Lpc2000

Hi Tim and Kris,

  Thanks for the info. Already performed all those things. This
particular board layout can have from 2 to 3 processors on it
(LPC2138) and we have several experts in EMI in Australia looking the
board over as a second opinion. We learned from other designs using
the PXAG49 that separate crystals with no vias is a must. We get
fundamental frequencies usually in the 110 to 150MHz range with a
29.4xxMHz clock when we had just 1 crystal line feeding the
processors. We will find out soon enough when one of the designs with
the new processors hits the testing range in the second week of
February. In terms of equipment, no. So its testing at a lab. and us
adding and subtracting parts.

regards,
Charles

--- In lpc2000@yahoogroups.com, "microbit" <microbit@c...> wrote:
> Hi Tim,
> 
> > knowing more about the board (and other components) it would be very 
> > hard to say what will turn out to be the source of the interference. 
> > Throwing ferrites and cap's on the board can be counter productive 
> > without knowing the frequencies that are causing 
> > the problem.
> 
> Exactly, and I've seen this in practice - engineers going "the safe
side" with
> 4 layers, and creating more bad than good.
> I'm not opposed to 4-layer per se, but for mixed a really careful
layout with
> 2 layers often suffices if you know the quircks a bit.
> 
> > Do you have access to a spectrum analyzer? This is 
> > probably the most critical tool needed to diagnosis the problem.
> 
> Indeed, with near field probe on speccie, you can tell a lot.
> 
> > Also be aware, that even though your crystal is a lower frequency, 
> > edges internal to the part(s)will occur much faster, as well as 
> > harmonics of the fundamental frequency.  One you have the 
> > frequencies and magnitude involved, then it's possible to formulate 
> > a strategy.
> 
> Half indifferent, you'd have to drive the Xtal pretty hard to cause
this,
> and premature aging/drifting/life time will be badly compromised.
> But nothing a good old series resistor won't fix, drive-wise.
> 
> > I believe most all the obvious fixes have been discussed already, 
> > but I did not see anyone ask about your trace lengths, width and 
> > spacing, as well as are there traces that run along the edge of the 
> > board.  
> 
> I rasied this issue a while ago on MSP430 group, and a heated discussion
> "fobbed" this all off, dismissing as "unnecessary" for digital < 30 MHz.
> Well you could have 4-8 MHz clock high slew rate bus signals on "long"
> tracks, that are completely the wrong impedance, terminate really
poorly,
> and cause heaps of EMI.
> If this is mastered, it's almost a challenge/game to do something on
2 layers
> instead of 4 :-)
> 
> 
> > Do you have any traces that change direction by more than 45 
> > degrees?  
> 
> Another important one, along with NOT using vias on clock/crystal
lines !
> 
> 
> B rgds
> Kris
> 
> 
> 
> [Non-text portions of this message have been removed]