Designer’s Guide Community

fern

New Member

Offline

Posts: 2

Running differential LNA simulation
May 31^st, 2006, 9:33am

Hi All,

I'm designing a differential LNA, how do i simulate it in cadence using just 2 ports?

From analogLib, i realised there are 2 components balun and a phase splitter. What is the difference and which component should i use.

P.S. I know that there was a eariler mail regarding this question, but i did not quite understand the solution.

Thank you.

Best Regards
Fern

Back to top

IP Logged

ACWWong

Community Fellow

Offline

Posts: 539
Oxford, UK

Re: Running differential LNA simulation
Reply #1 - May 31^st, 2006, 10:29am

Try looking at

http://www.designers-guide.org/Analysis/diff.pdf

for general differential circuit simulation guidance.

you can use 1 port for the input differential, and a 2nd port as the differential load. But there is no real constraint as to the number of ports allowed when simulation is cadence (sp, or spectreRF), so long as you number and include them correctly in the simulation you can also use ports to the common-mode generation.

hope it helps.

Back to top

IP Logged

mikki33 Community Member Offline Analog/Mixed Signal/High Speed Posts: 57 Israel	Re: Running differential LNA simulation Reply #2 - Jun 1^st, 2006, 1:38am You may use VCVS (voltage controled voltage source, gain=-1) to split the single ended input signal into the differential one. The same is useful at the output to combine diff. signal into the single ended.
Back to top	If you don't have time to do it good you will find time to do it again IP Logged

ACWWong

Community Fellow

Offline

Posts: 539
Oxford, UK

Re: Running differential LNA simulation
Reply #3 - Jun 1^st, 2006, 8:20am

mikki33 wrote on Jun 1^st, 2006, 1:38am:

You may use VCVS (voltage controled voltage source, gain=-1) to split the single ended input signal into the differential one. The same is useful at the output to combine diff. signal into the single ended.

VCVS has zero source impedance, so although you can do this in the LNA design you must account for finite source and load impedances (both differential & CM).... ie you nedd to add them in to insure loading/matching effects are included. This is why the port/balun method is better.

Back to top

IP Logged

fern New Member Offline Posts: 2	Re: Running differential LNA simulation Reply #4 - Jun 4^th, 2006, 9:15pm Hi All, Thanks for the help, it was very helpful.. I'm most grateful... Regards Fern
Back to top	IP Logged