On 2/7/2014 8:08 AM, Larry Bradley wrote:
> This is partly EE and partly PIC.
>
> I have a sailboat with a large battery bank (450 amp-hours) for house
> use (lights, refrig, etc). There is a large alternator and regulator
> that can easily charge the batteries at 70 amps or so. Unfortunately,
> the batteries are rather far from the alternator, and with the wire size
> I was able to use, when charging at high currents, the voltage coming
> out of the regulator can be around 15 volts. This does not present a
> problem, except when I also want to charge  engine starting battery,
> which is close to the alternator.
>
> At present I solve the problem manually. When charging the house
> batteries at a high rate, I manually disconnect the start battery so
> that it does not get too much voltage. When the alternator voltage drops
> down to less than 14 volts, I turn the switch back on.
>
> I also have a home-built battery monitor system based on a PIC 18F4260
> (measures battery voltages, currents, amp-hours, ...)
>
> I was thinking about using it to control the start battery charging. I
> can use a PWM module to drive a transistor switch, using the existing
> capability I have to measure the start battery voltage. This requires me
> to write code for a voltage regulator. Not a great problem, I expect.
>
> However, I was wondering if I have to get that complex. I was thinking
> about just using a "bang-bang" type regulator. Measure the battery
> voltage. If above 14 volts, turn off switch. If below, turn on switch.
> As I recall, this is the way that OLD car regulators used to work when I
> was a kid (the old mechanical regulators).
>
> This would be REAL simple. Unfortunately, the boat is 100 miles away,
> and buried in snow (this is Canada, after all!) and I don't have a 12
> volt battery here to experiment with.
>
> Thoughts from you EE types would be appreciated
>
>
>
Pulse-charging lead-acid batteries works very well, as long as the cells=20
don't overheat and the peak voltage is not exceeded. Proper charging=20
_IS_ simply a matter of monitoring peak voltages and cell temperatures=20
of ALL rechargeable batteries, but each type requires a different charge=20
current, a different cell temperature, and a different peak voltage.

In your case, having a long, smaller wire to the battery just means that=20
you must charge the second battery slower, to keep the wire from=20
overheating. A heavy electrolytic cap near the second battery might keep=20
the voltage up enough to pulse-charge the second battery. I'd use=20
something like 10,000+ Mfd at 25vdc or so, then make the pulses 1/3 to=20
1/4 duty cycle, which will allow the cap voltage to charge back up=20
during the off interval.

We do this with solar panels during the winter when solar level is at a=20
mnimum. The PV arrays  generate voltages high enough but the current=20
level is too low to charge the battery. To solve this, we charge a=20
capacitor until its level is high enough to send a proper pulse to the=20
battery. We then send a burst of energy to the battery.

We use a heavy P-MOSFET to deliver the charge to the battery.  EBay=20
sellers from China have the best prices for some reason, at 1/3 the=20
price of Digikey & Mouser. I am distressed at ordering from China but=20
being poor, I am forced to seek the lowest price.

Bob A

--=20

The only place success comes before work is in the dictionary.

VINCE LOMBARDI

--=20
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.