HBAC and HBXF for Oscillator Sensitivities Simulation

Dear sgcad,

I cannot provide a specific answer as I am not sure of your specific simulator and source settings. However, if you are trying to compare the HBAC and HBXF result, the latter using the same excitation source as HBAC analysis and the only active AC source in the schematic, I am not understanding why you chose sidevec =[-1] in the HBXF analysis and sidevec = [1] in the HBAC analysis.

sgcad said:

Now it would actually be nice to use HBXF, as with one simulation run I could derive other sensitivities of interest such as supply pushing and tuning gain. I'm using:

hbxf ( vosc VSS ) hbxf sweeptype=relative relharmvec=[1] start=1m stop=1M dec=5 sidevec=[-1] annotate=status

Comparing this sweep (harmonic -1) to the HBAC result as above I see:

I cannot provide a specific answer as I am not sure of your specific simulator and schematic source settings. However, if you are trying to compare the HBAC and HBXF result, the latter using the same excitation source as HBAC analysis and the only active AC source in the schematic, I am not understanding why you chose sidevec =[-1] in the HBXF analysis and sidevec = [1] in the HBAC analysis.

There is an example of the simulator settings to determine the power supply sensitivity of an oscillator using HBXF starting on page 629 of the "Spectre Circuit Simulator and Accelerated Parallel Simulator RF Analysis in ADE Explorer User Guide". In the example, there is no mention of setting sidevec to -1. The only time this setting is mentioned is when simulating a down-converter and looking at its lower sideband.

I might suggest you try to follow the process and simulator settings shown in this Cadence example.

https://support.cadence.com/apex/techpubDocViewerPage?xmlName=spectrerfinexplorer.xml&title=Spectre%20Circuit%20Simulator%20and%20Accelerated%20Parallel%20Simulator%20RF%20Analysis%20in%20ADE%20Explorer%20User%20Guide%20--%20Frequency%20Domain%20Analyses:%20Harmonic%20Balance%20-%20Conversion%20Gain%20From%20Power%20Supply%20to%20the%20Output%20Frequency&hash=pgfId-1373220&c_version=21.1&path=spectreRFinExplorer/spectreRFinExplorer21.1/Chap3.html#pgfId-1373220

sgcad said:

so far I've thought of the XF/HBXF/PXF analyses simply as "reversed" AC/HBAC/PAC analyses that should give exactly the same result if one picks the correct transfer function.

I think you are intending to suggest that the HBXF analysis differs from the HBAC analysis in that the former uses the input to stimuli to determine the transfer function to the defined output while the latter assumes an output and determines the defined input to provide the resulting output. If so, I think your comment is basically correct.

I know this may not specifically answer your question, for which I apologize, but it is the best I can do given my experience and the limited amount I can gather about your simulation from your Forum post!

Shawn

support.cadence.com/.../techpubDocViewerPage

Dear sgcad,

I cannot provide a specific answer as I am not sure of your specific simulator and source settings. However, if you are trying to compare the HBAC and HBXF result, the latter using the same excitation source as HBAC analysis and the only active AC source in the schematic, I am not understanding why you chose sidevec =[-1] in the HBXF analysis and sidevec = [1] in the HBAC analysis.

sgcad said:

Now it would actually be nice to use HBXF, as with one simulation run I could derive other sensitivities of interest such as supply pushing and tuning gain. I'm using:

hbxf ( vosc VSS ) hbxf sweeptype=relative relharmvec=[1] start=1m stop=1M dec=5 sidevec=[-1] annotate=status

Comparing this sweep (harmonic -1) to the HBAC result as above I see:

I cannot provide a specific answer as I am not sure of your specific simulator and schematic source settings. However, if you are trying to compare the HBAC and HBXF result, the latter using the same excitation source as HBAC analysis and the only active AC source in the schematic, I am not understanding why you chose sidevec =[-1] in the HBXF analysis and sidevec = [1] in the HBAC analysis.

There is an example of the simulator settings to determine the power supply sensitivity of an oscillator using HBXF starting on page 629 of the "Spectre Circuit Simulator and Accelerated Parallel Simulator RF Analysis in ADE Explorer User Guide". In the example, there is no mention of setting sidevec to -1. The only time this setting is mentioned is when simulating a down-converter and looking at its lower sideband.

I might suggest you try to follow the process and simulator settings shown in this Cadence example.

https://support.cadence.com/apex/techpubDocViewerPage?xmlName=spectrerfinexplorer.xml&title=Spectre%20Circuit%20Simulator%20and%20Accelerated%20Parallel%20Simulator%20RF%20Analysis%20in%20ADE%20Explorer%20User%20Guide%20--%20Frequency%20Domain%20Analyses:%20Harmonic%20Balance%20-%20Conversion%20Gain%20From%20Power%20Supply%20to%20the%20Output%20Frequency&hash=pgfId-1373220&c_version=21.1&path=spectreRFinExplorer/spectreRFinExplorer21.1/Chap3.html#pgfId-1373220

sgcad said:

so far I've thought of the XF/HBXF/PXF analyses simply as "reversed" AC/HBAC/PAC analyses that should give exactly the same result if one picks the correct transfer function.

I think you are intending to suggest that the HBXF analysis differs from the HBAC analysis in that the former uses the input to stimuli to determine the transfer function to the defined output while the latter assumes an output and determines the defined input to provide the resulting output. If so, I think your comment is basically correct.

I know this may not specifically answer your question, for which I apologize, but it is the best I can do given my experience and the limited amount I can gather about your simulation from your Forum post!

Shawn

support.cadence.com/.../techpubDocViewerPage

You might also want to look at this rather old app note (it's talking about PSRR for oscillators, but you can use the same idea to look at tuning sensitivity). It's in the SPECTRE installations at: <SPECTREinstDir>/tools/spectre/examples/SpectreRF_workshop/Legacy/PSRR_Osc_AN.pdf .

(to be honest, I didn't fully read it through to make sure it covers what you're trying to do - this is just a quick pointer for something that might be relevant).

Andrew