Solar Panel to charge CPAP LI-Ion Battery - Voltage Regulator Problem

A little explanation of the "why" before I get into the details of the "what":

My parents both require CPAP machines to sleep, as they have each been diagnosed with sleep apnea. They both also love to camp, though, so they purchased batteries to run the machines and enable them to do so. The batteries alone will get them through ~3-4 nights. I bought them a solar panel so they could recharge the batteries during the day at their campsite and extend their stays in the wilderness. The gift is unfinished, though, as I will now explain.

I'm looking for the best solution to regulate or step down the voltage from the solar panel- from it's peak 18V to the battery's 12 volt input requirement.

I believe the battery has a protection circuit, so I'm not worried about cutting the current when the battery is full. The way I see it, the AC Adapter that would normally be plugged into the battery to charge it doesn't need to be unplugged when it's full, so why would my solar panel?

I tried to create a simple voltage regulator circuit using two capacitors (100μF and 10 μf) and the following regulator (part# 29300 12WT):

http://www.micrel.com/_PDF/mic29150.pdf

The circuit I constructed is, I believe, exactly like the one found on page 14 of that data-sheet, and is pictured below.

However when I tried to connect a variable power source to the input and measured the output with my voltmeter, the output voltage would vary with the input, even beyond 12 volts. At 18 volts input, the output was above 15 volts, which is too high for my application, and signifies I'm likely doing something wrong, or I have the wrong parts, or something. I attempted to change connections to different pins on the regulator, to no avail.

I constructed a similar circuit with a different 5V regulator in an electronics lab at my school, and the same thing occurred. No matter what voltage I gave as input, the output would change with it, far beyond the 5V level. A circuits teacher in the lab thought I had everything connected correctly, but couldn't figure out why the output was varying, and didn't have enough time to look into it.

Here's a picture of the current setup, please excuse the horrible soldering (I've had to resolder a LOT while trying to find the problem, not to mention the tip on the only soldering iron I have available is very fat). I am aware that the positive input wire is black, it does not confuse me and I am not mixing it up.





The solar panel I have is 30W and outputs an average of 18V (depending on sunshine). Edit: Product page for the solar panel (30W Model):

http://www.solarcynergy.com/Solar-PV-Panels.html
Important info:
Max Power: 30W
Open Circuit Voltage (Voc): 21.6V
Short Circuit Current (Isc): 1.93A
Optimum Power Current (Imp): 1.74A
Optimum Operating Volatage (Vmp): 17V

Here's the product page for the battery:

http://www.batterygeek.net/v/vspfiles/Super_CPAP_Battery_Pack_222Wh.asp
Important info:
charge current: 2500mA
charge voltage: 12V (read off of charger itself)

I would greatly appreciate any help!

 

That's a Li-Ion battery; you don't charge them from a constant voltage source like you're trying to do. Please don't try using your circuit to charge the battery.

The specifications on the page you linked to weren't very clear.

Exploding Li-Ion batteries are not a good thing.

 

a simple google search for "li ion solar charger:"

http://www.google.com/search?hl=en&rlz=1C1GPCK_en___US315&q=li+ion+solar+charger&btnG=Search

leads me to think that it is possible to charge the batteries I have with a solar panel. If I can't do it in the manner described above, how can I?

 

Especially not at five hundred bucks a pop.

 

I agree with both of you that expensive batteries exploding wouldn't be good. However people seem to be using LI Ion batteries with solar panels all over the place. For example, Tesla Motors offers solar panel installation along with their LI Ion powered cars to charge them completely off-grid. I know there's a way to do this, can someone please point me in the right direction?

Should I create a separate post specifically for my inquiry about the battery being charged by the solar panel, leaving out my attempt to use a voltage regulator? I would still like to know why the regulators are putting out unstable voltage, though.

 

Lithium Ion batteries are extremely picky about charging. The heat and terminal voltage need to be monitored with a constant current charge, then the current is monitored with a constant voltae charge for a very short period, switch back to the current at a different level.

The batteries have "Charge Protection" built in, which is nothing like a charge controller. The protection is meant to kill the battery in a non-explosive way when charged improperly. A Couple notebook manufacturers bought into the marketing that the new Lithium packs had charging built in, so they skimped with their onboard charge control. We all remember the flaming notebooks, right?

There are a wide variety of Charge control/battery maintenance ICs on the market that require power in, and some support components. One of those should have the ratings you are looking for.

 

The specifications also say "CPAP Battery Protection Built in circuitry protects against thermal run-away and overheating."

Judging from the photo of the battery and its charger, I'd be surprised if the actual charger isn't inside the battery and the "charger" is simply the same type of power supply that would be used to power a laptop. It sure looks like one, at least. If the "charger" says it's 12 volts and so many amperes and doesn't mention a thing about being an Li-Ion charger, then I wouldn't be too worried about replacing it with something else.

I would test it by connecting test probes to it while it charges the battery and watching its voltage as it charges the battery. If it's constantly 12 volts, and remains 12 volts even after the battery is fully charged, then I'd be very surprised if it isn't simply a 12 volt power supply.

My concern is how well the charging circuit will function if the solar cells aren't producing the current that the charging circuit expects, and so the input voltage falls below 12 volts. A solar cell charger which damages a $500 battery would be a terrible gift.

Anyway, the datasheet (which strangely doesn't seem to contain a pinout) has this to say:

Code ( (Unknown Language)):

1.  
2. Minimum Load Current
3.  
4. The MIC29150–29750 regulators are specified between
5. finite loads. If the output current is too small, leakage cur-
6. rents dominate and the output voltage rises. The following
7. minimum load current swamps any expected leakage current
8. across the operating temperature range:
9.  
10. Device                               Minimum Load
11. MIC29150................................................5mA
12. MIC29300................................................7mA
13. MIC29500..............................................10mA
14. MIC29750..............................................10mA
15.  

So try connecting a 1K resistor as a load and see if that helps. Assuming it does, I would definately make it part of the circuit rather than rely on the connected load to always draw the minimum current.

Anternately, I don't believe the 7812 has a minimum load.