On 02:46 PM 20/02/2002 -0800, JaMi Smith said:
>Well ~
>You're almost right . . .

Rant On

What the hell is your point - are you deliberately being  prickly?  My turn 
to be pedantic and pick up on your misconceptions. Am I peeved - yes.  JaMi 
(I persist in this weird spelling since you do but I must admit that I 
thought it was a typo when I first saw it)

>A typical RF Choke on a PCB is a certain length conductor (typically 1/4
>wavelength), which is acting as an inductor, from one active part of an
>RF circuit (typically the Collector of an RF Transistor) to another part
>of the circuit, typically power (or sometimes ground), such that it is a
>"short" at DC, and isolated (or open) at a given frequency.

A typical RF CHOKE is lossy (deliberately) and hence one would not normally 
use a relatively high-Q PCB track to implement a choke.  Inductor yes, 
choke no.  At least not below a few GHz where line edge accuracy, surface 
finish of the copper, losses in the dielectric etc etc start to lower Q 
into the choke realm.

Now I use chokes to reduce the chance of oscillations in amps etc often in 
the base cct, usually in the collector feed (before the collector 
inductor).  Since I have not designed many truely broadband amps in the 
recent years I have not placed a choke in a collector cct for many 
years.  Yes, I use them for supply filtering before I get to the hot 
collector cct.

(To elaborate, many old HF and VHF amplifiers used chokes in the collector 
cct to provide bias and then use high-Q input and output matching ccts to 
give the freq response desired.  Moderm design for many RF engineers 
consists of narrow fractional bandwidths, low power operation and miniature 
design.  The use of RF collector chokes is not really an option for many of 
us.  Hence my beef with the word "Typically" in JaMi's rant.)

1/4 wavelength line is not what I would call a choke.  Choke to me is a 
soggy inductor.  A 1/4 wavelength line is more likely to be a moderate to 
High-Q resonator and has a different function to the old-stye broadband, 
flat response chokes.

>Thus one can
>pass power thru the Choke to the Transistor, but it does not appear to
>be a short at RF frequencies. There are other applications, but this is
>probably the most common one you will find on a PCB, and is actually
>quite common today.
>Now - Go ask Protel how to make it . . .

Use a track.  Protel supports tracks, you know.  Used a curved 
meander.  Protel supports curved meanders (autorouter ripping them up not 
withstanding).  Been there done that as have many other on this list.  You 
are by no means the only one with significant RF experience.  I know I am 
counting 15 years and others here have more than that.

Brian G.  and others wrote a spiral track generator for Protel.  Use 
that.  Up to the knowledgeable designer to figure out the details of the 
spiral though, not that PCB CAD package.  Now, some of the other CAE 
packages will help in designing RF ccts with copper elements.  Agilent, 
Eagle and others.  But Protel is not an RF CAE package, still it certainly 
supports the track-based inductors that I need; I design them, I place 
them.  Bingo!

If you want a EM simulator then Protel is not it.  You can use Eagle etc to 
design the layout from a RF POV.

>No, don't give us your version (which we have already seen here in the
>list), but go ask Protel how to make it.

What exactly is your point?  Protel knows very well how to put copper onto 
a PCB.  It is the designers job to do the design, you know.

JaMi, you can take the above in two ways.  1) You think I am an bastard or 
2) a warning that this is usually a very polite place, welcoming people 
from everywhere, and you have simply niggled enough people to piss me off 
sufficiently to write the above. Pull your head in or go away.

I welcome your reasonable, friendly input in the future - but I am close to 
adding a JaMi filter to my email...

Rant Off

Ian Wilson
Considered Solutions Pty Ltd     mailto:[EMAIL PROTECTED]
ABN: 96 088 410 002
5 The Crescent
Ph: +61 2 9411 4248   Fax: +61 2 9411 4249

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
* Contact the list manager:
* Forum Guidelines Rules:
* http://www.techservinc.com/protelusers/forumrules.html
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *

Reply via email to