LNA Performance Simulations: Difference between revisions

From VLSI
Jump to navigation Jump to search
Lixiang (talk | contribs)
No edit summary
 
(29 intermediate revisions by 2 users not shown)
Line 1: Line 1:
==1. Power Consumption==
==Power Consumption==
*Ground both the input and output of your LNA. If there is no DC blocking capacitor at the output then leave it open-circuited.
*Ground both the input and output of your LNA. If there is no DC blocking capacitor at the output then leave it open-circuited.
*Using the Analog Design Environment choose to run a dc simulation and make sure that Save DC Operation Point is selected.
*Using the Analog Design Environment choose to run a dc simulation and make sure that '''Save DC Operation Point''' is selected.
*Run the simulation. When it is finished click Results -> Print -> DC Operating Points
*Run the simulation. When it is finished click '''Results -> Print -> DC Operating Points'''
*Click on the DC voltage source and note the delivered power.
*Click on the DC voltage source and note the delivered power.
==2. S-Parameter Simulations==
 
*You must first instantiate ports at both the input and output of the LNA (analogLib -> Sources -> Independent -> port).
==S-Parameter Simulations==
*You must first instantiate ports at both the input and output of the LNA '''(analogLib -> Sources -> Independent -> port)'''.
*Edit the properties of the input port  
*Edit the properties of the input port  
**Set the Resistance to 50 Ω and the Port number to 1.  
**Set the '''Resistance''' to 50 Ω and the '''Port number''' to 1.  
**Set the Source type to dc.
**Set the '''Source type''' to dc.
*Edit the properties of the output port
*Edit the properties of the output port
**Set the Resistance to 50 Ω and set the Port number to 2.
**Set the '''Resistance''' to 50 Ω and set the '''Port number''' to 2.
**Set the Source type to dc.
**Set the '''Source type''' to '''dc'''.
*In the Analog Design Environment and choose the sp analysis type.
*In the Analog Design Environment and choose the '''sp''' analysis type.
**Click the Port Select button and select the input and output port on the schematic.
**Click the '''Port Select''' button and select the input and output port on the schematic.
**Set the Sweep Variable to Frequency
**Set the '''Sweep Variable''' to '''Frequency'''
**Set the Sweep Range to Start-Stop and enter 1.5G for the Start value and 3.5G for the Stop value.
**Set the '''Sweep Range''' to '''Start-Stop''' and enter 1.5G for the '''Start''' value and 3.5G for the '''Stop''' value.
**Set the Sweep Type to Linear and the Number of Steps to 1000.
**Set the '''Sweep Type''' to Linear and the '''Number of Steps''' to 1000.
**Leave the Do Noise set to no.
**Leave the '''Do Noise''' set to '''no'''.
**Click OK
**Click '''OK'''
*Run the simulation
*Run the simulation
*To plot the results click Results -> Direct Plot -> Main Form ...  
*To plot the results click '''Results -> Direct Plot -> Main Form ...'''
**Make sure that the Analysis is sp and the Function is SP
**Make sure that the '''Analysis''' is '''sp''' and the '''Function''' is '''SP'''
**Select Plot Type as Rectangular
**Select '''Plot Type''' as '''Rectangular'''
**Set the Modifier to dB20
**Set the '''Modifier''' to '''dB20'''
**Now simply click the button corresponding to the parameter you wish to plot.
**Now simply click the button corresponding to the parameter you wish to plot.
==3. Noise Figure using S-parameter Simulation==
 
