I have built a li-ion 18650 7S pack as my solar backup. It is being charged by 3*250W solar panels in series (about 105V 8A.)

My charger doesn't have an option to monitor individual cell voltages. It just charges the battery pack to the given voltage (4.2V*7 = 29.4V.)

I have a BMS monitoring my pack and if a cell goes above 4.2V it generates a signal which I monitor using an ESP32.

What I want to do is turn off solar side if cell goes above 4.2V with that signal.

Can Mosfet do this job or is there a better way to achieve this?

Note: The BMS has overvoltage cutoff but if I connect the solar charger through the BMS, it cuts off the battery power to the controller while PV side is on. The controller manual clearly say that cut PV before battery.


What you are asking for would ideally be handled by a properly implemented charger and balancing BMS.

However, a somewhat informal but entirely doable way of shutting down solar input is to use a suitable rated MOSFET to short circuit the solar array - or a portion of it. The MOSFET MUST be either fully on or fully off.
The solar array is not harmed by short circuiting and dissipates minimal energy.

Power dissipation = I^2R. At 8A and say a 10 milliOhm MOSFET power in the FET
= 64 x 0.010 = 0.64 Watt.
10 milliohm Rdson MOSFETS are common and cheap and even lower or much lower Rdson devices are readily available. At that power dissipation heatsinking is trivially easy.

A relay could be used for the same purpose. However, opening a 100V 8A DC contact requires switches designed for the task. At those power levels the system is potentially lethal - both due to the voltage alone and also due to the arc potential.

Unless there was a good reason to run the panels in series at 100V+ I'd favour operating them in parallel at 35V.


Using 3 x 35V panels in series to charge a 29.4V max 7S battery pack sounds very strange UNLESS you have an MPPT controller. IF you have an MPPT controller that's (potentially) fine.

A BMS usually both detects cell overvoltage AND drains current from the charging source to prevent cell overcharging. As long as the battery is reasonably well balanced a balancing BMS will be able to handle the current needed to maintain balance.

  • \$\begingroup\$ Hmm - are there induction or resistive-region issues with this solution? The classic failure of e.g. motor driver schemes is that they work in the static analysis and then blow up during switching. Perhaps put a standard 100V bulb in there as a dump resistor. \$\endgroup\$
    – pjc50
    Jan 22 '20 at 11:37
  • \$\begingroup\$ @pjc50 Right, the MOSFET might absorb significant energy during the switching. I'm not sure how an extra resistor would help with that. Maybe you could add a snubber network so that the current through the MOSFET won't rise much until the MOSFET is fully on and the voltage across it won't rise much until it's fully off. \$\endgroup\$
    – user253751
    Jan 22 '20 at 11:42
  • \$\begingroup\$ IF you can keep away from arc formation then shorting/opening PV panels is fairly benign. There "should be" minimal inductance. If you do have inductance you need to snub the worst case energy spike. [I can "arc weld" (very carefully) using a single 35V, 300W panel! That rather amazed me - drawing and holding an arc with suitably tiny electrodes (or wire ends) is a warning of what can happen. A friend of a friend stupidly pulled open a connector on a 400V ish string on a sunny day. He was hospitalised. \$\endgroup\$
    – Russell McMahon
    Jan 22 '20 at 11:42
  • \$\begingroup\$ @RussellMcMahon Out of curiosity, what was his injury? \$\endgroup\$
    – user253751
    Jan 22 '20 at 11:43
  • \$\begingroup\$ yes charger controller is a MPPT one. when i looking through the solutions, i found that DC-DC Solid state relays available. can this do the job? \$\endgroup\$ Jan 22 '20 at 11:44

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.