Forum Home
PC World Chat
Thread ID: 55858	2005-03-21 06:24:00	Sealed Lead-acid battery charging in a PC	george12 (7)	PC World Chat

Post ID	Timestamp	Content	User
336454	2005-03-21 06:24:00	Hi, As some of you know, I made a PSU using linear regulators to run a PC off 12v. It works just great, drawing a total of around 5 amps. But how do I keep the batteries charged etc. If I stick a "12v" unregulated power supply across the battery bank, it draws around 1.8 amps no matter how charged or discharged. This is fine when they're reasonably depleted, but when they're full they get hot and start boiling :o. I have no experience with smart charging, so need some help. By "12v" I mean it says 12v, but when unloaded it is 17v, and when the bank is across it it's around 14v. 95% of time the PC will be running from AC power. By that I mean a power supply makes 12v DC and inputs it into MY power supply, powering the PC. I want it in this state to just keep the battery bank topped up. But when the AC power disappears, the PC needs to be able to instantly begin running off the batteries. When external power is reconnected, it needs to charge up the batteries nice and fast, then go back to the start. Can I do it? How do I?	george12 (7)
336455	2005-03-21 06:43:00	I have built many chargers for SLA's over the years. All had very long life on the batteries so seemed to be well suited. Best results were by using a constant voltage source, set at about 14.4 - 14.7 or so volts. 14.7 is technically the correct maximum voltage from memory I used 3 terminal 12 volt regulators, with the ground lead offset by forward biased diodes, each diode adds about 0.6 volts (depending on diode characteristics) to the output voltage. That usually determines how fussy you get with the output voltage. 4 diodes = 14.4v You can use a resitive divider and make it continously adjustable. You need to protect against reverse biasing the regulator when no power is applied to it, but a reverse shunt diode across the regulator works. Temperature shutdown of a 3 terminal regulator nicely limits the charge rate, but SLA's are usually only designed for a slow charge. If you boiled a SLA, it sure will have a short life.	godfather (25)
336456	2005-03-21 06:47:00	www.rcbatteryclinic.com	godfather (25)
336457	2005-03-21 07:12:00	My collection has had an extremely tough life, but lasted fine. They came from electric scooters for kids up to 14 yrs old, so they're desiugned for a little abuse I think. When I was ten I used one to make sparks. Yeah, shorting it... I had one across a circuit of mine, that was off. But the circuit went bad about a week later without me knowing, and somehow managed to draw about 6 amps from the poor thing for about a week. It was down to three volts unloaded when I realised. And there's the (not sealed lead-acid) motorbike battery that got down to five volts, and had no acid at all left when I found it. It came back fine too. But they've all come through fine despite all this. Even though most of the stuff should have killed them. Maybe I'm just lucky. So if I apply a constant 14.4 volts to them, they will never overcharge, even after weeks? In that case, good. That's great. Next q, will my hard drive complain with 14.4 volts across it instead of 12? Because if I don't have the circuits separated, I won't have to have it "Switch" to batteries if the power goes out. The current voltage of the batteries is 13.46 volts. Is that overcharged? Because I read that 13 volts is 100%, and 12 volts is 0%. This therefore is 146%!? Thanks for the help!	george12 (7)
336458	2005-03-21 08:24:00	the problem with lead acid batterys if you connect them to a basic charger they over charge and they get stuffed very quickly. a good charger will detect when ful and revert to trickle charge. however it to will overcharge them but it takes days instead of hours. the best is a fully auto chrger which turns off after the battery is charged. these can be left on almost permantly. frankly it would be cheaper to buy a ready made charger than to make a decent charger useing a computer psu. open votage of battary is 13.46, it will be 12v under small load.	tweak'e (69)
336459	2005-03-21 08:39:00	As shown in the link I gave: "Lead acid (gel cells) should be charged with a constant potential charger specifically designed for these batteries. These are sometimes referred to as a CVC charger. You can charge them with a constant current charger but you must terminate charge when the voltage reaches 14.7 volts. You should not exceed the C/10 charge rate. If you have a 7 Ah battery in your field box the maximum constant current charge rate should not exceed 700 mA. It will take about 14 hours to charge from a fully discharged state (voltage less than 12 volts). A CVC (Constant Voltage Charger) is exactly what the name implies. It is clamped at a certain voltage and puts out all the current it can until the battery reaches the clamp voltage, usually something around 14.5 volts, then the current drops off to maintain it at this voltage. A constant voltage charger is characterized as one having a current capability of supplying a fixed voltage to whatever load is applied." In my standard design, 4 diodes in the ground leg of the 3TR gives 14.4 volts. Using 3 diodes = 13.8 volts. 12v 3TR exact output vary by device, you need to tailor each system accordingly	godfather (25)
336460	2005-03-21 11:34:00	Oh goodness, they're only 4AH batteries! Looks like my unregulated, uncontrolled 1800mA charging rate is a bit high! Now the thing is, how do I best integrate it all with the 12v PSU I made? I can't put 14.4v into my hard drives, but I need it to quickly switch to batteries. I can easily dedicate a data bit of the parallel port to detecting if mains is on, and switching over if it turns off, but is there a better way? I'm happy to build the regulating circuit you mention, but to lower the current do I just lower the voltage, or do something more complex?	george12 (7)
336461	2005-03-21 11:50:00	the problem with lead acid batterys if you connect them to a basic charger they over charge and they get stuffed very quickly. a good charger will detect when ful and revert to trickle charge. however it to will overcharge them but it takes days instead of hours. the best is a fully auto chrger which turns off after the battery is charged. these can be left on almost permantly. frankly it would be cheaper to buy a ready made charger than to make a decent charger useing a computer psu. open votage of battary is 13.46, it will be 12v under small load. Yeah, when loaded to 5A by the PC running it drops to 12.60v. But the thing was talking about the open voltage, hmm. They were cetainly fizzing in the morning. The issue here is that when powered I need 5a being supplied to the system at as close to 12v as possible, and anything between 1.2A (400mA per battery) and a few mA supplied to the batteries - at 13.8 or 14.4v. Maybe like a similar type of charging system to a car is needed?	george12 (7)
336462	2005-03-21 12:45:00	The fizzing is oxygen coming off the +ve plates and hydrogen off the -ve plates don't light up a smoke anywhere near them. This means they are in over-charge and should now be held at a constant voltage after reaching this state at a high current rate of charge, CVC as GF said. If you want to get pedantic you should have a negative temperature coefficient on this voltage for maximum battery life, I forget the value now, but that is going too far anyway for your modest system, IMHO.	zqwerty (97)
336463	2005-03-21 12:52:00	Go here to get an idea of what you could do: www.myra-simon.com	zqwerty (97)
1 2