Undercharging batteries on solar with Victron MPPT

[noelex 77]

Originally Posted by Tessellate

Sorry I'm not sure I understand - when you say "the battery monitor is just counting the AHrs and assuming, incorrectly, the batteries are becoming more charged" are you saying that additional 30-35 AH was not additional charging? That the batteries were already 100% and 3.5Ah per day just dissipated into keeping the batteries on float?

Yes, that is exactly correct.

If a fully charged, 100% SOC battery is attached to a battery charger it will still accept some current at a float voltage. So more energy is being put into the battery, but the capacity of the battery is not increasing and no more energy can be taken out of the battery.

The current that a fully charged battery will accept at the float voltage depends on the type of battery and its condition. The current will only be small, but this accumulates to many amp hours over long periods of time.

The extra energy is mostly dissipated as heat.

If the battery is left on float for months or years many battery monitors will just keep counting up the amphours associated with the small current entering the battery. Your 200Ahr battery could well accumulate 1000 Ahrs or more if left on float for a long time. In some ways the battery monitor is correct: a 1000 Ahrs of energy has been put into the battery, but this does not mean you can withdraw 1000Ahrs of energy. If you discharged the battery, assuming it met the specifications and it had not deteriorated after the long period on float, you would still only recover 200Ahrs.

This is why battery monitors reset themselves to zero at the first discharge. They recognise that the excess AHrs measured after long periods on float are not contributing to the true battery capacity that is available.

The important number is the amount of energy we can can recover from the battery, not the amount of energy we can squeeze into the battery. The amount of energy that has been put into the battery will always be greater than the amount that can be recovered, sometimes much greater.

Note: The above us a slightly simplified view. We consider a battery 100% full after the absorption phase, but strictly speaking the capacity will slightly rise during the float stage. For the most part this slight extra capacity is best ignored. For most practical purposes 100% SOC can be regarded as the termination of the absorption phase if the correct end amps are reached. This is especially true with the very aggressive end amp settings normally recommended on forums such as this.

[john61ct]

Originally Posted by Tessellate

How do I find endAmps for my batteries?

Use .005C then, details above.

Good luck getting tech support. That and published tech spec docs are IMO dealbreaker prereqs in choosing a bank.

When you replace, call the listed makers and ask for local distribution. For those bargain Deka GCs Sam's and B+ are well distributed in many places.

> endAmps as one word, with end in lowercase and Amps capitalized (sentence case). Is that a standard symbol / acronym?

Used by some charge source maker(s), dunno, what I use for shorthand.

Alternative was Deka, sorry Defa a typo, and it's not about a specific current rate, certainly not

< 0.1a

but about the trailing amps stopping declining, go read it again carefully.

Any precision 100% Full definition only **needs** to be reached (before dropping to Float) a few times per week for longevity, but if your gear / situation can automate getting there daily that would be the ideal.

With variable solar and tweaking crude AHT measures, "most days" is often best we can shoot for.

And really excessively long over-charging can create its own problems, especially with AGM / GEL.

Best of course is BM coordination / shunt based charge termination

[mitiempo]

Originally Posted by noelex 77

Yes, that is exactly correct.

If a fully charged, 100% SOC battery is attached to a battery charger it will still accept some current at a float voltage. So more energy is being put into the battery, but the capacity of the battery is not increasing and no more energy can be taken out of the battery.

The current that a fully charged battery will accept at the float voltage depends on the type of battery and its condition. The current will only be small, but this accumulates to many amp hours over long periods of time.

The extra energy is mostly dissipated as heat.

If the battery is left on float for months or years many battery monitors will just keep counting up the amphours associated with the small current entering the battery. Your 200Ahr battery could well accumulate 1000 Ahrs or more if left on float for a long time. In some ways the battery monitor is correct: a 1000 Ahrs of energy has been put into the battery, but this does not mean you can withdraw 1000Ahrs of energy. If you discharged the battery, assuming it met the specifications and it had not deteriorated after the long period on float, you would still only recover 200Ahrs.

This is why battery monitors reset themselves to zero at the first discharge. They recognise that the excess AHrs measured after long periods on float are not contributing to the true battery capacity that is available.

Not necessarily true in this case. The Link 10 resets to full when the voltage exceeds 13.2 and the current is 2% or less of stated bank size for 5 minutes. The Link, if set properly, should never keep counting float amps once these conditions are met.

See page 16 of the manual.

