4
$\begingroup$

In the March 2000 QST there was an article on the Tuna Tin QRP transmitter, which I thought I might build: Tuna Tin schematic

The text describes Q2 as a C Class amplifier, but I just can't see how that can be true.

"It's ouput (of Q1) tickles the base of Q2 (lightly) with a few mW of drive power, causing Q2 to develop approximately 450 mW of dc input power as it is driven into the Class C mode."

The base biasing on Q2 maintains a quiescent base voltage of 1.4 V, which is plenty to bias Q2 on. I simulated the circuit in LTSpice and then disconnected the oscillator portion and sure enough there was a steady 8.9 mA of collector current. So I conclude that Q2 is in fact in Class A mode.

So, either the article has been wrong for many years (seems unlikely), or there is something here I have misunderstood (more likely :-) ).

So, my question is: Is Q2 operating in Class A or Class C?

Improve this question
$\endgroup$
1
  • 1
    $\begingroup$ But when you don't disconnect the oscillator portion, does Q2 ever turn off? If it does, it's AB/B/C, not A. The diagram indicates that the magnitude of the AC superimposed on the bias is greater than the bias itself, so C makes sense. $\endgroup$
    – hobbs - KC2G
    Jul 8, 2022 at 14:35

1 Answer 1

Reset to default
6
$\begingroup$

So, my question is: Is Q2 operating in Class A or Class C?

This is an OOK transmitter (on-off keying) where both oscillator and final amplifier are keyed. Keying only the final amplifier (with oscillator running continuously) would likely result in harsh keying, spreading key-clicks over a too-wide bandwidth.

This rising-amplitude oscillator, driving a biased power amplifier starts out Class A at the beginning of key-down. RF power output is tiny. With amplitude still rising, the power amplifier soon bottoms out (as @hobbs mentions). This means that for a portion of the RF cycle, current through Q2's collector is zero. Upon reaching full amplitude, Q2 is likely OFF more than it is ON - class C.

Hopefully, when key is dis-engaged, oscillations die out slowly too, so that key-up doesn't "click" as well.

Keep in mind that this is a very low-power transmitter. This bias arrangement where power amplifier conducts some small DC current (albeit a fraction of RF peak current) is not appropriate for higher power...a high-power RF amplifier often has its base DC grounded through a choke or step-down transformer winding. Such an amplifier rests with no DC current at all during key-up. Other methods are used to mitigate key-clicks.

Here's my LTspice simulated circuit, that excludes the RF output matching/filter circuit...it affects the key-down transient to a small extent.
Notice that the oscillator's build-up (Q2 in this schematic) is the more important part of a slow click-free rise of amplitude on key-down. LTspice circuit fragment tuna-tin

key-down transient from LTspice TRAN

Improve this answer
$\endgroup$
2
  • 1
    $\begingroup$ Now this makes sense. I always assumed a Class C amplifier is intermittently turned on, but your answer shows that we can equally have Class C by intermittently turning off the biased amplifier. A new insight for me. $\endgroup$
    – Ian_B
    Jul 8, 2022 at 23:33
  • $\begingroup$ @hobbs-KC2G astutely observers that base drive to the final stage has a large amplitude (he compares RF drive amplitude to the DC bias voltage). This is the mark of Class C, D, E...the transistor gets turned on quite hard on the +ve peak, and fully off on the negative peak of the RF cycle. Notice that collector current bottoms-out at 0mA above, when drive amplitude rises (at the 1.7ms mark). Collector current becomes a series of pulses @ 7 MHz. $\endgroup$
    – glen_geek
    Jul 9, 2022 at 1:01

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .