Check is voltage is present (potential-free)


I like to build a check circuit for my rc charger if a battery if connected as I running it with a generator. If there are problems there the FETs will explode if a battery is connected to the charger.

I can control the PSUs with I2C so the last point is to check if voltage is present. I do not need to measure the voltage simple check if a battery is connected to the charger and block the PSU to powerup the output.


  • 0.5 .. 50V
  • detection also if polarity is wrong (not connected correctly)
  • no measurement necessary
  • potential-free

I think if using a Optocoupler but did not know how to build it for the whole voltage range

I like to build it potential free as I do not know the circuits of the charger and also it is up to 50V.

Many thanks for any tip with ICs to use or simply a circuit to buy.


Search for "constant current source" circuits. Eventually you can find LED with integrated current source for use as current limiters to the optocoupler.

Can you go into more detail regarding your question: becaue as I understand it, power diodes between the charger and the battery might also solve the issue of current going back into your charger when it is not fed by the generator, but I could be wrong.
Second: whatever you make, will it be fast enough to react once the generator fails?


Using power diodes would not solve this completely as I cannot easily discharge a battery anymore.

I use server PSUs in series which has some seconds puffered on main output if the generators failed. Also the PSU outputs an error pin (it watches 220V input) and I can simply switch on/off main output by arduino digital output.

1st case: generator fluctuating because of overload. Error pin on, switch off PSU main output. Switch on PSU if batteries not detected anymore.
2nd case: System is off and connected to power. PSUs powering the arduino and the arduino is switching on main power until no batteries are detected anymore. (mostly similar to 1st)

The main issue is to avoid that the main power is going back on if batteries connected.

Could you provide an electrical drawing of your setup? It is difficult to understand what you are doing without a drawing of the setup.

Sorry for the very simple drawing, normally I use eagle to prepare most of my (dummy) circuits but actually I am on holidays now and prepared it with a simple free IPad app. I hope you can connive that.

That is my situation: I like to measure high voltage of a rc charger output. I do not know how the circuits in the chargers are build but a friend told me I should better not connect the arduino directly as it could damage the chargers converter circuits. As also the voltage is higher than usual I like to decouple the measurement of both Lipo batteries connected to the chargers.

Arduino has its own supply line with 5seconds backup and always on. Over I2C or control pin I can switch the PSUs main power. I like to detect if batteries connected, no voltage measurement is necessary. So using an optocoupler with a supply range of 3v to 54v should be the best way in my opinion.

How are the chargers activated?
Can you check the voltage or current of the charger input?

The chargers are two of these and I can control the output of the PSU via arduino. The PSU will also deliver everything via I2C but that information is not necessary. I have to avoid that the main power switches on if batteries connected to the chargers only.

It looks like this but lacks the security circuit.

With the cable ports on the front and the lid, just curious if you're planning on operating this with the lid closed?

No. Batteries are too large so I put the ports outside. Used it for years but chargers get sometime broken so I need this security circuit. It’s a design flaw of this chargers (and many else) that fets will damage if a battery if connected when switched on.

I've got a hiccup problem with batteries and MOSFETs too. A few talented users helped me to mitigate it though. I wonder how similar they are.

I do not see a generator nor a load on your drawing, but in your text you claim that diodes cannot be used to solve your problem, please explain.

This topic was automatically closed 120 days after the last reply. New replies are no longer allowed.