LIFPO4 14500 ...Unregulated Solar Charger Repercusions?

[nickelflipper]

Have been playing with LIFEPO4 AA batteries in a garden/landscape light. Have tried an unregulated solar charging approach (the mini panel is rated at 5.5V 100ma), other than a schottky blocking diode. On a bright sunny day I have found the battery upwards of 4.5 to 5.1V max. The battery has not exactly been hot, and is trying to push less 1 ma of current.

Thoughts on longevity or anecdotes of such an application?

Just got a high voltage limit circuit fired up with a LM393 comparator, but it has not been sunny enough to see it function "in circuit".

 

[timmmm864]

I'd be interested to hear how you go with your voltage limit circuit.
I have been fiddling with a simple solar charging setup also.
A couple of 5.5v 180ma rated panels connected in parallel along with a blocking diode. They effectivley charge up a couple of nimh AAs(differrent chemistry batterys to your LIFPO4 I know) but I haven't been able to figure out a simple way of judging cutoff (other than timing & approximating amount of charge delivered - not much use if the AAs are only partially discharged to begin with)

 

[nickelflipper]

I'd be interested to hear how you go with your voltage limit circuit.
I have been fiddling with a simple solar charging setup also.
A couple of 5.5v 180ma rated panels connected in parallel along with a blocking diode. They effectivley charge up a couple of nimh AAs(differrent chemistry batterys to your LIFPO4 I know) but I haven't been able to figure out a simple way of judging cutoff (other than timing & approximating amount of charge delivered - not much use if the AAs are only partially discharged to begin with)

To do it correctly, it's not exactly simple. I wish I had a schematic drawn up for it, but I don't at the moment. A LM393 comparator circuit similar to this one here in post #13 would be a start.

For my setup, I had a forward voltage drop of about 0.44V on the 1n4148 diode, which is used as the reference voltage for one of the comparator circuits. A voltage divider was calculated for the cutoff voltage, which in my case was approx. 3.6-3.7V.

So Vin equals the cutoff voltage 3.6V, Vout equals the Vref or 0.44V. Apply that to the voltage divider equation of Vo = (R1/(R1+R2))*Vin or R2 = R1/((Vin/Vo)-1). My divider worked out to be R1=100k, and R2=13.7k. The same could be applied for the nimh cutoff voltage.

I am using a pfet (like the FDN338) as a high side switch between the solar panel output and the blocking diode. The pfet gate is hooked up to the comparator output, which is normally low, and thus turned on. When the cutoff voltage is met, the comparator goes high and the pfet is turned off, and charging is stopped. It will might take 2 or 3ma to run this circuit.

 

[VegasF6]

Sounds like an interesting project. Do you think you could integrate a proper charge control IC such as the MAX1555? I am looking at it at sparkfun, but it shows out of stock. But Maxim still shows it in production so should be availible from the usual suspects.

 

[nickelflipper]

Sounds like an interesting project. Do you think you could integrate a proper charge control IC such as the MAX1555? I am looking at it at sparkfun, but it shows out of stock. But Maxim still shows it in production so should be availible from the usual suspects.

Doesn't look like that chip would work for me because it's the wrong chemistry and needs a working voltage of 3.7-7.0V. That, and my app is for a garden light, which is cost sensitive, and perhaps a little more forgiving being LIFEPO4. Otherwise,that chip looks good for Li-ion, as long as the power requirements can be met.

Looking back, using a comparator circuit for a charging circuit, needs careful consideration, no matter the chemistry. In my case, it is one step above an unregulated circuit, and budget related.

 

[VegasF6]

I did find other ICs at Mouser that are lifepo4 specific.
http://www.mouser.com/Semiconductors/Power-Management-ICs/Battery-Management/_/N-wnwk?P=1yzxj8e
$1.22 each for the MCP73123 or MCP73124.

You supply voltage is 5.5 at best, correct?

Obviously 4.5 to 5.1V that you have found your batteries at are severely over charged. Though lifepo4 is considered a safER chemistry, they can certainly vent, as can any battery. At the very least the life cycle of that cell could probably be counted on 1 hand

Have you seen this example of a safe chemistry venting?
http://www.candlepowerforums.com/vb/showthread.php?t=262234
Those are the Sony Konion cells.

 

[KiwiMark]

On a bright sunny day I have found the battery upwards of 4.5 to 5.1V max.

Ooh, that is pretty high. I doubt that it will fail in any kind of spectacular way, LiFePO4 cells are pretty safe. But even if the cell lasts for a year or 2 I can't see the performance being what it should be even after only a month or 2. The cells performance will surely be headed on a downward slope quite quickly with that level of over charging.

 

[nickelflipper]

I did find other ICs at Mouser that are lifepo4 specific.
http://www.mouser.com/Semiconductors/Power-Management-ICs/Battery-Management/_/N-wnwk?P=1yzxj8e
$1.22 each for the MCP73123 or MCP73124.

You supply voltage is 5.5 at best, correct?

Those maxim chips still presume a DC voltage that is well above the intended battery (voltage) that is to be charged. That works if you're strapped to a wall wart, USB port, or say a car battery. There is a microcontroller involved in the project, and if absolutely necessary, it could be upgraded to takeover charging responsibilities.

The mini solar panel can at best produce maybe 5.2V connected to the battery because of the forward drop of the blocking diode. For the most part, that range would normally be between say 3.0-3.4V. On a warm bright sunny summer day it can go above the 3.4V. No signs of venting yet, and by all means I intend to keep it that way with a 3.6V limit circuit in the future.

Wow....that's a lot of power in a battery pack like that.

There is one battery that has been probably abused in an unregulated way for a couple of weeks. I intend to keep track of it to see how it keeps up with the others for comparison.

Thanks to everyone thus far for their suggestions, and insights.

 

[timmmm864]

To do it correctly, it's not exactly simple. I wish I had a schematic drawn up for it, but I don't at the moment. A LM393 comparator circuit similar to this one here in post #13 would be a start.

For my setup, I had a forward voltage drop of about 0.44V on the 1n4148 diode, which is used as the reference voltage for one of the comparator circuits. A voltage divider was calculated for the cutoff voltage, which in my case was approx. 3.6-3.7V.

So Vin equals the cutoff voltage 3.6V, Vout equals the Vref or 0.44V. Apply that to the voltage divider equation of Vo = (R1/(R1+R2))*Vin or R2 = R1/((Vin/Vo)-1). My divider worked out to be R1=100k, and R2=13.7k. The same could be applied for the nimh cutoff voltage.

I am using a pfet (like the FDN338) as a high side switch between the solar panel output and the blocking diode. The pfet gate is hooked up to the comparator output, which is normally low, and thus turned on. When the cutoff voltage is met, the comparator goes high and the pfet is turned off, and charging is stopped. It will might take 2 or 3ma to run this circuit.

Thanks for the response - I am a bit of a noob with electronics so I will need to nut out how the comparator circuit works, and then see if I can apply my new found knowledge -Tim