Fra avatar image
Fra asked

Best absorption method

hello, having a smartshunt on the net, what is the best method to calculate the absorption time?
I have 178ah of AGM.
Is it better for the various Chargers to have adaptive absorption?
Or is it better to set a very long fixed time and a tail current (sent to all chargers by the smartshunt) of 1.8% of the capacity? So 3.2A.

I have seen some problems with adaptive absorption.

1) if the batteries are at 12.6v at the start of the charge, the absorption lasts 1 hour and is too little for AGM batteries discharged at 80% soc.
2) if absorption has been completed during the day and the batteries discharge slightly during the day, the charge cycle restarts but absorption lasts 1 second.. and in my opinion this is not correct. This defect does not occur with the non-adaptive tail current method.

My doubt is:
Is a tail current correctly set and sent by the smartshunt always a correct method to end the absorption phase? Even at different depths of discharge?
charge current limit
2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

6 Answers
JohnC avatar image
JohnC answered ·

Hi @Fra

In my view, the mppt Tail function (fed by shunt) is by far the best way to terminate Absorb with Pb batteries. It works on rebulks too.

It's in Amps of course, but 1.8% is a bit low for my floodeds. I use ~3%, and that gives me ~99.0% SOC. I Float in the last 1% SOC.

Absolutely recommended. A great feature.

2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Fra avatar image
Fra answered ·

Ah ok thank you. I put 1.8% of capacity because are AGM. And ROLLS (not mines)

recommends prolonging absorption 1 hour after the current drops to 2%. With his agm.

Do you think with 1.8% risk overcharged?

however I have seen that the end-of-charge current varies a little depending on the circumstances (temperature and above all previous discharge). After repeated light discharges the tail current tends to be lower, even reaching 2.6A.

After a deep discharge, at some point, around 3.3A, it stops dropping.

Curious phenomenon that I don't understand what it depends on.

And also the working voltage of the battery improves after deep discharge compared to repeated light discharge.

my AGMs are at 99/100% when the current is between 1.5 and 2%.

Do you think tail current with shunt is the best method for bluesmart ip22 too?

eliminating adaptive absorption based on bulk phase?

I'm not sure if tail current with shunt is able to lengthen the absorption time after deep discharges..

as victron recommends. (but rolls for example recommends a fixed absorption time regardless of the depth of discharge..

and always quite long).

In your opinion?

2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

JohnC avatar image
JohnC answered ·


You're delving deeply now. Ask 20 people the same things and you'll get 20 different opinions.. so gospel this isn't, based only on experience with my own GC2 pattern batts..

Temp has a big effect. Cold batts are sluggish, sag badly in V under load, and really need that realtime boost in charge V's. I use -100mv/C (48V), and at that the charge seems to need much the same Abs time seasonally. They may taper in Tail a little more slowly in the cold (which usually happens for me when discharge is a bit deeper too), but I haven't found a need to change the Tail setting.

I don't hold ultimate faith in batt makers settings. Lab conditions aren't always the same as in the 'wild'. Up to the user. They'd need to actually provide me with a feasible reason why I should add another hour to Abs. My Abs time is like 1-1.5hr to 3% Tail, but I'm 'sneaking up' on Abs with a solar day. I don't use 'shore' charge to go beyond Abs.

And on a couple of occasions on really long cloudy summer days I've actually reached 100% SOC without reaching Abs on the way. That's rare, but the shunt and V/Tail agreed.

The lab boffins don't test like that. There may of course be other reasons like stratification, but I'd need convincing..

2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Fra avatar image
Fra answered ·

I understand exactly what you mean.

I also get to 100% without ABS sometimes.

I believe that the higher the bulk charge current, the longer the ABS will be.

if you charge a 200ah bank at 2A, at beginning of the ABS voltage they will already be charged, without having to keep the ABS voltage for a long time.

Rolls said that their thick plate AGM will sulphate if the ABS times are not respected.

But I gave the same example of a 200ah bank loaded at 2A and they told me that in theory this way, at the start of the ABS it will actually already be charged.

You can check your flooded batteries carefully with the densimeter to know the real 100%.

I can't do that with AGMs.

Victron says a deeply discharged battery also needs 6 or 8 hours of ABS.


but it probably depends on the bulk current. The higher it is, the more inefficient it is.

