# Voltage divider circuit getting hot

I recently got into arduino and salvaged some N-type MOSFET Transistors. I'm trying to power some motors with a 12v battery and control them with my arduino. I found that the transistor gate and source need to have a common ground. So instead of powering my arduino from a 5v phone charger battery, I'm powering it from a 12 volt battery and using a voltage divider circuit to cut the voltage to 6, and still be able to have a common ground with the battery and the arduino. I'm using two 220 ohm resistors in series to divide my voltage. For a few seconds they work just fine and I'm getting about 6v out of the resistors, however after a few seconds they both begin to burn up and smoke.

I'm wondering if maybe I'm doing something fundamentally wrong in the circuit, my resistors can't handle the .25 mA the arduino is pulling, or I'm just stupid.

• What is the maximum power your resistors can dissipate? I suspect that you are exceeding this limit. Commented Jan 25, 2017 at 2:34
• And a side note: why not use a 7805 voltage regulator? A voltage divider is not a good power source. Commented Jan 25, 2017 at 2:35
• Share your connection setup. The resistors shouldn't be burning out if they are 1/4 watt ones At least. Commented Jan 25, 2017 at 2:52
• If something shorted across (or anywhere nearly did so) one of the resistors, the other one would experience significant heating. You also probably aren't even close to taking the right approach. Might be better to spend a little time discussing what "some motors" means and what you are trying to do, overall. The (.025 mA) thing doesn't sound right, either. The arduino wants to pull more than that.
– jonk
Commented Jan 25, 2017 at 3:12
• @jonk, I'm sure he means .025A, not mA. Author should edit the question to correct that. Commented Jan 25, 2017 at 3:14

Voltage dividers work great if and only if all of the current that flows through R1 also flows through R2.

simulate this circuit – Schematic created using CircuitLab

However, as soon as another current path forms between the resistors, the divider no longer works. Since some of the current is diverted away from R2, the voltage drop changes. What's worse, if the current to the load is variable, the voltage of the divider will be variable too:

simulate this circuit

It's for this reason that resistor dividers are almost never used as a power source. Instead, you would use a voltage regulator of some type. Voltage regulators have active feedback to maintain the output voltage regardless of current draw (up to a point, of course). And some regulators, such as SMPS, can have very high efficiencies, so you don't have to burn off so much heat to drop the voltage.

For your simple circuit, a linear regulator is probably best suited. They're very simple to use (typically a single IC chip plus two capacitors) and inexpensive. They're great for most low-current digital applications. The 7805 is a very common linear regulator that outputs a fixed 5V, however, it is ancient technology and there are plenty of modern devices available as well.

simulate this circuit

• If the current really is 25 mA this should work fine. If it is much more, it might be wise to use a buck, just because there is a 5V drop from input to output. Commented Jan 25, 2017 at 4:04
• True. Assuming dissipating a watt in the linear regulator is acceptable, you could source up to 1W/(12V-5V) = ~143mA Commented Jan 25, 2017 at 4:36
• A heatsink would be a nice addition to the above circuit. Commented Jan 25, 2017 at 4:58
• Thanks! I will either pull a voltage regulator off an old circuit board if I can find one, or buy one online. Commented Jan 25, 2017 at 15:55
• @patrickjackson Mouser is your friend. www.mouser.com Commented Jan 25, 2017 at 19:38

From your description, you're dissipating 160 mW in each resistor, before you ever start supplying current to your Arduino.

What is the power rating on your resistors?

This is where you really want to use either an old linear regulator or (better) a baby switcher, or even a small linear regulator and a stout pass transistor. I'd be sorely tempted to use something like a 78L06 and a 2N3055, just because 2N3055 is DIRT CHEAP and DIRT COMMON.

The more voltage you reducing with the voltage divider, the hotter the resistors get, because the whole voltage actually goes through them, only the line to load actually has the reducted volts on it.