inrush current damaging microcontroller

Thread Starter


Joined Nov 4, 2020
Hi, I have been experimenting turning on heavy loads with arduino using transistors and relays with varying success.
The main problem I have is the in rush of current to start a high power HID spotlight is enough to drain all power from microcontrollers and other power supplies, causing them to reset or damaging them.
I would like to know what I can do to keep the steady power to my micros and power supplies while the heavy load starts up.
do people use batteries to keep micros alive during this process ?IMG_2661.jpgor is there a circuit that i can implement to get the job done?
I have attached a basic schematic of my project.


Joined Jan 23, 2018
If the power supply voltage dropping because of the heavy additional load on the power source is the problem, and not some inductive voltage spike from switching OFF the relay, which seems to be the basis for post #2, then the first thing is to know the current capability of the power supply and the current requirement of the load. It may be that the supply is not able to power both loads.
If the current capacity of the supply is great enough to power both loads with adequate margin, then the problem may be in the wiring, which, if it is wired like the diagram shows, has a flaw in that the high load current passes through the same conductors as processor supply current. That would be my fist guess as to the problem, based on the information that we are given.
So one question is how much current does the "high power spotlight" require?? That matters. And, what is the source of the twelve volts?
Last edited:

Thread Starter


Joined Nov 4, 2020
sorry yes i do have a resistor on tip 120 and have diode on relay.
As misteBill2 said its a current capability problem on start up.

Thread Starter


Joined Nov 4, 2020
the spec on the lamp is HID 70w 12-24v input it draws 8 amps when running but during start up i am not sure.
I have powered it via fully charged 12v battery or my 30 amp power supply and have had troubles with both.
I seem to think I need to separate both circuits?


Joined Aug 7, 2020
Is it a hot-restrike HID? If so, there‘s probably 20kV in the igniter circuit.
Need to keep that well away from your circuit!


Joined Mar 14, 2008
Buffer the buck converter power as Beau Schwabe suggested (may need a larger capacitor), and make sure there is a connection for the load ground to the voltage supply, that is completely separate from the electronics ground.

Beau Schwabe

Joined Nov 7, 2019
" Not disagreeing with you, but what would be the mechanism for damaging the chip with a badly routed ground connection? " - Noise entering the substrate of the IC can induce negative voltages which can cause a latch up condition.


Joined Jan 29, 2010
hi Beau,
The TS states his lamp takes 8 amps when running, that could be upto 5 times the current while lamp element reaches its operating temperature. ie: 40 Amps.

A poorly routed ground/0v line could easily develop a high negative spike that could take out his MCU.