The relation between the pwm frequncy and the rpm of a dc motor

Thread Starter

aragon1971

Joined Apr 7, 2008
84
1604499704208.pngDear Friend

i setup the following : a dc motor controlled by a mosfet In the Gate of mosfer applied a PWM signal . I change the frequency of PWM signal ( the width of the pulse is const) and the rpm is increase . The rpm begin to decrease with the frequency increase when the f began very high (a few kHz) , How a can explain this behavour ?



thank you
George
 

sagor

Joined Mar 10, 2019
479
Also, depends a bit on the capacitance on the gate of the MOSFET. 100k gate to ground may be a bit too big, and not allowing the gate to discharge fully between cycles. Same for the 270 ohm resistor, you may not need it. Again, it depends somewhat on what kind of MOSFET you are using.
If frequency is too high, MOSFET may never turn off fully. But, if width of pulse is too small, MOSFET may never turn on fully. Too many unknows.
 

Thread Starter

aragon1971

Joined Apr 7, 2008
84
The Mosfet is the IRF520 and
i connect the gate of the mosfet in a pulse generator ( from elvis platform ) the resistance 270Ω and 100 kΩ IS NOT CONNECTED The motor is a small fan from pc with 12V max voltage 1604503699155.png
 

Papabravo

Joined Feb 24, 2006
16,773
Motors are controlled by current. It is the duty cycle of the PWM rather than the frequency which determines the current in the motor coil(s). There is a caveat to this statement. The frequency cannot be too low and it also cannot be too high. It must occupy a middle range of frequencies determined by other factors, like moment of inertia and winding inductance.
 

Thread Starter

aragon1971

Joined Apr 7, 2008
84
Motors are controlled by current. It is the duty cycle of the PWM rather than the frequency which determines the current in the motor coil(s). There is a caveat to this statement. The frequency cannot be too low and it also cannot be too high. It must occupy a middle range of frequencies determined by other factors, like moment of inertia and winding inductance.
thank you i m looking for a model if there
 

LesJones

Joined Jan 8, 2017
3,477
The key to the answer is the fact that you state " ( the width of the pulse is const) ". For example if the frequency is 1 Khz and the duty cycle is 50% then the on pulse will be 500 uS. If the frequency is increased to 1.9 Khz the on time is still 500 uS but the off time will only be 26 uS ( As the period of the waveform has changed from 1000 uS down to 526 uS) So now the duty cycle is now 500/526 = 0,95 = 95%. PWM control normally keeps the frequency constant and varies the duty cycle of the PWM waveform.

Les.
 

Alec_t

Joined Sep 17, 2013
12,197
As you say, Les. So for a fixed 'on' pulse width the duty cycle increases with frequency. You'd expect rpm to do likewise, but the TS is reporting the rpm decreases as the frequency increases. :confused:
 

shortbus

Joined Sep 30, 2009
8,797
The motor is a small fan from pc with 12V max voltage
What type of fan? Many(most?) of the computer fans sold today are BLDC motors with a built in motor drive. If yours is that type PWM probably isn't going to work with it.

Also your link to the mosfet is very confusing, compared to the actual data sheet. The link says it can be used with micro controllers because of the low threshold voltage, that is just a dumb statement. Threshold is where the mosfet is barely on. While the 8V your putting on the gate in your case is close to the fully on gate voltage of 10V, it still isn't turned fully on, so that may be effecting your readings too.
 

crutschow

Joined Mar 14, 2008
27,697
As you say, Les. So for a fixed 'on' pulse width the duty cycle increases with frequency. You'd expect rpm to do likewise, but the TS is reporting the rpm decreases as the frequency increases. :confused:
The problem may be that the circuit is missing pulses as the frequency increases beyond the fixed pulse-width.
Without looking at the waveform it's a guess as to what's happening.
 

LesJones

Joined Jan 8, 2017
3,477
Hi Alec,
I re read the original post. He starts off by saying when he changes the frequency the speed increases. I ASSUMED that he meant increase the frequency. I can't think of an explanation for the speed to decrease when the frequency is above a certain value. The TS would need to tell us the fixed duration of the ON pulse and the range of frequencies he used.

Les.
 

Audioguru again

Joined Oct 21, 2019
3,505
We do not know of your signal generator has an output coupling capacitor to produce +4V and -4V. If it does then the Mosfet might only barely turn on.
 

Thread Starter

aragon1971

Joined Apr 7, 2008
84
i use a bru
What type of fan? Many(most?) of the computer fans sold today are BLDC motors with a built in motor drive. If yours is that type PWM probably isn't going to work with it.

Also your link to the mosfet is very confusing, compared to the actual data sheet. The link says it can be used with micro controllers because of the low threshold voltage, that is just a dumb statement. Threshold is where the mosfet is barely on. While the 8V your putting on the gate in your case is close to the fully on gate voltage of 10V, it still isn't turned fully on, so that may be effecting your readings too.
I use a pulse with max value 5V (GATE THERSOLD IS 4 V) and the MOSFET IS ON ... The rpm is increase with the width of pulse it is beter to use a normal dc motor (brushed motor)?
 

Audioguru again

Joined Oct 21, 2019
3,505
The 4V max gate-source threshold voltage of a Mosfet is when it is barely turned on (that is what the word "threshold" means). Its drain-source current is only 0.25mA so it is almost turned off, far from being turned on. It turns on only a little more with 5V.
The datasheet of the IRF520 Mosfet shows that it is fully turned on when its Vgs is 10V.

If you like a"520" Mosfet then you should use an IRL520 (see the L?) that is fully turned on when its Vgs is 5V.
 

Thread Starter

aragon1971

Joined Apr 7, 2008
84
yes but when the maximum of PWM is 5 V the mosfet is ON and the motor is rotating
when in is less than the motor is stopped !
 

djsfantasi

Joined Apr 11, 2010
7,825
yes but when the maximum of PWM is 5 V the mosfet is ON and the motor is rotating
when in is less than the motor is stopped !
Did you miss Audioguru Agains post?When the MOSFET has 5V on its gate, it is almost turned off. Less than 5V, it is turned off. The MOSFET you are using needs 10V to fully turn on. Hence, you’d be better off with a logic level MOSFET.
 
Top