Scribd
Upload
51 views3 pages

90° Phase Difference Network Design

This document discusses the calculation of poles for 90-degree phase difference networks to achieve a desired phase error over a given bandwidth. It outlines Weaver's method for calculating the poles in 5 steps: 1) select the lower and upper frequency band edges, 2) calculate a coefficient, 3) determine the minimum number of poles from a graph based on allowed phase error, 4) distribute the poles between two networks, 5) compute the pole angles and positions for each network. It also discusses considerations for determining required accuracy of multipliers and 90-degree phase difference networks based on carrier rejection and unwanted sideband rejection.

Uploaded by

Spiros Loutridis
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
51 views3 pages

90° Phase Difference Network Design

This document discusses the calculation of poles for 90-degree phase difference networks to achieve a desired phase error over a given bandwidth. It outlines Weaver's method for calculating the poles in 5 steps: 1) select the lower and upper frequency band edges, 2) calculate a coefficient, 3) determine the minimum number of poles from a graph based on allowed phase error, 4) distribute the poles between two networks, 5) compute the pole angles and positions for each network. It also discusses considerations for determining required accuracy of multipliers and 90-degree phase difference networks based on carrier rejection and unwanted sideband rejection.

Uploaded by

Spiros Loutridis
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

CALCULATION OF THE POLES OF 90 PHASE DIFFERENCE NETWORKS [after Heaver ]

1.

Select F

2.

Calculate:

and F , the lower and upper ends of the bandwidth.

k =

[1 -(

Q' = L + 2L5

+ 15L9

. . .

ira/loge(Q')

3.

Select a phase error that can be allowed, and consult the graph below to
determine the minimum corresponding number of poles "n".

4.

Choose two networks, A and B. If n is even, there will be n/2 poles in each
network.
If the number of poles is odd, put an extra one in network A.

5.

Compute the angles for A and the poles p :


''
<J> (r) = (45/n)(4r - 3)
for r = 1, 2, . . . (n/2) or [(n+l)/2]
$'(r) = ARCTAN((Q2-Q6) Sin 4<j> (r) / [1 + (Q2+Q6) Cos 4* (r) ])

pa(r) = [l/AF^/F^] TAN[<|>a(r) - ^(r)]

and the angles for B with its poles p, :


*b(r) = (45/n)(4r - 1)

for r = 1, 2, ...

(n/2) or [(n-l)/2]

^(r) = ARCTAN[(Q2-Q6) Sin 4^(r) / [1 + (Q2+Q6) Cos 4<j>b(r)])


Pb(r) = [l//(F/Fh)] TAN[((.b(r) - *b(r)]

6.

The above poles are normalized in terms of F . To get the actual poles
for the networks, multiply by F .

6a

(14)

Phase
Error &
.110 i

Bandwidth
10

(f./fJ
"'
100

1000

0.1

0.01

11

80

Design Chart for Number of Poles (n) for a Given Normalized Bandwidth
and Allowable Phase Error. The Unwanted Sideband Rejection is also
Plotted along with the Phase Error. [After Bedrosian]
DETERMINING THE REQUIRED ACCURACY:
When it comes to determining the required accuracy of multipliers and 90PDN's,
the two major considerations are "carrier rejection" and "unwanted sideband
rejection." Carrier rejection is largly a matter of the properties of the multipliers.
This comes in when you apply two signals f and f to a multiplier and
find that in addition to f-L+f2 and ' f .^f2 you also get a little f

and a little f,.

Generally, the higher of the two frequencies will be the one that is noticed, and
this is the reason for the term carrier rejection.
The point is that if this gets
through the multiplier, it will find its way to the output of the frequency shifter.
For example, if it is the program signal that is the highest, this will come through
the multiplier as a Sin(u t + 4.) in one case and a Cos(u t +<(>) in the other case
where if is the phase shift across the multiplier which is the same for either of
the multipliers and for either of the quadrature signals since they are the same
frequency.
There is nothing that the summers do that will get rid of the program
signal once it is through, since they are in quadrature, and no combination of
addition or subtraction will cause them to cancel.
Thus, carrier rejection is
important in multiplier selection for frequency Shifters.
6a

(15)

-^

Normalized Pole Positions for a bandwidth f./f. = 1000.


Note the symmetry
that shows up in this plot of Log(Pole positions) vs. Linear n.
The poles
were calculated according to Weaver's method.
6a

(16)

You might also like