diy solar

diy solar

Tiny low power "charger" (next to nothing)

DThames

Solar Wizard
Joined
Nov 22, 2019
Messages
2,669
There have been a couple (first second) of threads in the past year or so, asking about the need for tiny solar charger with no idle/dark consumption current. In both threads the poster asked/desired to charge with a tiny solar panel and no charge controller. Both cases had very low load or very intermittent loads. On both threads I have suggested a simple Zener regulator circuit. Some have suggested that it wouldn’t work for one reason or another. I think many have failed to consider it because they don’t understand the concept. So this thread is to shed some light and give anyone interested in learning a chance to look into this possible solution.

Note, this is not a battery “charger”. It is a circuit that will allow the battery to be charged from a solar panel and not be charged above a specific voltage. The circuit also can limit the charge current or the panel’s output can be the limit, but the current must stay well within the range of the zener’s ability to deal with the heat. I am using a 5watt Zener but without a heatsink and cooling it might handle 1 watt…maybe.

The basic circuit is explained on this web page. R(load) is the battery. Vin is the solar panel. I purchase some of these 12v zeners to do my demo with. I am using a 4watt panel. I have 50 ohms for Rs(current limit) to keep the current limited to about 150ma.

Principle of charging: In the case of using a 12v Zener, when the battery is not connected, all current from the panel would be shunted to negative/ground by the Zener. That is why it needs to be properly heat managed. When the battery is not below 12v and the battery is connected, the battery will not share the current and again the Zener will shunt all current to negative. When the battery is less than 12v, the battery charging load will cause the voltage of the I(load) point (see diagram on web page) to be pulled down to the battery voltage. When this happens, the Zener will not be above its 12v threshold and will not conduct current. The battery will slowly charge based on the battery’s equivalent circuit and the power source current limit (panel or resister Rs). As the battery gets charged to 12v, when the battery voltage might want to go above 12v, then the Zener starts hogging the current, shunting the power to negative, and stopping further battery charging. This state continues as long as the power source is active (until dark) or until something causes the battery voltage to drop below 12v.

Things to note, the battery will never be “fully charged” but will rest at the Zener/circuit set voltage. If you set the Zener to a desired “float” voltage, that is where the battery will be charged to. This may not be ideal for some battery type, but the circuit will prevent voltage from going beyond the Zener voltage point. I have looked and it is hard to find the exact Zener value and watts rating that we might desire. In the 12v example, if you would put a Schottky diode in series with the 12v Zener, you could raise the charge cutoff by about 0.2-0.3 volts, to about 12.2-12.3v. If you put a silicone diode in series with the Zener, you can add 0.6v to the cutoff giving you 12.6v.

My testing/demo is with a 3 cell Li-ion pack so 12v will give me 4v per cell, which is a pretty good state of charge. I discharged the pack down to 11.3v and put it in the sun yesterday morning. This morning the battery was at 11.75v, so it is not “there” yet. The sun is now on the panel and the meter says 11.81v with 0.08 amps.

More later.
 

Attachments

  • Zener-next-to-nothing-charger.jpg
    Zener-next-to-nothing-charger.jpg
    194.5 KB · Views: 16
Update:
In about 4 hours (after charging all day yesterday) the battery was slightly above 12v. I have discharged it enough that it would have to do some work to get back to full voltage. In and out of the sun through the afternoon, the max voltage showed 12.2v on the meter, which would be what that specific zener's voltage is.
 
Hi you can use mosfet and zener diode but the voltage cutoff is not occurate and will depend of temperature.
you can remplace R1 by a POT to adjust the trigger voltage
 

Attachments

  • OverVoltageProtect.png
    OverVoltageProtect.png
    12.2 KB · Views: 11
Be very careful as I see no indication that the li-ion (not specified, but they don't appear to be LFP) battery bank is balanced, and a "pack level" voltage indication is not enough.

Have you at least run these cells through an individual charger and tested for both capacity and internal resistance such as an Xtar, Opus, or other similar hobby charger?

While it seems insignificant, this can be recipe for a "vent with flame" incident that could grow much larger. Especially if someone thinks it is so simple, nothing can go wrong.

Be safe out there...
 
Be very careful as I see no indication that the li-ion (not specified, but they don't appear to be LFP) battery bank is balanced, and a "pack level" voltage indication is not enough.

Have you at least run these cells through an individual charger and tested for both capacity and internal resistance such as an Xtar, Opus, or other similar hobby charger?

