H-bridge Question

Thread Starter

yassser

Joined Jul 25, 2011
91
Hello,

I wanted to control a DC motor speed (but not direction) , so I made a circuit of only one switching transistor to be able to transform the low power PWM signal into a high power one to feed it to the motor .



now , if I want to control the direction with that same circuit , why wouldn't I take the output of the ct. (connected to motor ) and use some kind of relay to be able to reverse the polarity of the motor terminals back and forth by a signal from the microcontroller and be able to control the motor direction with one switching transistor .

why is that not correct , and if it is possible , why do we need 4 switching transistors in the H-bridge configuration to be able to control the motor speed and direction .
 

praondevou

Joined Jul 9, 2011
2,942
Why are you still using 1k as a gate resistor? Since you used the 2110 you can lower it now.

now , if I want to control the direction with that same circuit , why wouldn't I take the output of the ct. (connected to motor ) and use some kind of relay to be able to reverse the polarity of the motor terminals back and forth by a signal from the microcontroller and be able to control the motor direction with one switching transistor .
It's possible, but I'd recommend you make sure that you don't switch the direction while the motor is running. The motor has to come to a stop, you switch the direction with the relay, then you restart the motor (which will turn into the other direction). Also, your relay contacts have to be rated for the motor current + a safety margin, so it's not going to be one of those small relays you may think.
 
Last edited:

iONic

Joined Nov 16, 2007
1,662
You can use a DPDT relay to switch directions. As far as "Why 4 switching transistors" in a H-Bridge config. I know of a couple of reasons: The obvious is the ability for a solid state solution to direction and the other is it is more efficient. I'm sure others can offer more in depth explanations both to my reasons as well as offer others.
 

Thread Starter

yassser

Joined Jul 25, 2011
91
Why are you still using 1k as a gate resistor? Since you used the 2110 you can lower it now.
I lowered it , I just forgot to change it in the schematic.

It's possible, but I'd recommend you make sure that you don't switch the direction while the motor is running. The motor has to come to a stop, you switch the direction with the relay, then you restart the motor (which will turn into the other direction). Also, your relay contacts have to be rated for the motor current + a safety margin, so it's not going to be one of those small relays you may think.
yes of course , I would prevent the mc from accepting the direction change signal while the motor is running .
 

Thread Starter

yassser

Joined Jul 25, 2011
91
You can use a DPDT relay to switch directions. As far as "Why 4 switching transistors" in a H-Bridge config. I know of a couple of reasons: The obvious is the ability for a solid state solution to direction and the other is it is more efficient. I'm sure others can offer more in depth explanations both to my reasons as well as offer others.
this is understandable , thank you.
 

SgtWookie

Joined Jul 17, 2007
22,230
Relays are power-hungry, slow, and noisy. They can be easier to use to implement an H-bridge. Rather than a single DPDT relay, I'd suggest using two SPDT relays, as then you can have a BRAKE function as well as FWD and REVERSE.

The motor is connected between the common terminals.
Then you connect +V to the N.O. terminals, and GND to the N.C. terminals.

You won't be able to use your reverse-EMF diode as you have it shown; you'll need four of them - two from each motor terminal, one to +V, one to GND.
 

iONic

Joined Nov 16, 2007
1,662
Relays are power-hungry, slow, and noisy. They can be easier to use to implement an H-bridge. Rather than a single DPDT relay, I'd suggest using two SPDT relays, as then you can have a BRAKE function as well as FWD and REVERSE.
How would the wireing work for a break?

The motor is connected between the common terminals.
Then you connect +V to the N.O. terminals, and GND to the N.C. terminals.


You won't be able to use your reverse-EMF diode as you have it shown; you'll need four of them - two from each motor terminal, one to +V, one to GND.
How would this be wired?
 

SgtWookie

Joined Jul 17, 2007
22,230
What is that schematic you posted supposed to be?
It isn't even close to what I was talking about.

Here, like this:



The diodes take care of the reverse-EMF when the relay contacts are "flying"/bouncing - without the diodes, the relay contacts will get burned much more quickly.

When the coils are both either powered or not powered, the motor leads are effectively shorted, which acts as a brake.

There is no combination of coils energized/de-energized that will cause problems.

Your schematic DOES have problems. You can't use individual relays like that. It might work in simulation, but if one relay gets sticky, or fails to toggle, you have a dead short across the supply.
 

Attachments

Last edited:

praondevou

Joined Jul 9, 2011
2,942
Wookie meant it the other way around "The motor is connected between the common terminals.Then you connect +V to the N.O. terminals, and GND to the N.C. terminals."

But I still would recommend stopping the motor (turning off PWM) before switching the relay and reapplying PWM to the MOSFET. I'm not sure about the braking function, because the energy of the motor (generator) is already dissipated in the EMF diode, so this helps to brake it, right?

I AM sure that we used relays in the way you shown and it caused a lot of trouble. I explained it in another post, when opening a relay contact through which current is flowing it will create an arc between the contacts you just opened, if the common pole then reaches the other pole the arc will essentially short all three poles, which in the OP's case will not be good since it will provide a current path from + through the relay contacts to the drain of the MOSFET and then to GND since he didn't turn the PWM off.
I'm sure the effect is worse with DC, and 15A at 160V is a lot.

The only way to get rid off this problem is to make sure the current ceased before switching the relay (turn off PWM) or to use a relay that has a bigger contact gap.

edit: oops, Wookie was faster.
 
Last edited:

iONic

Joined Nov 16, 2007
1,662
What is that schematic you posted supposed to be?
It isn't even close to what I was talking about.

Here, like this:



The diodes take care of the reverse-EMF when the relay contacts are "flying"/bouncing - without the diodes, the relay contacts will get burned much more quickly.

When the coils are both either powered or not powered, the motor leads are effectively shorted, which acts as a brake.

There is no combination of coils energized/de-energized that will cause problems.

Your schematic DOES have problems. You can't use individual relays like that. It might work in simulation, but if one relay gets sticky, or fails to toggle, you have a dead short across the supply.
Yeah, I know it's not what you explained to the OP. I will open another thread
so that I do not hijack this one as my questions are tangent to the thread topic.
 
Top