Amplifier Question.

Feedback and tighter component tolerances.

Or to put it another way...better design.

That's terrible quality control there.

I don't see 20% tolerance resistors much these days. If you design an amplifier with 20%-tolerance resistors, you'd get these large variances in gain and other issues (distortion, reduced gain, increase power consumption, etc).

I'm with Dell on the usage of lower tolerance components. Another suggestion is the proper sizing of the heat sinks for the power transistors or ICs.

Create a very high gain amp (using say transistors with a wide range of gains), then use feedback to set the actual gain. Just like you do with op amps. Jeff

yes, you can add deep negtive feedback loop to reduce the gain change.

I'd hate to suggest it for a real production line, but, maybe the lecturer is looking for the use of a pot or select on test components.

Questions like these do always appear when lecturers or teachers
1) have no industrial background as (sadly) usual when they`ve got the obligatory education,

2) want to lead you into a trap (when giving no more advices as you told above),
or just

3) want to know if you have an idea (or better more) of possible influences.

As a minimum hint you should know (or have an idea of) whether its a conceptional problem (basic design, selection of OP or transistors, temp-co of resistors, unwanted input-currents, mismatch of impedances), a production problem (tolerances of incoming components too big, bad offsets etc.), or a quality control problem (stronger selection, in this case not useful if quality of preproduction is so bad as it seems).

Regards Uwe