Solar charges at low progressive current during bulk.

there are many mppt manufacturers that don't even allow abs longer than 3 hours. probably for security reasons.

In addition, solar charging if it doesn't complete one day, it will do it the next day or 2 days later...

Even as a schunt, I reach about 100% in 1.5-3 hours of ABS. and this corresponds to a tail current of about 1.5-2%.

What you say about your Flooded makes sense, as the agms have lower internal resistance I think.

One day I want to try starting from 60% soc, loading from bluesmart immediately at 20A bulk, how the behavior changes compared to solar.

I expect ABS to start at a much lower SOC and consequently longer ABS. Compared to the progressive current bulk of the mppt (I think more efficient).

a question:

If one day, after a relatively deep discharge, I see that the current stops dropping below 3.3A (2% of my batteries), can I try to insist on ABS for another 2 or 3 hours to see if it drop again?

It shouldn't do any harm right?

In any case, every 21 days I do 1 hour of equalization at 14.75v at the end of the absorption at 14.45v.

It's probably useless, but I don't think it could hurt.

On AGMs the actual equalization would be harmful (over 16v). But some schools of thought allow for a mild EQ from time to time.

I have floating at 13.6v and -24mv/C°. For safety. Yesterday with battery at 36 C° the floating was 13.4 v!!! And abs at 14.2! I charged with alternator for 1 hour at almost 0.3C current. 50A for 178ah! I make these only occasionally.

On your flooded batteries (are they OPZS or traction type??) do you equalize?

2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

JohnC avatar image JohnC ♦ commented ·


I agree with all your observations, conclusions, and 'anecdotes'. You've been there. For your questions, which you can probably answer yourself anyway..

You can test a longer Abs to see what happens. But it's what you do every day that's important. A sealed batt devoid of moisture isn't much good.

I don't bother with equalization. I've done it several times to see if I can see any difference, but never did. I don't think daily cyclers need it really, they get a stir-up daily.

Your 13.6V/-24mV is essentially the same as my 48V bank. My -100mV is 0.1V per degC, so I see a big V change too over the year with ~35 degC range. Temp Comp is very important, and batt maker's recommendations may also include an 'allowance' for users whose chargers lack it. So take care using their gospel figures.

My GC2's are a standard pattern that all the big makers supply. 6V Golf Cart, they're cheap and rugged, been around for yonks. LiFePO4 beckons when these croak, 6yo now. Still have some real oldie survivors in a 12V bank, one of which is at least 13yo, ha.

0 Likes 0 ·
Fra avatar image Fra JohnC ♦ commented ·
I wrote a nonsense earlier.

i.e. that the agm have a lower tail current than Flooded due to lower internal resistance.
actually a lower internal resistance should have a higher tail current.

But I have noticed in these 2 years of cyclical use a slight increase in the tail current.

I didn't quite understand the last part you wrote about lifepo4, golf car batteries and 13 years of life.
My English is quite poor.
If you can, write in a more "scholastic" way please.

For example when you wrote

" o take care using their gospel figures."

What do you mean exactly?

My AGMs have more than 2 years of daily cycles (average 20/30% dod, occasionally 30/40%). Maybe 840 cycles.

so I believe they are mid life (1500 cycles 30% dod in theory).
I have noticed a slow but progressive lowering of the voltage in these 2 and a half years.
For each SOC I think I lost 0.15/0.2v.

However the capacity still seems to be full or nearly so. So , the lowering of voltage maybe is normal?

At the next round I am very undecided whether lifepo4 or traction flooded lead acid.

No more AGM.

a lifepo4 made by a serious company (torjan, exide, victron, mastervolt) costs a capital .. And I would trust.

But I have some friends who have taken cheap lifepo4 from 300 euros for 100ah, or 550 euro for 200ah ..or assembling cheap EVE cells from China with AliExpress bms.
Well, at least 3 have had a broken cell in just over a year.
I don't have much faith in cheap lifepo4.

a trojan cyclic AGM or flooded by Trojan traction costs more than cheap lifepo4 ... is there a reason?
0 Likes 0 ·
JohnC avatar image JohnC ♦ Fra commented ·


I'll try to be a bit more 'scholarly, at the risk of sounding like an AI bot. :)

My 'gospel' thing means (to me) something that is taught and accepted as real, even though it may not quite be. Has historical religious overtones, I won't use it again. Sorry.

