Battery voltage - safe discharge levels on AGM deep discharge batteries

[Solarexploits]

Had a good hunt round the forum as I thought this must have been covered b4, but if it has I didn't find it.

 

There are a no. of charts around t'internet showing battery voltage as giving a good guide to state of charge of the batteries, but there appears to be some discrepancy/alternative figures when it comes to AGM & Gel batteries

 

I have 4 x Powersafe 12V 170FS deep discharge batteries, which are described by Powersafe as valve regulated Lead acid - elsewhere in their literature they refer to them as AGM, looking at the charts at 50% discharge, which I would prefer to stay above, for sealed or flooded it gives 12.2V & AGM at 12.3V. The batteries have got battery balancers on them & are normally within .1 of a volt difference, possibly a spread of maybe 0.15 between highest & lowest

 

Several questions I hope you guys can help with as can't find it given anywhere in Powersafe's spec

 

1. Is that discrepancy of 0.1V reasonably accurate, might seem a little petty, but given that a drop of just 0.2V takes them from being 50% to being 25%, that's quite significant

 

2. Whether you measure with a multimeter at each battery, or take the reading from the inverter input, unless all loads have been removed (which would include the inverter, so you wouldn't be reading anything from that!) then the voltage you record is going to be under load, rather than the unloaded volts given in the charts - how significant is this likely to be, my load when no solar charge available is typically around 3 to 5% according to the inverter? - no idea how accurate that figure is, nor where it calculates it from or how. There does seem to be a slight discrepancy between what the inverter displays & measuring with a multi meter, for instance inverter displays 54.0V (it has increments of 0.1V), individual battery readings are 13.35, 13.38. 13.43 & 13.46, so I would expect a reading of 53.6V at the inverter?

 

3. As soon as the batteries are no longer being charged with solar, the battery voltage seems to drop quite quickly into the 49 to 49.9V level, load would be around 300 - 400Watts at this period, over the next couple of hours load rises to around 600 - 700W for a couple of hours b4 dropping back to around 200 Watts, by this point, currently around 4 hours after the sun's gone down, battery voltage is down in the 48 - 48.9V area, If I'm wanting individual battery voltage of 12.2V as a min, this is marginal, if it should be 12.3, then it's already too low & even with just 200W, a 4A draw, by the time the charger kicks in, another 2 & 1/2 hours later, will have dropped further - I'm not around to monitor this as already asleep by that point! Am I letting battery voltage drop too low, bearing in mind these are "on load" volts?

 

4.Having been charged on the overnight cheap leccy for 4 hours, by around 7am, still drawing around 200W at this point, there is already some (very small) incoming solar & voltage is  up to 49.5 - 50.5 range - does that sound sensible?

 

5. How resilient are deep discharge batteries to dropping below that 50% & are the voltages you see in the common charts accurate for this type of battery, or are they taken from typical car battery sizes & what I'm looking at I should expect to be different?

 

These were SH batteries, but are dated 2021, I started using them earlier this year & unfortunately as I didn't understand balancing & had some other hiccups, I did have one or 2 of them down at below 11V a couple of times, when they arrived with me they were showing 13.0V off load, I'm wondering if I've done some damage, but the pack as a whole has worked so much better  since having the balancers on. From the charts anything below 11.8V is completely discharged, however, as I accidentally discovered, this is most certainly not the actual case as i defo measured at least one unit below 11.0V & others not far above & inverter was still working fine & supplying power as normal,  it's safety cut off voltage is far lower at 42V or 10.5V per battery???

 

Sorry for all the Q's, hoping someone can help!

 

 

[billt]

There's a forum dedicated to this sort of thing and you're more likely to get a good answer there. In fact you'll probably find the answer in their FAQs or with a search.

 

diysolarforum.com

 

It's US based, but most of the info is applicable elsewhere.

Edited October 11, 2022 by billt

[Miek]

If the batteries are secondhand it's likely that they have slightly different internal resistances so not surprising they don't show identical voltages. Taking a voltage under load is not ideal. Really you need each one disconnected and ideally after having a few hours of no load to get a good stable accurate voltage reading. 

 

Recommended float charge voltage:

2.29Vpc at 20°C -from the datasheet. That's quite low for lead acid,  so make sure you are not overvolting them during charging, a lot of chargers go to 2.36Vpc at least. 

 

Do you measure your load? If so just let them have a full absorption charge and go In to float, then discharge them without any charging input till they hit 2.0Vpc and see how many Watt hours you used. If it tallies with the datasheet more or less then they are healthy enough. 

 

 

 

[Solarexploits]

I know from earlier testing the batteries have differing internal resistance, (are any 2 batteries actually identical???) hence the battery balancers, which usually keep things within 0.1V of each other, but surely you would still expect that if you add the individual battery voltages checked with multi meter (confirmed with separate meter) they would come to the same figure (perhaps within 0.1 or so) of what the inverter is recording

 

2.29Vpc would give an overall charge voltage of 55V, the inverter is currently set on it's standard AGM setting which is 54V, so would it be worth changing this to 55V as otherwise presumably I'm not actually getting a full charge in? Actually, having just reread the datasheet it says both freshening charge & fast charge should be 2.29 - 2.4Vpc, so does that mean in fact I should have the charging voltage set to 57.6? If that's the case then I'm not charging them fully in the first place. I obviously want to get a full charge in whenever available, but certainly don't want to "cook" them!

