Help me understand what I'm looking for: extremely low amp charge circuit.

eh4

Flashlight Enthusiast
Joined
Oct 18, 2011
Messages
1,999
I want to find a very small, hopefully very inexpensive charging circuit for lithium ion batteries which has either such a low overhead that it will trickle charge well under 100mA, or else keep the circuit intact if the mA were below the threshold for the charge circuit to operate.

Basic low power (100mA max)intelligent charging with the ability to intermittently trickle charge even a few mA, that's what I'm looking for.

Initially this is for a 1 watt, 6 volt CIGS panel. The panel is only 1 oz, and the idea is to keep the weight, volume, and complexity down, and be able to pot the circuit for full waterproofing.
Re: http://www.candlepowerforums.com/vb...-charging-18650-on-the-AT-light-weight-hiking

For a 4 oz., 5 watt version I'm hoping that the same can be done if the max charge were 1,000mA, while still allowing intermittent, very low mA charging.
 
Last edited:

eh4

Flashlight Enthusiast
Joined
Oct 18, 2011
Messages
1,999
Googling around, it looks like maybe this could be done with two parallel circuits: a dumb resistor and diode, or just a diode with enough forward voltage drop to limit the voltage to 3.9-4v for the float, and an intelligent charge circuit that would activate when the amperage rose to a considerable % of the max output in direct sunlight.
 
Last edited:

PeterH

Newly Enlightened
Joined
Aug 31, 2010
Messages
52
My understanding is that charging LiIon cells without a full charge cutoff is asking to overcharge, with very bad results.
 

hiuintahs

Flashlight Enthusiast
Joined
Sep 12, 2006
Messages
1,840
Location
Utah
Googling around, it looks like maybe this could be done with two parallel circuits: a dumb resistor and diode, or just a diode with enough forward voltage drop to limit the voltage to 3.9-4v for the float, and an intelligent charge circuit that would activate when the amperage rose to a considerable % of the max output in direct sunlight.
I've designed a few lithium ion chargers but they've all been based off stable input power such as 5v USB or 12vdc. That way I could maintain constant current.

I have seen a number of requests on CPF for something that just runs off of the raw solar power. The problem as you have recognized, is that to take advantage of the maximum energy delivered by a small solar panel without a storage device is to have some intelligent electronics that can modulate the current on an as need basis depending on the amount of sunlight. I know of nothing out there. You probably will have to come up with your own circuit of some type. It's an area that I have been thinking about for my next lithium ion project.
 

eh4

Flashlight Enthusiast
Joined
Oct 18, 2011
Messages
1,999
Thanks for the replies, I'm going to solve the easy part experimentally, with a battery that I'm ok with losing, though I don't think there will be any big problems.
It'll either work, not work, or there will be some weird feedback effect from the lower voltage circuit being intact while the charge controller is active. - that'll be beyond me.

Peter S., from what I've read so far, a continuous charge with a fully charged Li-ion battery is bad, but a trickle of less voltage than full charge should not do any harm, if the battery is at 4.2 then a 4.1v should deliver 0 current.

hiuintahs, I'll keep you updated here if anything useful comes of it.

Anyways I just wanted to see if someone could tell me before I made a mess. This is probably going to be a pretty yawn inducing experiment, whatever the results, unless one is excited by the prospects of slowly charging a single battery with a featherweight panel.
 
Last edited:

eh4

Flashlight Enthusiast
Joined
Oct 18, 2011
Messages
1,999
My understanding of electricity is largely limited to analogies about water. Funny thing is, if electricity really behaved like water then I could easily solve this problem, all that's needed is a valve that feeds a Y.. one of the upper arms of the Y is a higher amperage path that feeds the charge controller, and the other one is for a trickle of current below the requirements of the charge controller. When there is insufficient pressure (current), the valve is closed -with a spring. The valve leaks when it's closed, it leaks to the other arm of the Y that goes through a diode, dropping the voltage below 4.1v.
When there is sufficient pressure (current), the spring is overcome, the valve opens to the wider arm of the Y - which powers the charge controller and at the same time plugs the "leak" that fed the smaller arm of the Y.
When the pressure drops below the requirement of the charge controller once again, the spring closes the valve, which resumes leaking through the smaller arm of the Y, through the diode which limits the voltage for safe (safer?) trickle charging.
 
