MPPT query

[nicknorman]

49 minutes ago, Tony1 said:

 

Thanks, yes I'm planning to connect the laptop next time I do a nearly-full charge.

They've been charging a bit today but its only 15-20 amps from the solar, so not really enough to push their voltage up much at the moment. 

 

The batteries are connected up in a chain using their AMP type connectors, and I did expect that plugging the laptop cable to the last battery would allow to me to read them all, but it didnt work out. I had to connect the laptop cable to each battery in turn.

It'll be something I'm doing wrong no doubt, but tbh just being able to monitor at cell level is a huge improvement, so I'm more than happy with that at the moment. 

 

You need to use the module ID drop down list near top right to select the battery you want to look at.

[Tony1]

19 minutes ago, TheBiscuits said:

 

This video might help you, he's a non-techie bloke doing exactly what you need.

 

Do you have a terminator on the other end of the daisychain?

 

Thanks again, I'll check that out. 

I don't have a terminator on the far end from the laptop connector, but I do have some unused connectors so I could maybe try making one.

As I said, just getting the software working and being able to view the cells feel like a big step forward so I'm over the moon with that anyway. 

 

[TheBiscuits]

2 minutes ago, Tony1 said:

I don't have a terminator on the far end from the laptop connector, but I do have some unused connectors so I could maybe try making one.

 

It's just a 120 Ohm resistor between the two data cables - probably pins 3 & 4 on the plug.

 

Don't put the resistor on the +/- 5V power though ...

[Tony1]

 

Just a quick update- I had a bit of a charge to see how the cell voltages changed on approaching the knee. and so far they are staying very similar.

 

After the charge started the battery below developed a voltage spread of about 10mv, with cell 1 lagging behind the other three, but as the SoC started to go up, the voltages formed a tighter grouping, as can be seen. 

 

The balance status button is saying they are not balancing, so I'll carry on going past 14.4v and see if they start to do any balancing then. 

 

In order to hold them above 14.4v you have to put a fair few charging amps in, and I don't want to break them by overcharging, but I'd like them to spend maybe 30 mins above that level and see if any balancing goes on. 

 

Well this is fun, I must say.

As long as nothing explodes.

 

 

 

 

[TheBiscuits]

2 minutes ago, Tony1 said:

Well this is fun, I must say.

As long as nothing explodes.

 

Give us a screenshot of the "Batt System Info" tab please.

[Tony1]

 

OMG, the balancing button is saying its active.

Praise be to the Lord

 

Who knew batteries could be this exciting? 

 

3 minutes ago, TheBiscuits said:

 

Give us a screenshot of the "Batt System Info" tab please.

 

Here we go- you can see cell 1 is rather astray

 

[Rob-M]

[TheBiscuits]

6 minutes ago, Tony1 said:

Here we go

 

Now press the "Scan" button and do it again ...

[Tony1]

2 minutes ago, TheBiscuits said:

 

Now press the "Scan" button and do it again ...

 

Here, we are, I've renamed them to 1 2 and 3:

 

 

 

[TheBiscuits]

1 minute ago, Tony1 said:

Here, we are, I've renamed them to 1 2 and 3:

 

Excellent, it can see and talk to them all without needing a terminator, unless you've already made and fitted one.

 

Are you getting active balancing on all three modules?  Just switch between them using the communications module selection.

 

 

 

Also, ALT+ PrtSc should give a screenshot of just the active window which would be easier to read than zooming the whole desktop.

[Tony1]

1 minute ago, TheBiscuits said:

 

Excellent, it can see and talk to them all without needing a terminator, unless you've already made and fitted one.

 

Are you getting active balancing on all three modules?  Just switch between them using the communications module selection.

 

 

 

Yes, it read them all without the terminator thankfully, but it has shown they need more balancing. 

Battery 3 is the only one that is not active balancing, but I think it will start to top balance if I leave it for a longer period, right?

But then I dont like leaving them at high voltages/high SoC for long periods...

 

Battery 2 has the widest voltage spread between cells, at 138mv - in fact the spread is getting wider as the voltage goes up, which is not what I was hoping for.

