PSS and Pnoise analysis in cadence for an injection-Locked VCO

Status
Not open for further replies.
M

MaryamHosseini

Newbie
Joined
Oct 2, 2021
Messages
4
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Visit site
Activity points
37
Hi everyone
I am designing a first harmonic injection locked VCO oscillating on 2.4GHz. To calculate PN, I ran PSS and PNOISE analysis in cadence, but there was an error "V0 is a periodic input signal, which is inconsistent with autonomous circuits" because of the existence of a periodic input voltage with the frequency of 100MHz. After setting PSS simulation according to beat freq.=100MHz, tstab=40ns, number of harmonics=30, not checking the oscillator button and PNOISE simulation according to sweeptype= default, the phase noise result is +30dBc/Hz at very low offset (10KHz) frequencies becoming flat for higher frequencies. could you please guide me with this problem? am I doing sth wrong?
 

Attachments

  • 20211002_142235.jpg
    2 MB · Views: 271

Sort by date Sort by votes
BigBoss said:
I believe you're seeing the first harmonic's Phase Noise but it should be 24th Harmonic's.
Did you do that ?
Click to expand...
I tried this one by setting the sweeptype in pnoise analysis to relative and relative harmonic to 24. The result was just a flat response rising for higher offset frequencies(the blue diagram)! I do not know if the whole settings of PSS and PNOISE like choosing the engine or beat frequency are right.
--- Updated ---

I tried different modes to see why I get positive phase noise, and it seems checking oscillator in PSS or having the same beat frequency as the frequency of output is important. which I can not have either of them in injection locked VCO!!!!
 

Attachments

  • 20211003_083711.jpg
    2.8 MB · Views: 293
  • 20211003_083844.jpg
    2.6 MB · Views: 290
  • 20211003_095428.jpg
    2.5 MB · Views: 280
Last edited:

Upvote 0 Downvote
Phase noise around 100MHz or 2.4GHz should not be different if the pss is done with a sub harmonic beat frequency, as phase noise is a sampled noise.

I guess something else is going wrong. Why do you have 100MHz tone? You are saying it is a first harmonic ILO. Do you mean to say you are using 100MHz to lock the VCO to 2.4GHz? It is impossible to do a 24th harmonic locking unless it is a proper PLL with divider. More info would be helpful.
 
Last edited:

Upvote 0 Downvote
The injected signal has the same frequency as the output has, but it is not a clean signal. So every 24 period, I replace the zero crossing of the unjected signal with the zero crossing of a clean reference to lower the jitter. In this case I have a reference with frequency of 100MHz.
(I need to inject that "not clean" signal, because it creats a topology that gives a better start up condition)
 

Upvote 0 Downvote
I think I was wrong when I said the phase noise around 100MHz and 2.4GHz is not different.

I guess in your case, some sort of aliasing of the noise must be happening when you measure the noise, which is artificially increasing the floor, since you are using a sub harmonic best frequency.
 

Upvote 0 Downvote
But according to the simulations, I can see that by changing the settings in pss and pnoise, the PN varies. I believe that the circuit has a good phase noise response but I can not find the right settings.
 

Upvote 0 Downvote
Status
Not open for further replies.

Similar threads

Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn more…