+ Post New Thread
Results 1 to 16 of 16
  1. #1
    Full Member level 2
    Points: 860, Level: 6

    Join Date
    Jan 2018
    Posts
    131
    Helped
    1 / 1
    Points
    860
    Level
    6

    Response of the TLINE for different frequencies and lengths

    Hi,
    1.I have tried to sweep the length of TLINE keeping frequency constant for the open circuit case and short circuit case but I don't really see the expected reactive behavior.
    Click image for larger version. 

Name:	transmissionLineAsReactive.JPG 
Views:	11 
Size:	27.2 KB 
ID:	153758

    2.I have tried to sweep the Frequency of TLINE keeping length constant for the open circuit case and short circuit case but I don't really see the expected resonance behavior.
    Click image for larger version. 

Name:	transmissionLineAsResonance.JPG 
Views:	9 
Size:	36.0 KB 
ID:	153757

  2. #2
    Super Moderator
    Points: 257,296, Level: 100
    Awards:
    1st Helpful Member

    Join Date
    Jan 2008
    Location
    Bochum, Germany
    Posts
    44,888
    Helped
    13657 / 13657
    Points
    257,296
    Level
    100

    Re: Response of the TLINE for different frequencies and lengths

    I agree that the relation between electrical length and input impedance are presented correctly in the diagrams. You should be able to see in in a correct simulation setup. Unfortunately you neither show your simulation setup nor the results, which answers do you expect?



    •   AltAdvertisement

        
       

  3. #3
    Full Member level 2
    Points: 860, Level: 6

    Join Date
    Jan 2018
    Posts
    131
    Helped
    1 / 1
    Points
    860
    Level
    6

    Re: Response of the TLINE for different frequencies and lengths

    In the short circuit case, as I increase the length when 0<length<lambda/4, I expected to see inductive behavior and lambda/4<length<lambda/2, I expected to see capacitive behavior. And the same behavior should repeat again beyond lambda/2

    In the open circuit case, as I increase the length when 0<length<lambda/4, I expected to see capacitive behavior and lambda/4<length<lambda/2, I expected to see inductive behavior.

    I attached my test setup and results




  4. #4
    Advanced Member level 5
    Points: 29,515, Level: 41
    BigBoss's Avatar
    Join Date
    Nov 2001
    Location
    Turkey
    Posts
    4,289
    Helped
    1298 / 1298
    Points
    29,515
    Level
    41

    Re: Response of the TLINE for different frequencies and lengths

    You are sweeping "freq" variable ?? If it's so, you're doing a mistake because freq is not a variable for Cadence Spectre..
    Instead, you should fix the frequency in s-parameters simulation in Spectre ( single point) then change the Line Length.
    Also, look at Zm ( Port Input Impedance ) instead of s-parameters.



  5. #5
    Full Member level 2
    Points: 860, Level: 6

    Join Date
    Jan 2018
    Posts
    131
    Helped
    1 / 1
    Points
    860
    Level
    6

    Re: Response of the TLINE for different frequencies and lengths

    Quote Originally Posted by BigBoss View Post
    Instead, you should fix the frequency in s-parameters simulation in Spectre ( single point) then change the Line Length.
    Also, look at Zm ( Port Input Impedance ) instead of s-parameters.
    Exactly I did the same but you gave me one idea, I was not sweeping the frequency parameter in the TLINE. Should I need to do that to get its response against frequency or is sweeping the frequency in s-parameter simulation enough?

    What exactly happens when we do s-parameter simulation, does it over write all other frequency variables in the schematic (in this case frequency parameter in TLINE component)?



    •   AltAdvertisement

        
       

  6. #6
    Advanced Member level 5
    Points: 29,515, Level: 41
    BigBoss's Avatar
    Join Date
    Nov 2001
    Location
    Turkey
    Posts
    4,289
    Helped
    1298 / 1298
    Points
    29,515
    Level
    41

    Re: Response of the TLINE for different frequencies and lengths

    In fact, it's more comfortable doing this test in ADS.Cadence Spectre is not so good for this type plottings.
    Click image for larger version. 

Name:	tl_test.png 
Views:	9 
Size:	139.7 KB 
ID:	153762Click image for larger version. 

