When charching batteries - and measuring amps

....

found it...

:)

You're overthinking it. Never measure anything if you can't do something useful with the answer. No need to measure solar input current because there's absolutely nothing you can do about it. No need to measure load current, because you have few devices and only need to measure each once. Turn off everything you don't absolutely NEED. That's the best you can do. Battery voltage is a reasonable indication of how much juice is in there. Put a 9V zener in series with an analog voltmeter and make a new expanded scale for it. If you're a purist, use a 6.2V zener in series with a silicon diode...or a temperature compensated 9V. Buy a battery operated thermometer. and most other very low current devices. Transistor radio uses way less current than a car stereo. Mp3 player even less.

If you're gonna deep discharge it, get a battery designed for such use. Print up a bunch of harbor freight coupons and go several times to get a pile of the free led flashlights. Take 'em apart and make the lights you need.

Use a diode in series with the solar panels to stop the battery discharging though the panels when it gets dark.

He probably ought to have a crude charge controller in series too - nothing fancy just tailing off the charge when it gets above 14v (and before thegassing threshold is reached). Although a 12W solar panel with peak charging rate of less than C/10 probably can't really hurt a 20Ah deep discharge battery over the longer term it might shorten its useful life.

You can get a bit more juice out of the solar panel by adding mirrors either side provided you have the space and can adjust its position. eg. \_/ where \ and / are plexiglass mirrors at 60 degrees.

If you are going to do it with a PIC and the odd high brightness white LED the current consumption could be made to be easily under 500uA. But this does seem to be overkill for what is required here.

When I last did this calculation to power my greenhouse irrigation system I realised it was more cost effective to have two slightly smaller batteries and take along a freshly charged one along as needed.

Regards, Martin Brown

You are right. Thx for reminding me. Still,I had a great night using my brain again, recalling theories and stuff.

A few questions here if I may: What do you mean by deep discharge? What is C/10 ?

I was thinking of a lead battery.

...

...

Yeah, I will think about that but on a caravan. I will remember the idea though.

Might be an option. I will consider that to.

Given the time, my overshoot and consulting my GF, we came up with this:

I will use a 3 1/2 panel meter (I have one from somewhere), and get an AD592 for measuring temperature. I will add a swicth/button, so I can change the "view". So, basically it will be a 100Kohm and 1Kohm for getting 1/100 of power, and and LM358/324 for the temperature. While at it, I might add the outdoor temperature.

That will probably be it.

Sonnich

Capacity divided by ten, for example, 250mA for a 2.5AH battery.

Old rules for LA flooded cell traction batteries was C/5 until battery reached gassing voltage then C/10 for three hours to push the last 20% in. Charges a flat battery in eight hours nominal. Providing the battery was not overheated before start of charge.

Grant.

If you wanna experiment and learn about electronic stuff, this is as good a project as any to waste some time. If you want to GET THE JOB DONE, you're still overthinking it.

Most battery technologies are compromises. Your car battery is designed to deliver a HUGE current for a short time to start the car. It doesn't like being discharged slowly to 10V. Lead batteries that have long life with deep discharge cycles don't like to deliver huge currents. Deep cycle batteries are often labeled for RV use or trolling motors. "deep cycle" is the thing to look for.

If you have a GF, I can recommend many MUCH better ways to spend some quality time. ;-)

we came up with

You probably want to use the series zener to make it expanded scale in the 9-15V region. A 15V meter won't give you sufficient resolution where you need it. and get an

You can buy a digital temperature readout with wired remote sensor for dirt cheap in any department store. At a garage sale, they're usually close to free.

Battery charging is tricky for any battery technology. I'll try not to give any exact numbers, 'cause that will just illicit a firestorm of name calling. Look up the battery you're gonna use and check the charging voltages and temperature coefficients. Make up a series string of zener diodes and regular diodes that match the voltage and tempco of the battery plus the isolation diode. Simple 3-terminal shunt regulator can work too. Set the volts at the float level. Another diode with a switch across it will let you manually raise the voltage to 14.something if you need to top off the system.

If you want it very simple/cheap, there's a tradeoff between charge efficiency and battery life.

I like learning, wasting time, building stuff. It's called a hobby. But I found I build many fewer things since I added another question to my planning...

