norrin
Newbie level 2
- Joined
- May 13, 2013
- Messages
- 2
- Helped
- 0
- Reputation
- 0
- Reaction score
- 0
- Trophy points
- 1,281
- Activity points
- 1,306
hi,
i'm designing an integrated k/ka-band power amplifier with cadence virtuoso which looks similar to this one:
in my case the load inductivities L1 of the cascode stage are about 100 pH. i simulated the s-parameter stabilityfactor kf. it is below 1 for a very wide frequency range from 20 GHz to 70 GHz. the minimum is kf=0.16. i also did a transient analysis but nothing oscillated in the amplifier.
when i increase my load inductivity to 500 pH or 1 nH or more then my stability factor decreases below 0 and i can see a common-mode oscillation at around 5 GHz in my circuit at the transient analysis.
i can remove this oscillation by adding series resistance to my inductivities but i really don't want to have resistivity in this circuit! i also tried many other approaches but none of those cancelled the oscillation.
so i think my circuit is always conditionally stable with load inductivities.
but how can i be sure that my circuit does not oscillate at a given load impedance at the output of about 100 ohm? is the transient analysis with a small maximum time step enough?
anouther question: how do i interpret the source/load stability circles that cadence can draw in its s-parameter analysis? can i use those to see the real stability of my circuit?
i'm designing an integrated k/ka-band power amplifier with cadence virtuoso which looks similar to this one:
in my case the load inductivities L1 of the cascode stage are about 100 pH. i simulated the s-parameter stabilityfactor kf. it is below 1 for a very wide frequency range from 20 GHz to 70 GHz. the minimum is kf=0.16. i also did a transient analysis but nothing oscillated in the amplifier.
when i increase my load inductivity to 500 pH or 1 nH or more then my stability factor decreases below 0 and i can see a common-mode oscillation at around 5 GHz in my circuit at the transient analysis.
i can remove this oscillation by adding series resistance to my inductivities but i really don't want to have resistivity in this circuit! i also tried many other approaches but none of those cancelled the oscillation.
so i think my circuit is always conditionally stable with load inductivities.
but how can i be sure that my circuit does not oscillate at a given load impedance at the output of about 100 ohm? is the transient analysis with a small maximum time step enough?
anouther question: how do i interpret the source/load stability circles that cadence can draw in its s-parameter analysis? can i use those to see the real stability of my circuit?