 

[TheBiscuits]

1 minute ago, Tony1 said:

Yes, it read them all without the terminator thankfully, but it has shown they need more balancing. 

 

That's expected - you haven't been balancing them.

 

2 minutes ago, Tony1 said:

Battery 3 is the only one that is not active balancing, but I think it will start to top balance if I leave it for a longer period, right?

But then I dont like leaving them at high voltages/high SoC for long periods...

 

Yes, but they will only start balancing above a set voltage.  Probably around 3.5 volts per cell ish.  They will then take many hours to balance, so you need to get them high enough to start and then leave the laptop connected so the balancer stays active.

 

If you are starting to panic, stop charging and see how long the balancing continues with no charging taking place but with the SoC at or above the balancing threshold.

 

5 minutes ago, Tony1 said:

Battery 2 has the widest voltage spread between cells, at 138mv - in fact the spread is getting wider as the voltage goes up, which is not what I was hoping for.

 

Again, that's to be expected on non-balanced batteries.  Nick keeps trying to tell you that in the midrange where you tend to keep your SoC they will always look balanced - it's the mistake Peter keeps making - but any cell imbalance will only show up at very high and/or very low cell voltages.

 

Get them all high enough to balance, keep a beady eye on all of them and let the balancers do their thing.  Keep the laptop connected but be ready to disconnect the charging if they start climbing too high on one cell.

 

If possible, make a long enough lead that you can keep the laptop plugged into the batteries semi-permanently, as this will activate the cell balancing every time you charge the batteries sufficiently.  It will take many hours - probably days - to balance the cells using their internal controls if they are a long way out.

 

Don't panic too much about the knee, but don't run off the top of it - keep watching the voltages on every cell.

 

[Tony1]

 

I'm not sure if the active balancing function is working very well. 

Battery 3, which isnt using it, has a much smaller spread of cell voltages than the others, which actually seem to have increased their voltage spread since the active balancing indicator came on. 

There might be a case for not connecting the laptop when high charging/balancing, if that starts the active balancing that doesnt actually balance

 

 

 

[TheBiscuits]

3 minutes ago, Tony1 said:

I'm not sure if the active balancing function is working very well. 

Battery 3, which isnt using it, has a much smaller spread of cell voltages than the others, which actually seem to have increased their voltage spread since the active balancing indicator came on.

 

Is battery 3 the one at the far end of the daisychain?  If so it might need the terminator after all.

 

Again, if you are fretting, stop charging and just watch for a while.  You have trained yourself to keep the voltages low enough that they can't balance because you couldn't see the cell voltages - fair enough - but now you can see them, you have become a human BMS ...

[Tony1]

5 minutes ago, TheBiscuits said:

 

That's expected - you haven't been balancing them.

 

 

Yes, but they will only start balancing above a set voltage.  Probably around 3.5 volts per cell ish.  They will then take many hours to balance, so you need to get them high enough to start and then leave the laptop connected so the balancer stays active.

 

If you are starting to panic, stop charging and see how long the balancing continues with no charging taking place but with the SoC at or above the balancing threshold.

 

 

Again, that's to be expected on non-balanced batteries.  Nick keeps trying to tell you that in the midrange where you tend to keep your SoC they will always look balanced - it's the mistake Peter keeps making - but any cell imbalance will only show up at very high and/or very low cell voltages.

 

Get them all high enough to balance, keep a beady eye on all of them and let the balancers do their thing.  Keep the laptop connected but be ready to disconnect the charging if they start climbing too high on one cell.

 

If possible, make a long enough lead that you can keep the laptop plugged into the batteries semi-permanently, as this will activate the cell balancing every time you charge the batteries sufficiently.  It will take many hours - probably days - to balance the cells using their internal controls if they are a long way out.

 

Don't panic too much about the knee, but don't run off the top of it - keep watching the voltages on every cell.

 

 

Thanks very much, I'm going to have to do a proper balance when we've got a full day of sun, I don't want to be running the engine for that long. 

