|
 |
Some distributed systems questions ... - RF Cafe Forums
|
taz
Post subject: Some distributed systems questions ...
Unread
postPosted: Thu Oct 07, 2004 12:12 pm
Offline
Lieutenant
Joined: Thu Oct 07, 2004 11:52 am
Posts: 2
hi,
I have
two basic but important questions, I hope you can help me:
1.
Why do we use distributed model (and not lumped) in frequency applications\problems
??? what are the advantages\disadvantages ?
2. Why there are incident
waves and reflected waves in a transition line ? this happen for both
sinusoidal and step function….why ???
Thanks,
taz
Top
Profile
Itay
Post subject:
Unread postPosted:
Thu Oct 07, 2004 12:36 pm
Greetings,
No clue for your
first question. If you will be more specific I might give you an answer.
Basically, distributed elements can be modeled as lumped elements for
the the purpose of analysis; for example, this is the way to derive
the telegraphic equations of a transmission line.
For your second
question: Incident waves are voltage waves that flows from a source
eg: Generator to a load. In an ideal circuit all the energy of these
waves is being absorbed in the load. In the practical world, a portion
of this energy is reflected back due to impedance mismatch. The impedance
of the load should be equal to the characteristic impedance of the system
eg Zo, which is usually 50 ohm in RF systems, 75 ohm in Video applications
and 600 ohm in Audio applications. The amout of the mismatch determines
the size of the reflected waves. The reflection coefficient determines
the size of the reflections. It is defined as: rho=(ZL-Z0)/(ZL+Z0).
From here you can also derive the return loss, which is the amount of
the reflected wave in relation to the incident wave: RL=-20 log |rho|.
That means how lower is the reflected wave comparing to the incident
one (It should be as low as possible of course)
Hope this helps,
Itay
Top
taz
Post subject:
Unread postPosted:
Thu Oct 07, 2004 1:02 pm
Offline
Lieutenant
Joined: Thu
Oct 07, 2004 11:52 am
Posts: 2
Hi Itay, thanks for your answer
!
I will start with your second answer, I agree with what you
said 100%, and for the present time I accept it as an axiom. This is
why I'm asking these questions …why mismatch is causing reflected waves
? why In the practical world, a portion of this energy is reflected
back ??? why match circuits don’t have reflected waves ???
I understand
the math, but I don't understand the physics behind it ….
And
for my first question, for my best understanding …these are two different
theories , but according to what you said – they are derived from another
. is that a correct interpretation ???
by the way, why in speakers
wiring there's no indication for impedance ??
again…thanks,
taz
Top
Profile
Itay
Post subject:
Unread
postPosted: Thu Oct 07, 2004 1:42 pm
Greetings Taz,
You
are welcome.
Yes the theory of transmission lines is derived
from an analysis of lumped model which defines the transmission line
as LPF circuit that includes L,G,R,C elements and also their values
per unit length. From this theory you also derive the Zo and Gamma coeffiencient:
Zo= sqrt (L/C) - In loseless tranmission lines and Zo= sqrt ((R+jwL)/(G+JwC))
in lossy transmission lines and many more definitions...
I advise
you to read more in text books regarding your questions and find better
answers. Posting questions like these in a forum won't bring the answers
you would expect, as there are many lines to write about it and people
aren't that patient to write them...
Regarding your question
about speakers, please read the following:
http://www.duncanamps.com/technical/dummyload.html
Speakers have a varying load impedance and there is no meaning for
the impedance of the wiring.
I hope this helps,
Itay
Posted 11/12/2012
|
|
|
|
|