==Noise Figure using S-parameter Simulation==
*Follow steps 1 – 4d from the S-parameter simulation instructions above.  
*Follow steps 1 – 4d from the S-parameter simulation instructions above.  
*In the sp Choosing Analyses window, set Do Noise to yes
*In the '''sp''' Choosing Analyses window, set '''Do Noise''' to '''yes'''
**Select the Output Port
**Select the Output Port
**Select the Input Port
**Select the Input Port
**Click '''OK'''
*Run the simulation
*To plot the results click '''Results -> Direct Plot -> Main Form ...'''
**Make sure that the '''Analysis''' is '''sp''' and the Function is '''NF'''
**Set the '''Modifier''' to '''dB10'''
**Click on '''Plot'''
==IIP3  and 1-dB Compression Simulation==
*You must first instantiate ports at both the input and output of the LNA (analogLib -> Sources -> Independent -> port)
*Edit the properties of the input port
**Set the '''Resistance''' to 50 Ω
**Set the '''Port number''' to 1
**'''Source type''' should be '''sine'''
**Fill in '''fund1''' for the '''Frequency name 1'''
**'''Frequency 1''' should be 2.4G
**This is the frequency of the desired signal
**Fill in '''prf''' for '''Amplitude 1''' (dBm)
**This is a variable name (to be defined later) of the power of the input signal
**Click on '''Display second sinusoid'''
**Fill in '''fund2''' for '''Frequency name 2'''
**'''Frequency 2''' should be 2.42G
**This is the frequency of the second tone or “blocker”
**Fill in '''prf''' for '''Amplitude 2 (dBm)'''
**This sets the power of the blocker equal to the power of the input signal
**Click '''OK'''
*Edit the properties of the output port
**Set the '''Resistance''' to 50 Ω and the '''Port number''' to 2
**Set the '''Source type''' to '''dc'''
*In the Analog Design Environment we need to enable the pss analysis
**Verify that fund1 and fund2 are displayed in the Fundamental Tones section.
**Select Beat Frequency and click '''Auto Calculated'''
***The beat frequency should be 20 MHz, this is the greatest common divisor of '''fund1''' and '''fund2'''
**Select Number of harmonics under '''Output harmonics''' and fill in the value of 123
***This field defines the number of harmonics of the beat frequency that the simulation will consider. For IIP3 tests we need to consider up to the frequency (2*2.42GHz – 2.4GHz = 2.44GHz). This means that we need 122 harmonics of the beat frequency (2.44GHz / 20MHz = 122). We use 123 harmonics to go one harmonic higher.
**Set the '''Accuracy Defaults (errpreset)''' to '''conservative'''
**Set the''' Additional Time for Stabilization (tstab)''' to 20n
***This allows any startup transients to settle before calculating the IIP3
**Click '''Sweep'''
**Choose '''Variable''' and check''' no''' for Frequency Variable?
**Fill in prf for the variable name
***This is the variable defining the power of both the input and blocker signals
**Check''' Start-Stop''' under '''Sweep Range''' and fill in -50 for '''Start''' and 0 for''' Stop'''
**Set the Sweep Type to Linear and set the '''Step Size''' to 5
**Click OK
**Click OK
*Run the simulation
*Run the simulation
*To plot the results click Results -> Direct Plot -> Main Form ...
*To view the results for IIP3 click''' Results -> Direct Plot -> Main Form ...'''
**Make sure that the Analysis is sp and the Function is NF
**Set the '''Analysis''' to''' pss'''
**Set the Modifier to dB10
**Set the '''Function''' to '''IPN Curves'''
**Click on Plot
**Make sure that Select Port ( fixed R(port) ) is set
4. IIP3  and 1-dB Compression Simulation
**Click Variable Sweep (“prf”) for Circuit Input Power
1. You must first instantiate ports at both the input and output of the LNA (analogLib -> Sources -> Independent -> port)
**Enter -25 for '''Input Power Extrapolation Point (dBm)'''
2. Edit the properties of the input port
***Some experimentation might be in order. You want the resulting extrapolated line to match well with the straight portion of the IM3 components at low power.
a. Set the Resistance to 50 Ω
**Select '''Input Referred IP3''' and '''Order 3rd'''
b. Set the Port number to 1
**Select either 2.44G (2*2.42G – 2.4G) or 2.38G (2*2.4G – 2.42G) for the 3rd Order Harmonic
c. Source type should be sine
**Select 2.4G for the '''1st Order Harmonic'''
d. Fill in fund1 for the Frequency name 1
**Select the output port on the schematic
e. Frequency 1 should be 2.4G
**You Should get a plot similar to Fig. 1 below
i. This is the frequency of the desired signal
***Note that I use AWD whereas the default waveform viewer in Cadence is Wavescan.
f. Fill in prf for Amplitude 1 (dBm)
*To View the results for the 1-dB compression point click '''Results -> Direct Plot -> Main Form ...'''
i. This is a variable name (to be defined later) of the power of the input signal
**Set the '''Analysis''' to '''pss'''
g. Click on Display second sinusoid
**Set the '''Function''' to '''Compression Point'''
h. Fill in fund2 for Frequency name 2
**Check that '''Select Port ( fixed R(port) )''' is set
i. Frequency 2 should be 2.42G
**Select''' Output Power''' for '''Format'''
i. This is the frequency of the second tone or “blocker”
**Enter 1 for '''Gain Compression (dB)'''
j. Fill in prf for Amplitude 2 (dBm)
***We are interested in the 1-dB compression point after all
i. This sets the power of the blocker equal to the power of the input signal
**Enter -25 for '''Input Power Extrapolation Point (dBm)'''
k. Click OK
***Again some experimentation might be in order
3. Edit the properties of the output port
**Select '''Input Referred 1 dB''' Compression
a. Set the Resistance to 50 Ω and the Port number to 2
**Under the '''1st Order''' Harmonics select 2.4G
b. Set the Source type to dc
**Select the output Port on the schematic
4. In the Analog Design Environment we need to enable the pss analysis
**The resulting plot should look similar to Fig. 2 below.
a. Verify that fund1 and fund2 are displayed in the Fundamental Tones section.
b. Select Beat Frequency and click Auto Calculated
i. The beat frequency should be 20 MHz, this is the greatest common divisor of fund1 and fund2
c. Select Number of harmonics under Output harmonics and fill in the value of 123
i. This field defines the number of harmonics of the beat frequency that the simulation will consider. For IIP3 tests we need to consider up to the frequency (2*2.42GHz – 2.4GHz = 2.44GHz). This means that we need 122 harmonics of the beat frequency (2.44GHz / 20MHz = 122). We use 123 harmonics to go one harmonic higher.
d. Set the Accuracy Defaults (errpreset) to conservative
e. Set the Additional Time for Stabilization (tstab) to 20n
i. This allows any startup transients to settle before calculating the IIP3
f. Click Sweep
g. Choose Variable and check no for Frequency Variable?
h. Fill in prf for the variable name
i. This is the variable defining the power of both the input and blocker signals
i. Check Start-Stop under Sweep Range and fill in -50 for Start and 0 for Stop
j. Set the Sweep Type to Linear and set the Step Size to 5
k. Click OK
5. Run the simulation
6. To view the results for IIP3 click Results -> Direct Plot -> Main Form ...
a. Set the Analysis to pss
b. Set the Function to IPN Curves
c. Make sure that Select Port ( fixed R(port) ) is set
d. Click Variable Sweep (“prf”) for Circuit Input Power
e. Enter -25 for Input Power Extrapolation Point (dBm)
i. Some experimentation might be in order. You want the resulting extrapolated line to match well with the straight portion of the IM3 components at low power.
f. Select Input Referred IP3 and Order 3rd
g. Select either 2.44G (2*2.42G – 2.4G) or 2.38G (2*2.4G – 2.42G) for the 3rd Order Harmonic
h. Select 2.4G for the 1st Order Harmonic
i. Select the output port on the schematic
j. You Should get a plot similar to Fig. 1 below
i. Note that I use AWD whereas the default waveform viewer in Cadence is Wavescan.
7. To View the results for the 1-dB compression point click Results -> Direct Plot -> Main Form ...
a. Set the Analysis to pss
b. Set the Function to Compression Point
c. Check that Select Port ( fixed R(port) ) is set
d. Select Output Power for Format
e. Enter 1 for Gain Compression (dB)
i. We are interested in the 1-dB compression point after all
f. Enter -25 for Input Power Extrapolation Point (dBm)
i. Again some experimentation might be in order
g. Select Input Referred 1 dB Compression
h. Under the 1st Order Harmonics select 2.4G
i. Select the output Port on the schematic
j. The resulting plot should look similar to Fig. 2 below.
 




