[Coco] NitrOS-9 newbie needs some help

[jdaggett at gate.net]

Mike

You are correct in that the sharpest waveform for the clocks and 
signal lines are not the best. Essentially the runners between ICs 
are transmission lines. One of the assignments I had in my EMI 
class was to model a clock with a given output impedance and vary 
the load impedance and plot the reflections at three different clock 
speeds. 

I shudder to see what a poorly laid out PC mother board signals 
look like with a FSB of 400Mhz. Runners on the PCB have to 
consider RF transmission line theory and almost require an RF 
engineer to overlook the layout for RFI and other nasties that occur 
with clock speeds that high. 

Even at 20 MHz you start to get into uglies on severely mismatched 
runners. 

james


On 10 Aug 2004 at 11:51, KnudsenMJ at aol.com wrote:

From:           	KnudsenMJ at aol.com
Date sent:      	Tue, 10 Aug 2004 11:51:41 EDT
Subject:        	Re: [Coco] NitrOS-9 newbie needs some help
To:             	coco at maltedmedia.com
Send reply to:  	CoCoList for Color Computer Enthusiasts 
<coco at maltedmedia.com>
	<mailto:coco-
request at maltedmedia.com?subject=unsubscribe>
	<mailto:coco-
request at maltedmedia.com?subject=subscribe>

> In a message dated 8/9/04 11:34:23 PM Eastern Daylight Time, 
> dgacke at ektarion.com writes:
> 
> > If this were true you ought to be able to see a deformed clock pulse
> > 
> >  using an oscilloscope. Have you or anyone else examined the clock
> >  lines 
> >    under the conditions described above?
> 
> ISTR putting my O-scope on the SCS lead out the cartridge port, before
> and after removing those "government emission control" series R and
> shunt C.  With the R/C removed, I saw a nice sharp rectangular
> negative pulse.  With those parts still in, it looked like a sagging
> bridge that barely touched bottom in the middle.
> 
> I don't recall ever scoping the RAM control leads or E and Q clocks,
> but the same probably applies.
> 
> ISTR that the proper, professional fix, that cuts the RFI emissions
> while not degrading the signal as you add more loads, is to replace
> the series R with a ferrite bead (inductor), and lose the shunt C. 
> Just thread the little bead ring on a short piece of wire that
> replaces the resistor.
> 
> BTW, RFI emissions aside (like trashing radio and TV all thru the
> house), having too sharp edges on the signals can cause ringing and
> reflections, which probably won't cause much trouble at our clock
> speeds, but just thought I'd mention that the swiftest, sharpest
> waveform doesn't always win. --Mike K.
> 
> -- 
> Coco mailing list
> Coco at maltedmedia.com
> http://five.pairlist.net/mailman/listinfo/coco