Passive Car Battery Trickle Charger

On Monday, 1 March 2021 at 08:57:29 UTC-8, bitrex wrote: ...

Often it is more the other way round that the system detects the battery is above about 80% charged and reduces the output voltage of the alternator (Honda reduces it to ~12.5V). If the battery is below 80% SOC the voltage will be ~13.8-14.5v.

Older cars used the same voltage (13.8-15V) independent of battery state-of-charge.

Modern maintenance-free batteries use a lower max voltage to avoid excessive gassing.

Correct.

I'm rich in diodes so it's not a problem - other than my prototype is going to resemble a total rat's nest by the time it's finally completed. I'll post a photo so everyone can have a good laugh at it. John L's design is looking better and better. :-D --

"Andrey Semyonovitch really was rather stupid; he attached himself to the progressive cause and 'our younger generation' from enthusiasm. He was one of the numerous and varied legion of dullards, of half-animate abortions, conceited, half-educated coxcombs who attach themselves to the idea most in fashion, only to vulgarise it and who caricature every cause they serve, however sincerely."

- Fyodor Dostoevsky

Yes. I had a brain fart. The diode is connected parallel to the panel, but it is not for reverse battery connection. A reversed battery would either blow the diode or any fuse you have in series with the panel. Draw yourself a schematic.

The diode is there so that multiple panels can be connected in series. If one panel in the series string gets shaded, the diode provides a path for the other series panels to continue to provide power. Draw yourself a schematic.

It is not a "kamikaze diode" as I mentioned above. If you do not have any other panels in series, the diode is not needed. If I were doing this with only one panel, I would change the diode wiring to be in series with the output so that it would eliminate any back feed from the battery. Conserve your mA-hours.

I don't have any of those bandgap diodes so I'll use one of my many TO-220 cased 5V regulators. Massively overspec at 2A, but rock solid with plenty of headroom voltage. --

"Andrey Semyonovitch really was rather stupid; he attached himself to the progressive cause and 'our younger generation' from enthusiasm. He was one of the numerous and varied legion of dullards, of half-animate abortions, conceited, half-educated coxcombs who attach themselves to the idea most in fashion, only to vulgarise it and who caricature every cause they serve, however sincerely."

- Fyodor Dostoevsky

Oh! Just spotted it's dual supply. No matter. I have a ton of those old 741s leftover and will just use one of those. --

"Andrey Semyonovitch really was rather stupid; he attached himself to the progressive cause and 'our younger generation' from enthusiasm. He was one of the numerous and varied legion of dullards, of half-animate abortions, conceited, half-educated coxcombs who attach themselves to the idea most in fashion, only to vulgarise it and who caricature every cause they serve, however sincerely."

- Fyodor Dostoevsky

Yes. His suggestion has far fewer components, easier to construct.

We use that bang-bang method in our commercial product but it is much more complicated. We use a micro-controller to do all the major work but its side job is just to keep the battery charged using the bang-bang method. We have programmed in thermal compensation, for example.

JL's suggestion may work fabulously for you if you decide to try it. Whatever you decide, have fun at it!

That should work. But you could also use one of those 6.2V zeners as well and adjust your divider accordingly. Almost any reference voltage should be usable as long as it is stable. I've seen people use a forward biased LED as a reference and it will show power is on as well. Not necessarily recommending that because I don't know the temperature drift of an LED.

My calculation indicates you will need 3 diodes per "series chain" to get to the desired 13.5 float voltage. That now raises the component count to 70 components.

Then you need to consider what the resistor value should be. Do you have a lot of resistors to choose from?

I wonder if it detects somehow whether there's a regular flooded lead-acid or an AGM battery in there. AGMs are sensitive to overcharging so maybe just assumes it. I see that all Hondas (and maybe all cars) that do auto-stop have AGM batteries stock.

Not really when you find you've destroyed the battery with a trickle charger. Seen it happen with electric golf carts. Guys put them on trickle chargers, go away for the winter, and come back to dead batteries. Then you are looking at $500 to replace them.

Particularly sensitive I should say

You don't do electronic or electrical design. I've known about it since I started getting paid for doing electronic engineering, some forty years ago now.

State of Charge. Wikipedia lists it as one of about a dozen meaning for the acronym - it should have been obvious from the context, but it wasn't for me. Authors always imagine that their favourite acronyms are obvious from the context, and their readers are rarely as deeply immersed in the subject as they are.

For plug-in applications, there are lots of cheap battery minders out there that won't damage anything. I have a backup generator that I keep on one of those, and it works like the bomb. Seems like it wouldn't be hard to make a solar-powered one--the voltage vs. temperature curves of lead-acid batteries are hardly a trade secret.

Cheers

Phil Hobbs

No, the diode is there to cover the case where you stack multiple panels in series. Often, you want a higher working potential before conversion.

For (physically) larger panels, you will often find the "cell array" treated as a set of arrays, each with a similar bypass diode present.

Two hundred bucks for a car battery? I don't think I've ever paid more than $85.

Cheers

Phil Hobbs

Bandgap. But any zener 5-8 volts maybe would work.

Also, are R3 and R4

R2/3/4 set the nominal battery voltage and the on-off hysteresis. Probably not equal. R3+R4 could be one pot, with a small hazard.

And I'm guessing any old jelly

Pretty much, as long as it will tolerate the maximum solar panel voltage and swing high enough to turn off the p-fet. Putting the opamp on the solar-cell side means it doesn't drain the battery and is not exposed to load-dump nasties.

Fun. Pity nobody else wants to play the circuit game... in an electronics design group.

e:

s design group.

The customer is Cursitor Doom. He hasn't described his problem all that cle arly and his proposed solution is a long way from anything sensible.

He isn't going to take any reasonable solution seriously - he wants to solv e his problem with junk from his spare parts box - and if he did take up a suggestion posted here he'd almost certainly get it wrong and resist any su ggestion that he'd done anything wrong.

Most people have enough sense to avoid him as a playmate.