3
\$\begingroup\$

I came across this 28 MHz transmitter design:

Transmitter design

I'm a bit puzzled by the PA stage, here's the circuit so you don't all have to follow the link above just to get an idea:

enter image description here

What puzzles me starts with the bias arrangement of Transistor T3. It seems that the transistor is turned hard off until the input signal swings more than 0.7 V above its mean.

Now, I believe this is, at it's heart, simply a "class C" amplifier, but I thought that class C amps usually had a parallel-tuned/tank circuit in the collector line, but this one simply has an inductor (L2) -- the tank that's just cut off the left of the circuit fragment is in the oscillator. That makes me think that the collector signal is "seriously ugly"(tm) and all the cleanup filtering is done by the low pass network that follows.

So, I guess my questions are:

  1. Have I understood it correctly?
  2. Is this approach fairly normal for this type of low power transmitter, or it it surprising that there's not a tank in the collector line?
  3. Is there some technique that I'm unfamiliar with in play here? (perhaps there is a tank in the collector and I simply am failing to see it!)

Thanks for any input!

\$\endgroup\$
4
  • 1
    \$\begingroup\$ (1)Understood basically correct. (2)Fairly normal. (3) Could be "Class-E" rather than "Class-C" due to series-tuning of C17, L3. We cannot tell without actual component values and waveform amplitudes. \$\endgroup\$
    – glen_geek
    Feb 21 at 17:54
  • \$\begingroup\$ @glen_geek thanks, very helpful. I shall now go do more reading on both classes C and E.. \$\endgroup\$ Feb 21 at 18:19
  • 1
    \$\begingroup\$ >>> all the cleanup filtering is done by the low pass network that follows.<<< Filter is not only a filter but also a "matching" impedance network to the antenna. \$\endgroup\$
    – Antonio51
    Feb 21 at 18:52
  • 1
    \$\begingroup\$ L2 is likely sufficiently large that it looks like a current source at RF, this is fairly common in such designs and has the virtue that the collector can swing up to well above the supply rail depending on the design of the output network. \$\endgroup\$
    – Dan Mills
    Feb 21 at 22:20

2 Answers 2

1
\$\begingroup\$

Now, I believe this is, at it's heart, simply a "class C" amplifier...

You are correct. There's not much else it could be with that biasing arrangement.

I thought that class C amps usually had a parallel-tuned/tank circuit in the collector line.

Is this approach fairly normal?

Parallel-tuned tank circuits are normal for tube amps, but the general practice that I'm familiar with (amateur radio transmitters starting from around the mid-1970's, and up to kW) uses this sort of arrangement.

Because of the inherent low-impedance, broadband nature of transistors, they "like" fixed-tuned lowpass (or relatively broad bandpass) filters on their outputs.

\$\endgroup\$
1
\$\begingroup\$

but I thought that class C amps usually had a parallel-tuned/tank circuit in the collector line ...

Not only.

Remember that an inductor has always a parasitics capacitor ... So it is well "tuned" ...

From this webpage

If one wants to simulate a similar PA ... here an example of one can get.
This is a PA having ~0.5 W as input. It can deliver ~ 5 -> 7.5 W.

The biasing is very similar.
It has another special particular (network R1-C1) often found in these amplifiers. See last simulation and compare to see what happens ...

enter image description here

And spectral result.

enter image description here

enter image description here

\$\endgroup\$

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.