[[image:Figure1.png|700px|center]]
Figure 1: Sample IIP3 plot


[[image:Figure2.png|700px|center]]
Figure 1: Sample IIP3 plot
Figure 2: 1-dB Compression Point
Figure 2: 1-dB Compression Point

Latest revision as of 23:45, 15 March 2011

Power Consumption

  • Ground both the input and output of your LNA. If there is no DC blocking capacitor at the output then leave it open-circuited.
  • Using the Analog Design Environment choose to run a dc simulation and make sure that Save DC Operation Point is selected.
  • Run the simulation. When it is finished click Results -> Print -> DC Operating Points
  • Click on the DC voltage source and note the delivered power.

S-Parameter Simulations

  • You must first instantiate ports at both the input and output of the LNA (analogLib -> Sources -> Independent -> port).
  • Edit the properties of the input port
    • Set the Resistance to 50 Ω and the Port number to 1.
    • Set the Source type to dc.
  • Edit the properties of the output port
    • Set the Resistance to 50 Ω and set the Port number to 2.
    • Set the Source type to dc.
  • In the Analog Design Environment and choose the sp analysis type.
    • Click the Port Select button and select the input and output port on the schematic.
    • Set the Sweep Variable to Frequency
    • Set the Sweep Range to Start-Stop and enter 1.5G for the Start value and 3.5G for the Stop value.
    • Set the Sweep Type to Linear and the Number of Steps to 1000.
    • Leave the Do Noise set to no.
    • Click OK
  • Run the simulation
  • To plot the results click Results -> Direct Plot -> Main Form ...
    • Make sure that the Analysis is sp and the Function is SP
    • Select Plot Type as Rectangular
    • Set the Modifier to dB20
    • Now simply click the button corresponding to the parameter you wish to plot.

