building a class AB amp

Thread Starter

jasnc2107

Joined Apr 27, 2009
23
i won't be needing 8 diodes but? Because i will be using tip132 and tip137 as the push pull and as an emitter follower. Instead of the MJE15032 and MJE15033. I've got a power supply of 5A...and it seems it is not enough for the circuit. How can i suit the power supply to the circuit?
 

Ron H

Joined Apr 14, 2005
7,063
I don't see anything in your circuit that could draw 5 amps. I think something else is wrong. Are you sure your power supply can provide 5 amps? Have you measured the current?
 

Thread Starter

jasnc2107

Joined Apr 27, 2009
23
its when i insert the power stage that the power supply acts like that...its LED starts flashing and fluctuating...and it gives an output of 0.5v max..and i am sure that the power supply is good. It gives a 5A 120w. Don't know what to do...
 

PRS

Joined Aug 24, 2008
989
There must be a short circuit somewhere on your power amp. Look around. As Ron says, this circuit, as drawn, does not use anywhere near 5 amps. Check your components one at a time to see if they are working right.

As for your Darlington configuration using discreet transistors rather than single Darlingtons, nothing is gained. And you need a total of 4 diodes, not 2, not 8. This is because there is about 0 volts at the emitter of Q3, then .6 volts at its base, then there is 1.2 volts at the base of Q2. The center of your diode string is also about 0 volts so using 2 diodes your bias Q2 at 1.2 volts.
 

Ron H

Joined Apr 14, 2005
7,063
You have a bad component, or something is wired incorrectly. Do you have a polarized capacitor on the power stage that is connected between +24V and GND? If so, could it be connected backward?
If one or more of the diodes is reversed, that could cause your problem.
 

Thread Starter

jasnc2107

Joined Apr 27, 2009
23
found out what was wrong...had the darlington pnp of the push pull blown...now its working although it seems that i have wrong biasing but...
 

Thread Starter

jasnc2107

Joined Apr 27, 2009
23
it seems that the problem occurred because i put the darlingtons on the same heat sink. Do they get shorted that way? The darlingtons are getting very hot and the 3watt resistors i got in between are getting too..
 

bertus

Joined Apr 5, 2008
22,270
Hello,

That depends what pin is connected to the housing of the transistor.
You better use an isolation material like mica to mount the transistors and also use isolation rings on the screws.

Greetings,
Bertus
 

PRS

Joined Aug 24, 2008
989
The transistors probably have their collectors in contact with the heat sink. So you had a short between Vcc and ground. By the way, the analysis I gave you above assumed you had a two ended supply, sorry. But you still need a string of 4 diodes if you use either Darlingtons or Darlington configurations.

By the way, to tell if the case of your xistors are connected to the collectors use an ohm meter between the case and the collector pin.
 

Thread Starter

jasnc2107

Joined Apr 27, 2009
23
what do you suggest me as wattage for the 700mohm resistors i placed...the 3watt and the 5watt resistors i placed...got heated in seconds
 

Ron H

Joined Apr 14, 2005
7,063
what do you suggest me as wattage for the 700mohm resistors i placed...the 3watt and the 5watt resistors i placed...got heated in seconds
If you mean that this is without any signal applied, then the quiescent (idle) current in your output stage is too high. It should not be more than a few tens of milliamps.
 

Ron H

Joined Apr 14, 2005
7,063
it also seems that i have very poor low frequency reproduction...
YOu should seriously consider the "rubber diode" that Bertus recommended for setting output stage quiescent current. The transistor needs to be thermally coupled to the output transistors.
Your poor low frequency response is due to the low values of your emitter resistor bypass capacitors.
 

PRS

Joined Aug 24, 2008
989
Here's a sketch of my power amp stage. I figure you'd be interested because it is similar to yours. I made a 2-ended supply for it because I didn't want an output capacitor. My low frequency response is less than 10 Hz and the high is about 200 KHz. It attenuates the input by 0.84 v/v.

Note the 4 diodes. These are necessary to elliminate crossover distortion. This amp reproduces sine, triangle and square waves perfectly, however I have 0.25 volts of ripple voltage riding my supply lines. I intend to use feedback to reduce this to 0.025 volts.
 

Attachments

Last edited:

Ron H

Joined Apr 14, 2005
7,063
Here's a sketch of my power amp stage. I figure you'd be interested because it is similar to yours. I made a 2-ended supply for it because I didn't want an output capacitor. My low frequency response is less than 10 Hz and the high is about 200 KHz. It attenuates the input by 0.84 v/v.

Note the 4 diodes. These are necessary to elliminate crossover distortion. This amp reproduces sine, triangle and square waves perfectly, however I have 0.25 volts of ripple voltage riding my supply lines. I intend to use feedback to reduce this to 0.025 volts.
You need an emitter bias resistor on the left end of the 22uF cap.
 

PRS

Joined Aug 24, 2008
989
Thanks, Ron. I had forgotten to draw it. Honest, the amp works very well and is on my bench right now while I work on the voltage amp.
 
Top