2
\$\begingroup\$

I am working on the output matching network of a commercial push-pull class AB amplifier based on LDMOS BLF188. The output matching network consists of a 1:4 transformer and a Toroid with 8 turns The secondary of the transformer will be connected to a 50-ohm load/antenna. The matching network works comparatively well up to 70 MHz. I am trying to modify the network to work up to 600 MHz.As of my understanding is the core material is the restricting factor to work at the desired frequency range. I would appreciate any suggestion regarding how can I achieve good output matching up to 600 MHz. Is there an alternative way to design the output matching network with lumped elements for the push-pull configuration? Thanks in advance.

\$\endgroup\$
  • \$\begingroup\$ I'm not altogether convinced that you might know why you are matching the output to the load. It's not clear from your question and if you don't need to then it saves you extra complexity. \$\endgroup\$ – Andy aka Dec 20 '18 at 18:22
  • \$\begingroup\$ Sorry to nay-say, but if the amplifier as a whole is only designed to work in the HF then chances are you'd need to change everything to work at 600MHz, not just the output matching network. \$\endgroup\$ – TimWescott Dec 20 '18 at 18:39
  • 1
    \$\begingroup\$ This may help with the basics. slideplayer.com/slide/3606271/13 \$\endgroup\$ – Sunnyskyguy EE75 Dec 20 '18 at 19:17
2
\$\begingroup\$

Going to a transmission line transformer will help extend the bandwidth, but probably not to 600MHz!

The issue is that for a transmission line transformer (Or balun) to work well at high frequency the lines must be electrically short at that frequency, while for the thing to work well at low frequency you need the ferrite to add magnetising inductance, achieving both is hard (Above maybe 100MHz you will find that a transmission line transformer works just fine without any ferrite at all).

Further, above maybe 60MHz or so the device parasitics start to become a real pain, and narrowband matching that can tune these out becomes somewhat desirable, you can build a broadband VHF/UHF amp, but it is always going to be a compromise, and a HF/VHF/UHF amp from a single device is really stretching it.

Consider for example that the gate impedance generally falls with frequency (at least until the leadout inductance becomes dominant), at HF it is high enough that a swamped gate design works, but eventually the gate impedance becomes annoyingly low, and a narrowband match with an L network becomes the way to play (You could transformer your way down to very low Z and do a swamped gate design but gain would be very low because all the drive would be winding up in your swamping resistors without developing much voltage at the actual gate. Similar considerations apply to the drain match.

\$\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.