When is a battery fully charged?

[nicknorman]

There have been a few threads recently where the question of what percentage of the AH capacity of the bank, expressed as charge current, reflects a fully charged battery bank, has cropped up. There has been mention of figures as high as 3%. We are currently out cruising with boat having been on float charge for a couple of weeks. Battery voltage with engine running is 14.5v and charge current is 0.7A for a pretty new 440AH bank of open wet cells. That represents 0.15% of bank size. Now THAT is fully charged, much more than that can mean gradual sulphation due to never quite getting fully charged. That is the end of my pontification!

Edited June 20, 2012 by nicknorman

[Paul C]

Have a read of this:

 

http://www.victronenergy.com/upload/documents/Book-EN-EnergyUnlimited.pdf

 

The answer is in there!

[nicknorman]

Have a read of this:

 

http://www.victronenergy.com/upload/documents/Book-EN-EnergyUnlimited.pdf

 

The answer is in there!

It is a good document which I have read before. I will read it again when I am not actually driving the boat! However one has to bear in mind that it is written in support of their products and the threads I mention have included people who have been caught out by believing their Victron battery monitors. In other words one has to bear in mind that, useful though it is, the document is not impartial.

[Paul C]

Yeah if you take the section on battery monitors with a pinch of salt, the rest of it which describes battery charging, physical differnces between batteries, etc is quite useful.

[nicknorman]

Have a read of this:

 

http://www.victronenergy.com/upload/documents/Book-EN-EnergyUnlimited.pdf

 

The answer is in there!

Having now stopped for lunch and skimmed the document, in fact I can't find the answer there. Have I missed it, or are you just presuming that this pont is covered?

[smileypete]

Also have a look at The Battery FAQ, the answer is also well explained in there:

 

http://www.batteryfaq.org/

 

For more information, the US Battery, Trojan websites for deep cycle batts, and of course the Smartguage website.

 

 

There have been a few threads recently where the question of what percentage of the AH capacity of the bank, expressed as charge current, reflects a fully charged battery bank, has cropped up. There has been mention of figures as high as 3%. We are currently out cruising with boat having been on float charge for a couple of weeks. Battery voltage with engine running is 14.5v and charge current is 0.7A for a pretty new 440AH bank of open wet cells. That represents 0.15% of bank size. Now THAT is fully charged, much more than that can mean gradual sulphation due to never quite getting fully charged. That is the end of my pontification!

Well kind of but not quite, this goes into some detail:

 

http://www.union-services.com/aevs/459-462.pdf

 

cheers, Pete.

~smpt~

Edited June 20, 2012 by smileypete

[nicknorman]

Also have a look at The Battery FAQ, the answer is also well explained in there:

 

http://www.batteryfaq.org/

 

For more information, the US Battery, Trojan websites for deep cycle batts, and of course the Smartguage website.

 

 

 

Well kind of but not quite, this goes into some detail:

 

http://www.union-services.com/aevs/459-462.pdf

 

cheers, Pete.

~smpt~

So, from your first reference I am told that my batteries are fully charged when current falls to 2%, which for me is about 9A. Seems a far cry from 0.7A which is actually fully charged! From your second reference I get that the final fully charged current may increase as the batteries age, which is fair enough.

 

However I can't help thinking that if I applied the 1st reference and always stopped charging at 9 amps, it would not be long before my bats were losing capacity with ref. 2 kicking in!

[smileypete]

So, from your first reference I am told that my batteries are fully charged when current falls to 2%, which for me is about 9A. Seems a far cry from 0.7A which is actually fully charged! From your second reference I get that the final fully charged current may increase as the batteries age, which is fair enough.

 

However I can't help thinking that if I applied the 1st reference and always stopped charging at 9 amps, it would not be long before my bats were losing capacity with ref. 2 kicking in!

 

Well there's also this:

http://www.sandia.gov/pv/docs/PDF/batpapsteve.pdf

 

cheers, Pete.

~smpt~

[Gibbo]

Assuming you want the proper answer...

 

They are never fully charged. You will always fall slightly short of a full charge. If you didn't, your batteries would last forever (think about it).

 

They are are as fully charged as they can ever be when the charge energy falls to match the level of the internal heating, electrolysis and other internal losses, which will takes weeks.

 

There is no such figure as "when the charge current drops to X%", because no matter what figure you choose, you can always put a bit more in, therefore they can't have been fully charged.

 

What you have to do is come to a sensible compromise between how long you want to spend charging, and how fully charged you really need them.

 

For non shorepower charging I tend to aim for a charge current of 4%, it works in practice. For testing and development I use 0.25%. The difference can be tens of hours.

