H-Bridge DC Motor Controller

Using Arduino Project Guidance
#1

Hi all,

I'm fairly new to electronics, but I have an interest in automotive control systems and I've been spending a while designing an H-bridge motor driver, similar to that used for electronic throttles.

Requirements:

  • 12A Peak Current, maybe 8A sustained.
  • ability to drive at 100% PWM duty (not throttle application, but others)
  • H-Bridge (reversible direction)
  • Frequencies from 1Hz to 100kHz

Half of the Full bridge schematic attached (apologies, not very good with Fritzing)

When using MOSFETs, I see that driving the high-side gate is not straightforward. I don't want to use a bootstrap capacitor design as it won't give 100% duty, so I've opted for a P-Channel High side.

I've used a single PWM pin and a digital output to reverse the direction. The P-Channel operation will be inverse so I've included an additional 'NOT' on the input of the gate driver IC.

Things I'm unsure about:

  1. The gate driver I'm using is designed for bootstrap n-channel high side. The spec sheet does not explicitly say I can use it like this but I don't see why not. Will it work?

  2. 1R series gate resistors. I remember this from a schematic somewhere but no idea how to actually size this one

  3. The capacitors have been copied from http://www.analog.com/static/imported-files/tech_articles/Inside_iCoupler_Technology_Driving_an_H_Bridge.pdf . I don't know why they would be needed. Is it just to help out the Zener reference in supplying the required transient current? (The gate drive IC in the link would obviously be a good choice, but RS don't stock it).

  4. Anything else obviously stupid? :roll_eyes:

Spec sheets:
Gate Driver: http://docs-europe.electrocomponents.com/webdocs/0dca/0900766b80dca3c7.pdf
P-Channel MOSFET: http://docs-europe.electrocomponents.com/webdocs/0791/0900766b8079129b.pdf
NOT Gate: http://docs-europe.electrocomponents.com/webdocs/0025/0900766b800255a3.pdf
AND Gate: http://docs-europe.electrocomponents.com/webdocs/0026/0900766b80026988.pdf

#2

There have been some notable discussions of DIY H-bridge designs. Below is one to read. The most cost effective approach for time and $$ is buy a design proven to work.

#3

Thanks for the post
I take your point, buying one would be more cost effective, but I'll learn nothing - not really the aim of this exercise

That thread makes some reasonable points, but I think I've already protected for the majority of the advice.
Some of the key points:

  1. Risk of shoot through - Using AND and NOT gates, I'm using fewer PWM outputs and creating an interlock against shoot-through. I'm still vulnerable to very short-duration shoot-through, but I intend to use a scope and add buffer/delay to my gate driver inputs to mitigate this.
  2. Using P-Channel to avoid high-side gate drive complexity - included
  3. Driving P-Channel gates - Gate should be tied to source to turn off the MOSFET - pretty sure my gate drive IC will be doing this. I'm hoping the IC will drive the P-channel gate between V_B and V_S

If anyone has any replies to the questions in post 1, all comments appreciated

#4