While it seems insignificant, this can be recipe for a "vent with flame" incident that could grow much larger. Especially if someone thinks it is so simple, nothing can go wrong.

Be safe out there...
If I put it in service I would put a BMS on it but seriously, have you done the math? Both threads referred to call for a 1.5watt solar panel.
 
This is why something like this is so dangerous without realizing it. How about this math:

Let's say your cells are unbalanced, and you have the "pack level" cutoff set to 12.3v with the zener circuit.

Cell 1 = 3.5v
Cell 2 = 3.5v
Cell 3 = 5.3v !!

But wait - even so, it's only a 1.5watt panel! Ever hear of thermal runaway? Cell #3 is getting destroyed chemically, and even with a puny panel current eventually it sets itself off.

This is what I'm talking about. It's a real-world issue.
 
Hi you can use mosfet and zener diode but the voltage cutoff is not occurate and will depend of temperature.
you can remplace R1 by a POT to adjust the trigger voltage
Did someone actually test out that circuit? Feeding that circuit with AC? P-CH MOSFET's body diode not shown and not come into play?
 
Did someone actually test out that circuit? Feeding that circuit with AC? P-CH MOSFET's body diode not shown and not come into play?
the little example work perfectly fine with input voltage under the max voltage of vgs capability of mosfet eg 18V.
if you reverse polarity at the input, it's protect your circuit no current at all.
if input is higher than normal, it will switch off.
it's just an example, this design has to be changed to check output voltage instead input voltage...
 
the little example work perfectly fine with input voltage under the max voltage of vgs capability of mosfet eg 18V.
if you reverse polarity at the input, it's protect your circuit no current at all.
if input is higher than normal, it will switch off.
it's just an example, this design has to be changed to check output voltage instead input voltage...
Are you sure that circuit is supposed to be fed with AC per diagram? Look at the Voltage source as shown in the schematic.
When the input is positive with ref to the circuit ground the MOSFET M1 is reversed bias because it is P-CH MOSFET so the current will flow through the Body Diode of the MOSFET M1, when the input is Negative with ref to ground, M1 will be bias on, the body diode will be bias off. What is the purpose of M1?
Link to the website of this circuit?
MOSFET spec:
 
Last edited:
yes it's a simulation as you can change Vin with a one mouse click :
let's try with 10VAC only positive pulse at the output
let try with 15VAC, nothing at the output
 
This is why something like this is so dangerous without realizing it. How about this math:

Let's say your cells are unbalanced, and you have the "pack level" cutoff set to 12.3v with the zener circuit.

Cell 1 = 3.5v
Cell 2 = 3.5v
Cell 3 = 5.3v !!

But wait - even so, it's only a 1.5watt panel! Ever hear of thermal runaway? Cell #3 is getting destroyed chemically, and even with a puny panel current eventually it sets itself off.

This is what I'm talking about. It's a real-world issue.
The thread was NOT about the construction of the battery pack but about the overall regulation of a panel without a charge controller. Anyone using a Li-ion without a BMS should know better. But in the interest of public safety. "Always use a BMS on Li-ion".Use-BMS.jpg
 
To the danger zone with this thread.... @svetz

Struggling to understand the point of the exercise.... (which is mostly why I stay out of the danger zone forum)
Thanks for showing us something you built that is useless in real world applications?
 
I know EEs each favor their own circuits so I will share one of mine which was ultimately used in thousands of real-world products.

When I was a young engineer (probably 30-35 years ago) my boss gave me the task of designing a solar charge circuit for a low current device and I came up with the simplified schematic shown. I believe the total circuit with the micropower opamp and bandgap zener drew about 100 uAmps. The actual circuit even had crude temperature compensation using a thermistor shunted by a resistor. The positive feedback around the opamp is to provide a few mV of hysteresis. Alternatively a comparator could be used I thought it was clever at the time as I had never seen other circuits which shunted the solar current when it was not needed to charge the battery.
 

Attachments

  • Simple Solar charger.pdf
    229.5 KB · Views: 6
I think that even at this level, it points out how when dealing with batteries, especially with li-ion, one needs to look at the large picture as a system, and not just a singular aspect. Good example actually.
 
I
To the danger zone with this thread.... @svetz

Struggling to understand the point of the exercise.... (which is mostly why I stay out of the danger zone forum)
Thanks for showing us something you built that is useless in real world applications?
If can charge enough to power a few led lights at night it works as a legal anchor light for a boat on anchor.
 
Back
Top