0
\$\begingroup\$

schematic

simulate this circuit – Schematic created using CircuitLab

recently trying to make a led strip that can be controlled over bluetooth. I need high currents up to 10A. for the test purposes I implemented the circuit above. and instead of leds I used 12 volts halogen capsule bulb, because led strip that draws 10 amps is very long. one of these capsules draws 1.6 amps and I connected 7 of them parallel. when I connected it directly to the power supply, it draws about 10 amps.

but the question is when I connect them to the mosfets drain, mcu instantly get burns. I am using nrf52810 SOC. the interesting thing is I have 2 more channels with the exact same circuit and both works fine. I tried to change bulbs, didn't work. what could be wrong with my circuit.

the light buls are theese sch

mosfet and bjt documents.

I am doing the test with the 0% duty and 100hz pwm signal. mosfet should be fully conducting.

ps: It works with low currents like 3A.

\$\endgroup\$
  • \$\begingroup\$ Cold lamp draws 10x current than hot, ground shift from inadequate wiring may cause issues. low ESR decoupling Caps near load 12V are essential. Twisted and EMC rules apply. \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Oct 14 '19 at 14:37
  • \$\begingroup\$ knew that, tried with the same string that works with another channel. should be draw less current because it is hotter. still damages the mcu. and if current is the issue mosfet should be damaged not mcu. am I wrong? \$\endgroup\$ – sinany1 Oct 14 '19 at 14:40
  • 1
    \$\begingroup\$ If interface voltage transients exceed supply voltage by +/- 0.x V then CMOS latchup”SCR” effect burns up IC but but should unlatch on power cycle if not too hot. this means EMI rules are being ignored for crosstalk. high current induced transient into signal out \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Oct 14 '19 at 14:51
  • \$\begingroup\$ Your MOSFET may be damaged leading to 12 volts on the gate and somehow getting back through the additional transistor. Or it could be ground current issues. Best to sort out the driver issues without the MCU first. \$\endgroup\$ – Chris Stratton Oct 14 '19 at 15:54
  • \$\begingroup\$ try using an optocoupler \$\endgroup\$ – jsotola Oct 14 '19 at 16:18

Your Answer

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

Browse other questions tagged or ask your own question.