# Need help with three-stage transistor amplifier in LTSpice

I am currently attempting to design an amplifier in LTSpice for an E1 project. My professor tends to emphasize analysis of these circuits, so he hasn't really given us any kind of a design process to follow, and he hasn't really done any example problems of circuit design from the ground up. So needless to say I can use some help with this.

The criteria are as follows:

• Needs to operate between 100 and 10000 Hz
• Input voltage .02Vpp
• Must deliver 2 watts to load
• Input impedance of 2200 Ohms, output 32 Ohms

Here is what I have so far:

And right now it does pretty much nothing at all:

Really not sure where to go from here; would greatly appreciate some help. Thanks!

• Without analyzing it too closely it looks reasonable. I would have to look more closely at the output stage. Have you checked your bias points? Have you verified the gain and frequency response of the intermediate stages (including the load of the subsequent stages)? What are your supply voltages. – Edgar Brown Dec 7 '18 at 5:30
• Wait. 200V? For 2W on a 32 ohm load??!!!! Even 20V would be excessive. What are you using as transistor models? – Edgar Brown Dec 7 '18 at 5:35
• Not being familiar with amplifier design, I've rooted out some basic documents on class A amplifiers. A quick look and a go at the calculator says the first two stages each have a gain of 160, for a combined amplification of 25600. 0.02Vpp * 25600 is 512Vpp. Is that correct? – JRE Dec 7 '18 at 6:32
• You state a specification of "Input impedance of 2200 Ohms, output 32 Ohms." However, your circuit diagram shows a source impedance of 2200 Ohms. That's not an input impedance for the amplifier. That's an output impedance of your source, in the schematic. So which is it? Does the amplifier need 2200 Ohm input impedance? Or is it the source that has a 2200 Ohm output impedance? Or is the 2200 Ohm resistor considered a part of your amplifier?? – jonk Dec 7 '18 at 6:53
• Are there any quality requirements? It seems to me that you would have a large amount of cross-over distortion with this configuration, and there's no feedback. Maybe it's not important in this assignment. – pipe Dec 7 '18 at 14:44

Each of your two gain stages are in saturation.

Increase the emitter resistors to 16 thousand ohms.

To produce 2 watts across 32 ohms, use Prms= Vrms/Zload.

Reduce your +- power from 200 volts to 20 volts.

Regarding that 10,000Hz (3dB point? 0.707 down??), what is the Miller Effect input capacity of stage1? of stage2? of stage3? That Cmiller, times the Rsource, sets the stage timeconstant and thus sets the 3dB frequency.

Regarding the output stage, I'd replace each of R13 and R14 with 2 series diodes.

And put 10 ohm resistors in emitter of Q4 and Q6.

• those 10 ohms looks like to be dissipating 0.5-0.6 watts on them, so if it will be used, this should be considered I think. Also, is it certain that Cmiller will be the dominant pole? – muyustan Dec 17 '19 at 17:38