0
\$\begingroup\$

I am trying to design a more linear amplifier by using less linear amplifiers in a push-pull configuration. IE run the signal through a 180deg splitter, each leg running through identical amplifiers then recombining through a 180deg combiner.

This should increase the OIP3, OIP2, and P1dB but what happens to the noise figure? I would think it should decrease by 3dB but I'm not sure that is right and am having trouble finding information on it.

Are there any other consequences of doing this with amplifiers besides added costs of more amplifiers?

EDIT:

Frequency Band: HF (3-30Mhz)

Power: Max 12dBm

For example

schematic

simulate this circuit – Schematic created using CircuitLab

As seen here in the second half, OIP3 and P1dB should improve by 3dB but no mention of the noise figure and that is what I am trying to figure out.

\$\endgroup\$
6
  • \$\begingroup\$ What is less linear amplifiers? If you mean fewer then it seems you are using more not less/fewer. \$\endgroup\$
    – Andy aka
    Commented Apr 1, 2020 at 15:54
  • \$\begingroup\$ As in the amplifiers are less linear, as in OIP3 is lower. \$\endgroup\$
    – TylerM
    Commented Apr 1, 2020 at 15:59
  • \$\begingroup\$ what powers, frequencies and applications are we talking about? (chances are it's easier to just linearize your non-linear amplifiers with other methods, but even if your approach is a sensible one, this makes a lot of difference when it comes to realistic approaches) \$\endgroup\$ Commented Apr 1, 2020 at 16:34
  • \$\begingroup\$ added more information \$\endgroup\$
    – TylerM
    Commented Apr 1, 2020 at 17:31
  • \$\begingroup\$ Is this for power amplification, or are you using this in the front end, or what? I don't think you'll improve (or degrade) the noise figure, but you should be able to calculate that yourself. \$\endgroup\$
    – TimWescott
    Commented Apr 1, 2020 at 19:40

1 Answer 1

1
\$\begingroup\$

The devil is in the details. To the degree that the two amplifiers are matched and the phase splitters & combiners match up, you should be able to reduce the 2nd-order effects nicely -- but "nicely" means something like 10 or 20dB.

You won't change the 3rd-order performance much, unless each amplifier ends up running at half power and if the amplifiers perform well at half power.

And don't be surprised if you build a push-pull power oscillator, or make matters worse -- your scheme depends on good isolation between the amplifiers.

\$\endgroup\$
1
  • \$\begingroup\$ I've added some more information to my original post \$\endgroup\$
    – TylerM
    Commented Apr 1, 2020 at 17:32

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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