Unusual Plan for Solar LEDs

[ghurd]

I put a couple tiny battery powered LED bulbs in a shed a year and a half ago.
Uses 4 alkaline D batteries.

The bulbs are very underpowered on 6V, by design, because the idea was for the batteries to last a lot longer.  The bulbs could operate at 12V and still be fine.

I want to solar power the system, but I want the system to work even if there is no sun for days, and somebody leaves the lights on for 3 days.

Here is where it gets interesting:

The solar power will be stored in a 5 cell 1300mAH NiCd pack.  6V nominal, about 7.2V max.
(Thanks Norm and Bruce!)

What happens when the NiCds get low?
The alkaline pack kicks in some help!

After everything gets "Broken In", the alkaline pack should kick in when the NiCd pack gets down to about 1.1V/cell.

The alkaline pack will supply enough to keep the bulbs running a long long time, which will keep the NiCd pack from getting so low as to be a danger of cell reversal.

The solar panel will probably be 16 cells, about 7.2Vmp, about 0.6Amp.
If the sun ever comes back out when I have time for quick testing, results may show it needs 18 cells.  Given typical solar cell power curves, and the ratio of charging amps to capacity, I expect 16 cells is enough even though it looks a little bit low.
More common ratios would be better served with a higher Voc (meaning 2 more cells in the panel, or 9V).

Charge controller is the caveman style 2 series diodes per cell type.  Hardly great, but functional.

The system is not state of the art.
It is something a regular guy could do without a lot of money, or a lot of experience and skill.  I think it is a neat project for a newbie to get his feet wet with something that actually works and is useful too.

I am in dire need of a distraction lately, and this is it.
G-

For size considerations, all the parts are not shown, like the diode count and cell counts are technically incorrect.  You get the idea.

[MadScientist267]

Not sure I'm following the purpose here, nor parts of the design.

What its obvious: LEDs get their power from one of 3 sources... PV, NiCd, or Alkaline.

What isn't so obvious: Why diodes across the NiCds, and a schottky in series with the alkaline. I get that you don't want to charge the alkaline, but...

Observations from my side of the table:

The diodes across the NiCds are presumably for voltage limiting during charge? NiCd don't need this provided chernobyl isn't the charging source. Diodes used in this way also are close to their Vf, and as the number gets close, current rises. Result? A nice steady discharge from the NiCds into the diodes.

Second, the schottky between the alkalines and the rail would prematurely load your D cells. A regular run of the mill 1N4001 would do the job nicely. 5x1.25V (full nicd) = 6.25V, 4x1.6V (full alkkalines) = 6.4V, -0.3V drop (at the currents you're pulling) would keep your D cells nice and fresh.

You can add a second 4001 in series if you really wanted to make sure the NiCds were toast before the alkalines took over, but....

Steve

[Norm]

Kinda what I did Glen....Have 3-4 LED lamps originally run by 3 D alkaline
each then as they finally ran down I switched to 3 Ni Cad each then I was
always thinking of making each lamp dual powered  ....the alkaline would be
a back-up in case the Ni-Cads didn't get charged that day (I usually PedGen
charge them .....a simple morning exercise of about 20 minutes) or which
could be charged by solar or small WT ?

Somwhere in here we could use a trickle charge to the alkaline....
that switches to the NiCads when they need charged and the LEDs
then switch over to the alkaline ...I like the alkaline running LEDs
with their internal resistance higher  than NiCads the LEDs don't or can't
suck a lot of current thru them.

I think you're on the right track...keep goin'
Norm.

[ghurd]

Steve,
It is Chernobyl!  0.6A into a 1.3AH NiCd pack.
Debating about doubling it to 2.6AH.

I always had great luck with keeping the charging amps low, like C/10 ~ C/5.
But it is solar, so it will not be like someone throws a switch and sends 600ma into the NiCds.
Current will increase as the sun comes up.

