The Cyber-Spy.Com Usenet Archive Feeds Directly
From The Open And Publicly Available Newsgroup
This Group And Thousands Of Others Are Available
On Most IS NNTP News Servers On Port 119.
Cyber-Spy.Com Is NOT Responsible For Any Topic,
Opinions Or Content Posted To This Or Any Other
Newsgroup. This Web Archive Of The Newsgroup And
Posts Are For Informational Purposes Only.
From: email@example.com (Bill Sloman)
Subject: Re: High-speed components/design
Date: 12 Dec 2002 09:05:10 -0800
NNTP-Posting-Date: 12 Dec 2002 17:05:10 GMT
"Rob Bowmaker" wrote in message news:<6bWJ9.firstname.lastname@example.org>...
> What kind of components do 100 megabit/second ethernet cards use? If they do
> 100 megabits/second, that would mean that a response time of at most 10ns
> would be required -- most components I've seen don't come anywhere near that!?
> I'm especially interested in transistors and/or comparators that can deal
> with such quick responses.
I've no idea what sort of components 100 Mbit/sec ethernet cards use -
they are high volume commodity products so I'd guess it is all buried
in one application specific integrated circuit designed and made for
the job in 100,000 part lots.
For low volume products there are comparators around that are plenty
fast enough for the job
but Maxim also have parts of adequate speed. For the the digital glue
you can always use Motorola ECLinPS parts, available ex-stock in small
quantities from Newark (via Farnell if you live outside the United
States), though fast programmable logic with an LVDS interface option
would be the way to go.
For the transistors look at low power wideband transistors from
Philips or Infineon - Farnell list about half a page with
gain-bandwidth products from 3.5GHZ to 10GHz - about 15 years ago I
was getting sub-nanosecond switching out of the equivalents of the
BFR92 (NPN 5GHz) and the BFQ92 (PNP 5GHz).
Just don't forget to put about 33R of base-stopping resistor right up
against the base, unless you are a lot cleverer at RF design than I am
- with that in place they don't self-oscillate.
Come to think of it, I published a two-transistor circuit using 5GHz
transistor in Journal of Physics E: Scientific Instruments back around
Bill Sloman, Nijmegen
Go Back To The Cyber-Spy.Com
Usenet Web Archive Index Of
The sci.electronics.design Newsgroup