Im trying to design a cascaded amplifier with 3 common emitter stages. As long as the input AC voltage is below about 20mV the amplification is fine, as you can see on AMP1. But the output of AMP2 is also maximum at 20mV and and goes much below zero. AMP3 is also maximum 20mV. The bias of each stage is at 773mV. How can I modify the bias of the second and third stage for higher input /output amplitude than 20mV?

Many thanks![CascadedAmplifier]1

  • \$\begingroup\$ What is the total gain needed for the circuit? \$\endgroup\$ – devnull Aug 28 '19 at 20:36
  • \$\begingroup\$ There is no special gain required. It's just an exercise to learn. Why is the upper input limit only about +20mV when connected to the following stage. The last stage with no more amplifier following has an output of about +2,6V. \$\endgroup\$ – betex Aug 28 '19 at 21:18
  • \$\begingroup\$ your frequency is way too high. Reduce your test frequency to 1000,000Hz. \$\endgroup\$ – analogsystemsrf Aug 29 '19 at 3:37
  • \$\begingroup\$ your circuit is a trans-impedance amplifier, with very low Rin. That Rin, in parallel with 1500 Rcollector, is the loadline resistance. \$\endgroup\$ – analogsystemsrf Aug 29 '19 at 3:43
  • \$\begingroup\$ @BEsmart Are you trying to learn how to design that single stage, itself? It frankly looks as though you just used a "boilerplate" approach to me, so I'm not even sure you understand one stage let alone three chained up together. Also, these have no local NFB (emitter degeneration, for example) and nothing to increase the input impedance (such as bootstrapping) and no global NFB (which you absolutely will need if you ground your emitters like that.) You should start by demonstrating just one well-designed BJT stage instead of asking why all 3 stages don't work right. Just a thought. \$\endgroup\$ – jonk Aug 29 '19 at 4:27

You can't.

A swing of 20mV seems like a pretty small signal -- but it's not small enough to have a reasonably linear current through a typical silicon diode. So the emitter current of one of those transistor stages will be severely distorted with that large of a signal.

You need feedback to linearize the circuit. Emitter degeneration would be easiest, but you may be able to do it using your existing shunt feedback and putting a resistor in series with each DC coupling capacitor.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.