Hello,

I'm using this short type solar panel regulator circuit, as shown in the schematic, to charge a 12V SLA battery. However, it's possible that my PIC could become non-functional and I'd like to avoid overcharging the battery by just letting the panel stay connected to the battery 100% of the time. Which is what happens if the PIC doesn't put 5V on gate of the MOSFET (when the gate is at 5V, the MOSFET conducts and shorts the panel).

So I added this 'safety circuit' that is supposed to apply 5V to the gate if the battery voltage ever gets above 15V.

My thinking is that if the battery voltage rises above 15V, the 15V zener will conduct and the voltage divider will provide the ~5V to the MOSFET gate, which will short the panel and hold the battery voltage to 15V. However when I test this circuit and the battery voltage rises above 15, I see only about 3V on the node between the 15V Zener and R2. I would expect to see closer to 15V. Also the gate voltage doesn't get above a volt or so. Any ideas as to why that doesn't seem to work the way I'm expecting?

Also, I'm pretty sure the PIC is not going to like sinking 500mA through that I/O pin but I'd rather ruin a PIC than a battery (it's cheaper and safer).

The calculations for the resistor value for R2 can be found here: https://docs.google.com/spreadsheet/ccc?key=0As0iaN-Y_KkNdHp0X3dvTngwVEVQOW9vNk5IMHBReVE

 

Did you get that from Ebay, China ?

 

Get 'what' from eBay or China? This is a circuit I've been working on for a while, the parts are from many sources. The panel comes from China.

Also I should add that for my experiment the wire labeled RC3 is connected to ground.

 

You need to learn about the voltage drop vs. current properties of your zener. The high voltage end is more predictable - it's gonna stay at 15v or less until it blows - but the voltage on the low side is more complicated. But the basic idea is valid. I have a solar light that uses this approach to limit overcharging. I've posted the circuit here. See the arrangement of D2 and Q1.

 

The zener value is too high, it needs to be less than 15v also the resistor R3 is too low value and makes a voltage divider to the FET gate, which is the reason your FET gate voltage stays too low.

Try a 11v or 12v zener and a R3 value of 2k2. If you replace R2 and R3 with a trimpot of say 1k that should help solve the problem and let you adjust the set point where the panel output is limited.

 

Thanks for all the replies here. I'm looking for opinions.

After looking into this issue (and what happens on the anode side of a Zener diode) I think that my circuit is not suitable. It seems to me that there will be voltage bleeding over to the gate of my MOSFET before the battery reaches the desired maximum of 15V. That will lead to messing up the clean switching I've got going on with that MOSFET.

So I'm thinking that this approach to a 'safety' circuit, to handle the case where my PIC is not functioning is not viable. Thoughts? Ideas?

 

Ever look at a "watchdog" circuit to reset your circuits in case the PIC goes off to see the wizard? I've used them in high power instrumentation to prevent meltdown of hot assemblies.

The watchdog chips are very common. You need to write a routine to periodically write to the chip. If the write does not come in the time window, the system gets reset.

 

I'm using the PIC16f1829, which has a watchdog timer built in. I'm using that. So I guess it's unlikely that a failure will occur where the PIC can't run it's PWM peripheral and limit the current from the panel. However it's possible and I don't like the idea of a panel connected to a battery that is unregulated. In my experience I've just seen the batteries become unchargeable but I have read that they can vent hydrogen and possibly explode.