Back 6 years ago I had the same choice to make. I didn't really think Li's (of all types) were quite ready. Too many unproven promises, compatibility issues, strange brands, and hellishingly expensive.

I was coming off a 12V bank of 6x parallel strings (an imbalance nightmare) to a 48V system. So I picked 8x of the best, which went well, at least for long enough to test. So I bought 8x new GC2's, for only a quarter the price of my new Victron system. And budgeted for 5 years, which they've exceeded (and still going).

When they fail, I'm ready for LiFePO4. Floodeds are tedious to top up, clean up the increasingly creeping sulphate on terminals, etc. Financially they may actually be better than Li, but I'm over the maintenance.

What I'll buy? Something on the Victron compatibility list (or proven clones). And Victron have stopped increasing that list for cost and other reasons.

Choosing a Pb? A recognized brand, but some you pay more for the name than the name warrants. I inherited some of my originals, and at one time had 5x different brands together. An emergency purchase of a USBattery lasted only a year, the 13+yo (still going) is an Exide and the only other ones I have left are Century(Yuasa). Not an objective comparison though, you get to choose your own fate. A 'made in prc' label would be off my list though.

0 Likes 0 ·
Fra avatar image
Fra answered ·

hahahha, I didn't know Gospel meant gospel at all. That is, the 4 books written by the alleged disciples of the alleged Jesus.

I thought GOSPEL was just the typical black singing of American churches.

That's why I didn't understand.. now it's clear what you meant.

Are you American or English? or are you not english speaking? I'm Italian and my English sucks. A good half is from Google translate.

Which battery lasted you 1 year?

How was the 12v system a balancing nightmare?

What is the victron compatibility list?

The lifepo4 cells are all Chinese and they are good quality too I guess.

But the cheaply assembled batteries don't convince me. I am very skeptical that a 600 euro for 200ah lifepo4 renogy will last longer than an 800 euro for 200ah AGM Trojan

And I'm afraid that taking cells and bms on alibaba, even if they are EVE or CATL etc, I would not get the same result as a lifepo4 battery made by victron or Exide or Trojan etc etc ..

but maybe I'm wrong.

While I know that an agm Trojan will last me at least 6 years of cyclic use and a flooded maybe even more than 10.

And I'm afraid that if I make a lifepo4 in the garage in 2 years it will be thrown away for some unknown reason.

But... I'll stop here otherwise we'll go off topic.

I'll leave you my e-mail if you want to leave me yours with a message. In the future I would like to ask you more about Lifepo4.. Thank you

2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

JohnC avatar image JohnC ♦ commented ·


To your questions..

I'm Australian, so of a culture in between Indigenous, US, England, Asia and Pacific Islands. I speak Strine (an unscholarly version of English). And have lapses. Pardon me.

That failed battery was a 'USBattery' brand. They compete with Trojan in the US. The one I got might have been in stock at the local store for years though. I inherited some Trojans which have passed too. They were ok, but too expensive now with the US exchange rate.

The parallel string imbalance comes naturally to Pb's, and is worsened by different brands and ages. You can't control the current through each string and it causes all sorts of charge issues. Like one string may get hot and destroy itself, and it may actually have been the best one. You won't see Victron anywhere suggesting more than 3 strings, and I think that's too many. Li's suffer this too, but they have a bms to help avoid it. And probably makes the bms the most important part of Li batts. Poor quality there isn't likely to go well. Having good local support for LI's is important I think. No good having an uninterested supplier who pushes you off to a Chinee maker's website for help..

Compatibility.. or just use the @ function and it will email me.

But I think you're on top of it fairly well. Your choice now.

0 Likes 0 ·
Fra avatar image Fra JohnC ♦ commented ·

Oh okay clear! Australian..the third chance!
The English I understand most easily is the one written by non-english people...

Sorry, I'm not clear about the parallel string unbalancing.

You mean between batteries in series in each string?

I mean...between elements in parallel the voltage is always the same (assuming a symmetrical and crossed connection).
What can be unbalanced is the charge (or discharge) current due to the difference between internal resistance.

But if the batteries in parallel are similar, then the charge will be approximately proportional since they are still at the same voltage.

