[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Differential voltage probes 1 -> 3



Original poster: "Jim Lux by way of Terry Fritz <teslalist-at-qwest-dot-net>" <jimlux-at-earthlink-dot-net>

The way this is dealt with in real life is to have the diff amp have such a
high impedance that the difference in source impedance is a tiny fraction of
the input impedance.  Unfortunately, this is tricky, since the stray
impedances (particularly reactive components, like capacitance) tend to be
large (a few pF, for instance).

The other thing to watch out for is the C in the probe cable.  Real probes
use very low C cable, and, in the most sophisticated designs, actually drive
the shield in a bootstrap configuration, so as to keep the voltage across
the dielectric as close to zero as possible (without oscillating!)  These
sorts of designs also use double shielded cable, so the guard (inner shield)
is itself shielded from the outside world.

----- Original Message -----
From: "Tesla list" <tesla-at-pupman-dot-com>
To: <tesla-at-pupman-dot-com>
Sent: Tuesday, July 01, 2003 5:23 AM
Subject: RE: Differential voltage probes 1 -> 3


 > Original poster: "Steve Conner by way of Terry Fritz
<teslalist-at-qwest-dot-net>" <steve.conner-at-optosci-dot-com>
 >
 > I'm not talking about the match of the resistors inside the HV probe, I
know
 > that can be good. I'm talking about the match of the source impedances
 > you're probing. For instance if your 'cold' probe is on a circuit ground
 > rail, that's coupled to AC ground, and your 'hot' probe is on a
 > high-impedance signal, like the cap of a 555 timer or something. The
source
 > impedance of the signal adds to the resistance of the hot side probe
divider
 > chain and screws up the CMRR.
 >
 > I only know about this because it's a well known problem with balanced
line
 > I/O in audio engineering. Anyway I just bought a Tek 7603 scope and I
found
 > the 25MHz diff amp/comparator plugin on EBay for $25 :) I'm going to try
 > that out with a pair of matched-ish 100x probes, which will give a system
 > similar to what you proposed, and will report back to the list.
 >
 > Steve C.
 >
 >
 >
 > Original poster: "Terry Fritz" <teslalist-at-qwest-dot-net>
 >
 > Hi,
 >
 >  > > Anyway, I'm going off this whole diff amp thing, since I realized
that
 > the
 >  > > CMRR depends on the matching of the two impedances (referred to the
diff
 >  >amp
 >  > > box ground) you're probing, and they're unlikely to ever be matched
in
 >  >real
 >  > > life. I'm going to investigate the optocoupler route instead.
 >  > >
 >  > > Steve C.
 >
 > Matching the resistors may not be hard at all since other values could be
 > paralleled with the 1K's to "tweak" the resistance.  I'll have to test and
 > see, but I think it would be pretty easy to get a 0.1%+ match this way.
 >
 >