Hello,
I am trying to level shift a PWM signal from 12 V TO 5 V i read that using BS170 MOSFET can be a solution, the voltage divider one didn't give a good result when tested out, can someone please give a clear solution.
Thanks

 

Hi 1234,
Welcome to AAC.
If you level shift using a transistor the PWM phase will be inverted.
E

 

Hello,
I am trying to level shift a PWM signal from 12 V TO 5 V i read that using BS170 MOSFET can be a solution, the voltage divider one didn't give a good result when tested out, can someone please give a clear solution.
Thanks

What was wrong with the voltage divider?

 

What was wrong with the voltage divider?

I suspect you know the answer to that.

 

I suspect you know the answer to that.

I know several possible answers.
Most likely: two grounds not connected together
Or: poor choice of resistor values

 

This might work:

https://learn.sparkfun.com/tutorials/bi-directional-logic-level-converter-hookup-guide/all

It uses the bss138 (digikey link). Someone double check me, but Vgs max is 20V, so it might be fine for your 12v to 5v shifting.

 

If you level shift using a transistor the PWM phase will be inverted.

that depends on way it is connected. in this case input voltage is larger than output so common drain connection is possible:


Vdd would be the 5V supply, not the 12V

 

And why are any of these better than a two-resistor potential divider?

 

And why are any of these better than a two-resistor potential divider?

The OP said a voltage divider didn't work, but did not specify why. Just taking a complete stab in the dark, maybe the inputs are not high impedance or have some capacitance, both of which have to be considered if using a simple voltage divider but would be less of an issue if using a transistor solution.

 

if sized correctly either solution can work. We just do not know what is this supposed to be driving and what values for resistor were used. if done correctly, transistor solution can produce good result over wider range

 

A single CMOS buffer chip would give a low-impedance push-pull output (LTspice sim below):
The six gates on the chip are connected in parallel to give a high drive capability, if needed.
The input resistor limits the current to below the 10mA max. the internal input protection diodes connected to Vdd can conduct.

 

A single CMOS buffer chip would give a low-impedance push-pull output (LTspice sim below):
The six gates on the chip are connected in parallel to give a high drive capability, if needed.
The input resistor limits the current to below the 10mA max. the internal input protection diodes connected to Vdd can conduct.

View attachment 315313

So, if the purpose of the level shift was to interface with 5V logic, then all that is needed is R1.

 

if the purpose of the level shift was to interface with 5V logic, then all that is needed is R1.

Yes, if it's CMOS logic, and that's what the TS is wanting to do.

 

Yes, if it's CMOS logic, and that's what the TS is wanting to do.

Maybe he might tell us, but I can't think of much else that needs 5V apart from logic.

 

I suspect he is trying to draw significant power from the circuit which is why the voltage divider did not work. Do not take “signal” seriously. Most people here refer to the output of a motor controller as a PWM “signal”.

We also don’t know if it is a typical open drain output of a PWM power controller, which means it could be used at 5V with no translation.