 

I think its the voltage that triggers them to balance, so when I turn off the engine I think the voltage will drop below the threshold, and the balancing will stop, but there's only one way to find out... 

 

 

 

 

[Tony1]

9 minutes ago, TheBiscuits said:

 

Is battery 3 the one at the far end of the daisychain?  If so it might need the terminator after all.

 

Again, if you are fretting, stop charging and just watch for a while.  You have trained yourself to keep the voltages low enough that they can't balance because you couldn't see the cell voltages - fair enough - but now you can see them, you have become a human BMS ...

 

Oddly enough, battery 3 is the one plugged into the cable. 

On the plus side, the spread of cell voltages is starting to reduce  a bit, and the higher ones are staying just under 3.7v, so it looks like the balancing will get there in the end, but it'll have to be done with solar I think, I dont have time for it today.

 

I dont mind being a human BMS, as long as the BMV712 is making the brews. 

As long as it doesnt take the beer....

 

[TheBiscuits]

2 minutes ago, Tony1 said:

the spread of cell voltages is starting to reduce  a bit, and the higher ones are staying just under 3.7v, so it looks like the balancing will get there in the end, but it'll have to be done with solar I think, I dont have time for it today.

 

That's exactly how active top balancing works.  It stops charging the high cell(s) and allows the lower cell(s) to catch up.  The circuit may also be able to do passive top balancing, and steal some charge off the higher cell(s) and feed it to the lower cell(s) when not getting an external charge.

 

Stop charging and leave the laptop connected.  If they aren't charging they won't go too high - nobble the solar as well of course - and you can then go about your day leaving the batteries to sort themselves out.

 

Give us an update in a few hours when you know if they'll carry on balancing  when not being charged.

[Tony1]

 

Just a quick follow up, after switching the engine off the solar also pretty much stopped, so the voltage dropped and the active balancing stopped.

But that said, as the voltage comes below 13.6v, the spread narrows a lot. 

 

Battery 2 now looks ok at this lower voltage, in the way you explained.  

 

Thats enough battery excitement for one day, I'm off the shops to get a beer so I can calm down 😂

 

[nicknorman]

The reason why the cell spread seemed to increase as the balancing progressed was because the cells were still charging. Entirely to be expected. Because of the non-linear nature of the "knee", for a given amount of Ah put into each cell, there will be a much bigger increase in voltage for the higher cells than for the lower one. Ideally if you adjusted the charging voltage so that the balancing was happening but virtually no additional current was going into the batteries, this would balance them without risking taking any one cell too high.

 

It does seem a somewhat non-ideal way to do balancing. I stole MP's scheme to measure the differences between the lowest cell, and the other 3 ones as the top cell hits 3.6v. The charging can then stop, meanwhile the balancing resistors are switched on for an amount of time depending on the difference between each cell and the lowest cell. Of course it requires a bit of guesswork to decide the relationship between cell voltage difference vs amount of Ah to take out, but an underestimate ensures that it gets there eventually after several charges to max cell voltage of 3.6 over weeks or months, without needing to hold the voltage high.

[Tony1]

1 hour ago, nicknorman said:

 

 ....if you adjusted the charging voltage so that the balancing was happening but virtually no additional current was going into the batteries, this would balance them without risking taking any one cell too high....

      

          --------------------------------------------------------------------------------------------------------------------------------------------------------

 

....the charging can then stop, meanwhile the balancing resistors are switched on for an amount of time depending on the difference between each cell and the lowest cell...... 

 

 

With all the B2Bs I have, I can certainly adjust the charging current down to about 10 amps if needed, but the snag with using them is the need to run the engine for the large number of hours it would take to fully balance the cells. So I want to avoid using the engine if possible.

The solar is a decent bet, except in changeable weather its very unpredictable. On a day like today, it can start overcast and giving 10-15 amps, then the sun escapes the clouds for half an hour and it jumps to 70 amps. But the solar looks the best bet- its just about choosing a day when its reasonably steady, and I can use either one or  both sets of panels to get the needed charging current. 