 

I do measure load, but have no way of tracking total load taken, the load is inconsistent as freezer cuts in/out, TV, lights or computer switched on/off etc etc - I always wonder how folks manage to come up with any accyrate figure for their "base load" as mine is certainly far from consistent even overnight as freezer would still be intermittent

[SteamyTea]

10 minutes ago, Solarexploits said:

are any 2 batteries actually identical

Depends how you round, or truncate the numbers.

10 minutes ago, Solarexploits said:

always wonder how folks manage to come up with any accyrate figure for their "base load" as mine is certainly far from consistent even overnight as freezer would still be intermittent

By high temporal sampling or Wh counting.

 

I use Wh counting and it is very revealing, even though I know I may be almost 1Wh out at any time.

3.6 kJ I can live with as accuracy.

[Radian]

1 hour ago, Solarexploits said:

I always wonder how folks manage to come up with any accyrate figure for their "base load"

As I understand it, you would be looking at average power draw overnight or during the day when the house was unoccupied. Meaning no lights or intermittent demands like TV's, washing machines etc. So between the last light going out at bedtime and before the alarm goes off in the morning, only the standby power of various gadgets permanently powered-up plus the intermittent on/off of fridge/freezers would be counted.

[Temp]

19 hours ago, Solarexploits said:

3. As soon as the batteries are no longer being charged with solar, the battery voltage seems to drop quite quickly into the 49 to 49.9V level, load would be around 300 - 400Watts at this period, over the next couple of hours load rises to around 600 - 700W for a couple of hours b4 dropping back to around 200 Watts, by this point, currently around 4 hours after the sun's gone down, battery voltage is down in the 48 - 48.9V area,

 

1 hour ago, Solarexploits said:

I do measure load, but have no way of tracking total load taken, the load is inconsistent as freezer cuts in/out, TV, lights or computer switched on/off etc etc

 

Try and find a decent energy monitor to measure the exact consumption.

 

In the absence of data.. If we assumed a load of 500W for 4 hours that's 2kWh. At about 48V thats about 2000/48 = 42 Ah.

 

What capacity are the batteries?  The Powersafe 12V 170FS is a range of batteries with a capacity from 31Ah to 190Ah depending on the exact model. So they could be anywhere from totally empty to 75% full depending on the model.  Clearly they aren't totally flat though.

 

What does it say on the batteries?

 

 

[Solarexploits]

18 hours ago, billt said:

There's a forum dedicated to this sort of thing and you're more likely to get a good answer there. In fact you'll probably find the answer in their FAQs or with a search.

 

diysolarforum.com

 

It's US based, but most of the info is applicable elsewhere.

Thanks for that, I've put a post on there

[Solarexploits]

"I use Wh counting and it is very revealing, even though I know I may be almost 1Wh out at any time."

 

How is Wh counting done?

[Solarexploits]

13 minutes ago, Temp said:

 

 

Try and find a decent energy monitor to measure the exact consumption.

 

In the absence of data.. If we assumed a load of 500W for 4 hours that's 2kWh. At about 48V thats about 2000/48 = 42 Ah.

 

What capacity are the batteries?  The Powersafe 12V 170FS is a range of batteries with a capacity from 31Ah to 190Ah depending on the exact model. So they could be anywhere from totally empty to 75% full depending on the model.  Clearly they aren't totally flat though.

 

What does it say on the batteries?

 

 

The batteries are 170Ah, if you look at the range nomenclature for Powersafe, a 190 is 190Ah etc

 

I realise now, that a lot of the info is actually within the data sheet, including the resilience to deep discharge, I just wasn't understanding it fullyPowersafe Datasheet.pdfPowersafe Datasheet.pdfPowersafe Datasheet.pdf 

 

I think all I really need to be sure on now, is , am I reading the datasheet correctly, see attached below, am I correct that the charge voltage can safely be set to 57.6V (2.4Vpc) without overcharging?

[SteamyTea]

1 hour ago, Temp said:

In the absence of data.. If we assumed a load of 500W for 4 hours that's 2kWh

Just had a look at my data for last week, base load is 30W.

1 hour ago, Solarexploits said:

How is Wh counting done?

 

[Temp]

2 hours ago, Solarexploits said:

I think all I really need to be sure on now, is , am I reading the datasheet correctly, see attached below, am I correct that the charge voltage can safely be set to 57.6V (2.4Vpc) without overcharging?

 

The short answer is no.

 

I've had a good read of the datasheet and they don't say anything about using cell voltage to limit charge time. eg It does not say you can simply charge them until the cell voltage reaches 2.4V (or any particular voltage). In fact it even has a warning that "Continuous charge at 2.40 volts/cell will significantly reduce the battery life".   