And because of where the Vmp is, in relation to the max battery voltage, the charging current should drop quite fast as the battery voltage increases.  Sort of the same idea as those "self regulating" 12V 30-cell solar panels that didn't work for crap on FLAs?  LOL

"the schottky between the alkalines and the rail would prematurely load your D cells"...
Not with my math.   Thats where 'After everything gets "Broken In"...' comes in.   
I went with 1.2V for the NiCds, or a hair less.
The alkalines won't stay at 1.63V very long.  I went with 1.5V and less.
Should keep the NiCds around 1.1V min.  I don't like taking them lower than that because it seems to me it takes more power (relatively, like it needs more AH in than it should [never charted it with a computer, just observations over the years]) to get them recharged if they get down to say 1.0 or less.


Norm,
It really won't be switching back and forth between packs.
When the NiCds are low, the alalines will start supplying part of the LED curent.
The 2 packs will share the load.

It would be a GREAT place for a tiny WT.
Not a project I want to get involved in now.
Time, parts cost, welder-less garage, etc.

Something else might be added too.
An idiot light.   
Crazy bright red LED with a 10K resistor right above the padlock would be enough to let whoever locks up the shed at night (usually midnight to 2:30AM) to remember to shut the lights off!
"Gosh.  It is pitch black out here.  Whats that red light 2" above my fingers?"
They'd forget to turn them off 3~4 days a week, and since they are not designed to be very bright, in the morning the guy who opens it (admittedly, that is usually me) may not notice they are on.

G-

[Norm]

My 4 LEDs +1 runs on 3 NiCads and will be also have the 3 "D"
cells....I just need it designed to switch to the "D" cells when the
NiCads drop to nominal voltage, and turn on a tiny red light when it's
running on the "D" cells.
Can you show me how?....I'd really like to know

I'm going to start working on  adding the 3 "D" cells right now

......It would be nice but not really necessary to
have a small indicator green light to tell when the NiCads
were charged.

Here's your thought...
  Should keep the NiCds around 1.1V min.  I don't like taking them lower than that because it seems to me it takes more power (relatively, like it needs more AH in than it should [never charted it with a computer, just observations over the years]) to get them recharged if they get down to say 1.0 or less.

 

How about if you also charge the alkalines?
the same rule seems to apply?
seemed like it was much easier to keep the alkalines charged
if I charged them every day .....alkalines don't use as much
current as NiCads to run the same set of LEDs.

I can only imagine how much less charging it would require
to replace the amount of electricity the alkalines use over
what is required to run my 4+1's with the NiCads.

I run my 3 lamps for about 6 hrs. and recharge all nine of
the NiCads at once (in series) in about 15-20 minutes (about 10
watt/hrs.)

So my theory is ....it's more efficient to recharge alkalines .....
use a little ...charge a little....use a lot.....charge a lot...lot...lot !

If you get the idea?
anyone?  or am I just rambling ?   LOL
Norm.

.....but I do think in the long run
using both kinds of batteries would give you the best of both worlds.

[ghurd]

Idicator lights are kind of a bugger to get working like what you want.
3 NiCds don't have a lot of voltage to work with, and that makes it even worse.

They make rechargable alkaline cells.
(wallyworld has them)

My Grandfather had a 1960s(?) battery charger that claimed to recharge anything.
Even had spot to put F... and "Z" sized cells?
It worked, but the batteries didn't last long after a recharge.

Problem is that with 2 sets of cells, the solar panel will be at the lowest voltage.
I suppose could add a diode and resistor to allow a few ma into the alkaline pack, but not sure if that is a great idea.  Seems I read somewhere it makes them leak?  Leaking could have been a result of trying to charge them too fast, or too much, instead of just a few ma to maintain them?

Another problem with that, in my setup, is the alkaline pack would be cycling as much as the nicds, and I don't think that is a good idea in my setup.  I don't want the alkalines to cycle because they are just there for backup.
G-

[ghurd]

For those who have not figured it out, or later when everything is old,
this post is related to the post here,
http://www.anotherpower.com/board/index.php/topic,707.0.html
G-