Continue to Site

Welcome to

Welcome to our site! is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Matching Network for Avago AMMP6408 amplifier

Not open for further replies.

Ahmad Mushtaq

Newbie level 5
Mar 3, 2011
Reaction score
Trophy points
Activity points
Avago ammp6408 is a broad band MMIC amplifier (6-18 GHz) and its RL is merely 3 dB.I need to use it in a narrow band application (9.4-9.6 GHz). i designed a single stub matching network from the given S parameters to improve its RL in the desired band. The things are working in Momentum simulation but practically the results are quite different.

can anyone help me on this ?? i think the introduction of matching network Affects the given S parametrs of the amplifier so you cann't simply do matching. It changes the characteristics of the MMIC IC.

I designed the CPWG layout with matching network at the input and then simulated in ADS using co simulation feature. then after fabrication process, i used VNA to to get Practical S params. there is as such no issue in fabrication or soldering the IC. it is taking nominal current 650 mA @5V. But i couldnt solve the issue why simple matching network was not working practically ??

Practical matching do work, question is if it is working as intended. A easy check is to measure with VNA when the stub is cut away, and if you then get same parameters as your simulation was based upon. Guess that these results will differ also.
It is easier to succeed with making a matching network practically if it is based on practically done measurements as it is complicated to get all parameters included in the simulation.
A common problem regarding this kind of matching is the VNA measurement setup calibration. Here is a VNA instruction for matching with discrete components but you can get some ideas about possible error sources: Vector network analyzer calibration, AnTune

i cut the stub away , and now the magnitude of S11 and S22 is almost the same as provided by the manufacturer.
okay now i want to ask: if u have an IC of a partially matched amplifier and u want to match it at a particular freq, can u do this or not ??

once i designed a LNA , there is compromise between RL , noise figure during input matching network design. now lets say some1 design the LNA for minimum Noise figure and then it provides u only S parameters of this module. now u say its RL is not good at this freq , can u draw a matching network only on the basis of S parameters, would it only be a conjugare match and addition of an extra matching network wouldnt change the provided S parametrs ??

S-paras is reliable. Need tradeoff among NF, RL, and Gain etc.

You write that magnitude of S11 and S22 are correct, is also the S11 angle for which the stub is designed correct?

Not open for further replies.

Part and Inventory Search

Welcome to