Amplifying MHz signals

[Infraviolet]

That separated handling of DC and AC will ensure that V_be and other temperature/current/manufacturing variabilities in the transistor are mostly made insignificant?

[TimFox]

The DC part reduces the effect of any change or difference in V_BE on the DC (quiescent) operating point.
Note that if you partially bypass the emitter resistor (either circuit), the required capacitor is governed by the (smaller) AC resistance needed, not the (larger) DC resistance.
Specifically, if the AC resistor is RAC and the transistor's internal resistance is r_E = (26 mV)/I_E , then the capacitor's reactance X_C must be substantially less than (RAC + r_E) to short out the DC resistor RDC at the lowest frequency of interest.

If you really need that much peak-to-peak output with that small a supply voltage, you are probably better off with a fast op-amp circuit, where the DC operating point can be set precisely by the usual circuitry, including capacitors where needed to obtain a lower DC gain (for stability) than the AC gain (for required amplification).

[Infraviolet]

I can get my quiescent V_e emitter voltage up to 1.2V this way without compromising gain when I pick appropriate values, at these levels am I in a situation where V_be variations and other variables properties of the transistor will no longer have much effect? Which equation specifically governs V_be's effect on the gain and the quiescent DC levels of the various transistor pins? Am I along the right lines thinking of:

V_b_qui is set by the biasing resistor ratio,
V_e_qui is always V_be less than V_b_qui by V_be,
R_c is controlling the current I_c,
and R_e*I_c (as I_c is virtually I_e) is controlling how high V_c_qui is

V_be changes will then alter V_e_qui, which would serve to shift V_c_qui around, but not affect gain unless V_c_qui 's shifting brought it to a value where clipping became possible?

Every 0.1V increase in V_be raises V_c_qui by 0.1V? 0.1V of drop to V_be lowers V_c_qui by 0.1V?

EDIT: might have just found what I needed
setting R_e>(2.5mV)*expected_temp_range/(proportion_error_tolerable*current) is supposed to ensure temperature variation won't interfere with a common emitter amplifier by more than the specified proportion.
http://hyperphysics.phy-astr.gsu.edu/hbase/Electronic/npnce3.html#c2

[mawyatt]

Do you know about cmos gate linear amplifiers?
The cheapest of them all:
CD4xxxUB series up to 15V Vdd
www.youtube.com/watch?app=desktop&v=e6q7graP-1Y

These CMOS inverters make wonderful amplifiers and limiters, we've used them in these modes since the early renditions of the RCA CMOS CD4000 series.

Best,

[Infraviolet]

I've got some 74HCU04 chips, the U indicates unbuffered so in theory a good choice. Will test soon.

[Infraviolet]

Just thought I'd add:

I'm ordering a PCB with the transistor version of this amplifier setup on it, all worked on breadboard. But even with SMD parts, 0603 passives and sot-23 and soic, the transistor one turns out to need a remarkably large area. I'm going to check all the practical results of how the system works with the transistor version, then later return to further testing of 74HCU04 CMOS inverters as amplifiers*, because it would be nice to get a future version of the PCB down to a more compact size.

* I had some odd behaviours and lots of noise on output signals where the transistor versions had looked to give clean signals for the same situations, may have been due to poor probing or breadboard layout at the time though, I'll try again after fully verifying the function of the design using the transistorised PCBs