high side driver

[foTONICS]

Hey guys,

So with a little help and research I came up with this circuit but Q2 keeps getting really hot. I've made sure at least one of the control signals is always high so I was wondering if anyone could give a little constructive criticism towards it,

-thanks for all the help

 

[(*steve*)]

It would be very interesting yo know what voltage is across Q2 when it is on and when it is off.

If you swap control signals 1 and 2 over, does Q2 still get hot?

Presumably no other transistor gets hot.

edit: neither Q2 nor Q3 can turn on fully. Using PNP transistors here (2955's) and altering how you drive them will make it easier to fully saturate them. Alternatively, biasing the bases from a voltage source above +12V would help.

Another alternative is to use N and P channel mosfets, but at this point we're straying a long way from your original design.

Also, the fact that neither of your "lower" transistors have base resistors could be a significant issue.

 

[foTONICS]

It would be very interesting yo know what voltage is across Q2 when it is on and when it is off.

When I have a chance to play with the circuit some more I'll let you know

Presumably no other transistor gets hot.

no just Q2

edit: neither Q2 nor Q3 can turn on fully.

the motor is only drawing at max about an amp and given the typical hfe values of MJE3055 is this a big issue?

 

[(*steve*)]

Considering that at a collector current of 1A, Q2 gets all its base current from the CB voltage drop and through a 470R resistor, you can do some calculations...

With a gain of 20, the base current required is 50mA. 50mA across 470 ohms is 23.5V -- OK, so the base doesn't get 50mA But also clearly the voltage drop across the transistor is going to be large. And that will be why its getting hot. Also the motor won't be running at full speed (or if it does, the torque will be low under load as the current will be limited to significantly less than 1A).

edit: using a PNP transistor for Q2 and its counterpart would be a great idea.

 

[foTONICS]

I noticed that I left out something that may be of importance. I origrinally tested half of this motor driver, comprising of only the transistors necessary to make the motor operate in one direction. This consisted of Q 1,2,6, and 7(I forgot to label a few of the transistors so assume 7 is the switching transistor for 6). The motor operated with just these, it's when I added the second half that things started to get hot.

 

[(*steve*)]

That would be the lack of base resistors on Q7 and Q8.

However it doesn't fix all the other problems

 

[foTONICS]

I'll whip together the other half and let you know about the voltage drop across Q2 and I'll add some base resistors

 

[foTONICS]

using a PNP transistor for Q2 and its counterpart would be a great idea.

The reason I'm using MJE3055T's is because they are the only power transistors I have on hand(I don't have any PNP power transistors). I rebuilt the circuit, it works as it should and nothing over heats. The only modification would be your suggestion of base resistors from the control signals, attached is the schematic I followed

Also: Is it a problem that only one resistor is feeding both Q6 and Q4 or should they both have their own resistor?

Also Also: thanks a lot for the help, you've been a massive help, you and the rest of the people on these forums

 

[foTONICS]

here is the breadboard

 

[(*steve*)]

Also: Is it a problem that only one resistor is feeding both Q6 and Q4 or should they both have their own resistor?

They should have their own resistors. If not, they won't share the base current evenly and one transistor may get turned on far more than the other.

I think you'll have problems if you increase the load on the H bridge. The symptom will be the high side transistor(s) getting warm/hot and the cure will be to provide the base bias from a higher voltage (this may be the easiest way) or redesign with PNP transistors.

 

[(*steve*)]

It's really weird to see a 2N3055 used to drive another transistor, but if it works...

Did you discover the wiring error (presumably) which lead to the transistor getting hot?