Is charger suitable for solar panels?

HKJ

Flashaholic
Joined
Mar 26, 2008
Messages
9,715
Location
Copenhagen, Denmark
Sometimes I am asked if a charger is suitable for solar panes and until now my only advice has been to get a simple charger, they are more likely to work with a solar panel.


But this do not say everything, another parameter is if the charger locks up when voltage dips (Brown out) and it is possible to test for that.
I did a script for my test station that will reduce voltage a variable bit, then restore to full voltage to see if the charger still works.
Charge%20Solar%20Raw.png

It looks like this, it is a bit hard to read.
Charge%20Solar.png

Using the data to draw a more simple curve it is much easier to see that this charger locks up when the voltage dips below 4.50V (Current = 0) and it will only reset if the voltage gets down to 2.0V. It do not change charge current but stays at the same current in all cases.


A charger that is good for solar usage would not have any brown out problems, but would always recover.


Are there any comments to this? Ideas to improvements? Ideas for a better title on the chart?
 

parametrek

Enlightened
Joined
Apr 3, 2013
Messages
578
This is great. Essentially you've scripted exactly what I do by manually twisting the PSU knob up and down.

However I don't think a graph is necessary for the final result. Just saying "The charger will not resume from voltages between 2.25V and 4.25V and isn't good for small solar panels." would be enough.
 

HKJ

Flashaholic
Joined
Mar 26, 2008
Messages
9,715
Location
Copenhagen, Denmark
This is great. Essentially you've scripted exactly what I do by manually twisting the PSU knob up and down.

I expect my test will include more voltages and more precise voltages and I do not get bored with twisting a knob a lot of times :D
One advantage with scripting is that each test is exactly the same.

However I don't think a graph is necessary for the final result. Just saying "The charger will not resume from voltages between 2.25V and 4.25V and isn't good for small solar panels." would be enough.

Maybe, you could also say the same for the charge curves: if I see they are fine I do not need to publish them.
I like publishing my data, making it possible for technical people to do their own evaluation.
 

Gauss163

Flashlight Enthusiast
Joined
Oct 20, 2013
Messages
1,604
Location
USA
Yes, that's a great idea. Yet more curves to caress our eyes (true charger porn).
 

IonicBond

Enlightened
Joined
May 2, 2013
Messages
221
Location
Southern California
Great timing! Not all chargers *really* operate properly under solar conditions - mostly recovery from passing clouds, temporary debris landing on panel (wet leaves etc) cutting panel output and so forth.

So to get a stable source of power, it might be a two-step process: Charge a smart power-bank, like an Anker bank that will properly reset when charging. Then use that bank to power chargers that may get cranky.

Example: I'm about to test a LiitoKala LII-402 directly attached to a 28 watt panel (using the 5v output). The LiitoKala defaults to knowing about nimh or li-ion. (THANK YOU for the excellent review!)

But, since my projects involve LiFeP04, the LiitoKala needs a manual button tap to set it for LFP. So, If I have panel shading, and the charger resets, will it default to the normal 4.2v nominal li-ion setting and overcharge my LFP if I don't babysit the charger in the sun?

I'll test, but my guess is that it will reset to the default and overcharge LFP to 4.2v, so I'm probably going to adopt a lossy two-step process by using an Anker power bank to gulp down the solar, and use the bank for a steady source to the LiitoKala at my convenience.

BTW, the *only* solar charger that properly applies a solar-specific algorithm to NIMH is the PowerFilm 4AA charger. Rather than constantly reset during passing shadows, and missing multiple dv/dt and other EOC triggers on each reset to already charged batts (very bad!) , it does a conservative slow charge after a rapid charge up to about 80% SOC. It is also smarter about dead / overdischarged batts too. Hard to find this kind of smarts in a solar AA charger, although unfortunately it charges in pairs. Too bad those models are hard to find these days.

So point is - be careful if testing nimh chargers directly - if they constantly reset due to unstable sun sources, and especially if the cells are already fully charged, that is more opportunity to miss the EOC signals and overcharge.

And, in the case of my LiitiKala - it may reset and overcharge LFP.

Directly charging nimh and li-ion batts can be a tricky business. That's why I just hammer down on an Anker power bank, which seems to survive unstable solar conditions most of the time.
 
Last edited:

HKJ

Flashaholic
Joined
Mar 26, 2008
Messages
9,715
Location
Copenhagen, Denmark
Example: I'm about to test a LiitoKala LII-402 directly attached to a 28 watt panel (using the 5v output). The LiitoKala defaults to knowing about nimh or li-ion. (THANK YOU for the excellent review!)

It is the Lii-402 I am using for my testing above.
 

HKJ

Flashaholic
Joined
Mar 26, 2008
Messages
9,715
Location
Copenhagen, Denmark
I have been improving a bit on the curves today, including fixing a bug and added more samples.

Charge%20Solar%20Raw%20LiIon.png


Charge%20Solar%20LiIon.png




Charge%20Solar%20Raw%20NiMH.png


Charge%20Solar%20NiMH.png




Maybe using lines are better now:

Charge%20Solar%20LiIon1.png


Charge%20Solar%20NiMH1.png



But when the two curves overlap it is difficult to see (I cheated with the dots, one curve has a 0.02V offset).
 