I suspect once they get up to 98% SoC or more, it only takes maybe 10-20 amps to keep the voltage above 14.4v so that the balancing stays activated the whole time. 

I can then hold them there for most of the day if needed, obviously keeping an eye on temp, and checking that the voltage of the higher cells doesnt go above say 3.7v. 

Having seen videos of lithium battery fires, I dont like the idea of them sitting there 100% full and over 14.4v for a whole day, but it looks like if I want to balance them properly I'll have to do that, albeit only once every few years.

 

On the other point, I dont think I have balancing resistors of the kind you mentioned, and if my charging stops, so does the balancing. 

 

My batteries do go up to high voltages like 14.4 and beyond every weeks, so the balancing is something I need to sort out. 

 

Edited March 10, 2022 by Tony1

[nicknorman]

28 minutes ago, Tony1 said:

 

With all the B2Bs I have, I can certainly adjust the charging current down to about 10 amps if needed, but the snag with using them is the need to run the engine for the large number of hours it would take to fully balance the cells. So I want to avoid using the engine if possible.

The solar is a decent bet, except in changeable weather its very unpredictable. On a day like today, it can start overcast and giving 10-15 amps, then the sun escapes the clouds for half an hour and it jumps to 70 amps. But the solar looks the best bet- its just about choosing a day when its reasonably steady, and I can use either one or  both sets of panels to get the needed charging current. 

I suspect once they get up to 98% SoC or more, it only takes maybe 10-20 amps to keep the voltage above 14.4v so that the balancing stays activated the whole time. 

I can then hold them there for most of the day if needed, obviously keeping an eye on temp, and checking that the voltage of the higher cells doesnt go above say 3.7v. 

Having seen videos of lithium battery fires, I dont like the idea of them sitting there 100% full and over 14.4v for a whole day, but it looks like if I want to balance them properly I'll have to do that, albeit only once every few years.

 

On the other point, I dont think I have balancing resistors of the kind you mentioned, and if my charging stops, so does the balancing. 

 

My batteries do go up to high voltages like 14.4 and beyond every weeks, so the balancing is something I need to sort out. 

 

I would try to keep the cell voltage no higher than 3.65v, which is 14.6v IF all the cells are balanced. This should be done by controlling the charge voltage, not the current. I’d start off by setting the charge voltage to say 14.4, then maybe notching up 0.1v at a time as the cells balance in order to keep the balancing function operating. In reality you only have to do this once in a blue moon - once balanced, my experience is that cells remain in balance for a long time.

[Tony1]

27 minutes ago, nicknorman said:

I would try to keep the cell voltage no higher than 3.65v, which is 14.6v IF all the cells are balanced. This should be done by controlling the charge voltage, not the current. I’d start off by setting the charge voltage to say 14.4, then maybe notching up 0.1v at a time as the cells balance in order to keep the balancing function operating. In reality you only have to do this once in a blue moon - once balanced, my experience is that cells remain in balance for a long time.

 

Great, thanks Nick. 

I think I'm going to struggle to keep the cell voltage below 3.65, but I'll give it a try. 

One of the problems is that if I'm only getting say 10 amps of solar charge, then it doesnt matter if I set the charging voltage to 14.4 on the MPPTs.

With only 10 amps of charge coming in, the charging voltage wont go above 13.5v, until well above 90% SoC is reached. 

 

Once I'm into the final 5% or so of the charge, it seems like you can keep the voltage over 14.4 even with a small charge like 20 amps, so thats where I'll try to fine tune the charging voltage on the MPPTs, so that it stays above 14.4 but below 14.6, and thus it keeps the balancing active.

 

It feels a bit like walking a tightrope, to keep the voltage high but avoid overcharging the batteries (or overheating them)- but at least its only once in a blue moon as you say.

Another complication is that if I set the MPPT charging voltage at exactly 14.4v, they'll go into float when they get to that voltage. So it will need to be tweaked as it increases.

In fact, the truth is that both MPPTs display higher battery voltages than the BMVs do (and the valence software as well). I dont know why they do that, but it means I have to set their charging voltages about 0.3v above what I actually want them to be, as a compensation.

