simultaneous homebrew batteries charging/use?

February 21, 2006 10:10 PM

Electronics question - can I recharge batteries while using them?

I'm building a bicycle generator and I'm trying to figure out the ideal setup. I'd like to be able to have a battery backup and then charge the batteries while I'm using them if possible. The setup is going to be a 12v dc motor used as a generator going to a 140w power inverter meant for a car ciggarette lighter. I have 3 12v 12ah SLA (sealed lead acid) batteries that were initially used for an electric bike but I might hook up in parallel. I also have this UPS but it likes to beep at me whenever its doesnt have enough power and I havent tested out how long it will last but I doubt it will power a laptop very long. I've stumbled across this which has terminals for charging and discharging, which gives me hope. I will probably use this generator to power a laptop/charge my cel phone/ run LED's/ maybe run a fan or a little stereo.

I would like to know if simultaneous recharge/use is possible with SLA, NIMH, NICD, or LION (kind of expensive but will consider) batts. Bonus points for explaining or pointing me to how to accomplish this. Lastly, is my idea for a setup ideal? this guy who is the master of bike generators says that my idea looses efficiency, but he's kind of hard to follow. here's a bunch of extra links about bike generators from 2000

posted by psychobum to technology (7 comments total)

If I understand you correctly, you're hooking up your bike to a 12VDC motor to power an inverter...to run a battery charger.

Since you're charging batteries, which by their very nature are DC components, you're probably better off by building a circuit to regulate the amount of current that's flowing into the batteries. There are intergrated circuits that do this for you (MAXIM-IC, is one good source). You want to do this (instead of what you propose) because you're losing efficiency by converting dc to ac then back to dc again.

Also, you can run stuff off your setup as you charge the batteries. You need to be sure that the power drawn by whatever that's running is less than the power supplied by the motor, otherwise you would be drawing power from your motor and your batteries.

Last point about batteries: there is a voltage-charge relationship that are specific to the different battery-types. Your circuit needs to follow this relationship closely when you charge them otherwise you risk damaging the cells.

The best source for info on how to charge batteries is the ARRL Amateur Radio Operators Handbook, which you can probably find at the public library.

FWIW YMMV, etc.
posted by scalespace at 10:58 PM on February 21, 2006

...the best source I know for info...

I'm sure there are better sources wrt technical accuracy, but the ARRL handbook is pretty straightforward and easy to understand.
posted by scalespace at 11:00 PM on February 21, 2006

As scalespace points out, that's a lot of expensive (efficiency-wise) conversion you've got going on there. Ouch!

It sounds like he totally has it. Apart from seconding what he said, and suggesting you look into whether it might be more efficient to use an alternator setup (I don't know -- EEs help out here) I really don't have much to add.
posted by Opposite George at 11:43 PM on February 21, 2006

IAAEE. What you want to do is eminently possible but you need to keep in mind that the charger is effectively supplying the load as well as filling the battery, so the battery will charge only very slowly. So you need to make sure the charger is grunty enough for a start. Off-the-shelf battery chargers may or may not be, more importantly, your legs may or may not be... Consider 200W a reasonable cruise power, lose 40% to the generator+electronics and you're left with not much more than 100W, use that as your energy budget. If the laptop supply wants 200W, the batteries will be going down not up, even if you're pedalling.

How practical it is will depend on the type of cells. NiCd and Li-Ion are very fussy with respect to their charge profiles, NiMH somewhat so, SLA (Pb-H+) not very much at all. That means that you can probably do it relatively simply if you stick to the SLAs.

The efficiency of the generator is maybe 50-80% but you can't avoid that. An inverter is probably 80-90% depending on load and power factor. A charger expecting 240V input is probably also about 70-80%, so charging the batteries via the inverter and charger is horribly inefficient. What you want to do is build (and I have no idea if you're capable of this, if not you might find a tame EE friend to do it for you for fun and cost of parts) a charger that runs directly from the generator.

The motor: check if it's the permanent-magnet or has a stator winding ("parallel-wound" or "series-wound"). If the latter, it cannot be used as a generator: it will spin freely and do nothing. A PM-brushless-DC can be used but requires major surgery. If it has separate stator and rotor windings you can do very tricky regulation of charging by varying the stator current; this is what car alternators do, so....

I would recommend that you don't try to design all this yourself; just get a complete charging system out of a small car from the wreckers. It should include alternator, regulator and battery and cost maybe a couple hundred $$. This does everything you require: charges a big lead-acid battery from rotary motion and assumes you can take power from it at the same time - that's exactly what happens in a car with headlights, radios, etc. If you want lots of storage capacity, just put a bunch of big truck batteries in parallel and be prepared for a few weeks of hard pedalling just to get them fully charged. Don't actually physically connect them in series until they're all fully charged already.