Noise Figure using S-parameter Simulation

  • Follow steps 1 – 4d from the S-parameter simulation instructions above.
  • In the sp Choosing Analyses window, set Do Noise to yes
    • Select the Output Port
    • Select the Input Port
    • Click OK
  • Run the simulation
  • To plot the results click Results -> Direct Plot -> Main Form ...
    • Make sure that the Analysis is sp and the Function is NF
    • Set the Modifier to dB10
    • Click on Plot

IIP3 and 1-dB Compression Simulation

  • You must first instantiate ports at both the input and output of the LNA (analogLib -> Sources -> Independent -> port)
  • Edit the properties of the input port
    • Set the Resistance to 50 Ω
    • Set the Port number to 1
    • Source type should be sine
    • Fill in fund1 for the Frequency name 1
    • Frequency 1 should be 2.4G
    • This is the frequency of the desired signal
    • Fill in prf for Amplitude 1 (dBm)
    • This is a variable name (to be defined later) of the power of the input signal
    • Click on Display second sinusoid
    • Fill in fund2 for Frequency name 2
    • Frequency 2 should be 2.42G
    • This is the frequency of the second tone or “blocker”
    • Fill in prf for Amplitude 2 (dBm)
    • This sets the power of the blocker equal to the power of the input signal
    • Click OK
  • Edit the properties of the output port
    • Set the Resistance to 50 Ω and the Port number to 2
    • Set the Source type to dc
  • In the Analog Design Environment we need to enable the pss analysis
    • Verify that fund1 and fund2 are displayed in the Fundamental Tones section.
    • Select Beat Frequency and click Auto Calculated
      • The beat frequency should be 20 MHz, this is the greatest common divisor of fund1 and fund2
    • Select Number of harmonics under Output harmonics and fill in the value of 123
      • This field defines the number of harmonics of the beat frequency that the simulation will consider. For IIP3 tests we need to consider up to the frequency (2*2.42GHz – 2.4GHz = 2.44GHz). This means that we need 122 harmonics of the beat frequency (2.44GHz / 20MHz = 122). We use 123 harmonics to go one harmonic higher.
    • Set the Accuracy Defaults (errpreset) to conservative
    • Set the Additional Time for Stabilization (tstab) to 20n
      • This allows any startup transients to settle before calculating the IIP3
    • Click Sweep
    • Choose Variable and check no for Frequency Variable?
    • Fill in prf for the variable name
      • This is the variable defining the power of both the input and blocker signals
    • Check Start-Stop under Sweep Range and fill in -50 for Start and 0 for Stop
    • Set the Sweep Type to Linear and set the Step Size to 5
    • Click OK
  • Run the simulation
  • To view the results for IIP3 click Results -> Direct Plot -> Main Form ...
    • Set the Analysis to pss
    • Set the Function to IPN Curves
    • Make sure that Select Port ( fixed R(port) ) is set
    • Click Variable Sweep (“prf”) for Circuit Input Power
    • Enter -25 for Input Power Extrapolation Point (dBm)
      • Some experimentation might be in order. You want the resulting extrapolated line to match well with the straight portion of the IM3 components at low power.
    • Select Input Referred IP3 and Order 3rd
    • Select either 2.44G (2*2.42G – 2.4G) or 2.38G (2*2.4G – 2.42G) for the 3rd Order Harmonic
    • Select 2.4G for the 1st Order Harmonic
    • Select the output port on the schematic
    • You Should get a plot similar to Fig. 1 below
      • Note that I use AWD whereas the default waveform viewer in Cadence is Wavescan.
  • To View the results for the 1-dB compression point click Results -> Direct Plot -> Main Form ...
    • Set the Analysis to pss
    • Set the Function to Compression Point
    • Check that Select Port ( fixed R(port) ) is set
    • Select Output Power for Format
    • Enter 1 for Gain Compression (dB)
      • We are interested in the 1-dB compression point after all
    • Enter -25 for Input Power Extrapolation Point (dBm)
      • Again some experimentation might be in order
    • Select Input Referred 1 dB Compression
    • Under the 1st Order Harmonics select 2.4G
    • Select the output Port on the schematic
    • The resulting plot should look similar to Fig. 2 below.


Error creating thumbnail: File missing

Figure 1: Sample IIP3 plot

Error creating thumbnail: File missing

Figure 2: 1-dB Compression Point