1
\$\begingroup\$

I am interested in constructing my own version of an RF power limiter. There exist models such as the 11930A Power Limiter, but are very expensive. The intended use for the device is to limit input power to 30dBm, the maximum rated input power for a spectrum analyser.

What are the internal components of a RF power limiter? Can I reconstruct it using a diode configuration circuit or something as such? The applications is from 100MHz to 2GHz.

\$\endgroup\$

2 Answers 2

2
\$\begingroup\$

If you are otherwise comfortable with RF PCB design in that frequency range, then you can buy diodes from MACOM and others that will limit the RF Power.

The simplest are just regular (low-capacitance) diodes back to back. But for GHz limiting you use PIN limiter diodes, which rectify the signal and use that to bias themselves into conduction. They function as power-variable attenuators, they don't just clip the wave.

Usually these diodes conduct at 0.5 V or so, so your limiter will start to limit at +5 dBm and saturate at +15 or so. Limiting to +30 and passing +20 will be much harder.

Limiters also have a maximum rating, for affordable ($100) ones it might be 1 W continuous and 10 W for a few microseconds. 100 W pulse are available for >$1000. More than that and you're on your own.

Limiters still have a finite speed, so they can't clamp a very wide band pulse with a short rise time. They will let through some RF for a nanosecond or two, if the pulse rises quickly enough. Check the datasheet if this might apply to you.

If you're not comfortable designing the boards, look at Mini-circuits, MACOM and other manufacturers for a limiter.

\$\endgroup\$
1
  • \$\begingroup\$ 1W does seem high but you should be able to find a very large pin diode that would do the trick (albeit very dirty). You could also split the signal 4 ways, limit each split, and then recombine.....although that's about as dirty as you can get. \$\endgroup\$
    – scld
    Commented Aug 19, 2015 at 13:11
1
\$\begingroup\$

For upto 100 MHz I would suggest: indeed build your own but you need up to 2 GHz. That's RF, resistors and diodes don't work so well there anymore.

I don't think diodes will help and also, it they clip you have a mixer ;-) Not something you want when looking at signals with a spectrum analyser (unless you want to intentionally mix down to a different frequency).

There are many other companies that supply RF/microwave components like minicircuits. If you Google around for sure you can find a suitable power limiter that will fit your needs and will be much cheaper than the Agilent part. Agilent charge a premium price for their products just because they can, many large companies buy Agilent anyway.

Here's a (maybe ?) suitable alternative: RF Explorer Power Limiter 0-3 GHz

And some more here

\$\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.