Transistors for HF amplifier

Semiconductor Discrete Semiconductor Products
, ,
1

Hello guys,

I’m working on a project for which i need 10W/30W transistors or power amplifiers for HF between 3-30MHz. I need this for my college thesis, so i need a transistor which can be simulated in Keysight ADS / Cadence AWR, kindly suggest something affordable by a college student.

Thank you!

2

Hello tpmanasi,

Welcome to the DigiKey TechForum.

That is a bit too vague. In single bi-polar transistors category for example, there are quite a bit to choose from within 10-30W range. Single Bipolar Transistors | Discrete Semiconductor Products | DigiKey

Look through those remaining parameters to help narrow the options down.

3

I would like to choose from GaN

4

Hello @tpmanasi,

Sounds like you are constructing a broadband HF linear amplifier. I have a few thoughts for your consideration:

  • Start with the ARRL handbook. This is the link to the latest edition Handbook 101. You could also review older editions as HF linear amplifier have been a staple for decades.

  • Leverage the reference designs for both layout and recommended components.

  • Consider lowering your power requirements. Mistakes with RF power transistors can be expensive. It’s like building motor controllers with MOSFETs; expect to release the magic smoke, and often. If I were your advisor, I would recommend shifting to 1 W. As silly as this sounds, see how far you can push the 2N2222. From an education and simulation standpoint, a 1W push-pull amplifier with parallel transistors isn’t that different than a 100 W amplifier. You still have all of the challenges of bias, thermal stability, heatsink, and survivability against power supply transients, VSWR, and the gremlins of self-oscillation.

  • The starting point for your specific request is this RF transistor. Recommend you contact MACOM to determine what can be simulated.

Happy soldering,

APDahlen

5

Hi tpmanasi,

Agreeing with @APDahlen on all of his points. Specifically, starting out with higher power just adds to the challenge and costs. Mistakes are generally less expensive at lower power levels.

If you specifically require GaN-based technology, here are a few more RF transistors to take a look at. As a note, both of the transistors described as “HEMT” (High Electron Mobility Transistor) are GaN-based as well.

6

Hey,

I’m already refering to the ARRL handbook.

Thank you for your advice, I’ll definetly look into 2N2222 . MACOM doesn’t have technical support for the amplifier i’m looking at. Will keep you posted

7

Hey,

I went through MACOM initially but they do not have technical support for the band, their non-linear models are valid only from 100MHz which is no way close to mine.

Thank you for the advice though, however I find that matching HEMT at this frequency is slightly difficult. Can you suggest something else as well.

8

That’s an interesting question, @tpmanasi,

Are you asking about transistor matching? If yes:

  1. The technique for matching small signal transistors is included in my video (below).

  2. The technique does not address high frequency operation.

  3. Strive to make your design tolerant of differences. Yes, there are times when matched pairs are important. However, it adds significant production cost.

If you are looking for something else, have you considered class E amplification? As a speculative and unverified idea, consider constructing a class E amplifier using a GaN device. From what I can tell, this is relatively new territory. Unfortunately, I cannot verify the validity of the idea or the applicability to your degree.

Sincerely,

Aaron