Last edited:

hiuintahs

Flashlight Enthusiast
Joined
Sep 12, 2006
Messages
1,840
Location
Utah
With the water analogy, pressure (voltage) is what moves the water (current) through a restrictive mechanism (diameter of the pipe). It would be a lot easier for your solar project if the voltage were to stay constant but that will vary with sun, angle, and the load (resistance) that is placed on it.

Ohms law is that V/R=I, or voltage divided by resistance = current. So in your case (Vsolar - Vbat)/R = I. For a constant R, the I will vary when the solar panel voltage changes. Then you still have to have a way to keep the battery from being over charged beyond 4.20v +/- whatever small tolerance is allowed. Also if a battery will be connected to solar panel at night or during cloudy times you should have a low voltage drop diode like a Schottky diode in series with your resistor. Thus (Vsolar - Vdiode - Vbat)/R = I.

A resistor is a fixed passive device. What you really need is an active device with feedback (intelligence as you mentioned). A buck switching regulator or LDO regulator with voltage and current feedback would be ideal for full control in conjunction with a microcontroller.........but only suited for an experienced electronics guy.

The best solution for you would be to use a USB powered lithium ion charger IC. That way you don't have to worry about a series resistor or series diode. Also you won't have to worry about the 4.20v cutoff point. There are a few of them from different IC manufacturers but from a cost standpoint look at the TP4056. I think the TP4056 max voltage input is 8v. That would work as long as your solar panel voltage isn't that high. You can pick one of those up on Ebay on a pcb for just a couple of dollars. Down load the data sheet from the internet and that way you can see how its used and you can see where to adjust the charge current to your own liking via a resistor that connects to the programming pin of the TP4056.

See if one of those will work for you.

http://www.ebay.com/itm/5Pcs-5V-Min...167396?hash=item1a0ff9a024:g:WLYAAOSwUuFWuAMA
https://dlnmh9ip6v2uc.cloudfront.net/datasheets/Prototyping/TP4056.pdf
 
Last edited:

eh4

Flashlight Enthusiast
Joined
Oct 18, 2011
Messages
1,999
:twothumbs thanks!
The controllers you linked will be great for the larger 5 - 10 watt version.
Thanks also for the customized recap of V/R=I, I get dyslexic with the velocity, pressure, volume analogies.
 
Last edited:

eh4

Flashlight Enthusiast
Joined
Oct 18, 2011
Messages
1,999
Good news, using two, 2.5 watt, 5v Allpowers PET encapsulated panels I can directly charge everything that I am needing to charge.
With April afternoon sunlight at about 45 degrees angle to the panels I'm getting about 1.5 watts with one panel and 2.3 watts or so with two panels in parallel.
These are 7$ panels.
The voltage stays between 4.5 and 5v.
Olight's flexible mini charger works,
Galaxy S7 works,
Suunto watch works,
Anker 16,000mAh external battery is the most particular, it's had some trouble re-initiating charge after a cloud, it's made to recharge at a reasonable rate after all so I'm not particularly surprised or disappointed. The idea is to skip the weight of the external battery in the first place.

The voltage stays constant enough for the charge circuits, so I'm letting them do the complicated part, I'll find the sunlight.

This doesn't solve the initial problem of trickle charging with a 1 oz, 1 watt max, 6v panel, but it's a step in the right direction.
Right now the problem is solved at .4 watts per oz for PET sealed mono crystalline panels, with the original goal of 1 watt per oz. for CIGS panel not really any closer to working. The little 6v panel that I obtained is more like 7.5v the only buck switching regulator that I have seems to eat all of the watts just trying to function.

Worth repeating; Iif anyone reads this and wants to skip ahead to nearly 2 watts per oz for 50$, there's already a 6 oz, 10 watt panel on Amazon that's using 22+% efficient Sunpower cells.
 
Last edited:
Top