Designer’s Guide Community

weber8722

Community Member

Offline

Posts: 95

Simulator for treating reciprocal mixing correctly?
Nov 25^th, 2011, 5:39am

Hi,

many modern harmonic balance simulators can treat circuit with both VCO and mixers correctly. So doing a noise analysis you should see the noise transfer from LO phase-noise due reciprocal mixing if a blocker is present at the RF port. I found it correctly in simulations, but only for fix-LO generator (with phase-noise), NOT for a real VCO circuit.
Ken Kundert mentioned in a great 1999 paper that many HB simulators cannot fully simulate noise correct in such cases with VCO+mixer, but now we have 2011....
What is the state of the art?

Bye Stephan

Back to top

IP Logged

sheldon

Community Fellow

Offline

Posts: 751

Re: Simulator for treating reciprocal mixing correctly?
Reply #1 - Nov 26^th, 2011, 5:30pm

Stephan,

� The issue is not the phase noise calculation. The issue is how do you
calculate the steady-state of a system that includes both driven, Mixer,
and autonomous, VCO, components. The solution has been available
for several years, semi-autonomous HB analysis. Spectre/APS RF's
semi-autonomous HB analysis will calculate the steady-state of a
mixed-systems, that is, circuits containing both autonomous and
driven components. This approach also addresses the issue of how
to simulate the effect of power supply noise on a VCO.

� � � � � � � � � � � � � � � � � � � � � � � � � � � � � �Hope this helps,

� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �Sheldon

Back to top

IP Logged

weber8722

Community Member

Offline

Posts: 95

Re: Simulator for treating reciprocal mixing correctly?
Reply #2 - Dec 2^nd, 2011, 5:13am

Hi Sheldon, thanks for your response, but let me clarify some things...

sheldon wrote on Nov 26^th, 2011, 5:30pm:

Stephan,

� The issue is not the phase noise calculation. The issue is how do you
calculate the steady-state of a system that includes both driven, Mixer,
and autonomous, VCO, components. The solution has been available
for several years, semi-autonomous HB analysis. Spectre/APS RF's
semi-autonomous HB analysis will calculate the steady-state of a
mixed-systems, that is, circuits containing both autonomous and
driven components...
� � � � � � � � � � � � � � � � � � � � � � � � � � � � � �Hope this helps,

� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �Sheldon

The problem is NOT to run an HB analysis on vco+mixer, the result from this is fine

. The problem is the HB-noise analysis on top of this! You can easily simulate the NF increase of the whole receiver using HB-noise, but it works only correct in simulators like spectrerf for fix-LO (with same phase-noise of course, driven mode), i.e. having no VCO circuit. Transient noise gives me consistent output noise in both cases. So there is no testbench problem.
I have never seen a science paper really showcasing reciprocal mixing simulation in HB, with true cross-check to tran-noise or well-known reciprocal mixing formulas. Do you know any?

Bye Stephan

Back to top

IP Logged