Lead-acid charging.

[Codebird]

I've been through a couple of cheap solar PWM chargers off of Amazon, and I've noticed some of them have 'unconventional' ideas about battery management. I'm pretty sure that under no circumstances should it go over 15V.

So, I built my own controller. Cobbled together. And it was good. Now I'm taking it one further: A fully-functional charge controller, open design which I shall release once I've worked the bugs  out. Cheap to make, over-current protection, logging, monitoring of input and output current, all the good stuff. The prototype is in my garage right now. But, there is one area that I cannot figure out: Optimal charging of a 12V lead-acid.

It's not that there's a lack of information. It's an excess, most of it contradictory. Some sites specify an exact optimal float voltage, some give a range, some say it depends on the type of battery, some say temperature compensation is essential, and a few say to see the manufacturer's datasheet. Some day saturation charging is essential, but don't agree on what circumstances it should be performed, or how, or when to stop. So, I'm seeking expert advice here.

My charger has the ability to monitor battery voltage to a precision of 0.04V. It can also monitor battery current, though due to my design for a super-low-cost current measuring circuit only to an accuracy of 20%. This should be good enough. It currently has no temperature monitoring ability, but this can be added if need be. Easy enough to put one of those nifty little one-wire sensors in.

My current approach uses a simple two-state machine:
- In the 'float' state, regulate a constant voltage of 13.4V. After 240 hours have elapsed in this state, or if the battery voltage falls below 11.5V, transition to 'saturation' state.
- In the 'saturation' state, regulate a constant voltage of 14.2V. When the current required to do this falls below 3A (With a peak-follower to eliminate the impact of clouds), transition to 'float' state.

This is for a flooded battery - it's a big one designed to power an RV. But the voltages are adjustable. For that matter, the whole charging process is governed by an arduino or a bare ATMega, so it could be adapted for other chemistries entirely if you wanted to.

Now, feedback, please. Am I doing it right, or is my approach going to shorten battery life? This charge controller is designed for low-cost, small-scale solar power installations up to 300W, so I really want to get the longest life practical from the battery - no point using this charge controller if it'll ruin your battery in a year. I'm quite hopeful about this design - it should cost about the same to build one as buying a low-cost PWM controller off the shelf, but it'll be superior to most of them.

[GoneTomorrow]

Temperature compensation is generally considered important, especially if the batteries are somewhat exposed to wild environmental temp changes. The voltages required to properly charge lead acid batteries actually vary substantially with different cell temperature. Compensation for FLA is about -4mV per °C per cell. So if your baseline absorption voltage is at 25°C, decrease by 4mV every degree above 20°C, and increase by 4mV for every degree below 25°C. An example of the effect is that at -15°C battery temp, your absorption voltage is going to be a whole volt higher than at 25°C. The temp sensing should ideally be remote, right on the batteries.

Maybe also add an equalisation mode, for flooded batteries that aren't moved for a few weeks, which can cause stratification of the electrolyte. Usually entails charging up to 14.8-15V for a few hours every month or so.

Probably not much of an issue for an RV that's gonna be moving around (stirring the electrolyte up), but definitely useful for static installations using FLA batteries.

[Codebird]

I was undecided about the the need for a temperature compensation sensor, but if you say it's important I can look into adapting the design - should just need a single component, though I don't know what kind of range I'm getting on one-wire bus with only a 3.3V driver. Right now my prototype is very slowly charging batteries, because December. I get about half an hour of direct sunlight a day on the panels.

I didn't use one initially because I notice the cheap-and-nasty controllers seldom have temperature compensation. This made me wonder just how essential it is. I'm going to leave it in for testing for a week or so (Once I manage to get the telemetry out - it's right on the edge of range for the 1200bps radio link I'm using), then decide what revisions it needs.

[GoneTomorrow]

If your batteries are in a fairly controlled environment, and you charge at low current (little battery heating), then you can probably get away without temp compensation. But otherwise you run the risk of over/undercharging when at extremes of temperature which will shorten battery life.

Yes, a lot of the cheapo chargers don't have it, because it costs money, and isn't essential for charging a battery, as long as you disregard lifetime. You'll find all the best solar chargers (Morningstar, Midnite, etc) are all temperature compensated, even at the low end.