Could it be set incorrectly?

[john61ct]

Pointing more to the bank being EoL?

Can that point be adjusted, say to 14.7 and .005C?

[noelex 77]

Originally Posted by mitiempo

Not necessarily true in this case. The Link 10 resets to full when the voltage exceeds 13.2 and the current is 2% or less of stated bank size for 5 minutes. The Link, if set properly, should never keep counting float amps once these conditions are met.

Thanks for that. I am not very familiar with the Link 10.

13.2v and 2% seem a strange choice, but these are the settings. The handbook also states the counter will not be reset to zero until the above criterion has been met and there is 10% discharge. This is not unusual most counters only reset to zero with a discharge although once again 10% seems a strange choice.

In this case there was no discharge, so my reading of the manual would suggest the monitor would not reset, which is what has happened in practice.

[Stu Jackson]

The Link 2000 manual says: "Some accumulation of overcharge amp hours is normal and harmless with systems continuously connected to a charger."


This is because float V is above battery voltage and amps can and will flow. I only wish I could c&p from a PDF, but it's too long a paragraph for me to retype, just go find the manual at Xantrex's site and read it.



The 13.2V and 2% For everyone installing a battery monitor: The "Gotcha Algorithm" thread, a "MUST READ"

Link-series Charging Algorithms -- The "Gotcha" Factor!

DEFAULTS are factory settings that are made to be modified to suit your setup.

Also read this one:

Programming a Battery Monitor (by Maine Sail)

Keeping Your Battery Monitor More Accurate Photo Gallery by Compass Marine How To at pbase.com

and his newer 'site

https://marinehowto.com/programming-a-battery-monitor/are defaults. Here's the Gotcha discussion yet again:

[john61ct]

Originally Posted by Stu Jackson

DEFAULTS are factory settings that are made to be modified to suit your setup.

Amen, preach it!

Couldn't believe a few weeks ago a member was actually arguing against that idea.

Even a shunt-based regulator will need tweaking.

[Tessellate]

Originally Posted by noelex 77

Yes, that is exactly correct.

...

Ah, I see what you're saying now. I was coming from the viewpoint that I had been undercharging at absorption stage, and the additional Ah put in at float / solar absorb was making up the difference (topping up the batteries from 95% to 100%). Now I think it's likely some combination of both. Some of those 35 Ah were probably additional capacity and some were just float dissipation.

It sounds like maybe we haven't been undercharging as much as I thought. We get a full charge about 5 of 7 days per week, as measured by end amps, Link 10 specs, and the solar controller itself. I was concerned that the 2 of 7 days we don't complete absorb (cloudy days) were killing battery capacity very rapidly. Likely it has decreased some, but maybe not EOL. A 20-hour test is pretty difficult for us to do.

Regardless this thread has been helpful because I'm learning that undercharging is likely more common and a bigger risk to battery life than overcharging. And that our batteries can take a longer absorption charge, which I didn't realize was possible (but is) through the Victron controller.

[Tessellate]

Originally Posted by mitiempo

Not necessarily true in this case. The Link 10 resets to full when the voltage exceeds 13.2 and the current is 2% or less of stated bank size for 5 minutes. The Link, if set properly, should never keep counting float amps once these conditions are met.

See page 16 of the manual.

Could it be set incorrectly?

That's not quite it, as I read the manual. The conditions to register a "full battery" are different from the conditions to reset to zero. After 13.2 and <2%, it starts flashing the 4th battery level bar which indicates full charge attained (these parameters seem a bit low IMO, and are configurable).

It still keeps counting amps though, as that is what an ammeter does, and Ah will continue tracking amps across the shunt. Conditions to reset to zero occur when discharge starts. If you're on float with no loads, discharge doesn't happen until you come back to the boat and turn some high load things on or turn off the shore charger.

[john61ct]

Originally Posted by Tessellate

It sounds like maybe we haven't been undercharging as much as I thought. We get a full charge about 5 of 7 days per week, as measured by end amps, Link 10 specs, and the solar controller itself. I was concerned that the 2 of 7 days we don't complete absorb (cloudy days) were killing battery capacity very rapidly.

No, the added benefit after 3-4 / week will not be huge, I used to think that way too.

> 20-hour test is pretty difficult for us to do.

True for 99%

> undercharging is likely more common and a bigger risk to battery life than overcharging

Scratch "likely", your setup would be in the top 1%