HKJ

Flashaholic
Joined
Mar 26, 2008
Messages
9,715
Location
Copenhagen, Denmark
Another charger:


Charge%20Solar%20LiIon.png



This looks fine for use with solar panels, the charger current will drop with voltage and then resume when voltage increases again.
The actual levels the voltage will drop at depends on the charge condition of the battery, for this test I started with an empty battery.


I prefer the dots for the graphic.
 

HKJ

Flashaholic
Joined
Mar 26, 2008
Messages
9,715
Location
Copenhagen, Denmark
The above charger did fine for LiIon, but NiMH is another story, I had to modify my script to include it on the curve. During testing I noticed some noise (Full load hum) from my power supply, this is not supposed to happen with a USB charger. I modified the test to include maximum input current, but will only include it in the curves is there is problems, like here:


Charge%20Solar%20NiMH.png



It looks like the charger has opened fully for the charge circuit due to the low voltage and is slow to turn the current down again. If there is enough current available it will probably destroy the charger (Not likely with a small solar panel). With the hole in the red line of dots, the charger is not suitable for solar panel usage when charging NiMH.
 

IonicBond

Enlightened
Joined
May 2, 2013
Messages
221
Location
Southern California
Wow - again thanks for doing this! Nobody documents these things like you do.

I suspected that nimh would be a problem. Most manufacturers don't really understand the real-world effects of solar charging and never turn their lab-supplies on and off simulating clouds etc.

PowerFilm is the one big exception, who recognized that with nimh under solar, resets are a big problem, and therefore just do a simple slow-charge once past the fast charge to about 80% soc. (In fact, when their led-light goes "solid", that is the indicator they have reached 80% soc, and actually need a few more hours of slow charge to fully charge). And of course they tell you not to leave the cells in the charger forever. It's the only solar charger I'll put my Eneloops in. I avoided the 2AA / USB version as the usb output is not efficient enough for modern day use, and like many products, if you burn up the usb portion, you also lose the nimh charger circuitry too. So I stick to the 4AA only version.

I put the LittoKala out in the sun with my LiFeP04's, and it initially did not like my usb cable that works with the SunJack charger. So I swapped for another brand, and now it works. Ugh - back to the varying cable pinouts like the old days. I haven't gone as far as trying to sacrifice any LFP yet, before I get some regular cycles in with the manual "LiFe" selection.

Nice work HKJ - I'm watching your lab results with high interest..
 

IonicBond

Enlightened
Joined
May 2, 2013
Messages
221
Location
Southern California
LiitoKala not suitable for LiFeP04 either:

Caught in the act - a large bird flew past the panels while I was babysitting it, the charger reset, and defaulted to the 3.7v nominal setting and started charging again. Which of course left unattended will take them to 4.2v eventually.

Heh, that answers that question. :)
 

parametrek

Enlightened
Joined
Apr 3, 2013
Messages
578
With the currents and the more interesting curves the graphs are much more important. Sorry for doubting you :)

IonicBond, the only charger I know of that does everything right is the Opus BT-C100 and possibly the larger versions of that single bay charger. It will remember chemistry and charge currents and has a 12V input too.

Multiple stage charging isn't really an answer. Charging batteries with batteries is inherently wasteful. And that is assuming the powerbank is solar friendly. If you are unlucky the bank will be using these same dumb circuits that can't get through a brown out.
 

Woods Walker

The Wood is cut, The Bacon is cooked, Now it’s tim
Joined
Jun 8, 2008
Messages
5,433
Location
New England woods.
I have done a bunch of field testing with portable solar charging. The best system I have found are solar chargers which auto restart on a cycle to clear out charge errors. There are multiple types of charge errors from reduced charge rates or a total failure The auto shut-off and restart is nearly bomb proof. Overall I find a powerbank to be the most error free. The simpler the charger the better. Some chargers will for example reset to a lower amp setting. The Nitecore F1 does this. The F2 likes to totally crash during power interruptions. The Olight UC magnetic charger works well. The Nitecore UM10 easily crashes. The Lii 100 and Lii 200 seems to charge at around around 1 amp 5 volts easily but can crash if the interruptions are numerous enough. Xtar sc1 is reliable but can charge at a slower rate in lower light than 1 amp chargers. However in good light hit 1.5 amps. The Ant MC1 plus is totally reliable however can default to .5 amps. The MC1 old school 1 amp model (they now sell a .5 amp) is totally reliable and doesn't default however will charge slightly slower than the new ANT but that can default to a .5 amps.

I have tested more.
 
Last edited:

HKJ

Flashaholic
Joined
Mar 26, 2008
Messages
9,715
Location
Copenhagen, Denmark
There are more than one way a charger can fail:

Charge%20Solar%20LiIon.png


This charger locks up at some voltage and discharges the batteries at other.
 

IonicBond

Enlightened
Joined
May 2, 2013
Messages
221
Location
Southern California
As always - thanks for this HKJ!

Sure enough, using a double-step of charging a battery from another battery is generally wasteful of energy. But if the charger reacts badly to resets, then that kind for forces one's hand.

I can and do charge LiFePo4 directly with solar - usually the smaller 20 and 40ah cells in a 4S configuration. Solar charge controller set NO HIGHER than 14.1v (fudging the typical "gel" setting), purposely disabling temperature-compensation, starting out with cells already balanced and grouped for capacity and internal resistance and all that.

But now we're getting waaay beyond flashlights and individual small cell chargers. Other forums for that. :)
 
Top