This afternoon for example, the MPPTs were giving the battery voltage as 15v, when the BMVs and the valence software were showing 14.4v. It really doesnt help matters, and at some point I'll have to try and find out why they arent more accurate.

 

Anyway, I've got your point here, which is to notch up the charging voltage bit by bit, allowing the lagging cells to catch up.

Judging by this afternoon, I think it might be worth trying this notching approach earlier in the process too. So instead of setting a charge voltage of 14.4 right out of the gate, maybe I should set it at 13.8, and go up in increments from there, hopefully after the lagging cells have picked up on the others. 

 

But if this catching up process (which sounds like balancing, really) is only ever going to happen over 14.4v then I might as well just blast them up to 14.4 asap, and then do the micro management. 

 

 

[nicknorman]

23 minutes ago, Tony1 said:

 

Great, thanks Nick. 

I think I'm going to struggle to keep the cell voltage below 3.65, but I'll give it a try. 

One of the problems is that if I'm only getting say 10 amps of solar charge, then it doesnt matter if I set the charging voltage to 14.4 on the MPPTs.

With only 10 amps of charge coming in, the charging voltage wont go above 13.5v, until well above 90% SoC is reached. 

 

Once I'm into the final 5% or so of the charge, it seems like you can keep the voltage over 14.4 even with a small charge like 20 amps, so thats where I'll try to fine tune the charging voltage on the MPPTs, so that it stays above 14.4 but below 14.6, and thus it keeps the balancing active.

 

It feels a bit like walking a tightrope, to keep the voltage high but avoid overcharging the batteries (or overheating them)- but at least its only once in a blue moon as you say.

Another complication is that if I set the MPPT charging voltage at exactly 14.4v, they'll go into float when they get to that voltage. So it will need to be tweaked as it increases.

In fact, the truth is that both MPPTs display higher battery voltages than the BMVs do (and the valence software as well). I dont know why they do that, but it means I have to set their charging voltages about 0.3v above what I actually want them to be, as a compensation.

This afternoon for example, the MPPTs were giving the battery voltage as 15v, when the BMVs and the valence software were showing 14.4v. It really doesnt help matters, and at some point I'll have to try and find out why they arent more accurate.

 

Anyway, I've got your point here, which is to notch up the charging voltage bit by bit, allowing the lagging cells to catch up.

Judging by this afternoon, I think it might be worth trying this notching approach earlier in the process too. So instead of setting a charge voltage of 14.4 right out of the gate, maybe I should set it at 13.8, and go up in increments from there, hopefully after the lagging cells have picked up on the others. 

 

But if this catching up process (which sounds like balancing, really) is only ever going to happen over 14.4v then I might as well just blast them up to 14.4 asap, and then do the micro management. 

 

Couple of things from that:

 

If the MPPT actual charging voltage differs from what is set, it seems likely that it is due to temperature compensation which is the domain of Lead Acid batteries, not Li. So there should be some means to kill that, either by telling the MPPT that it is charging Li batteries, setting the temperature co-efficient to zero, or maybe even just by disconnecting the temperature probe. This latter is what I have done with our Mastervolt Combi, because it is quite old and doesn't have a Lithium battery type in the settings.

 

Secondly as a temporary measure whilst you are doing this balancing, you could set the MPPT float voltage to be the same as the charge voltage. That way, when the MPPT goes to float ... nothing happens. Don't forget to set the float to a nice low safe voltage once you have done the balancing. Again this is kind of what I've done with my Mastervolt Combi - it is in "forced float" mode whereby it is always in float, and I (or rather my BMS) adjusts the float voltage to suit the circumstances. Obviously since our boat is connected to shore power all the time we are not there, it was important to ensure that the Combi charged at a fixed voltage all the time, and didn't get some mind of its own.

