Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com 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.

What is Common-mode resonance in power amplifier?

Status
Not open for further replies.

ChaoFan

Full Member level 2
Joined
Dec 14, 2004
Messages
125
Helped
7
Reputation
14
Reaction score
4
Trophy points
1,298
Location
China
Activity points
745
Is there any article about Common-mode resonance? and how to avoid it in PA?
Thanks.
 

This occurs in push-pull types. If you need two transistors to get the power, use two single ended stages an quadrature hybrids on the input and output.
 

Not sure about an article, but for any differential amplifier it is important to look at the stability circles in both differential and common mode (i.e. tie your inputs together and drive it with Port 1, and tie your outputs together and connect it to Port 2).

Make sure to look at all your stability circles up to Fmax of your device. Make sure the models of your ENTIRE design are valid up to that frequency. Otherwise you'll get a nasty surprise like:

1. 20 GHz differential limiting amplifier design oscillating at 87 GHz. Engineer bailed before it was found, his replacement couldn't squelch it after multiple turns and was laid off.

2. Redesign of 20 GHz differential limiting amplifier design oscillating at 90 GHz. My foul up, but I fixed it in the next turn and discovered the important error in the fab's models for the process.

3. Sister design of an 8 GHz differential limiting amplifier oscillating at 65 GHz. Same other engineer was laid off, new engineer avoided it using newly corrected models.

4. Off the shelf 6 GHz general purpose amplifier from a pretty well known outfit oscillating at 52 GHz (I had to find it for them, as they didn't even have equipment near there, and didn't even realize their fab's process was that hot). Design is 5 years old, massive shipping delay for us, now the vendor has to do wafer quals to keep hot wafers from causing this. Likely thousands in the field that will oscillate if they get cold (it is designed into many 5.8 GHz widgets). In this case the issue was not common mode vs. differential (it was a single ended design), but simply that if your design is only a 6 GHz design, you still have to look ALL THE WAY to Fmax, or you can get royally blindsided.
 

DEAR Garbonzo,
Thank you very much!
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top