Re: Is there an electronics design newsgroup?

From: Mike Monett (no_at_spam.com)
Date: 09/22/04 

Date: Wed, 22 Sep 2004 18:21:52 -0400

John Larkin wrote:

[...]

> Why not start with a passive divider, thin-film or something?

The capacitance of widely-separated electrodes is asymptotic to the
free-space value. This means the capacitance may be very much larger than
calculated from standard formulas. Here's an analysis of the field lines
on widely-separated circular plates. See the section at the end, titled
"Plates relatively far apart":

    "As the distance between the plates increases even further, the
    ovoids become increasingly symmetrical, until they approach
    ellipsoids. In the limit, the plates cease to interact at all, and
    each plate holds a charge based only on its potential with respect
    to the surrounding space. Thus, as the equipotentials approach
    ellipsoids, the capacitance approaches a constant value that is
    independent of further increases in plate distance."

    http://www.ttc-cmc.net/~fme/captance.html

So a thin film resistor may or may not reach the ~0.05 pf stray
capacitance needed. I tried using a 2D field solver, but ran into the
problem of accounting for the fringing fields off the sides of the
contacts. Obviously the problem is solvable - both companies did it. But
it's not clear how.
 
> A small PHEMT might have Cgs of 0.2 pF (naked chip) and Cgd even less,
> so it doesn't sound exotic... just a lot of hard work to keep the
> response clean.

Agree - solvable, with a lot of effort. Of course, you need the toys to
play with this stuff:)
 
> I still think a dual-gate mesfet follower could be a neat, cheap
> active probe. Bootstrap G2 and the input resistor, and servo it to
> Idss to kill the offset and LF dispersion. Some of the available IC
> buffers/followers would make a pretty nice probe, too.

2 to 6GHz is easy. We need a minimum of 20 to 30GHz bandwidth to handle
Gigacom, plus any exotic stuff you might come up with:)
 
> John

Regards,

Mike Monett

Relevant Pages

  • Re: Photodiode TIA
    ... John Larkin wrote: ... The stray capacitance across a 600K resistor ... so you'd need later gain stages to get up to the volts ... One of the Sirenza SiGe mmics would be even better. ...
    (sci.electronics.design)
  • Re: Photodiode TIA
    ... John Larkin wrote: ... The stray capacitance across a 600K resistor ... so you'd need later gain stages to get up to the volts ... What's your photodiode capacitance? ...
    (sci.electronics.design)
  • Re: Free Electricity !
    ... essential components - a material with a high capacitance which stores ... the charge, and a material with a very high resistence that keeps the ... The way I had thought of it is that the charged plates exert ...
    (uk.legal)
  • Re: eer
    ... >>I've never seen you explain HOW to make the plates smaller. ... group why this solution will not provide the energy density you hope ... The science you're using (more surface area = more ... capacitance) breaks down at the scales you are proposing. ...
    (sci.electronics.misc)
  • Re: CDROM metal stripper
    ... The plates have an ... plates and some reasonable estimation of the air gap, ... Polycarbonate has an er of 3, and the CDROM plates are about 50 mils thick. ... The capacitance of the brass plate to the aluminum surface is something on ...
    (sci.electronics.design)