Methodology for Common Emitter Amplifier Design

Thread Starter

SamR

Joined Mar 19, 2019
5,024
This is a start for me. I'm new at amplifiers and trying to lock down a good method for designing the basic core common emitter amplifier using an NPN BJT. This is what I have come up with so far. Right off the bat is taking half the supply voltage to determine the collector resistor correct? Comments or corrections please.

Common Emitter Amplifier Design
* Determine load line Q-point from datasheet for Transistor
* VC=VS/2?
* RC=VC/IC IC from load line
* RE=(VC/10)/IC (or RC/10)
* Use transistor tester to determine HFE
* IB=IC/HFE (not used here, info only)
*VB=VE+0.7V (or Vbe from tester?)
* R1= Use voltage divider Vout=Vs(R2/R1+R2) to resolve for =VB
* R2= From voltage divider above

No feedback to the base, no emitter capacitor, just the basic core so far.

Thanks, Sam

Here is what I'm coming up with.

ComEmitAmp.PNG
 
Last edited:

Ylli

Joined Nov 13, 2015
1,086
Think you have the collector and emitter voltages reversed. But otherwise it looks about right. Don't really need the low values or resistors for the base bias, as base current is fairly low. You could use 2.2K and 10K and have the same results.
 

Thread Starter

SamR

Joined Mar 19, 2019
5,024
Oops! I had the xstr reversed when I measured. I was scratching my head a bit about that. Yeah I'm trying to learn to use spice so playing with the values a bit there before breadboarding any changes.
 

AnalogKid

Joined Aug 1, 2013
10,971
Something to consider - For a single stage amplifier like this, you don't want the stage gain to be anywhere near the transistor's "open loop" beta. Just like with an opamp, you want the gain to be decided by something else. The function of the emitter resistor is called "degeneration", a form of negative feedback.

So with a typical small transistor and an Hfe of 100 or so, if you go for a stage gain of 10, you can ignore the transistor's beta and base current in your calculations and come very close to real-world performance in the first pass. In this case, the stage gain is Rc/Re. R1 and R2 set the base voltage, which sets the emitter voltage at zero signal. With a gain of 10, for every volt of collector change there will be 0.1 V of emitter change in the opposite direction. If you've got lots of headroom, like wanting a 1 Vpp output with a 9 V Vcc, you can pretty much ignore the changing emitter voltage.

ak
 

Thread Starter

SamR

Joined Mar 19, 2019
5,024
Are the basic calculations correct? The target is for a gain of 20 and it's nowhere near that yet. Voltage can go up if necessary. Next step is to add input and output capacitors and inject a signal. Then put a Colpitts oscillator on it. Need to know if I'm going to have the correct base to start. I think it's correct but would like some confirmation.
 

AnalogKid

Joined Aug 1, 2013
10,971
For the circuit in post #1, the gain is approx. 4.5. Do you want to work through this by hand? If not, there are many online calculators for this.

And, I'm pretty sure this site has a tutorial for this somewhere.

ak
 

Thread Starter

SamR

Joined Mar 19, 2019
5,024
Actually, did the calcs and I built it and measured and had a gain of 3.5. Then I put an emitter bypass cap on it with input and output caps and injected it w/ 10kHz @ 1Vpp and getting 6.5Vpp out. A bit weak but not bad for just messing around. At least it works. Now to improve it and add the Colpitts to it.


amp.PNG SDS00010.png IMG_0435.JPG
 

Thread Starter

SamR

Joined Mar 19, 2019
5,024
Strange... I put the other scope channel on the input which is displaying on the sig gen as 1Vpp and the scope is reading it as ~82mV. That gives me a gain of 77... Dang I had the probe on 10X. Never mind.

SDS00011.png
 
Top