[Yansi]

Even the simplest LM317 based Pb charger can be taught to compensate for temperature: Just stick two Si diodes inside the feedback loop. 

[KD0CAC John]

There are a number of reasons that the recommended charging spec.s are all over the place - that is because the spec's are all over the place , like all the ones you left out of the the question - the details of the battery / bank , the environment they are living , the use the are performing , etc.
To do on the cheap , then designing a charge controller with a very narrow usage - only one specific / type of battery , in a controlled environment  [ lets say 70* F. ] - so not as much of a need for temp control , as long as over charging is kept in mind .
I've had solar for several decades , and have had my Trojan L16 battery bank - 4 total 2 pair each in series for 12v & then those in parallel for higher capacity , and these ran my RV full time living with AC , furnace everything for 14 yrs.
These batteries were designed for commercial floor scrubbers with an intended life of 4 yrs. - so a well designed system makes for much longer life .
When buying an off the self charge controller , and quality I choose one that is programable [ that means that it needs the specific battery spec.s and orientation of the number of batteries to correctly charge , float , equalization charge etc. for max life ] also to be able to go to a different type of battery bank in future , like now , I need a new battery bank , and considering Nickel Iron batteries this time .
Or to make simple , 1st decide on the battery you are going to use - then build a charge controller around the intended battery , unless you want future options .
Good battery info - https://www.solar-electric.com/deep-cycle-battery-faq.html/#Battery%20Voltages 

[Codebird]

I can't refine it much more at this time of year, so I've not added temperature compensation yet, but here's an extract from the writeup:

----
Lead-acid batteries are a moderately difficult battery to charge properly. Doing so requires precise measurement of the battery voltage, and measurement of current with less precision. This firmware uses a two-state model: A saturation charge that maintains a battery voltage of Vcharge until the current required to do so (subject to a peak follower to discount the effect of passing clouds) falls below a set theshhold Icharged, at which point it transitions to a float charge that maintains a second, lower voltage of Vfloat. Transition to saturation charge comes when the voltage falls below 11.5V or after 240 hours, whichever comes first. This charge process is based upon that described on page 73 of the GNB Industrial Power Handbook for Stationary Lead-Acid Batteries, Part 1.

The exact values of these depend upon the battery. Though all lead-acids use the same electrochemistry, variations in such factors as electrolyte concentration and plate alloy material can affect the optimal voltages. If possible, these should all be set in accordance with the datasheet provided by the battery manufacturer in order to maximise battery life - but, if this information is not available, the source code includes 'preset' suggestions that will be approximately correct for flooded batteries, AGM batteries and gelled batteries, based on figures from a BBL guide to battery charging.
Preset   Vcharge   Vfloat
Flooded   14.4   13.4
AGM   14.6   13.4
Gel   14.2   13.6

Icharged should, in most cases, be set to 1/20th of the battery capacity in Ah.
---

Thoughts, anyone?

[Seekonk]

Here is why it doesn't matter.  Unless you are the telephone company, most active systems never really get out of the bulk charge mode.

[mos6502]

Regarding your charger, it would be easier to buy an Imax B6 and mod the firmware to suit your needs. The Imax B6 has all the necessary hardware, including a port for a temperature sensor. It already has a Pb charge mode that works quite well. The charger costs less than $20 and there's an open source firmware  that would be very easy to modify and add your own parameters:

https://github.com/stawel/cheali-charger

[Codebird]

Regarding your charger, it would be easier to buy an Imax B6 and mod the firmware to suit your needs. The Imax B6 has all the necessary hardware, including a port for a temperature sensor. It already has a Pb charge mode that works quite well. The charger costs less than $20 and there's an open source firmware  that would be very easy to modify and add your own parameters:

https://github.com/stawel/cheali-charger

I may look into it, but this isn't just a charger. It's a solar charge controller - it includes low-voltage load disconnect, and over-current load disconnect. A charger alone wouldn't be suitable, because you can't monitor the current draw of the battery when there is a load in parallel with it.

[mos6502]

Well that changes things. But still, check out the B6, maybe it can give you some ideas. Or maybe you can use it with a small add-on board. It has a buck-boost converter inside that can do 50W/6A max. Schematics are here: https://www.rcgroups.com/forums/showthread.php?1362933-IMAX-B6-Schematic