Motor Drive Circuit

[Kingpin094]

Does this circuit (see attachment) look like a good way to drive a 12V gear motor with stall current of roughly 700mA? (Vcc is 12V)

I built the circuit and it seams to work..... sorta. The motor spins quit nicely when the PWM is 100%.
There is little drop accross the H-bridge. When the PWM is set to below 100% ie. 90% the voltage
on the bridge drops from 11.5 to less then 6 and the transitors begin to get hot. The input logic
seams to function correctly and the bridge works if a'b = Vcc and ab' = GND (counterclockwise
rotation) and if a'b = GND and ab' = Vcc (clockwise rotation).

Anyone have any ideas on what could be wrong? Is this a good circuit or should I start over with
a different one?

Thanks ahead of time.

 

[ljcox]

Your circuit is wrong. I have numbered the transistors Q1 for upper left, Q2 lower left, Q3 upper right.

When a' and b are High, transistor Q2 is on and Q3 and Q4 are going on/off alternatively as the PWM signal goes H/L.

When Q4 is on, the motor is acting as a generator and so current is flowing through Q4 and the lower left diode. Since there is not much resistance in this loop, a relatively high current will flow and so I suspect that Q4 will rise out of saturation and thus be dissipating significant energy.

You need to ensure that Q4 is off when Q3 is off. Likewise for Q1 & Q2.

Also, the 1N914 diodes are not suitable. I would use power diodes such as the 1N4004.

Len

 

[Russlk]

The heating problem is due to the slow turn-off time of the transistors. When Q1 turns on, Q2 below it has not turned off so there is a short circuit across the power supply. You need to turn q2 off first, then turn Q1 on with a dead time of several microseconds.

 

[Styx]

why not use FET's ??? they are alot faster switching and dont need as much to drive tehm on. Also pretty much all FET's come with an anti-parallel diode this will reduce your component count.

Also you need interlocks. ie to stop a leg shootthrough.
Standard numbering for a single phase setup is

Q1 Q3
Q4 Q2

and for 3ph

Q1 Q3 Q5
Q4 Q6 Q2

 

[Sebi]

I agree Styx. The main problem: with logic gates drivers the PNP transistors cannot close properly.

 

[Kingpin094]

The reason I was using BJT's is that I had a bunch of them lying arround. I don't have
any FETS that can souce enough juice for a motor drive.

What if I put the logic gates on the 5 votl side and used the 23904's to drive the bridge. Do
you think that would help any? If i need to get some different parts I could but I was hoping
to make a working circuit with what I have on hand.

 

[Sebi]

Of course You can use also BJT-s. Just try out the circuit, it work.
A logic gate can't give enough current to base driving, and the high-side PNP never closed properly.

 

[Kingpin094]

Sebi will your mosfet based motor drive circuit work if I use bjt's?

 

[ljcox]

The attached circuit will correct the problem with your logic circuit.

Len