Name:	tl_test_out.png 
Views:	12 
Size:	68.3 KB 
ID:	153763



    •   AltAdvertisement

        
       

  7. #7
    Full Member level 2
    Points: 860, Level: 6

    Join Date
    Jan 2018
    Posts
    131
    Helped
    1 / 1
    Points
    860
    Level
    6

    Re: Response of the TLINE for different frequencies and lengths

    Quote Originally Posted by BigBoss View Post
    Click image for larger version. 

Name:	tl_test_out.png 
Views:	12 
Size:	68.3 KB 
ID:	153763
    Beautiful. I am not really good with ADS, as I never worked on it. Well, now I will try it in ADS.



  8. #8
    Advanced Member level 5
    Points: 16,001, Level: 30
    pancho_hideboo's Avatar
    Join Date
    Oct 2006
    Location
    Real Homeless
    Posts
    2,433
    Helped
    651 / 651
    Points
    16,001
    Level
    30

    Re: Response of the TLINE for different frequencies and lengths

    Quote Originally Posted by circuitking View Post
    Beautiful. I am not really good with ADS, as I never worked on it. Well, now I will try it in ADS.
    Both 1 and 2 are possible in any simulator.
    Simply still you can not understand parameters of tline.



  9. #9
    Full Member level 2
    Points: 860, Level: 6

    Join Date
    Jan 2018
    Posts
    131
    Helped
    1 / 1
    Points
    860
    Level
    6

    Re: Response of the TLINE for different frequencies and lengths

    Quote Originally Posted by pancho_hideboo View Post
    Simply still you can not understand parameters of tline.
    Yes, for sure, I still don't understand not only about tline but also many other things. That's why I am here to learn from experts like you.



  10. #10
    Advanced Member level 5
    Points: 13,414, Level: 27

    Join Date
    Apr 2014
    Posts
    2,082
    Helped
    844 / 844
    Points
    13,414
    Level
    27

    Re: Response of the TLINE for different frequencies and lengths

    Quote Originally Posted by circuitking View Post
    Yes, for sure, I still don't understand not only about tline but also many other things. That's why I am here to learn from experts like you.
    Don't feed the troll.

    - - - Updated - - -

    Quote Originally Posted by circuitking View Post
    I was not sweeping the frequency parameter in the TLINE. Should I need to do that
    No, the frequency + phase delay in TLINE is only another way to specify the line length. So this should be constant during your frequency sweep.



  11. #11
    Advanced Member level 5
    Points: 16,001, Level: 30
    pancho_hideboo's Avatar
    Join Date
    Oct 2006
    Location
    Real Homeless
    Posts
    2,433
    Helped
    651 / 651
    Points
    16,001
    Level
    30

    Re: Response of the TLINE for different frequencies and lengths

    Quote Originally Posted by circuitking View Post
    2.I have tried to sweep the Frequency of TLINE keeping length constant
    for the open circuit case and short circuit case
    I use Synopsys HSPICE from Cadence ADE.
    You may use ".LIN Analysis" instead of ".AC Analysis".

    "test_circuitslave_1.spi"
    Code:
    ** Generated for: hspiceD
    ** Generated on: Jun 17 11:47:47 2019
    ** Design library name: My_RFDE_Test
    ** Design cell name: test_circuitslave
    ** Design view name: schematic
    .PARAM f0=1G nl0=0.25
    
    .PROBE AC
    +    I(v1) IP(v1)
    +    I(v0) IP(v0)
    .AC LIN 1001 1e-3 4e9
    
    .TEMP 25.0
    .OPTION
    +    ACCURATE
    +    ACOUT=0
    +    LIST=3
    +    ARTIST=2
    +    INGOLD=2
    +    INTERP
    +    METHOD=GEAR
    +    NOMOD
    +    PARHIER=LOCAL
    +    PROBE
    +    PSF=2
    +    RUNLVL=6
    