[nb Innisfree]

But as we all know calculating batt capacity in the first place is impractical. I've tried all this percentage stuff but as I can't rely on shunt based monitor or smartgauge (batts have less than 50% capacity left) it's all a big stab in the dark. So before we were forced to give up ccing and batts were new we did a daily bulk charge to acceptance (one hour) and a weekly absorption via Victron until half a tank of genny petrol had expired (approx 8 hrs) and that I think was the best and most practical we could reasonably achieve. What percentage of capacity that final charge rate represented I have no idea, a bit pointless really as batts are now a bit old so I did a daily increasingly longer bulk charge to acceptance and extended into absorption until 2 hrs were up (waste of fuel after that)

Anyway it's all a bit academic as we can't boat anymore due to family commitments so I don't worry about charging probs anymore, though one advantage of being on a landline and not having to do partial state of charging is an extended battery life.

[bottle]

I quite simply charge to at least 95% (Smartgauge, new batteries) usually about three hours on average

 

Sometimes to 100%

 

Ran a little , check the other day, nothing scientific. used Smartgauge and monitored Victron via usb interconnect.

 

Took readings every 15 minutes, will not bore you with them all.

 

Setup: Victron combi, 24v, 2Kva, 50a. Batteries: Victron AGM (6) 24v, 330Amp.hrs.

 

Time..............%.....................V....................A

 

16:45.............69...................24.6..............0

 

16:45.............69..................26.4...............44

 

17:45.............77...................29.2..............19

 

18:45............91....................29.2...............7

 

19:45............100.................29.2................4

 

 

ps. adaptive charging.. on. I know, I know Gibbo.

[swift1894]

sooooooooooooooo!

What's confusing me is that sometimes we are talking about percentage charge/discharge and in another section the actual voltage reading of batteries.

12.8 volts =100% charge etc

 

Is there a table showing when 50% charge equals a certain voltage?

The Smartguage site advises that "Discharging deep cycle batteries to 50% results in the most economical use of the batteries in terms of battery life and monetary outlay."

So what does 50% discharge translate into voltage?

Or am I being naive?

[swift1894]

sooooooooooooooo!

What's confusing me is that sometimes we are talking about percentage charge/discharge and in another section the actual voltage reading of batteries.

12.8 volts =100% charge etc

 

Is there a table showing when 50% charge equals a certain voltage?

The Smartguage site advises that "Discharging deep cycle batteries to 50% results in the most economical use of the batteries in terms of battery life and monetary outlay."

So what does 50% discharge translate into voltage?

Or am I being naive?

 

Ah Hah!

Think I've found it (Section 3.1.1 Victron booklet)

12.65v+ = 0% discharged (100% charged?)

12.45v = 25% " " ( 75% " " )

12.24v = 50% " " ( 50% " " )

12.06 = 75% " " ( 25% " " )

11.89 = 100% " " ( 0% " " )

 

The bit in brackets is my own assumption. Am I right?

[nicknorman]

Assuming you want the proper answer...

 

They are never fully charged. You will always fall slightly short of a full charge. If you didn't, your batteries would last forever (think about it).

 

They are are as fully charged as they can ever be when the charge energy falls to match the level of the internal heating, electrolysis and other internal losses, which will takes weeks.

 

There is no such figure as "when the charge current drops to X%", because no matter what figure you choose, you can always put a bit more in, therefore they can't have been fully charged.

 

What you have to do is come to a sensible compromise between how long you want to spend charging, and how fully charged you really need them.

 

For non shorepower charging I tend to aim for a charge current of 4%, it works in practice. For testing and development I use 0.25%. The difference can be tens of hours.

Clearly it is asymptotic so there is no "right answer" but to me 4%, which would be 18A for my bank, seems very high. Personally I would not consider that to be "fully charged" and I could probably get at least another 10AH in without too much difficulty. As I eluded to before, a better measure is probably to look at the rate of change of charge current and stop charging when it is approaching zero.

 

Incidentally we have a new bank of cheapo 4x110AH bats, a Mastershunt with AH -counting SoC indication, and a Smartgauge. The batteries are on about their 4th cycle so capacity should be pretty much as rated. Spent yesterday tied up (in the rain) and it was interesting to note that the SG and Mastershunt (set on 440AH) agreed to within a couple of % about the SoC on various discharge rates from nothing much to electric kettle. My point is that for those that think SG is snake oil, you are wrong and it really does work pretty well, an amazingly well considering it is only measuring voltage and its time derivatives. It does lag the MS on charge, but catches up after charging is finished. Of course the SG will continue to be that accurate, the MS will not unless I slowly reduce the AH capacity and fully charge quite often.

[bottle]

Ah Hah!

Think I've found it (Section 3.1.1 Victron booklet)

12.65v+ = 0% discharged (100% charged?)

12.45v = 25% " " ( 75% " " )

12.24v = 50% " " ( 50% " " )

12.06 = 75% " " ( 25% " " )

11.89 = 100% " " ( 0% " " )

 

The bit in brackets is my own assumption. Am I right?

 

Yes.................. but whether you can test to hundredths of a volt will depend on you voltmeter and the accuracy of your voltmeter.

[Bob-18]

Voltage, with knowledge of cell temperature is a far better guide than current.