If I were to get a job at minimum wage. After subtracting taxes, transportation and all the other costs of taking the job, How many hours would I have to work to be able to go buy the thing I'm about to build? When I compare that number to the expense and number of hours it will take to build the gadget, it always comes out that I should buy it. The only stuff I build now is the stuff you can't just go buy. And that doesn't even consider the risk of hurting yourself during the building process or burning the place down cause you did the math wrong and the battery exploded.

And based on the questions asked already it needs to be very simple with a low and cheap component count.

A 100uA moving coil meter will be more effective and draw less current (10uA one would be better still but more expensive unless you can find one in a junk box). Expanded scale to show 10v as 0 and 15v as 100 is a very good one.

Or if you insist on doing it digitally how about a totem pole of 10v zener and diodes with high brightness LEDS off each tap each driven at

20uA (junction just visible indoors in daylight). Simple indicator. You can get finer divisions at the bottom end with Schottky diodes.

No contest here. Ready built thermometers are dirt cheap.

Switches get forgotten and left in the wrong state.

I think the simplest possible crude charge controller with no temp compensation or safety features would look something like

___________ PVpanel +14-18v raw @ ~2A | | R1 | | /

+------| Q1 | \ | | | D1 Z1 | | Battery | | OV-------

D1 Schotty diode rated >2A. Q1 NPN transistor chosen for Imax > 2A V>25v on a small heat sink/box. Gain at 2A likely to be ~20 so base current 0.1A at 16v from the panel R1 ~ 10 & 1W (or 0.3A at 18v). This means the Zener is 15v rated at 5W.

When the battery gets fully charged at around 15v-Vbe-0.3 the charge current drops to zero. You would have to experiment empirically with an extra Si diode and possibly a small resistor in series with the nominal

15v Zener reference to find the best final charge point to avoid gassing and maximise stored charge.

Might be easier using a power Darlington with gain ~600 so that R can larger ~ 330 (& 0.5W zener + Si series diode).

In an ideal world a charge controller should also drop the load if the battery terminal voltage gets below the safe discharge set point.

It is as well to put a low voltage automotive fuse in series with the storage battery close in to it and make sure that the terminals are both properly insulated. Most rechargeable batteries pack enough punch to make wire red hot in seconds if accidentally shorted out.

I doubt if the PV can really do much damage to a heavy duty battery.

Worth pointing out that you can buy ready built and tested modules from hobby shops for about £15 ($20). Unless you already have the bits that will be cheaper and also more reliable than beginners home brew.

Regards, Martin Brown

Hint from long time ago: a 5.1V zener has the right lead acid battery tempco.

Grant.

I'd like to see some supporting data on that.

1n751 is as close as you get to zero TC for a simple/cheap/ordinary zener diode.

Hah hah

So who said you wanted zero tempco? Go learn about batteries first

Grant.

Well, if you read what YOU wrote, you recommended a 5.1V zener as the right tempco. Now you say that the ~0TC of a 5.1V zener ain't right... which was EXACTLY my point.

Asking for supporting data was my nice way of hinting that your suggestion might have been in error and allowing you to correct it gracefully without the usual derision and name calling. You sunk that ship.

Thanks for the suggestion that I go learn about batteries. I'll get right on that.

Gah, was tired, needed sleep, had nine teeth out last Monday, rotten week.

Isn't 5v6 nominal zero tempco? So 5v1 was a little negative and supposedly matched flooded LA batteries way back in the mid-70s. Maybe processing has changed since then? Dunno. Sorry for narkiness. Pain does that.

Grant.

Oh, and we used two 5V1 in series, scaled that to 6 cell voltage, then there was string of voltage select resistors up to 36 cell from negative, into opamp summing node. I could just about draw that circuit from memory after 34 years ;) Weird phase control chip ran mains commutated SCRs for single or three phase, half controlled bridge but that okay 'cos batteries like a short circuit with a voltage offset, big DC inductor about half the size of the mains transformer. Not viable these days. Noisy too, harmonics must've been terrible. GMH ran rows of these with second shift batteries for their fork trucks. Too much information?

Battery was isolated from earth but the charger control ran on +/- 15V rails 'grounded' to battery positive. So current shunt was near circuit ground.

Old boss wont be annoyed I tell you 'cos he's dead. My first electronics job made big impression on me 'cos we used same control boards for up to

500A traction battery chargers.

Grant.