How to Connect Common Emitter with Emitter Follower to Speaker

[mbird]

Hi --

I was learning (just on my own via Internet for hobby) about transistors and got interested in the Common Emitter Class A amp.

I realize that I cannot connect a speaker directly to this (gain would drop severely). I have seen many articles etc. on the Internet that show two concepts for overcoming this impedance mismatch problem. One is an audio transformer in place of the CE collector resistor. The other is using an emitter follower.

I tried the Emitter Follower...

I did my calculations and was able to make a Common Emitter amp with Emitter Follower stage that when simulated did what I expected. I have attached my circuit.

Now I want to replace RL2 100K with a speaker as I have seen many places on the Internet. But when I do that the simulation shows a disaster (see attached pic with circuit and simulations).

It seems like from everything I have read that I should be able to do the following:

[Mic or Computer] --> [CE] --> [EF] --> [Speaker]

Can anyone provide guidance on this problem.

p.s. I understand this design may not be practical for real world applications -- but I am really just enjoying the CE and EF amps for fun and would like to go from simulation to breadboard and speaker. Then when that is learned maybe try some more practical amplifier circuits.

Thank you!!!

 

[duffy]

Class A with a 2N2222 just ain't got no oomph, so don't expect a heck of a lot, but...

Try reducing RB2a,b and RE2 by the same proportion. Cut the values in half first, then cut them in half again.

Also increase C6 to 250µf. That 25µf cap you're using has a whopping 16Ω Xc at that frequency - there's more voltage drop across the cap than the speaker.

 

[flat5]

A speaker has a resistance of 8 ohms, not 100,000 ohms.

 

[duffy]

He knows that. Read the post.

 

[audioguru]

The output transistor of a class-A amplifier must have a very high current for it to drive an 8 ohm speaker. Your little 2N2222 has a current of only 10mA.

For an 8 ohm load then the current in the output transistor must be about 1.25A so the output transistor must be a power transistor with heatsink (6.25W) and be driven by a driver transistor.
The resistor RE2 must be about 4 ohms and the output capacitor C6 must be 1000uF.

The output power at clipping is only 0.77W.

 

[mbird]

Class A with a 2N2222 just ain't got no oomph, so don't expect a heck of a lot, but...

Try reducing RB2a,b and RE2 by the same proportion. Cut the values in half first, then cut them in half again.

Also increase C6 to 250µf. That 25µf cap you're using has a whopping 16Ω Xc at that frequency - there's more voltage drop across the cap than the speaker.

Hi --

I tried what you said and I am now able to get about 50mW which is all I need to just hear the speaker so that is great!!! The problem is that the AC is clipped at the bottom and I can not figure out why? I attached graphic to show.

Thanks for your ideas and help!

 

[audioguru]

The bottom is clipped because your 200 ohm resistor cannot provide enough current into the output capacitor and the speaker.
My circuit uses only 4 ohms and its bottom is also clipped if the input level is increased. I even biased it at a higher voltage to help a little.

 

[mbird]

The output transistor of a class-A amplifier must have a very high current for it to drive an 8 ohm speaker. Your little 2N2222 has a current of only 10mA.

For an 8 ohm load then the current in the output transistor must be about 1.25A so the output transistor must be a power transistor with heatsink (6.25W) and be driven by a driver transistor.
The resistor RE2 must be about 4 ohms and the output capacitor C6 must be 1000uF.

The output power at clipping is only 0.77W.

Hi --

I was thinking (most of which is probably wrong so please correct!) that if I could just get 50mW I could hear something from the speaker. I just want to hear something even if low so I can take stuff I learned and calculated and put it on the breadboard. The speaker says it is a 1/4 speaker so I am hoping that even 50mW will be loud enough to hear (but that may be wrong?)

I saw so many articles on the internet where they used just a simple CE with Emitter follower that I was hoping to use just the two transistors for this experiment. The 2n2222 is small but maybe there is a better transistor that I could substitute and/or I don't have to limited to 10mA -- maybe I could redo the calcs, changing RE, and get more mA.

I am learning by doing calcs, then simulation with OrCAD trial v 16 so I am limited to what transistors I can simulate (although looks like Zetek and others have libs available for download).

I'd really like to get it to work just using the CE and EF stages (I know that is not real world but it's just for the hobby!)

Anyway I may be way off in my thinking so any nudging in the right direction would be appreciated.

Thanks!

p.s. I though of just using an audio transformer in place of the collector of my CE amp and hearing the speaker that way -- is that done or is that just not realistic?