 

If you want a voltage that the cells can be charged at indefinitely without overcharging that would be the float voltage 2.29V but that won't fully charge the battery. 

 

What charger do you have and what are its capabilities ? 

 

Instead of using cell voltage they recommend using a BMS system that measures the energy withdrawn from the battery on load and puts back in about 103% of that amount. They describe two ways that can be done in section 2.3..

 

1) Using a charger that measures the exact Ah being put into the battery. In this case it says the charger should output a constant 2.4V. Then while charging the charger/BMS should measure the energy put back in and when it reaches the 103% figure it should stop fast charging and switch to float charging or reconnect the load. In float charging mode the charger should output a constant 2.29V. 

 

2) Using a charger limited to a current of 0.5C for a calculated time. The charger should output a constant voltage that depends on temperature as per figure 4. Once the calculated time has expired it should stop fast charging and switch to float charging or reconnect the load. In float charging mode the charger should output a constant 2.29V.  

 

22 hours ago, Solarexploits said:

There are a no. of charts around t'internet showing battery voltage as giving a good guide to state of charge of the batteries, but there appears to be some discrepancy/alternative figures when it comes to AGM & Gel batteries

 

The reason is that the cell voltage varies depending on the type and make of the cell, its age and temperature.  Cell voltage can be used but its really not a great indicator unless you have manufacturers data which they haven't provided. 

Edited October 12, 2022 by Temp

[Solarexploits]

"What charger do you have and what are its capabilities ? "

 

The charger is the inverter, so it's either charging from the PV array or, overnight for 4 hours, it's charging from the built in charger within the inverter, in both cases the only control I have is the voltage setting on the inverter, which it would appear is definitely set too low, as you're saying that 2.29Vpc would be OK as a charge voltage, (tho' wouldn't give a full charge) which equates to 55V across the 4,they are currently only getting 54V. There is also a current limit setting whivh I think I have at 30A max for both charge systems, I would have expected that to be fine with 170Ah of batteries, would get a 50% charge in around 3 hours

 

I'm slightly mystified that it gives 2.4Vpc for both "freshening" & fast charge, but you're saying that would be too high? In my ignorance I'm thinking that I'm simply giving them a fast charge if I up the voltage2.4Vpc

 

I must admit I got somewhat lost by the whole thing about measuring amps taken & recharging to 103% etc - unless you had a dedicated charger that you could both measure what's taken (or Watts taken!) & had complete control over what the charger puts in, that just doesn't seem a realistic proposition in the real world

 

If I get several years of battery life I'd be happy with that as they would have easily paid for themselves & I would hope to replace with Li  in a few years time, I'm not worried if I shorten battery life by, say, a year, what I don't want to do is either undercharge so they keep dropping below 50% (or I have to remove some load), or overcharge so I cook them, which in both cases would presumably shorten life significantly more!

 

It would appear the batteries aren't ideal for charging from solar, but I did get them very cheaply, if I went for a voltage between the 2.29 & 2.4Vpc, say, 2.35, so 56.4V across the pack, is that going to get significantly more charge into them without being likely to do significant damage?

 

 

 

 

[Temp]

16 minutes ago, Solarexploits said:

I'm slightly mystified that it gives 2.4Vpc for both "freshening" & fast charge, but you're saying that would be too high? In my ignorance I'm thinking that I'm simply giving them a fast charge if I up the voltage2.4Vpc

 

No, about the refresh charge it actually says..

 

Quote

The freshening charge should be conducted using constant voltage in the range of 2.29 to 2.40 volts per cell for a period of 24 hours. The charge current should be limited to a value expressed numerically in Amps as 10% of the 10 hour capacity rating of the battery (ie. 19 Amps for SBS 190F).

 

In other words the voltage should be set to whatever value (between 2.29 and 2.4V) produces a charging current of 19A (for 190Ah cells). Charging at the 10% rate used to be called a trickle charge. Theses days they give it names like a balancing charge, refresh charge, conditioning charge etc.

 

[Temp]

42 minutes ago, Solarexploits said:

I must admit I got somewhat lost by the whole thing about measuring amps taken & recharging to 103% etc - unless you had a dedicated charger that you could both measure what's taken (or Watts taken!) & had complete control over what the charger puts in, that just doesn't seem a realistic proposition in the real world

 

That is what modern laptops and EV do. 

 

 

 

 

[Temp]

What's the inverter? Perhaps I can find a manual that gives more info on its capabilities.

[Solarexploits]

12 hours ago, Temp said:

What's the inverter? Perhaps I can find a manual that gives more info on its capabilities.

It's a Powland 4/5k, I'm told it's a Voltronic clone.

 

I've had some replies on the diysolar.com forum which are suggesting bulk charge at 57.6 & float at 54.9/55, I thought I might go slightly under those voltages for safety's sake

 

I now understand your point re the freshening charge, would it make sense to reduce the max amps setting to 20A which would be pretty close to their suggested 19A & should therefore prevent any damage

[Temp]

4 hours ago, Solarexploits said:

It's a Powland 4/5k

 

Looks like they make numerous different models. One I looked at briefly said it had "Smart battery charger design for optimized battery performance".