Once you've got that working, you can attach your inverter directly to the battery terminals and use it. Don't over-discharge your battery.

It is possible to build or buy a thing called a DC-DC convertor; if you can find one with 12V in and 18V output (usually), you can use that to directly power your laptop. Most laptops want 18VDC (look on the back of the brick), so inverting up to 240VAC and then using the laptop power supply to drop back down to 18VDC is inefficient. The DC-DC converter works on the same principle (switched regulation) as a laptop power supply and the inverter, it just does only one conversion (the one you want) instead of you composing that conversion out of two off-the-shelf but otherwise inappropriate ones.
posted by polyglot at 2:24 AM on February 22, 2006

Thanks for the responces so far. I'm definetly going to check out the AARL handbook tomorrow (providing I don't owe the library too much), from what I'm hearing I should figure out how to make a decent voltage regulator. I saw the DC-DC converter on batteryspace.com for $30 which would also be useful for running my newish 18v laptop off of those SLA's when I do time lapse. It sounds like I'm best to hedge my bets with choosing the most energy efficient things to run like an old school toshiba satellite pro vs. my newish gateway. Also figuring out the best/safest way to stick with DC as opposed to running it through an inverter. I'm going to invest in a kill-a-watt which will tell me how much power my appliances want to use (I've found them online as low as $21). I also think I need a "watts up" for diagnosing the batteries/charging situation but they are $60 which seems a little steep- there's probably a multimeter that will do most of the "watts up"s reporting for much cheaper tho i'll admit i havent looked yet. As far as the car alternator setup - so far this guy has been on the money and he says the car alternator is not a good idea :( . I haven't blown too much money (about $25 for the motor, bike trainer, and a cheap mulimeter) and it sounds like I could save the most if I could do the electronics myself so I'm going to try that route. Thanks again.
posted by psychobum at 3:58 AM on February 22, 2006

Ask more specific questions about the pedal powered generator article. I can see why you say it is hard to follow, but the author seems to have the right idea.

polyglot's answer is great. In addition to the concerns about the efficiency of the electronics, the batteries themselves will lose power as heat while charging (called coulometric charging efficiency, NiMH are only 66% efficient).

Remember that the voltage of the battery terminals will be a higher while charging than discharging. So the load will see a higher voltage while the battery is charging, this might not be a problem, but it should be considered.

The powerstream site I just linked seems to have all kinds of useful information, here is their article on Sealed Lead Acid Battery Charging Basics. There is also a lot of useful information about charging in the Xantrax Truecharge FAQ.

Here is a page that discusses hacking a car's alternator/regulator. The pedal powered generator article suggests that a car alternator probably isn't the best idea though, for pretty convincing reasons.

Finally, that battery pack you link to has two sets of terminals strictly for convenience. It's main use is probably powering RC cars/planes. You would want to charge it without having to disconnect or remove it, so an extra set of terminals is provided. In this context the load would be turned off with a mechanical switch while the battery pack is charging.
posted by Chuckles at 4:25 AM on February 22, 2006

psychobum: check out Chuckles' alternator-hacking link which describes how the regulation works - this is critical. There is no reason why alternators need to be run at high speed; in fact they can work quite happily at engine idle (700 rpm, about 1200rpm on the alt). At 1200rpm you're not looking at very significant windage losses; down near 600 they'll be bugger-all. By adjusting the field current (fiddling the regulator) you can achieve the output you desire, even at 600rpm (6x pedal speed). You may not even need to adjust the regulator at all for it to produce 14V at this speed if it has enough headroom.

I wouldn't discount an alternator at all. Yes, it may be a few percent (2-5% would be my stab in the dark) less efficient than a PMDC motor but that is more than made up for by the fact that all the electronics is there and ready for the using.

To use a PMDC motor effectively and get its full efficiency, you're going to be building a buck mode switching regulator. That is a difficult task even for someone who has built such things before, it would be an extremely difficult problem for a fresh EE grad to get it to run efficiently (let alone not burning down when turned on) and if you have only a passing knowledge of electronics, then forget it. You can do without the regulation electronics, but you'll throw away 20-30% in efficiency because you're running the batteries at too-high voltage and heating them while charging. In that case, you're worse off than with an alternator.

One last option is to go mechanical. If you include a derailleur system that allows you to change ratios and a voltmeter on the motor, you can change gears & cadence to keep the charging voltage near 14V. That would be efficient.
posted by polyglot at 1:26 AM on February 26, 2006