    ** Library name: My_RFDE_Test
    ** Cell name: test_circuitslave
    ** View name: schematic
    t1 net03 0 net02 0 ZO=50 F=f0 NL=nl0 IC=0 , 0 , 0 , 0
    t0 net08 0 0 0 ZO=50 F=f0 NL=nl0 IC=0 , 0 , 0 , 0
    v1 net03 0 AC 1
    v0 net08 0 AC 1
    .END
    "test_circuitslave_1.ocn"
    Code:
    simulator( 'hspiceD )
    design(	"/AHO/simulation/test_circuitslave/hspiceD/schematic/netlist/netlist" )
    resultsDir( "/AHO/simulation/test_circuitslave/hspiceD/schematic" )
    analysis('ac ?sweepType "linear"  ?numPoints "1001"  ?fstart "1e-3"  
    		?fstop "4G"  ?points "1G"  )
    desVar(	  "f0" 1G	)
    desVar(	  "NL0" 0.25	)
    option(	'TEMPDC  "25.0" )
    save( 'i "/V0/PLUS" "/V1/PLUS" )
    temp( 25.0 ) 
    run()
    
    Xin_short = imag((-1 / getData("/i(v0)" ?result "frequencySweep")))
    Xin_open = imag((-1 / getData("/i(v1)" ?result "frequencySweep")))
    
    plot( Xin_short ?expr '( "Xin_short" ) )
    plot( Xin_open ?expr '( "Xin_open" ) )


    1 members found this post helpful.

    •   AltAdvertisement

        
       

  12. #12
    Advanced Member level 5
    Points: 16,001, Level: 30
    pancho_hideboo's Avatar
    Join Date
    Oct 2006
    Location
    Real Homeless
    Posts
    2,433
    Helped
    651 / 651
    Points
    16,001
    Level
    30

    Re: Response of the TLINE for different frequencies and lengths

    Quote Originally Posted by circuitking View Post
    1.I have tried to sweep the length of TLINE keeping frequency constant
    for the open circuit case and short circuit case
    I used Cadence Spectre from Cadence ADE.
    You may use "sp Analysis" instead of "ac Analysis".

    "test_circuitslave_2.scs"
    Code:
    // Generated for: spectre
    // Generated on: Jun 17 11:59:19 2019
    // Design library name: My_RFDE_Test
    // Design cell name: test_circuitslave
    // Design view name: schematic
    simulator lang=spectre
    global 0
    parameters f0=1G NL0=0.25
    
    // Library name: My_RFDE_Test
    // Cell name: test_circuitslave
    // View name: schematic
    T1 (net03 0 net02 0) tline z0=50 f=f0 nl=NL0
    T0 (net08 0 0 0) tline z0=50 f=f0 nl=NL0
    V1 (net03 0) vsource mag=1 type=dc
    V0 (net08 0) vsource mag=1 type=dc
    simulatorOptions options psfversion="1.1.0" reltol=1e-3 vabstol=1e-6 \
        iabstol=1e-12 temp=25.0 tnom=25.0 scalem=1.0 scale=1.0 gmin=1e-12 \
        rforce=1 maxnotes=5 maxwarns=5 digits=5 cols=80 pivrel=1e-3 \
        sensfile="../psf/sens.output" checklimitdest=psf 
    ac ac freq=1G param=NL0 start=1e-3 stop=1 lin=1001 save=all \
        annotate=status 
    save V0:p V1:p 
    saveOptions options save=selected saveahdlvars=all
    "test_circuitslave_2.ocn"
    Code:
    simulator( 'spectre )
    design(	"/AHO/simulation/test_circuitslave/spectre/schematic/netlist/netlist" )
    resultsDir( "/AHO/simulation/test_circuitslave/spectre/schematic" )
    analysis('ac ?freq "1G"  ?param "NL0"  ?start "1e-3"  
    		?stop "1"  ?lin "1001"  ?dec ""  )
    desVar(	  "f0" 1G	)
    desVar(	  "NL0" 0.25	)
    envOption(
    	'analysisOrder  list("ac") 
    )
    save( 'i "/V0/PLUS" "/V1/PLUS" )
    temp( 25.0 ) 
    run()
    
    Xin_short = imag((-1 / getData("/V0/PLUS" ?result "ac")))
    Xin_open = imag((-1 / getData("/V1/PLUS" ?result "ac")))
    
    plot( Xin_short ?expr '( "Xin_short" ) )
    plot( Xin_open ?expr '( "Xin_open" ) )
    I can use ADSsim and Goldengate Simulator from ADE in addition to Spectre and HSPICE.
    Any simulator can give same results.
    Of course, LTspice have "tline".
    Last edited by pancho_hideboo; 17th June 2019 at 13:59.


