3
\$\begingroup\$

I read about distributed amplifiers. I know they are used in MMICs, but I wonder whether it would be possible to build a distributed amplifier with discrete components.

In my opinion, it would have the advantage that the output power of the amplifier is distributed to the individual transistors. Would it be possible to construct a distributed amplifier with discrete transistors for, say, 2 GHz?

\$\endgroup\$
7
  • \$\begingroup\$ 2 GHz wavelength in free space is around 6 inches (150 mm). Less on circuit board material, due to the dielectric. It might be a challenge to distribute it and still keep everything in phase. \$\endgroup\$
    – user57037
    Mar 22, 2017 at 3:42
  • 1
    \$\begingroup\$ Depends on how many taps you want. At 10 taps/wavelength, on FR-4, using 'mkeith' value of 150mm/2 or 75mm, you need a transistor every 7.5mm. Does you understanding of distribution suggest 360/10 = 36 degree phasing will produce good results? \$\endgroup\$ Mar 22, 2017 at 4:04
  • \$\begingroup\$ yes I read about that, however, I have not yet found any good books or papers which describe what the key parameters are for the design of a distributed amplifier. My idea is the following: I don't need a wide bandwidth, but I need a high power. So, instead of using a single high-power transistor, I think one could use a distributed amplifier and use cheaper transistors. This should be possible because the output power is distributed between the individual transistors. \$\endgroup\$
    – T. Pluess
    Mar 22, 2017 at 7:25
  • 1
    \$\begingroup\$ First you need to create a specification. Output power, frequency, bandwidth. Complex impedance of load. You say you only need narrow bandwidth - in that case a distributed amplifier may not be the best approach. \$\endgroup\$ May 6, 2017 at 2:53
  • 1
    \$\begingroup\$ I need a bandwidth of 100 MHz in the 2.4 GHz ISM band. However, I need an output power of 5 Watts. I thought a distributed amplifier would be an interesting approach since the 5 Watts could be distributed along many transistors. \$\endgroup\$
    – T. Pluess
    May 8, 2017 at 19:24

2 Answers 2

1
\$\begingroup\$

distributed amplifiers are well common up to 6 GHz. To bypass the wavelength problem you use transmission lines balun's and outer wide-band elements. 100 MHz in the 2.4 GHz is only 4% bandwidth that means that you will have a chance to use mostly desecrate passive devices for the matching.

To be more practical, lucky for you, you need to build an amplifier in a Wifi frequency, there are some transistors companies that offer a reference design and an assembled EV board matched to that frequency band if the quantities are right.

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

Using metamaterial (MTM) theory and Transmission Lines (TL) to conjugate the lead inductive impedances using composite right-left handed transmission lines (CRLH TLs) have enabled a whole new spectrum of applications in microwave design.

This type of TL supports an anti-parallel group and phase velocities (backward wave) , infinite wavelength propagation at the 0-order resonance, equivalent negative constitutive parameters ε_r and µ_r , among others.

Discrete pad and lead inductance makes this impossible but with well controlled impedances and 3D shielded isolation , MMIC use is acceptable to 5GHz and up with more skill. ref

\$\endgroup\$

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.