Facebook
Google
GitHub
Linkedin
Is changing the supply voltage to a 12V, 80mm brushless fan effective and safe?
Control 12V brushless fan speed by varying the supply voltage?
- Thread starter AlbertHall
- Start date
I am thinking mostly of reducing the speed.
vandveuser16776
- Joined Feb 21, 2026
- 232
I am using one now. It is for a fan that is supposed to run 24/7. It is rated 12v but I power it up with 9v. Speed is lower, but I think the fan's gonna live longer.
From google AI:
2-Pin or 3-Pin Fans (No Dedicated PWM Pin): Applying PWM directly to the power supply (chopping the DC voltage) is not recommended for standard brushless DC (BLDC) fans. Because BLDC motors have built-in commutation circuitry and capacitors, external power-pulsing can starve the internal controller, cause audible whining, increase ripple current, and damage internal components over time. Instead, use a variable DC voltage regulator or a proper 4-pin PWM-controlled fan.
This discussion will lead my choice of fan.
schmitt trigger
- Joined Jul 12, 2010
- 2,276
Regardless of what you decide, I would first start the fan at full 12V, and after a couple of seconds reduce the voltage.
Yes, and I understand why that is a good idea.
As mentioned there are BLDC fans with 4 pins, where one of the pins is a PWM input, designed for speed control.
The other BLDC fans with 2 or 3 pins (the 3rd pin is sensor feedback) have a rated voltage (usually 12V) but you can play with the speed by adjusting the voltage. So it is up to you, depending of the rest of the circuitry, what type of control would be easier to implement.
The other BLDC fans with 2 or 3 pins (the 3rd pin is sensor feedback) have a rated voltage (usually 12V) but you can play with the speed by adjusting the voltage. So it is up to you, depending of the rest of the circuitry, what type of control would be easier to implement.
Below are some numbers for a 12 Volt 80mm typical fan:
| VOLTAGE APPLIED (VOLTS) | CURRENT DRAW (mA) | FAN SPEED (RPM) |
| 12.078 | 157.4 | 2,739 |
| 9.013 | 115.2 | 2,139 |
| 6.025 | 75.0 | 1,460 |
| 3.011 | 31.3 | 114 to 125 |
Generally I prefer using PWM to control fan speed but varying supply voltage works. Keep in mind using using a pot that at 9.0 volts a current draw is 115.2 mA you want at least a 2.0 watt pot allowing for some headroom.
Ron
ZhangSheng
- Joined Sep 25, 2026
- 5
One distinction that matters here is that a brushless fan is not a plain DC motor with a supply across a winding. Inside the frame there is a small commutation controller, a Hall device and driver transistors, and that electronics has its own behaviour as the supply falls:
1. Start voltage is not the same as run voltage. Most of these fans have an undervoltage cut-out and need a higher voltage to start reliably than they need to keep turning once they are running. A fan that happily runs at, say, 5 V after being started at 12 V may sit still, twitch, or start only sometimes if it is switched on at 5 V - and starting gets worse with cold bearings, dust, a filter adding static pressure, or the fan mounted in its least favourable orientation. That hysteresis is why the suggestion to start at full voltage and then reduce it is sound practice rather than a nicety.
2. The lower end of the range is therefore not really usable control range. The speed/voltage curve is reasonably well behaved in the middle, then collapses quickly as the controller approaches cut-out, as the figures in the post above show at 3 V. In that region a small change in load or temperature is the difference between running slowly and stalled, with no useful airflow either way.
3. How the voltage is dropped matters. A series resistor or diodes drop a voltage that depends on fan current, and fan current changes with speed and load, so the setting drifts and the control interacts with the load. A linear regulator at least holds a voltage, but it dissipates the dropped voltage times fan current as heat and still leaves the start-up problem unsolved.
4. Chopping the supply is different from using a proper PWM input. On a 2-wire or 3-wire fan, PWM on the supply repeatedly powers the internal controller up and down; depending on frequency it can produce audible noise, unreliable starting, and a tachometer output that is meaningless during the off periods, because the circuit that generates the tach pulses is unpowered then too. A 4-wire fan avoids all of that because the supply stays at a steady 12 V and only the separate logic-level control input is modulated.
So, in practical terms: voltage reduction is effective and safe for slowing a 2-wire or 3-wire fan provided you never exceed the rated voltage, you stay well clear of the stall/cut-out region, and you test the worst case - a cold start, at the lowest voltage you intend to use, with the fan stopped, in its actual mounting and airflow conditions, several times in a row. If it must restart unattended after a power interruption, that start-up test, rather than the running speed, is what should set your minimum voltage.
1. Start voltage is not the same as run voltage. Most of these fans have an undervoltage cut-out and need a higher voltage to start reliably than they need to keep turning once they are running. A fan that happily runs at, say, 5 V after being started at 12 V may sit still, twitch, or start only sometimes if it is switched on at 5 V - and starting gets worse with cold bearings, dust, a filter adding static pressure, or the fan mounted in its least favourable orientation. That hysteresis is why the suggestion to start at full voltage and then reduce it is sound practice rather than a nicety.
2. The lower end of the range is therefore not really usable control range. The speed/voltage curve is reasonably well behaved in the middle, then collapses quickly as the controller approaches cut-out, as the figures in the post above show at 3 V. In that region a small change in load or temperature is the difference between running slowly and stalled, with no useful airflow either way.
3. How the voltage is dropped matters. A series resistor or diodes drop a voltage that depends on fan current, and fan current changes with speed and load, so the setting drifts and the control interacts with the load. A linear regulator at least holds a voltage, but it dissipates the dropped voltage times fan current as heat and still leaves the start-up problem unsolved.
4. Chopping the supply is different from using a proper PWM input. On a 2-wire or 3-wire fan, PWM on the supply repeatedly powers the internal controller up and down; depending on frequency it can produce audible noise, unreliable starting, and a tachometer output that is meaningless during the off periods, because the circuit that generates the tach pulses is unpowered then too. A 4-wire fan avoids all of that because the supply stays at a steady 12 V and only the separate logic-level control input is modulated.
So, in practical terms: voltage reduction is effective and safe for slowing a 2-wire or 3-wire fan provided you never exceed the rated voltage, you stay well clear of the stall/cut-out region, and you test the worst case - a cold start, at the lowest voltage you intend to use, with the fan stopped, in its actual mounting and airflow conditions, several times in a row. If it must restart unattended after a power interruption, that start-up test, rather than the running speed, is what should set your minimum voltage.
