Difference between revisions of "O-device (Lossy Transmission Line) and T-device (Lossless Transmission Line) modelling issues"

From LTwiki-Wiki for LTspice
(Created page with "The O-device (Lossy Transmission Line) in LTspice does not seem to correctly model dc behavior. This is troubling and could lead to very confusing simulation results in some c...")
 
m (corrected spelling of PSpice)
 
Line 11: Line 11:
 
graphic - there is no dc connection between the input pins and the
 
graphic - there is no dc connection between the input pins and the
 
output pins (this is why both ends must have a path to ground).
 
output pins (this is why both ends must have a path to ground).
An equivalent circuit for this device can be found the Pspice
+
An equivalent circuit for this device can be found the PSpice
 
manual and looks like it would be easy to build in LTspice using
 
manual and looks like it would be easy to build in LTspice using
 
b-sources to provide the controlled sources with delay.
 
b-sources to provide the controlled sources with delay.

Latest revision as of 16:01, 3 November 2019

The O-device (Lossy Transmission Line) in LTspice does not seem to correctly model dc behavior. This is troubling and could lead to very confusing simulation results in some cases. Also the O-device does not allow its loss elements to be frequency dependent so that skin effects (which are very real and sometimes very important) may not be modeled. I think that these limitation are generic to SPICE and are not just specific to LTspice.

It is worth noting that the T-device (Lossless Transmission Line) behaves with connectivity counter to that suggested by its symbol graphic - there is no dc connection between the input pins and the output pins (this is why both ends must have a path to ground). An equivalent circuit for this device can be found the PSpice manual and looks like it would be easy to build in LTspice using b-sources to provide the controlled sources with delay.

A while ago, EE Times online? published a series of articles by Roy McCammon that developed an easy-to-follow model for a frequency dependent lossy transmission line complete based on the so-called telegrapher's method.

http://i.cmpnet.com/rfdesignline/2010/06/C0580Pt1edited.pdf
http://i.cmpnet.com/rfdesignline/2010/06/C0580Pt2edited.pdf
http://i.cmpnet.com/rfdesignline/2010/06/C0580Pt3edited.pdf

His final subcircuit is reproduced schematically in the article and the text version is copied below:

.param Kft=1 ; 1 Kft = 1000 feet
.param Lcon=10n ; convergence inductance
+ C=15.72e-9 ; the value of capacitance at dc
+ Gdc=0.5n ; the value of conductance at dc
+ Rdc=52.50 ; the value of resistance at dc
+ Ldc=0.1868e-3 ; the value of inductance at dc
+ Linf=0.133e-3 ; inductance at infinite frequency
+ Ldel=(Ldc-Linf) ; inductance parameter
+ Zinf=(Linf/C)**0.5 ; characteristic impedance at infinite frequency
+ Yinf=1/Zinf ; characteristic conductance at infinite frequency
+ F2=5e6 ; the highest frequency in Hz
+ W2=6.28318*F2 ; the highest frequency in rad/sec
+ G1=23u ; the value of conductance at F1
+ G2=36u ; the value of conductance at F2
+ Rac=304.62 ; the value of resistance at F2
+ F1=3e6 ; the second highest frequency in Hz
+ A=1.6 ; inductance parameter
+ k=Log(G2/G1)/Log(F2/F1)/2 ; conductance parameter
+ WL=6.28318*161000 ; inductance parameter
+ WR=W2*(Rdc**2)/(((Rac**4)-(Rdc**4))**0.5) ; resistance parameter

.subckt single_mode_xline L1 R1
G1 N1 0 N1 0 Laplace=(((((Gdc+G2*(-(s/w2)^2)^k)+s*C)/
+ ((Rdc*(1-(s/wR)^2)^0.25)+s*(Linf+Ldel/(1+A*((-(s/wL)^2)^0.5)-
+ (s/wL)^2)^0.25)))^0.5))-Yinf
G2 0 N1 N2 0 1
G3 0 L1 N2 0 1
V1 L1 N1 0 Rser=0
H1 N4 0 V1 1
G4 N6 0 N6 0 Laplace=(((((Gdc+G2*(-(s/w2)^2)^k)+s*C)/
+ ((Rdc*(1-(s/wR)^2)^0.25)+s*(Linf+Ldel/(1+A*((-(s/wL)^2)^0.5)-
+ (s/wL)^2)^0.25)))^0.5))-Yinf
G5 0 N6 N5 0 1
G6 0 R1 N5 0 1
V2 R1 N6 0 Rser=0
H2 N3 0 V2 1
R1 N6 0 {Zinf}
R2 N1 0 {Zinf}
G7 0 N2 N3 0 Laplace= Exp(-Kft*( (((Rdc*(1-(s/wR)^2)^.25)+
+ s*(Linf+Ldel/(1+A*((-(s/wL)^2)^.5)-(s/wL)^2)^.25))*
+ (Gdc+G2*(-(s/w2)^2)^k+s*C))^.5))/(s*Lcon+1)
L1 N5 0 {Lcon} Rser=1
G8 0 N5 N4 0 Laplace= Exp(-Kft*( (((Rdc*(1-(s/wR)^2)^.25)+
+ s*(Linf+Ldel/(1+A*((-(s/wL)^2)^.5)-(s/wL)^2)^.25))*
+ (Gdc+G2*(-(s/w2)^2)^k+s*C))^.5))/(s*Lcon+1)
L2 N2 0 {Lcon} Rser=1
.ends single_mode_xline