Emitter follower cannot sink current

Thread Starter

Andrea666

Joined Jan 28, 2021
9
I'm reading "The art of electronics" and at page 106 the author states ".... an npn emitter follower
cannot sink current and a pnp follower cannot source current.". What does he mean with this statement ? Why an emitter follower cannot do that?
So, he continue "The result is that a single-ended follower operating between split supplies can drive a ground-returned load only if a high quiescent current is used". How is the last statement related to the first?
I really find hard to understand what the author means in this paragraph.
 

dendad

Joined Feb 20, 2016
4,452
One way to look at it, see the circuit of the NPN emitter follower, the transistor does not connect to the 0V line, just the collector to +ve, so it can source current from the +ve to the load that goes to 0V. It has no way to sink current to the 0V rail as it has no connection there.
And the reverse is true for the PNP emitter follower. Its collector connects to 0V so it can sink current from the load connected to the +ve. The PNP transistor cannot source current from the +ve as it has no connection to do so.
I'm not sure what the second point is about. I think a circuit of what the author means may help.
 

Thread Starter

Andrea666

Joined Jan 28, 2021
9
I'm not sure what the second point is about. I think a circuit of what the author means may help.
I think the author refers to an emitter follower with one side of Re connected to one negative voltage supplier and a load connected between ground and the emitter side of Re. I don't understand the link between the two statement.
 
Last edited:

dendad

Joined Feb 20, 2016
4,452
This is probably a "Class A" amplifier. So for instance, if the max load current is 10Amps, then the 0 input signal gives an output current of 5Amps. Then, with a full swing audio input, the output can swing around the 5Amps center from 0Amps to 10Amps.
Class A amps can give very low distortion but are not efficient.
 

ericgibbs

Joined Jan 29, 2010
18,767
hi A666,
Consider you had an 'electronic module' that stated it has an Open Emitter output.
Could that output pin Sink the current in an External load, reference to ground.?

E
 

dendad

Joined Feb 20, 2016
4,452
Another way to look at it is an NPN transistor is a switch between the +ve line and a load to 0V.
And a PNP is a switch from the 0V line and a load to the +ve.
 

Thread Starter

Andrea666

Joined Jan 28, 2021
9
Then, with a full swing audio input, the output can swing around the 5Amps center from 0Amps to 10Amps.
Then the point is that I need at least 10A bias current to feed the load at its max voltage drop (i.e. 10A), is it correct ?
So, the problem is that the npn-emitter follwer can only source the current ? But why a npn different (not emitter follower) circuit which can sink current doesn't have that problem? I think there is still something I didn't get.
 

dendad

Joined Feb 20, 2016
4,452
Draw the circuits out and have a look at them.

An emitter follower can only conduct current one way.
But, if you put a NPN and PNP transistor pair of emitter followers in series, the NPN sources the current and the PNP sinks current. So, the load can have current driven both ways.
This is how a complimentary pair audio amp output works.
https://www.allaboutcircuits.com/textbook/experiments/chpt-6/class-b-audio-amplifier/

1655805396654.png
 
Last edited:

MisterBill2

Joined Jan 23, 2018
18,179
The statement about not sinking current is a basic reality because there is no path for the current to flow through. This is basic circuit theory.
 

Thread Starter

Andrea666

Joined Jan 28, 2021
9
An emitter follower can only conduct current one way.
But, if you put a NPN and PNP transistor pair of emitter followers in series, the NPN sources the current and the PNP sinks current. So, the load can have current driven both ways.
Ok now is clear about the first athor statement, thank you a lot.
But as for the second statement, the simple emitter follower must have a high bias current because an npn emitter follower can only generate current, while the last circuit you present does not require?
 

MisterBill2

Joined Jan 23, 2018
18,179
That is the way "class A"amplifiers work. They are set so that the no-signal output voltage is in the middle of the span of linear control. So the truth is that a class "A" amplifier can never exceed 50% efficiency, and usually has less than that. This is why they are usually limited to low power sections of equipment.
 

Thread Starter

Andrea666

Joined Jan 28, 2021
9
That is the way "class A"amplifiers work. They are set so that the no-signal output voltage is in the middle of the span of linear control.
Ok now is all more clear. The disadvantage in the simple npn emitter follower is due to it can only source current. Then, it must have high DC current in order to guarantee the whole output swing. The circuit with npn and pnp in cascade can both sinking and sourcing so it doesn’t need an high DC current and then is more efficient
 

MisterBill2

Joined Jan 23, 2018
18,179
Exactly. It will need a bit of forward bias so that there is not distortion in the area where the transistors are barely into conduction. That part is a bit tricky, really.
 
Top