I have 2.agm in parallel from 110ah C20 and one from 68ah c20.
Different brands but the same type of construction and solar cyclic typology.
Same recommended charging parameters. Same age. Same expected life.

well, on an 18A load, you will see the 110ah battery give about 11A and the 68ah one give about 7A.
The same thing happens when reloading.

Some days one battery seems more performing and some days the other .. But I think it depends on the random imbalance of the individual cells in series that make up the batteries.

I don't see how either of them can be overloaded.

Obviously if batteries of different construction are used, such as a gel and a flooded one, the situation changes because the charge parameters are different and the internal resistance in relation to the capacity is different.
So at the same voltage of the parallel, the current contribution will not respect the proportion between the capacities. And it will balance only when the battery with the least internal resistance has been discharged enough.

for batteries in series, on the other hand, I understand the imbalance well because in that case it is the voltage that is distributed. And it will never be distributed equally with a strong risk of overcharging. But there are active balancers for this also for lead-acid batteries.

I don't know if I understood your point correctly.

Ok I didn't understand... I know US BATTERY and I thought they were high quality.
It is important to have fresh batteries. I don't like buying batteries online, but going to the store and measuring the voltage and possibly knowing the production date.
Better if the shop has a strong turnover of material.
0 Likes 0 ·
JohnC avatar image JohnC ♦ Fra commented ·


I was talking parallel batteries, and like you say, it's all about resistance. This might be fine with new batteries, hopefully all the same, so the current through all is the same. But this can change as they age for various reasons. Not necessarily an old age either.

It's actually quite a low resistance, and even how you wire them can impinge. This link is quite old now, back when Pb's were king. And the guy who wrote this knows his batteries..

(and thank you to the people who keep that site up).

So just the wiring can do this imbalance thing.

Then the unaware who wants a max 0.2C charge rate multiplies that out by the number of parallel strings, sets that, and thinks he's right. But not necessarily so.

The different currents manifests itself with the string with the least resistance getting warmer faster than the others. That brings it's resistance down, the current increases further, and you have some level of 'thermal runaway'. And possibly even destroying the best battery - that's the cruel part.

It's the sole reason I changed from 12V to 48V. Nightmares gone. And with a bonus that I only need 8 rather than 12 batteries.

I still run a smaller 12V bank with parallel strings, but they're lightly loaded and deliberately charged at only 13.6V flat. That helps a lot, but not all people have such an option.

Please don't underestimate this, it's real..

0 Likes 0 ·
Fra avatar image
Fra answered ·

ah ok I understand what you are referring to.
Perhaps it is more evident when there are many units in parallel.

For now, monitoring my 2 agms in parallel (of very different capacity), I have not noticed these problems.
The current is always divided APPROXIMATELY based on capacity...sometimes with a little advantage for one and sometimes for the other.
In any case, I charge at 0.1C on average, or less.. only sometimes with the alternator charged at 0.2/0.3C but for a short time.

Ah..the chargers are cross-connected. Positive on one and negative on the other.
They are not connected "in a cascade".

The cross connection helps so that no other unbalanced resistances are added between the batteries.

I have made installations on other campers with 2 agm in parallel (even different capacity) and I have not seen big differences in current (in relation to the capacity I mean).
However, even in these cases the charging current is quite low.

My batteries when charging increase the temperature by 2 or 3 degrees net of the increase in ambient temperature. Only if I recharge beyond 0.2 C does the temperature increase more noticeably.

In the world of campers there are a couple of Italian companies that produce "parallel managers for 2 batteries".
In practice they only put them in parallel if the load is large. With small loads they use them separately.

These devices do not exist in the industrial sector. I find it wrong to use lead acid batteries separately. Stress increases in all circumstances. and the available energy is reduced.
These things also swap batteries under chargers during the charge cycle..Which is often based on time and in my opinion this is wrong.

Last thing...using 12V in a domestic environment doesn't make sense for a whole series of reasons (huge currents, huge cable sections, low inverter conversion efficiency, expensive high current mppts...).
Unfortunately in the camper I am almost forced by the stupidity of the (European) automotive industry, which still insists on 12v despite the enormous increase in consumption and solar power.

You were right to switch to 48v.. For me, switching to 24v on the camper is too fiddly.

But if I were to camper a van myself, I would DEFINITELY use 24v.
2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Related Resources

Additional resources still need to be added for this topic