Lpc2000

Thanks, Tom!

--- In lpc2000@yahoogroups.com, Tom Walsh <tom@...> wrote:
>
> Guillermo Prandi wrote:
> 
> >--- In lpc2000@yahoogroups.com, "Guillermo Prandi" 
> ><yahoo.messenger@> wrote:
> >  
> >
> >>Hi. Sorry for the offtopic, guys. If anybody can point me out 
some 
> >>    
> >>
> >good 
> >  
> >
> >>forum/group to post this question, I'll be glad.
> >>
> >>I need to calculate trace witdh and cleareance to achive an 
impedance 
> >>of nearly 50 ohm for an RF trace surrounded by a GND polygon. My 
> >>    
> >>
> >board 
> >  
> >
> >>is SINGLE layered, so I have no power or GND planes. There's 
plenty 
> >>    
> >>
> Since RF is RF and not digital or analog, you may want to ask 
around 
> some of the lists which deal with RF.  For example, Ham Radio 
(Amateur 
> Radio).  I have an old copy of the: G.R. Jessop,G6JP "VHF UHF 
Manual", 
> ISBN #0-900612-92-4
> 
> Jessop goes into a lot of detail about microwave and one section of 
the 
> book deals with microstripline transmission circuits.  Both using 
double 
> clad PCB as well as transmission line charactoristics of various 
types 
> of materials: slab lines (PCB traces), parallel wires, wire 
parallel to 
> infinite plane, wire in rectangular trough, etc.
> 
> For slablines:
> 
> =============== begin formula ======================
> 
> 
> Where B is the overall length of the two parallel slabs, and, A == 
> inside distance between the slabs:
> 
> Zo ~= 377 * (A / B)
> 
> (IOW Zout is approximately equal to 377 times A / B).
> 
> And, given that A << B then "edge effects are neglected"
> 
> 
> ================== snip =====================
> 
> 
> That example is for two slablines in freespace.  However, depending 
on 
> the cross-sectional area of the slabs, I would suspect that the 
effect 
> of the PCB surface would be negligable.
> 
> If you present the line with a known impedance, design the slab 
line for 
> that impedance and terminate in the same impredance, then you do 
not 
> have to worry about 1/4 wavelength effects.  Radically change any 
of the 
> three impedances and the wavelength of the line becomes a factor.
> 
> My copy is rather dated, 1991, but it is an excellent reference.
> 
> Regards,
> 
> TomW
> 
> -- 
> Tom Walsh - WN3L - Embedded Systems Consultant
> http://openhardware.net, http://cyberiansoftware.com
> "Windows? No thanks, I have work to do..."
> ----------------------------------------------------
>