|
Home > Archive > Electrical Engineering > March 2008 > Trace impedance calculations
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
Trace impedance calculations
|
|
| blarggstar 2008-03-20, 5:25 pm |
| I am attempting to layout a PCB based off an existing "reference
design". I have a "Bill of Materials", physical copy of the board,
and a few design documents etc..
I'm stumpped on one section:
I have an 50 ohm "single-ended RF port" coming off a Balun going to an
SMA connector with antenna attached.
The SMA connector and antenna are both "50 ohm", so I am assuming that
the trace needs have 50 ohm impedance.
All the trace impedance calculators I am using are basically telling
me to use 100 mil width! That is huge!
The reference PCB I am copying has roughly an 8 mil wide trace and is
nowhere near 100mils wide!
What am I missing here?
I am actively reading a few different trasmission-line / PCB design
books and I only have a very basic understanding of these concepts.
Am I just crazy to even attempt this? Yes, of course, but that's not
stopping me.
I'm basically going send my PCB off and try it anyway...but if anybody
can point me in the right direction, I'd really appreciate it.
| |
| blarggstar 2008-03-20, 8:26 pm |
| At certain microstrip lengths and signal rise times, a microstrip
would NOT need to be considered a transmission line right?
Possibly the reference board trace is short enough?
(I'll admit that I'm pretty embarrassed for even asking these
questions. This is not my field, so I must look like a bumbling
idiot! )
| |
| J. B. Wood 2008-03-25, 9:25 am |
| In article
<977b3531-da2f-48d3-98dd-30cfdfcf593f@e6g2000prf.googlegroups.com>,
blarggstar <blargg27@gmail.com> wrote:
> At certain microstrip lengths and signal rise times, a microstrip
> would NOT need to be considered a transmission line right?
>
> Possibly the reference board trace is short enough?
>
> (I'll admit that I'm pretty embarrassed for even asking these
> questions. This is not my field, so I must look like a bumbling
> idiot! )
Hello, and as a rule of thumb you don't have to consider EM wave
propagation for conductors whose length is no greater than about 1/10 of a
wavelength. An EM modeling program like the Numerical Electromagnetics
Code (NEC) considers RF currents through no-loss conductors of length 1/20
wavelength to be constant in amplitude and phase (no time delay).
Sincerely,
John Wood (Code 5550) e-mail: wood@itd.nrl.navy.mil
Naval Research Laboratory
4555 Overlook Avenue, SW
Washington, DC 20375-5337
|
|
|
|
|