0
\$\begingroup\$

I'm trying to send PWM signal to MOSFET gate to control 12VDC 2A motor. The NE555 circuit is built as shown in this article (https://www.build-electronic-circuits.com/555-pwm-circuit/) with the same values+ some extra decoupling capacitors.

Schematic from link above

I want to send it over 3m cat5 cable in a twisted pair, the other pairs would be 5V+GND, 12V+GND and 12V logic on/off for the PSU.

So on one end there is NE555 and on the other mosfet with the motor.

Can I run into some problems here? I find this a bit tricky, how do I deal with the ground? Is there going to be a shift that could cause problems

\$\endgroup\$

1 Answer 1

5
\$\begingroup\$

Yes, this is going to be problematic:

  • you want to use twisted pair. To make any use of twisted pair, you'll need to drive it differentially. You don't have a differential driver! Your 555 output is single ended.
  • A 555's drive strength isn't great – so your gate signal's edges aren't incredibly steep (depending on gate capacitance) even without longer transmission line between. With transmission line, you will experience dispersion of the signal, and that leads to slower edge, meaning the MOSFET turns off and on slower, and that means it's longer in "half on" mode, which means it gets hotter. This is very undesirable. You will need to implement a "edge steepening" gate driver / line receiver to solve that.
  • Since you should be using differential signalling over twisted pair anyways, a differential receiver with some kind of Schmitt-trigger input behaviour would be desirable anyways.
  • Correlated with why you would want differential signalling: your MOSFET assumes identical ground potential as GND on the 555. Nope! That is usually not the case for motors in other places than the controller. Can't rely on that. You will need a differential receiver, a gate drive for your MOSFET that refers to local ground there, and gets its power locally.

Generally, the idea of transmitting a high-frequency signal (your PWM has very high frequency content – many times the switching frequency, because you need your PWM edges to be sharp, and not nice and sinusoidal!) over a long distance is a bit … bad. Instead, just communicate the PWM duty cycle you want (which changes much slower, necessarily!), and generate the PWM at the motor. Much easier!

Together with:

  • The 555 is a rather bad source for PWM anyways. I'm not sure why you'd ever want one to control a motor – the cheapest microcontroller can more flexibly, more reliably and easier to adjust create better PWM control.

Really, this calls for a microcontroller driving the MOSFET close to the motor, and a separate microcontroller controlling that microcontroller over a digital bus that's adequate for your distance and environment (and that bus might run very possibly be transported over twisted pair – or something else, again, depending on your distance and environment).

\$\endgroup\$
3
  • \$\begingroup\$ Thank you for your thorough answer! Really appreciate that. I was thinking that it could be easier to just wire potentiometer on one end through that cat5 cable to get the duty cycle+ for 5V use step down converter from 12V rail. As you pointed out at the end. That would be better? \$\endgroup\$ Commented Nov 13 at 20:18
  • \$\begingroup\$ hm, resistance measurements over long cables sounds bad. Go with two microcontrollers, quite honestly. Doing this with a NE555 is really not a great start. I don't know why anyone is still recommending that as PWM generator – it's literally a component from the 1960s, and so many advances in PWM generation have been made since then, even ignoring micrcontrollers (which essentially are special-purpose chips for controlling motors and the like, so exactly what you need here). Your problem here is one very classically solved with two microcontrollers – in fact, "I need to control a motor, but \$\endgroup\$ Commented Nov 13 at 22:29
  • \$\begingroup\$ … my control entry is over there" is pretty much one of the original things for which microcontrollers got invented (in the 1970s). \$\endgroup\$ Commented Nov 13 at 22:34

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.