building a class AB amp

Thread Starter

jasnc2107

Joined Apr 27, 2009
23
hi...i am new to electronics...i am designing a class AB amp...i am posting the design of this amp...i don't know why i am still getting crossover distortion with this setup...?? I also got problems with the calculated stage gains...which do not equal to the gain i got in the simulation..
Thanks
 

Attachments

Ron H

Joined Apr 14, 2005
7,063
hi...i am new to electronics...i am designing a class AB amp...i am posting the design of this amp...i don't know why i am still getting crossover distortion with this setup...?? I also got problems with the calculated stage gains...which do not equal to the gain i got in the simulation..
Thanks
Without looking at the rest of the circuit, you are getting crossover distortion because your diodes are shorted out.
 

Thread Starter

jasnc2107

Joined Apr 27, 2009
23
ok.. i resolved that problems shortly thanks...i don't knw why i am supposed to get a final stage voltage of 6v and i am getting only 3.8v??...is it possible to achieve a better gain?
 

Thread Starter

jasnc2107

Joined Apr 27, 2009
23
Hi all, For the output stage of the amplifier i am trying to use a darlington configuration. Which transistor do you suggest me?
 

PRS

Joined Aug 24, 2008
989
Try running your C5 output to the center of the two diodes, but leave the lines going to the bases where they are.
 

Thread Starter

jasnc2107

Joined Apr 27, 2009
23
Yes it helped me out.. i am having no more crossover distortion. I haven't understood yet if the output stage, the push pull stage amplifies or not?
 

PRS

Joined Aug 24, 2008
989
The push pull stage is a power amp, it has no voltage amplification. So is your amplifier working well? Are you dissatisfied with the voltage gain? I believe the concept of your design is very well done.
 

Thread Starter

jasnc2107

Joined Apr 27, 2009
23
I designed the circuit on a bread board...the first two stages worked fine...strangely when i inserted the output stage (i used tip132 and tip137) the dc power supply LED starts fluctuating and the amplifier gives no output...Can anyone tell me what is happening?/
 

PRS

Joined Aug 24, 2008
989
Typically, the output amp sucks up alot of current and so you include a stage called a driver between your power amp and the voltage amp. This could be a Darlington emitter follower.
 

balisong

Joined Feb 26, 2008
27
Unless you have an ideal power supply (none of us do), you will need decoupling caps for the output stage. Place several caps between 24V and GND as close as possible to the output transistors. You want a mix of capacitors to minimize the distortion over a broad range of frequency. Exact values don't matter, but try something like 1000uF + 100uF + 10uF + 1uF. Use ceramic caps if you can for the smaller values. Don't worry about overkill, more is better.
 

balisong

Joined Feb 26, 2008
27
An AB amplifier has both the high and low side transistors on at the same time. This is pratically a dead short to the power supply. You need to limit the shoot-through current with some resistors, Re in the attach image. They need to be smaller than you speaker impedance so they don't rob too much power from your output. And they need to be power resistors, your typical 1/4W resistors will self-destruct.
 

Attachments

Thread Starter

jasnc2107

Joined Apr 27, 2009
23
They result to be active both transistors because of the diodes keeping the NPN on for a little while in negative. So what resistors do i need to place instead of the 4ohm resistors i placed between transistors? PRS i placed an emtter follower in between voltage and power stage..should it be like that? see attached. What is the scope of that exactly?
 

Attachments

balisong

Joined Feb 26, 2008
27
I'd try .33 or .47 ohm resistors rated at 2W on the output.

Does your software allow you set the series resistance of the voltage source?
This will make the simulation much more accurate when dealing with large current/power.
 

Thread Starter

jasnc2107

Joined Apr 27, 2009
23
balisong so by placing those resistors i will reduce the current that the output stage will suck. don't think the my software got that feature. I placed capacitors as you suggested...on the collector side of the output transistors. Want that kill the dc biasing on the transistor?
 

Thread Starter

jasnc2107

Joined Apr 27, 2009
23
I tried to reduce load resistance to a bigger value...that is supposed to effect the current that it is sucked by the output stage? Still the dc power supply seems like overloaded. The power supply is rated 6A
 

balisong

Joined Feb 26, 2008
27
The caps go from 24V to GND, in parallel with power source.

In the simulation (not the actual circuit) raise the voltage on the source to 25V and place a 1 ohm resistor in series with it. This is a simple approximation of a current limit source (like batteries). A power-supply is more complex with voltage regulation and over-current circuit breakers.
 

PRS

Joined Aug 24, 2008
989
Balisong is right about the need for small valued resistors at your emitters. .3 ohms is good. If you don't have them make them using larger value resistors in parallel. Using 2 resistors of 1 ohm you get .5 ohms. I did this for a 7 watt amp and 1/4 watt resistors combined to form a single .5 ohm 1/2 watt resistor.

As for the emitter follower being used to drive your power amp, its purpose is to match the low impedence of your power amp to the high impedence of your voltage amp. For the sake of maximum output swing bias the base of the emitter follower halfway between you Vcc and ground. You can do this easily by using two similar resistors to bias the base.
 

Thread Starter

jasnc2107

Joined Apr 27, 2009
23
i inserted an emitter follower stage before the both transistors of the power stage. The dc supply still is not working because i thing is overcoming the 5A limit it has. What can i do to reduce current...?
 
Top