 

[duffy]

Try reducing RB2a, mbird, you need to bring that bias point up a little.

An audio transformer will certainly help. If you can get one with a primary resistance around 200Ω, you could replace RE2 with it, get rid of the cap completely, and maybe get a half-watt out of your circuit.

 

[audioguru]

An audio transformer will certainly help. If you can get one with a primary resistance around 200Ω, you could replace RE2 with it, get rid of the cap completely, and maybe get a half-watt out of your circuit.

60 years ago amplifiers used output transformers.
If you use a 200 ohm audio transformer then the voltage swing into it will be about 8V p-p which is 2.83V RMS. Then the power is only 40mW.

To get half a watt then the supply voltage will need to be 30V.

 

[flat5]

"I did my calculations and was able to make a Common Emitter amp with Emitter Follower stage that when simulated did what I expected. I have attached my circuit.

Now I want to replace RL2 100K with a speaker as I have seen many places on the Internet. But when I do that the simulation shows a disaster (see attached pic with circuit and simulations)."

Duffy, what did I miss when I DID read the post?

 

[mbird]

Duffy, what did I miss when I DID read the post?

Flat5, Duffy -- the fault is all mine for not explaining my situation correctly -- I'm learning ! -- I appreciate all your ideas and help!

mBird

 

[duffy]

Duffy, what did I miss when I DID read the post?

This, obviously.

I realize that I cannot connect a speaker directly to this (gain would drop severely). I have seen many articles etc. on the Internet that show two concepts for overcoming this impedance mismatch problem.

 

[duffy]

Then the power is only 40mW.

Thank you Mr. Duck. Sorry, mbird - my dumb estimate was off by a whole order of magnitude. I don't usually do that.

 

[mbird]

The output transistor of a class-A amplifier must have a very high current for it to drive an 8 ohm speaker. Your little 2N2222 has a current of only 10mA.

For an 8 ohm load then the current in the output transistor must be about 1.25A so the output transistor must be a power transistor with heatsink (6.25W) and be driven by a driver transistor.
The resistor RE2 must be about 4 ohms and the output capacitor C6 must be 1000uF.

The output power at clipping is only 0.77W.

Hi --

When I add my CE as input to this circuit and simulate I see that R5 and Q2 have a large amount of wattage(>5 watts each). I only need 200mW for my speaker (maybe less -- can speaker be heard OK at 100mW?). Is there a way to drive my 200mW speaker without having to loose so much power (or is this due to the fact that the design I have is simply not reasonable and can not drive a speaker without this need of the power transistor and the huge power on R5?) Doing it this way I see I cannot use a 9V battery (since I assume I can't get enough current for Q2 and R5).

Another idea...
Is there some audio amp in a chip that I could feed my simple CE (or my CE/EF) amp to and then that chip to speaker. I really just want to "hear" the calcs I made to get my CE amp working -- just for fun.

Thanks!

Thanks!

 

[kchriste]

You can use an LM386 for a simple audio amp. Or you could do something like this:

 

[audioguru]

You can use an LM386 for a simple audio amp. Or you could do something like this:

Your amplifier does not have any negative feedback so it will be very distorted.
The 10 ohms emitter resistors cause a huge loss of signal.

 

[audioguru]

A class-A amplifier like yours wastes a huge amount of power by getting hot.
A class-AB amplifier like an LM386 is much more efficient and works very well.

The datasheet for the LM386 amplifier shows that its output at clipping with a 9V supply is 450mW into 8 ohms with low distortion. With a 6V supply the output at clipping is 200mW. Its idle current is only 4mA.

 

[kchriste]

Your amplifier does not have any negative feedback so it will be very distorted.
The 10 ohms emitter resistors cause a huge loss of signal.

The modified schematic was never intended to be "HiFi". It was merely intended to illustrate the concept of a push pull emitter follower output and demonstrate its advantages over the original circuit posted by mbird. None of the circuits posted above, yours included, used negative feedback.

 

[audioguru]

The modified schematic was never intended to be "HiFi". It was merely intended to illustrate the concept of a push pull emitter follower output and demonstrate its advantages over the original circuit posted by mbird.

Your circuit is like many simple circuits except they all have negative feedback.
Your 10 ohm emitter resistors should be 1 ohm.

None of the circuits posted above, yours included, used negative feedback.

Each of my emitter-follower transistors has 100% negative feedback.
My simulation has the level almost as high as clipping and the distortion is shown to be very low.