Distributed Amplifiers

Click here to go to our main page on amplifiers.

Fellow microwave Dudes, we sure could use some pictures to go along with this text! Send us a picture and you'll receive a cool Microwaves101 pocketknife! Don't make us draw stuff with Microsoft Word, that gets ugly for everyone.

A distributed amp is a clever way to provide enormous bandwidths, as much as 100 GHz. Some distributed amplifiers can operate down to DC as well, so they are used as opto-electronic amps. The theory behind the distributed amplifier is that a number of FETs (at least two but more typically four, five or six) are fed by a periodic structure at the input that resembles a terminated transmission line. The combination of FET capacitance with the high-impedance connection lines resembles a lumped-element version of a fifty-ohm line. This trick is used to impedance match the input and output.

One limitation of distributed amplifiers is they don't make efficient power amplifiers. The load that each FET sees is not even close to optimum, and the signal distribution is such that some FETs get far greater voltage stress than others. They make mediocre LNAs for the same reasons.

New for May 2020!  Here's a note from Brecht Machiels, who wrote his PhD thesis on distributed amplifiers.  When you download his thesis, skip down to page number  91 to learn about traveling-wave transistors, a releated topic that one day desrves a page here on Microwaves101...

I happen to have designed a couple of distributed amplifiers in the past, and you're welcome to place the attached photo on your homepage. The picture shows a low-power LNA with 16dB gain and a 22 GHz bandwidth implemented in a 90nm CMOS process. Details about the design are described in my PhD thesis, which is freely available at  http://www.mos6581.org/distributed-amplification-in-cmos. It also contains an introductory chapter on distributed amplification that might be interesting to readers.

Distributed amplifier in CMOS, courtesy of Brecht Machiels

 

Here are some pictures of other distributed amplifiers. The first one is from Agilent and is a broadband effort used in test equipment, fabricated presumably at their Santa Rosa facility.

Distributed Amplifiers

Here's a cascode distributed amp, made by Bookham:

Distributed Amplifiers

Vintage distributed amplifier!

Bonus for November 2018!  Bruno from Italy sent us these pics of a pretty old distributed amplifiers... but we'll let him explain!

Please find attached the front and back photos of a quite old DA that was designed and built here at RAI (Italian public radio/TV broadcasting company) R&D, where I have worked since 1988.

I had picked up this DA  from the scrap heap many years ago, but I ignored the exact diagram and its features. It is marked “power amplifier for 20MHz link”. Probably it was a kind of video distribution amplifier. Why distributed? Probably to obtain “high”, linear, power at “wide” bandwidth was not easy in that era (end of the sixties – beginning of the seventies?).

A couple of parts of the DA had been salvaged in the output circuit. Maybe one day I will manage to find some spare time to fix it and to test it with a VNA just out of curiosity to see how it performs!

 

 

Editor's note: considering that capacitors in this unit are fifty years old, you might want to replace some of them before you fire it up.  Especially if it was stored in a place with high humidity or was hotter than room temperature (outdoors, basement, attic). Any electrolytic or paper caps should be replaced before turn-on, but the ceramic caps (the discs) in the RF paths should be OK.  Here's a website on this subject...Good luck and send us some data!  UE 

 

Author : Unknown Editor