From the schematic diagram above, we can see that what we imagine is to control the fan speed through PWM. This time, a 5V two-wire DC brushless fan is driven. It is not unexpected that the fan did not start up. Let's observe the waveforms everywhere:
The first point is the PWM output of the MCU:
10kHz, 50% duty cycle meets the setting;
The second point is the output of the primary triode:
This is a little different from what we want. The peak value is 5V, and the address should be close to 0. The problem is R27. The partial voltage after the 1K resistance is turned on is too large. Here we temporarily modify it to 10 Ω;
The third point is PMOS control:
PMOS is controlled by a pair of triodes to improve efficiency and reduce rise time. However, such waveform obviously does not meet the requirements. The peak voltage is reasonable, while others are unreasonable. The problem should still be the resistance R28, which is temporarily revised to 10 Ω;
Fourth, look at the output without load and the output with load:
No load
Loaded
There was a waveform, but the fan did not move;
Finally, let's see the effect of directly increasing the duty cycle:
The MOS output is directly pulled down. It should not be.
Let's change what we thought should be changed and try again.
The second point after modification:
Modified MOS control:
No load after modification:
When there is load after modification, the fan still does not rotate
However, when the duty cycle is changed to 100%, the fan turns, the MOS output is not pulled down, and the 5V voltage can make the fan turn.
By constantly testing the PWM at 1kHz, the duty cycle of 50% is also activated, and the speed is reduced. However, with obvious noise, it is found by changing different frequencies and duty cycles that non 100% duty cycle control can be achieved at low frequencies. However, the lower the frequency, the more obvious the noise, the lower the duty cycle, the more obvious the noise. It seems that only the switching control of the DC fan can be carried out. To achieve speed control, the control mode of this circuit still has obvious defects. For this DC fan with only two lines, it is better to control the speed by reducing the voltage, and the noise of the alternating signal is not easy to eliminate.
After this debugging and modification, the fan can be turned on and off, and stepless speed regulation can not be achieved. PWM output is not required, and GPIO control can be used directly.
This article is from the forum, click to view the full post content.