    1 members found this post helpful.

  13. #13
    Full Member level 2
    Points: 860, Level: 6

    Join Date
    Jan 2018
    Posts
    131
    Helped
    1 / 1
    Points
    860
    Level
    6

    Re: Response of the TLINE for different frequencies and lengths

    Quote Originally Posted by pancho_hideboo View Post

    "test_circuitslave_2.scs"
    Code:
    // Generated for: spectre
    // Generated on: Jun 17 11:59:19 2019
    // Design library name: My_RFDE_Test
    // Design cell name: test_circuitslave
    // Design view name: schematic
    simulator lang=spectre
    global 0
    parameters f0=1G NL0=0.25
    
    // Library name: My_RFDE_Test
    // Cell name: test_circuitslave
    // View name: schematic
    T1 (net03 0 net02 0) tline z0=50 f=f0 nl=NL0
    T0 (net08 0 0 0) tline z0=50 f=f0 nl=NL0
    V1 (net03 0) vsource mag=1 type=dc
    V0 (net08 0) vsource mag=1 type=dc
    simulatorOptions options psfversion="1.1.0" reltol=1e-3 vabstol=1e-6 \
        iabstol=1e-12 temp=25.0 tnom=25.0 scalem=1.0 scale=1.0 gmin=1e-12 \
        rforce=1 maxnotes=5 maxwarns=5 digits=5 cols=80 pivrel=1e-3 \
        sensfile="../psf/sens.output" checklimitdest=psf 
    ac ac freq=1G param=NL0 start=1e-3 stop=1 lin=1001 save=all \
        annotate=status 
    save V0:p V1:p 
    saveOptions options save=selected saveahdlvars=all
    You are fantastic. Which editior do you use to write these codes. Also does connecting to "NoConn" instance means it is open circuited?



  14. #14
    Advanced Member level 5
    Points: 16,001, Level: 30
    pancho_hideboo's Avatar
    Join Date
    Oct 2006
    Location
    Real Homeless
    Posts
    2,433
    Helped
    651 / 651
    Points
    16,001
    Level
    30

    Re: Response of the TLINE for different frequencies and lengths

    Quote Originally Posted by circuitslave View Post
    Which editior do you use to write these codes.
    What do you want to mean ?
    I don’t use any editor.
    Do you surely understand Cadence ADE ?
    Can you understand netlist ?

    Quote Originally Posted by circuitslave View Post
    Also does connecting to "NoConn" instance means it is open circuited?
    No.
    It is not circuit. No more than symbol.
    Surely see netlist.
    Last edited by pancho_hideboo; 24th June 2019 at 19:56.



  15. #15
    Full Member level 2
    Points: 860, Level: 6

    Join Date
    Jan 2018
    Posts
    131
    Helped
    1 / 1
    Points
    860
    Level
    6

    Re: Response of the TLINE for different frequencies and lengths

    Quote Originally Posted by pancho_hideboo View Post

    Surely see netlist.
    As pet T1 (net03 0 net02 0), net02 is connected to T1 on one side. Where is the other side of net02 connected?. In the image you added, it is connected to "NoConn" instance. How did you get that? does connecting a net to NoConn also mean leaving it as open?



  16. #16
    Advanced Member level 5
    Points: 16,001, Level: 30
    pancho_hideboo's Avatar
    Join Date
    Oct 2006
    Location
    Real Homeless
    Posts
    2,433
    Helped
    651 / 651
    Points
    16,001
    Level
    30

    Re: Response of the TLINE for different frequencies and lengths

    Quote Originally Posted by circuitking View Post
    As pet T1 (net03 0 net02 0), net02 is connected to T1 on one side.
    Where is the other side of net02 connected?.
    Nothing.

    Quote Originally Posted by circuitking View Post
    In the image you added, it is connected to "NoConn" instance.
    How did you get that?
    What do you want to mean ?

    Quote Originally Posted by circuitking View Post
    does connecting a net to NoConn also mean leaving it as open?
    "basic/NoConn" is a symbol, that's all.

    Compare Views of "analogLib/res" and "basic/NoConn".
    And confirm cell property "nlAction" of "basic/NoConn".

    Then consider and learn by yourself.



--[[ ]]--