That said, both current and voltage do depend on the exact type of battery being looked at (AGM/flooded/gell etc). Most "decent" battery manufacturers publish data giving charge current/voltage/percent charge curves (in various formats...)

 

But all agree on watching the rate of change gives as good a guide as any once you are approaching the limit - and its a LONG time, not just a few minutes as getting the last few percent in takes a very long time, and a lot of careful watching of rate and temperature.

[nicknorman]

Voltage, with knowledge of cell temperature is a far better guide than current.

That said, both current and voltage do depend on the exact type of battery being looked at (AGM/flooded/gell etc). Most "decent" battery manufacturers publish data giving charge current/voltage/percent charge curves (in various formats...)

 

But all agree on watching the rate of change gives as good a guide as any once you are approaching the limit - and its a LONG time, not just a few minutes as getting the last few percent in takes a very long time, and a lot of careful watching of rate and temperature.

Rested open circuit voltage perhaps, but that is rather an inconvenient thing to measure! Voltage under charge just depends on what is being used to charge, it's regulation circuit gain etc.

[Gibbo]

Clearly it is asymptotic so there is no "right answer" but to me 4%, which would be 18A for my bank, seems very high. Personally I would not consider that to be "fully charged" and I could probably get at least another 10AH in without too much difficulty.

 

Agreed, but when running the engine just to charge the batteries it's down to a compromise. At 18A going back in, and considering that's falling, it's probably going to take another hour to get another 10Ahrs back in. That doesn't make sense to me. I'd rather buy new batteries more frequently than run an engine just to get a few more percent SoC. It's all compromises and personal choice. You have to spend the money somewhere. It's a choice whether you spend it on diesel and engine maintenance or more frequent battery replacement. Having a vintage engine I go for the latter.

 

 

Incidentally we have a new bank of cheapo 4x110AH bats, a Mastershunt with AH -counting SoC indication, and a Smartgauge. The batteries are on about their 4th cycle so capacity should be pretty much as rated. Spent yesterday tied up (in the rain) and it was interesting to note that the SG and Mastershunt (set on 440AH) agreed to within a couple of % about the SoC on various discharge rates from nothing much to electric kettle. My point is that for those that think SG is snake oil, you are wrong and it really does work pretty well, an amazingly well considering it is only measuring voltage and its time derivatives. It does lag the MS on charge, but catches up after charging is finished. Of course the SG will continue to be that accurate, the MS will not unless I slowly reduce the AH capacity and fully charge quite often.

 

Can't disagree with any of that. Where it gets interesting is when (assuming the amp hour counter is being kept in sync) they start to run adrift as the batteries age. You'll notice SG reaching a lower %SoC faster than the amp hour counter. That's because the batteries are losing capacity as they age (sometimes much faster than people want to believe). The new one uses this info to automatically adjust the capacity so that the amp hour counter can track the capacity better.

 

You've found that it lags the MS on charge. I'd expect that. As the batteries age and become less efficient, you'll find it gets closer, until they meet (at roughly half the batteries' normal life expectancy), then SG will go the other way and lead the MS. Unfortunately it's the way it works and there's nothing that can be done about it due to the way it operates and not having a shunt. But at least, when it gets it wrong, it puts itself right on discharge instead of getting worse and worse as time goes on. Which was actually one of the main initial design criteria, the others being simple to understand and simple to install. I hope I met them!

 

Edit to mend broken finger typing.

Edited June 22, 2012 by Gibbo

[nicknorman]

The new one uses this info to automatically adjust the capacity so that the amp hour counter can track the capacity better.

 

 

So, is the "new one" out yet, or if not what is its ETA (and price!)? Sounds like it will be a good device, provided it is not too expensive (bearing in mind the SG does pretty much what you need, except for evaluating the actual remaining capacity.)

[Gibbo]

So, is the "new one" out yet, or if not what is its ETA (and price!)? Sounds like it will be a good device, provided it is not too expensive (bearing in mind the SG does pretty much what you need, except for evaluating the actual remaining capacity.)

 

No, it's not out yet.

 

But it will be worth the wait

 

I reckon about November will see the first production units, but I'm a mere employee engineer so I'm not allowed to say that (though I probably know better than management seeing as I'm designing it).

 

It would have been out earlier, but I kept getting dragged onto other projects/products.

 

A huge proportion of it is completed and debugged. The most difficult part was integrating three different SoC calculations and allowing the device to automatically choose the one that is most appropriate for the prevailing conditions and then "gliding" between the three as conditions change. That is all finished, debugged and works perfectly.

 

The only remaining parts are the exact style of enclosures (it comes in two parts) and agreeing with marketing exactly which features are now left in and which are removed.

[bottle]

which features are now left in and which are removed.

 

 

 

Can we contact you for a way to put them back in.

[Gibbo]

Can we contact you for a way to put them back in.

 

 

If I was dishonest I could start a whole new side-line industry here.