charger with constant current

--
Why not?

As long as the regulator can dissipate the power which it\'ll develop
with the voltage across it and the current through it there should be
no problem.

You'd probably do better dropping 18V down to 12V with a buck regulator - not a boost converter. The market is full of chips that do this with a minimum of external components.

A simple start point for a home brew if you want to regulate current, start with a pass transistor and buck coil, then use a ground sensing comparator with a little hysteresis to detect the volt drop on a current sensing resistor in the -VE rail, use the comparator to flip a S/R flip-flop which switches the pass transistor on and off.

You should approach the problem in a totally different way. Mostly by deciding what you need and pursuing it, instead of flopping around trying this and that and the other thing and wondering what's going wrong.

First, you should think about the battery chemistry. Since you didn't mention it, I assume you're not thinking hard about it. If you have a

6-cell lead-acid battery, then you almost certainly want a constant voltage charger, or a current-limited constant voltage charger. If you're charging a 10-cell NiCd pack then you can get by with a constant current; if you're charging a 10-cell NiMH pack then you probably want a constant current with peak detection, or a _really_ low current constant current charger and an acceptance of reduced battery life.

Once you've figured out what battery chemistry you're dealing with you should look to the flavor of battery that you're using -- automotive batteries have slightly different requirements than sealed lead-acid, high-capacity NiCds often can't be charged as fast as high-current ones, etc.

Get yourself a copy of "Rechargeable Batteries Applications Handbook", it discusses chemistries and charging methods (although it was written before Li-Ion and Li-Po batteries came along, so you're on your own with those).

Finally, once you know what the charger _should_ do, _then_ start thinking about how to make it do the right thing.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Do you need to implement control loops in software?
"Applied Control Theory for Embedded Systems" gives you just what it says.
See details at http://www.wescottdesign.com/actfes/actfes.html

Hi,

I am using a 12v sealed lead acid battery (being charge). My aim is to charge this battery then the circuit will charge a second 12v battery. Yes I want to do this way. I have a Pic chip that determines the battery voltage and transfer the charging to the desired battery. the problem now is that the pic chip sees a fully charged battery when in fact it only begun charging. The reason for pulse charging, I read it was better for batteries.. nevertheless, my pulse charging seems not to be working as it sees the maximum charge at the pole of the battery. How do they do it so that the battery only retains the average voltage on the charging battery..

K

Detecting the state of charge of a battery from it's terminal voltage alone is just about impossible. You can have a near-death cell that, when you give it a rest and measure it with a high-impedance meter, will read full charge.

Your best bet at detecting charge would be to look at the current that goes into the battery when it's getting its constant-voltage charge -- current below some threshold indicates that the battery is charged.

When you're all done charging both batteries, you probably want to put them on a float charge.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Do you need to implement control loops in software?
"Applied Control Theory for Embedded Systems" gives you just what it says.
See details at http://www.wescottdesign.com/actfes/actfes.html

You could add a switchable load resistor to test the battery for state of charge. Once the voltage under a moderate load is measured, you can apply the recommended charging current for a period of time, or until the target voltage is reached, at which point the charger would switch to a float voltage charge. You could switch in the test load from time to time to check the battery's condition.

Pulse charging should be OK, although I do not know if it has any benefits or drawbacks. But you need to measure the average voltage of the battery to determine its condition while charging, and the pulses may interfere with an accurate measurement. You can use many A/D readings and perform software averaging, but you will also need some RC integration to reduce spikes and noise. If the PIC is supplying the PWM, you may need to synchronize the A/D readings to include the voltage during the charging pulse as well as that when it is off.

There should be design notes on the Microchip website, or you could check the user forums at:

There are 63 hits for "Battery charge":

And there is a "Battery management design center":

Paul