Vous êtes sur la page 1sur 7

# 1/26/2005

1/7

## The Reflection Coefficient Transformation

The load at the end of some length of a transmission line (with characteristic impedance Z0 ) can be specified in terms of its impedance ZL or its reflection coefficient L . Note both values are complex, and either one completely specifies the loadif you know one, you know the other!

L =

ZL Z0 ZL + Z0

and

ZL = Z0

1 + L 1 L

Recall that we determined how a length of transmission line transformed the load impedance into an input impedance of a (generally) different value:

Z0 ,

in

Z0 ,

Jim Stiles

Dept. of EECS

1/26/2005

2/7

where:

## Z L cos A + j Z 0 sin A Zin = Z 0 A A + Z cos j Z sin L 0 Z + j Z 0 tan A = Z0 L + A tan Z j Z 0 L

How can we express the input impedance in terms its reflection coefficient (call this in )?

Z0 ,

in = ?

Z0 ,

## A: Well, we could execute these three steps:

1. Convert L to ZL:

ZL = Z0

1 + L 1 L

Zin = Z 0

Jim Stiles

Dept. of EECS

1/26/2005

## The Reflection Coefficient Transformation.doc

3/7

3. Convert Zin to in :
in =

Zin Z 0 Zin + Z 0

easier way?

## A: Actually, there is!

Recall in an earlier handout that the input impedance of a transmission line length A , terminated with a load L , is:
e + j A + L e j A V ( z = A ) Zin = = Z0 + j A I ( z = A ) L e j A e

Note this directly relates L to Zin (steps 1 and 2 combined!). If we directly insert this equation into:
in =

Zin Z 0 Zin + Z 0

Jim Stiles

Dept. of EECS

1/26/2005

## The Reflection Coefficient Transformation.doc

+ j A + L e j A ) (e + j A L e j A ) Z 0 (e in = Z 0 (e + j A + L e j A ) + ( e + j A L e j A )

4/7

2 L e j A = 2e + j A = L e j A e j A = L e j 2 A

## Q: Hey! This result looks familiar. Havent we seen something

A: Absolutely! Recall that we found that the reflection coefficient function ( z ) can be expressed as:
( z ) = 0 e 2 z Now, for a lossless line, we know that = j , so that: (z ) = 0 e j 2 z Evaluating this function at the beginning of the line (i.e., at z = z L A ): (z = z L A ) = 0 e
j 2 (z L A )

## = 0 e j 2 zL e j 2 A But, we recognize that:

0 e j 2 zL = ( z = z L ) = L

Jim Stiles

Dept. of EECS

1/26/2005

## The Reflection Coefficient Transformation.doc

5/7

And so:

( z = z L A ) = 0 e j 2 zL e j 2 A = L e j 2 A

Thus, we find that in is simply the value of function ( z ) evaluated at the line input of z = z L A !

in = ( z = z L A ) = L e j 2 A

Makes sense! After all, the input impedance is likewise simply the line impedance evaluated at the line input of z = z L A :

Zin = Z ( z = z L A )
It is apparent that from the above expression that the reflection coefficient at the input is simply related to L by a

phase shift of 2 A .
In other words, the magnitude of in is the same as the magnitude of L !
in = L e = L If we think about this, it makes perfect sense!
Jim Stiles The Univ. of Kansas Dept. of EECS
j ( 2 A )

= L (1)

1/26/2005

6/7

in Pabs =

V0+

2 Z0

1 in

## while that absorbed by the load L is:

L Pabs =

V0+

2 Z0

(1 )
2

Z0,

in Pa b s

Z0,

L Pa b s

A
Recall, however, that a lossless transmission line can absorb no power! By adding a length of transmission line to load L , we have added only reactance. Therefore, the power absorbed by load in is equal to the power absorbed by L:

in L Pabs = Pabs

V0+

2 Z0

(1 ) = 2 Z (1 )
2 2

V0+

in

1 in

= 1 L

Jim Stiles

Dept. of EECS

1/26/2005

## The Reflection Coefficient Transformation.doc

7/7

Thus, we can conclude from conservation of energy that: in = L Which of course is exactly the result we just found! Finally, the phase shift associated with transforming the load L down a transmission line can be attributed to the phase shift associated with the wave propagating a length A down the line, reflecting from load L , and then propagating a length A back up the line:

Z0,

in = e

j A

Le

j A

= A
To emphasize this wave interpretation, we recall that by definition, we can write in as:

V (z = z L A ) in = ( z = z L A ) = + V (z = z L A )
Therefore:

V ( z = z L A ) = in V + ( z = z L A )
= e j A L e j A V
+

(z

= zL A )

Jim Stiles

Dept. of EECS