Teensy 4.0 Brushed Motor and Mosfet get hot

W

wasserwiesel

Well-known member
Hello there,
I use the Teensy 4.0 to control a brushed motor (8.5x20) with a PWM signal (20khz).
I created a PCB for this. It works, but the motor and the mosfet get very hot quickly. I use a 1S Lipo Battery.

Even at 1A it gets hot. If, on the other hand, I connect the motor directly to a voltage supply, it only gets slightly warm.
So it must have something to do with the circuit. The Mosfet should not get warm with such a small load.
The motor driver is actually designed for 4.5V, not 3.7V. Nevertheless, I still use it because you can see in the data sheet that it
should output around 100mA even at such a low voltage. This is sufficient as a gate current for the Mosfet at that frequency. Even 50mA should be fine.

I've already tried to measure the current at the output of the driver, but my measuring device seems to be broken, at least the fuse^^
Could the mosfet get warm because the gate current is too low?
I don't know what to do at the moment...

Anyone have an idea?

Driver: https://datasheet.lcsc.com/lcsc/2005070932_Microchip-Tech-MIC4416YM4-TR_C515508.pdf
Mosfet: https://datasheet.lcsc.com/lcsc/1806130217_UTC-Unisonic-Tech-UT4404G-SO8-R_C84908.pdf

This is the schematic:

Unbenannt.PNG
 
On the thread yesterday I wanted to make sure that the PWM signal fits. Now it's mainly about the temperature and possible reasons for it.
I would appreciate any help in this regard
 
The high temperature is not due to the pwm frequency. Yesterday I also explained why I need at least 10khz. Of course, I also reduced and increased the frequency as a test, but without success, the motor and Mosfet got hot just as quickly. 500hz 50000hz it doesn't matter.
 
Does it get hot without PWM? (0Hz?)

You said you want the motor to react quickly. With 100, or 20 or 10kHz it is not faster than with..20Hz.
Physics, fundamental laws.

Massenträgheit. Da kommst du nicht drum rum. Auch mit 100MHz nicht.
Wenn er auch ohne PWM zu warm wird, ist er zu klein dimensioniert.
 
So... and you still think its not the PWM frequency?

Logic?

There are coils in the motor.
Each time the PWM switches off thei coils just do what coils do and you get current back which is short circuit by the diodes.

The more often you do that, the higher the current.
That translates to heat.
 
Yes I do. If it were that easy, I wouldn't have opened an extra thread for it. The PWM signal itself is OK. It must have something to do with the driver or mosfet. I think I made a mistake elsewhere. Maybe I'll try a different mosfet tomorrow. Checking the gate current wouldn't be wrong either, but I'm still waiting for my new voltmeter.
 
Oh, you mean the real analog voltage at the gate. I haven't tried that yet. But I assume that it won't get that warm. After all, switching it on and off is stressful for him.
 
I have never used the mosfet before. But the datasheet didn't look bad to me. But maybe I missed something. You pretend it is something completely unnatural to control a motor with 20khz?! I have a PID loop, I can't get very far with a handful of hearts. 1khz would be sufficient for sure, but as far as I know, that wouldn't be very healthy for the motor.
 
Depends on the motor. Some are ok with fast PWM, others not. Your motor seems not to like 20kHz.
You can even control a heater with PID 0.01 Hz or less.
 
The one I use is definitely for pwm. I used it on other projects as well, but with Arduino. There I have 40mA for the gate directly from the mk. But with Teensy I had to use a driver.
 
I don't understand. You mention a PID loop yet I don't see any possibility of running the motor in reverse direction in the schematic.
 
The motor gets hot even without my pid loop. I know a motor with a high D value can easily get hot or damaged quickly. So please do not deviate from the topic now. Do you guys see other failures with my schematic or components?
 
wasserwiesel said:
A motor can be fast and slow ��
Click to expand...
A typ PID loop will cause + and - deviations, especially if you want fast response and use a high pwm freq. Regardless, the MIC4416 spec for VS is 4.5 minimum. There is nothing that indicates otherwise. A blown fuse on your current meter seems the likely result. Try the spec. minimum V and maybe it will work. Not sure the driver is even necessary with an appropriate FET.
 
Please post full details of the motor involved, ie link to datasheet or product page please.
 
wasserwiesel said:
The motor driver is actually designed for 4.5V, not 3.7V. Nevertheless, I still use it because you can see in the data sheet that it
should output around 100mA even at such a low voltage.
Click to expand...
My bet would be the driver. You are formally operating the chip outside its voltage supply range and thus:
†† Notice: The device is not guaranteed to function outside its operating ratings.

Nevertheless, I still use it because you can see in the data sheet that it should output around 100mA even at such a low voltage.
Click to expand...
Did you read that from figure 2-22? I wouldn't draw that conclusion...

Paul
 
Thanks Paul, yes, it was from that figure.
Of course there was also some speculation. But in the end it works, at least the motor does what it should and has power. Even at 3.5V, I "see" no Problem with the driver. I just don't like the heat...
To what extent could the driver still be at fault?
 
If the driver isn't driving the FET gate with nice fast edges due to the too low power supply voltage, I can imagine that the FET gets warm.

Paul
 
Assuming the output from the driver is too small or smaller than I assumed, could that lead to the heat? Maybe because the MOSFET will never turn completely on and off?
I'd be very interested in that.
 
Yes, that could lead to the heat. A FET stays only cool if you switch the gate fast - hence the use of a driver to take care of that gate current. Simply said, a FET gate is a capacitor that you need to charge and discharge quickly.
Do you happen to have an oscilloscope?

Paul
 
You must log in or register to reply here.
Back
Top