Edited March 10, 2022 by nicknorman

[Tony1]

32 minutes ago, nicknorman said:

Couple of things from that:

 

If the MPPT actual charging voltage differs from what is set, it seems likely that it is due to temperature compensation which is the domain of Lead Acid batteries, not Li. So there should be some means to kill that, either by telling the MPPT that it is charging Li batteries, setting the temperature co-efficient to zero, or maybe even just by disconnecting the temperature probe. This latter is what I have done with our Mastervolt Combi, because it is quite old and doesn't have a Lithium battery type in the settings.

 

Secondly as a temporary measure whilst you are doing this balancing, you could set the MPPT float voltage to be the same as the charge voltage. That way, when the MPPT goes to float ... nothing happens. Don't forget to set the float to a nice low safe voltage once you have done the balancing. Again this is kind of what I've done with my Mastervolt Combi - it is in "forced float" mode whereby it is always in float, and I (or rather my BMS) adjusts the float voltage to suit the circumstances. Obviously since our boat is connected to shore power all the time we are not there, it was important to ensure that the Combi charged at a fixed voltage all the time, and didn't get some mind of its own.

 

Thanks again Nick- I do have the temp compensation disabled on the two MPPTs, but I do use a user-defined profile, not their preset one for lithiums, so maybe that has an effect somehow.

 

I have done that before, i.e. set the float voltage to be the same as the bulk charge. Especially in winter, because in winter you never actually want the MPPTs to go into float- you usually want them full on for the whole day. 

This is where profiles with float for me seem to be obsolete for lithium batteries, even though you still see them in all the available chargers and charging profiles. 

Sometimes with the B2Bs, it seems the entry into float is related to the time spent charging in the bulk phase, no matter what SoC or voltage is reached during the bulk phase. So there are times when a charger will go into float when the SoC is only say 70%, which is no use. 

Personally I want the batteries to be charged flat out by the MPPTs- after all they can only give maybe 80-90 amps absolute max.

And as soon as the batteries reach the target SoC, I want the MPPTs to just stop charging. I feel I have no real use for a float phase. 

 

Anyway, that moan aside, I do set the float to 14.4 when I need to synchronise, otherwise I do find the MPPTs pack in too early.

 

If I was leaving the boat alone for a month or two, I think I'd just disconnect the lithiums. The only thing that really matters is the power to the bilge pump, and that runs off the lead acid start battery. So for me a long term solution might be some kind of split charge type relay that would allow me to send some power back from the lithiums to the starter battery- assuming thats even possible. 

 

But in fairness, you dont see lots of boats sunk because the bilge pump ran out of power and too much rainwater drained into the engine bay.

Well I dont think you do....

 

Edited March 10, 2022 by Tony1

[peterboat]

13 hours ago, nicknorman said:

 

Hmmm 96 and 69 as mentioned above are identical ... if for one of them you are reading upside down ... (hint!)?

Anyway I wouldn't worry to much about the internal BMS SoC, maybe it's a bit like those SoC indications that you get with cheap solar chargers.

 

You should bear in mind that cell voltages around 3.2 to 3.3 are not relevant in terms of balancing. Something would be horribly wrong if there was a noticable difference at that sort of voltage, because the curve of SoC against cell voltages is extrememly flat. Out of balance is only detectable once the cell voltages rise up the knee, where any differences are massively magnified, say above 3.5v. I don't know if my cells are abnormal but I find that one really needs to get to 3.6v to be sure which is the "lead " cell, because the leader at 3.5v can become a lower one by the time 3.6v is reached. And of course it is what happens at 3.6 to 3.65v that is the important thing. This of course is why auto-balancing doesn't kick in until the cell voltages are fairly high.

 

As MtB says, I think far too much emphasis is placed on interconnect topology - always assuming that the interconnect cables are adequately beefy.

On your last para I would disagree - there are thousands of people like you who have a very limited understanding of the principles involved - and yet they set themselves up as marine electricians and take an unreasonable fee for screwing other people's systems up!

He has to take the batteries up to 14.6 volts with the puter connected to synchronise the SOC on the batteries. I did it with all of mine when I first bought and charged them. Though as you say as long as he keeps it 20 to 80% the cells will always be balanced enough not to matter.