconnecting when reaching target voltage), which is overheating. I'm actually more worried about that than the overcharge, because like you say I have my old 110Ah LA for disconnect protection (even if it has only got around 40% of its original capacity). We as a group have all well established though that ideally a) the charge source controls itself and shuts itself down, and b) you need more than one layer of protection, and I don't want to be routinely relying on my cell monitor to do that. 

 

I'm not using the relay from my BMV though. So perhaps that is the right option. AFAIK the BMV won't control a bistable relay, so I'd not be able to use the Tycos here. I think rather than try to build a circuit to power a bistable relay from a NC/NO signal, I'd just use a NC relay and have the BMV open it if the voltage exceeded X.

 

1 hour ago, Dr Bob said:

Before you embark on any project to control your alternator, try it first. It might work without any intervention. I dont know how much you run your engine but you may find over winter if not moving, you just run your engine and hour in the morning and an hour in the afternoon to put back in 180Ahrs (or so) so the alternator will not overheat or overcharge ie that is exactly what I am finding with our new boat.

I will try it and see. It will definitely overcharge if I don't have a disconnect because it's regulated to 14.5ish volts, but I do have the cell performance monitor board which will disconnect me if necessary. I don't know whether overheat will be a problem though, so yes I will try it out and keep an eye with my laser gun. If it doesn't overheat, then yeah, maybe a relay on my BMV is all I need.

 

1 hour ago, Dr Bob said:

You need a charging source to top balance. You then need to be in the voltage knee to balance which means not far of fully charged. Perhaps the clever balancers are working out how much to take out of cell 1 to put into cell 2 and then doing this once the charge has stopped and the voltage gone back to normal but what is sure that if a balance circuit is only transferring low currents (ie 1A ) then you would have to be charging in the knee for days if badly out of balance.

From my experience so far, I don't believe that any balancer could be clever enough to do any post-charge-processing because I find that voltage in the mids is not a very good predictor of which cells are going to hit the knee first. It could just be that my secondhand cells vary a lot in capacity.

 

But yeah, this makes sense. I was just wondering what @peterboat was trying to achieve with his active balancer keeping the mids balanced. As far as I know having the mids balanced is basically pointless.

[Dre]

Hi Ivan&Alice

 

That's great stuff

 

I have left a lead acid connected to the alternator to which I connected my sterling b2b to charge the Lifepo4 when needed. The sterling can autosense and start charging automatically but I have used the manual function via a little switch so I start it when I want. My model charges at about 50 amps with the option to go half power at the press of a button. It works just fine. 

For top balancing I used a cheap (but pretty nifty) power supply that works with anything up to 32V DC. Like you I got the bank charged with solar and stopped when the first cell got up to about 3.5v (that's when the out of balance increased quite dramatically). Then connected the power supply to the lead acid and top balanced the whole bank. I'm also full time liveaboard so no mains either!

I also looked at the various options for battery protection/bms but went for the simplest. All my DC goes through a 120A bms that also has low temperature protect. The bms has a little Bluetooth dongle that allows me to monitor all cells and check and change all the parameters I could possibly need. It also has a good coulomb meter. But I've actually disconnected the Bluetooth now that I'm happy the setup is working as I want a fit and forget system and I didn't want to keep looking at the app! 

Luckily I got a 2kw victron inverter super cheap on ebay that works a treat and it has remote relay inputs. So the inverter is connected directly to the battery as I didn't want all those amps going through the bms and programmed the hall effect battery monitor's relay to turn off the inverter if ever the battery voltage dropped too low. I think the Bmv also has a relay for this function. To be honest that's just for failsafe as I have never gone low enough to worry. 

I've just installed a 1kw element in my calorifier and been getting free hot water (draws about 80amps) while we're getting a lot of solar. 

So my system is pretty simple. Top balancing and setting up the mppt was the main concern as with solar the biggest risk is overcharging. A good top balance allowed me to set the mppt to 14.2 and still be perfectly balanced. Undervoltage failsafe is there but as continuous cruisers we're always aware of where our batteries are at so not much of an issue. As for alternator charging the Sterling is plenty good. It probably doesn't have the best profile for Lifepo4s but unlike solar I wouldn't be leaving it charging all day so it probably will never reach full charge anyway. Infact I hope to use it the least possible even in winter. 

I'm about to order it all again for my friend's boat and at less than a grand for the whole of the above system (not including the solar and inverter of course) is crazy good value considering the capacity (300a) and the performance if lithium! 

I keep reading the inefficiencies boaters have to endure with lead acids and wonder WHY!? ?

[peterboat]

2 hours ago, Dr Bob said:

I was late to this comment but it is interesting that the Marine know how site suggests a different approach. Take a +ve and -ve from a charged up 12V lead acid and connect across the +ve?-ve of the 3.6V 'super' cell of the 12 cells all in parrallel and disconect when the super cell gets to 3.6V. Leave 24 hours and then repeat. This is the way you can balance if you are not on shore power but it is a right faff to dismantle the bank to get them all in parallel. I didnt bother.

 

OK, well done. They are balanced enough. No need to do any more other than check it in 6 months time. Go back to charging to 80%.

 

Why not just put a 110Ahr Lead acid in parallel to accomodate the voltage surge on disconnection or buy one of the Sterling devices that allows a disconnection.

 

 

Why not use the relay on your BMV to terminate at a voltage just in case the alternator goes too far. You may need a circuit to activate the tyco latching relay. My system seems pretty idiot proof with a BMV relay and a cell monitor cut off.

 

You need a charging source to top balance. You then need to be in the voltage knee to balance which means not far of fully charged. Perhaps the clever balancers are working out how much to take out of cell 1 to put into cell 2 and then doing this once the charge has stopped and the voltage gone back to normal but what is sure that if a balance circuit is only transferring low currents (ie 1A ) then you would have to be charging in the knee for days if badly out of balance.

 

Ivan, I think you are now fine if you have a spread of 3.5 - 3.6V. No need to go over 3.45V for another 6 months.

 

Before you embark on any project to control your alternator, try it first. It might work without any intervention. I dont know how much you run your engine but you may find over winter if not moving, you just run your engine and hour in the morning and an hour in the afternoon to put back in 180Ahrs (or so) so the alternator will not overheat or overcharge ie that is exactly what I am finding with our new boat.

Bob the blurb with my boards say they balance at 5 amps so I am happy to continue with them rather than a BMS. I am not concerned about cut offs because the charger cuts off before 60 volts anyway.

After contacting the maker they say the boards power themselves and constantly balance the cells which for me is exactly what I want

[peterboat]

1 hour ago, ivan&alice said:

 

 

 

 

 

I

 

But yeah, this makes sense. I was just wondering what @peterboat was trying to achieve with his active balancer keeping the mids balanced. As far as I know having the mids balanced is basically pointless.

Its not pointless Ivan I can 400 amps from the batteries in the car plus it has regen so their is no guarantee that cells will stay balanced even in the mid range, so something that balances the cells constantly is the best thing possible for me. It can move 5 amps around which is more than enough for new quality matched cells like mine, and for 125 pound for the 3 banks why not its peanuts in comparison to the price of the batteries

Edited August 27, 2020 by peterboat