Maker Pro
Maker Pro

class A push - pull amplifier ?

Status
Not open for further replies.

anne ranch

Aug 2, 2021
17
Joined
Aug 2, 2021
Messages
17
I am looking into building a simple push--pull amp using 2N2222.
I need to verify that push-pull configuration COULD work just fine without heat sinks.

Also - most designs use resistor in emitter and it is USUALLY 10 Ohms.
Some designs use two or more transistors in parallel and reduce the emitter resistor down to 1 Ohm.

What exactly is the purpose of such small resistor in first place ?

I can supply schematic when somebody gets interested to discuss this .
 

Audioguru

Sep 24, 2016
3,656
Joined
Sep 24, 2016
Messages
3,656
The emitter resistor in the output transistor stage is never as high as 10 ohms, it is usually 0.1 ohms to 0.33 ohms.
The resistors allow matching of the Vbe and current gains of the output transistors.
Here is a simple class-AB amplifier using little transistors that are near their maximum allowed current so they work poorly producing distortion. The output power is so low that the transistors do not need heatsinks.
 

Attachments

  • low power audio amplifier.png
    low power audio amplifier.png
    25.4 KB · Views: 11

anne ranch

Aug 2, 2021
17
Joined
Aug 2, 2021
Messages
17
The emitter resistor in the output transistor stage is never as high as 10 ohms, it is usually 0.1 ohms to 0.33 ohms.
The resistors allow matching of the Vbe and current gains of the output transistors.
Here is a simple class-AB amplifier using little transistors that are near their maximum allowed current so they work poorly producing distortion. The output power is so low that the transistors do not need heatsinks.

1. The schematic I have looked had 10 Ohms emitter resistor with a note that "the size depends on power desired " and ended up with 1 Ohm resistor.
I am just asking for more technical explanation - this "try this and that " is little vague.
2. I have some understanding how callas A operates , however , this operating on half of the wave does not really explains why there is no need for heat-sink.
In my case I will be using 2N2222 (metal can - TO8 ?) and want to have maximum gain without smoking the transistor .
3. Now I am looking to have CW HF ( 14MHz) only , so should I be (too) concerned about distortion?
 

VenomBallistics

Aug 30, 2018
285
Joined
Aug 30, 2018
Messages
285
1. The schematic I have looked had 10 Ohms emitter resistor with a note that "the size depends on power desired " and ended up with 1 Ohm resistor.
I am just asking for more technical explanation - this "try this and that " is little vague.
2. I have some understanding how callas A operates , however , this operating on half of the wave does not really explains why there is no need for heat-sink.
What I read specified a NEED for a heat sink, which makes perfect sense as the transistor is never "off"
 
Last edited by a moderator:

WHONOES

May 20, 2017
1,217
Joined
May 20, 2017
Messages
1,217
What do you want to use your creation for. The link you provided seems to be for r.f. amplifiers.
 

Audioguru

Sep 24, 2016
3,656
Joined
Sep 24, 2016
Messages
3,656
Distortion is produced by harmonics. The harmonics of 14MHz are at 28MHz, 42MHz, 56Mhz, 70MHz and higher that cause interference to licensed radio communication and will make your circuit illegal.

A Metal case 2N2222 is expensive, rare and old but is still available as a replacement. Its maximum output power without a heatsink is only 0.5W. Maximum gain is voltage gain or current gain and is different from output power since you can have maximum gain with a very low output power (and very low heating).
 

VenomBallistics

Aug 30, 2018
285
Joined
Aug 30, 2018
Messages
285
Push-pull is not Class A, it is Class AB or B.
in the example given, true. I think our subject is terminally confused on the subject matter.
The 2N2222 is not any sort of power transistor. It wont sustain any serious duty WITH a heat sink, let alone without.
The article even goes so far as to say a heat sink is required.
 

anne ranch

Aug 2, 2021
17
Joined
Aug 2, 2021
Messages
17
I am busy trying to load / install spice circuit analysis software .
Hopefully it will actually answer my questions.
Cheers
 

anne ranch

Aug 2, 2021
17
Joined
Aug 2, 2021
Messages
17
it might answer some of them, but it does not show you heat
I am still operating under assumption that push-pull , in whatever class, doesn't IN theory produce heat - until proven otherwise.
Currently using ltspice and I have the basic schematic ready to be analyzed.
 

WHONOES

May 20, 2017
1,217
Joined
May 20, 2017
Messages
1,217
Class A can also be push / pull. Heat is generated in the majority of circuits, it just depends on the level. It is a by product of power being consumed.
 

VenomBallistics

Aug 30, 2018
285
Joined
Aug 30, 2018
Messages
285
I am still operating under assumption that push-pull , in whatever class, doesn't IN theory produce heat - until proven otherwise.
Currently using ltspice and I have the basic schematic ready to be analyzed.
a quote from the page you linked to ....
"The big disadvantage of this circuit is that TR1 must have a high standing DC current, which usually means the transistor must be fitted with a heat-sink. "
A class A amp, by definition, works on the principal of of it's active components being biased to a half way point in it's quiescent state. It's never off, current is always flowing, thus it's making heat, even at no signal.
 
Last edited:

anne ranch

Aug 2, 2021
17
Joined
Aug 2, 2021
Messages
17
a quote from the page you linked to ....
"The big disadvantage of this circuit is that TR1 must have a high standing DC current, which usually means the transistor must be fitted with a heat-sink. "
A class A amp, by definition, works on the principal of of it's active components being biased to a half way point in it's quiescent state. It's never off, current is always flowing, thus it's making heat, even at no signal.
You could read "the rest of the story " especially about "push-pull"
 

VenomBallistics

Aug 30, 2018
285
Joined
Aug 30, 2018
Messages
285
You could read "the rest of the story " especially about "push-pull"
I've built class A push pull amps, even without heat sinks.
But they were tube fired.
Do you men Class AB? If so, these are not a true class A in that they are not biased to midpoint, but rather, just inside the range of conductance. They are really just class B with a dirty shortcut at the bias generator.
 

anne ranch

Aug 2, 2021
17
Joined
Aug 2, 2021
Messages
17
I am not that concerned what is the class - the objective is to build and analyze using spice .
As long as the smoke stays in the TO can , I am happy camper.

At this point I have my basic schematic in EAGLE .
My next step is to add spice info to it - seems a little challenge.

As far as hardware - I found suitable strip board and got the transistor sockets mounted .
I did not find 1 Ohm resistor - so I got 2.7 Ohms in parallel and I can reduce it by adding more 2.7 in parallel.
Little messy but will do for the purpose of this experiment.

Now the question is - which direction to wind the load torroids?

Should the load winding start at the collector ? hence wound OPPOSITE to each other?
I am busy googling, but so far nothing...
 

WHONOES

May 20, 2017
1,217
Joined
May 20, 2017
Messages
1,217
Class AB transistor amplifiers are always biased to the mid point otherwise they would not work correctly. Class A transistor amps are just class B (or AB) biased to have a large current flowing in the output stage in the quiescent state. A class AB would have the same output stage topology but would be biased so that the output stage passes just a couple of mA in the quiescent state.
 
Status
Not open for further replies.
Top