Yes, sure, never design or build something you can buy for a fraction of the price.
That is often the case with things like amplifiers, TV's, NiMH chargers, what not.
Still it may be fun.
Anyways, I just took the gel battery specs, it told me the max charge current,
charge time, charge voltage, and a lot of other interesting stuff.
I already had the power MOSFET switch part (you cannot buy that), the PIC
with plenty of analog inputs, so on suggestion of somebody here, I used the
voltage drop over the MOSFET to measure current (it is almost exactly 20 milli Ohm).
And the rest is just basic math in the same PIC (that also drives a 4 line text LCD display).
I dunno total cost, but I think I can match that 600 $ price tag onhttp://
www.blueskyenergyinc.com
This gel battery has now had about 1200 full charge / discharge cycles,
with very little decrease in performance I can see, so I must be doing something right,
If something does not work, then I can fix it, an other advantage of DIY.
And in this project, using proto board, the design was practically rebuild 2 times..
If I have some idea I can just try it out, added a second PIC a few weeksago.
It has remote control from the PC via RS232, and manual control too.
I can read battery status from anywhere in the world, and control it too.
(It powers an amateur radio transmitter).
I can even make it play audio files from anywhere in the world, as the audio
comes via the PC, digital processor.
Or have the receiver record something....
Also on a timer (using crontab).
Now that is something you cannot buy anywhere...
Any item you buy has implicit some liability insurance associated with
it. Thus if it causes major damage, you can sue the manufacturer. Of
course, this explains why all the wall warts comes from China. "Honest
judge, the ACME manufacturing company doesn't touch the AC power line.
Blame XXXX for that dangerous wall wart."
Hopefully you have a watchdog timer in your charger. You can't take
the risk that the PIC goes off in a loop and whacks the battery. Also,
current sense using the power fet seems way too inaccurate.
Generally most manufacturer provide an independent battery protection
scheme as a safety, i.e. independent of electronics. In nicad packs,
it could be a PEPI. Having designed charger chips, I know you need
multiple protection schemes.
While you may have considered the possible states of a healthy gel
cell, you also need to consider the state of a gel cell on it's last
legs. You don't want to charge a battery that is not healthy.
I want to set up some gel cell charging scheme where the 3rd row of my
SUV resides when folded. [That is, yank out the seat and have a shop
build a lid, then recover the space for electronics.] There are plenty
of battery isolators on the market, but most charge controllers from
the solar and wind market can't handle the low drop out of being
charged from an alternator. However, the Blue Sky products look
promising:
http://www.blueskyenergyinc.com/
The trick with all these chargers is to interpret their nebulous
datasheets.
You want to run that car on gel batteries?
Or just use the car to charge those?
I want to run accessories on the gel cells. It's a bummer to have your
notebook computer or whatever drain your car battery when you are in a
remote area. So with the car not running, the gel cells are the power
source. With the car running, the alternator is the power source. That
is mostly what battery isolators do, and they would be fine if the
auxiliary battery could take the alternator voltage. [The isolator
uses relays to prevent reverse power, i.e. the auxiliary battery does
not get used to start the car.] Gel cells cant take as much voltage as
"regular" car batteries since they don't vent.
To make it more complicated, I'd like to not use switchers. I want to
be able to use ham radios and scanners from these gel cells, so I
don't want the RF hash from a switcher. So basically I need to provide
the float voltage to the gel cell from the alternator, i.e. high
current low dropout regulator with a bit of tempco compensation for
the gel cell. Reverse power needs to be prevented. Now this regulator
will see the load of the electronics, which could be high if a HF
transceiver were attached. Thus you wouldn't current limit this
regulator, but put a bidirectional current limit on the gel cell.
Probably something like a polyswitch would do the trick, though they
have really wide specs.
What I do now is for long periods in the boonies, I bring 5 gel cells
that I charge prior to the trip. For charging, I use an off the shelf
3 state charger since I don't care about switching noise in my garage.
These are 50 to 70 AHr batteries picked up on the surplus market (data
center pulls). When a battery reaches 10V, it gets retired and I use
the next battery.
