I use a 4N25 optocoupler in series with the solar panels and a 1K resistor to sense solar voltage to switch on the outside LED flood lights when it is dark. But now the optocoupler fails. With the old linear solar battery charger it worked fine. I use the PWM5 design and it works very well.
For protection a 1N4007 diode is reverse connected across the optocoupler's diode. What else can I do ?
PWM - Project.pdf (34.1 KB)
PWM - Project.pdf (34.1 KB)
I am sorry but I followed the instructions to post a image in the body but it did not work.
Curious... What is the voltage and wattage rating of the panel?
With a 1K resistor, even if using 24V Panel, you are within the nominal operating conditions of the 4N25 LED.
Something else is happening here.
Personally, if I were going to add an additional diode, I would have put a diode INLINE with the 4N25 LED not across it... its not like its going to see reverse polarity from the panel.
Interesting circuit - no idea what its supposed to do, curious diode/capacitor matrix at the top left,
very slow FET switching circuit with two unneeded transistors (a logic level FET would do), and mysteriously
shorting the solar panel to ground (seemingly no other function than to short out the panel).
Perhaps something is missing from that diagram? Could you explain what its supposed to do in detail?
Could you explain how the output of the opto coupler is used by posting the code - at the very least it
is going to be changing due to the PWM now, so a simple digitalRead() isn't going to be reliable.
The solar panel is rated at 80W at 20V.
What is strange is that I connected a LED in series with the opto diode to see when the opto fails and so far it is still working.
The PWM5 circuit utilizes a Dickson charge pump to get the gate voltage up to 20V as the source of the Mosfet is connected to the battery positive.
So you have confirmed that you are somehow breaking the internal LED in the OPTO?
Yes.
All this circuit does is check for voltage from the solar panel which either outputs a low when it is day or high when it is night to a digital input of the Arduino.
It must be the PWM in a way that breaks the opto diode. But so far so good with the LED in series. Good thing too - This is my last optocoupler.
aj777:
The solar panel is rated at 80W at 20V.
What is strange is that I connected a LED in series with the opto diode to see when the opto fails and so far it is still working.The PWM5 circuit utilizes a Dickson charge pump to get the gate voltage up to 20V as the source of the Mosfet is connected to the battery positive.
That's far less than clear in that circuit - draw the battery and panel in the diagram, with polarity clearly labelled.
You are using Arduino outputs as the charge pump? That's going to be overloading them severely at
start up, and risking burning out the protection diodes on the pins with the high voltages around - 100nF is
big enough to let big spikes through and damage logic.
I see no details of the PWM frequency or duty cycle.
The panel may well be rated 80W at 20V - but open circuit it's voltage could rise way above that and that may be why the opto dies.
You don't show the load on the opto photodiode - but say it's 10k and so needs to sink about 1/2 mA to pull 5v to zeroish. Hence you need 5x this to drive the LED ( 4N25's aren't very good)
2.5 mA
at 20V drive that means a series resistor of about 8k to the 4N25's LED, which would give you a huge safety margin.
Using a 1K it dissipates nearly 20^2/1000 watts or 400mW - pushing it for a standard 1/2 watt resistor.
No comments on the rest of the circuitry. I don't wish to be rude.
Allan