[Amps] Direct rectification of AC mains to drive the amp,

[Manfred Mornhinweg]

Bill,

I can see us two 30 years from now, VERY white haired, still discussing 
the same topic! ;-)

> Think about it:  With a battery powered scope and using the dual probe ADD +
> INVERT mode, 

Oh well, you were thinking about BOTH battery-powering the scope AND use 
the channel difference mode. I was thinking about either one OR the 
other method.

 > the only connection between the HV and the scope chassis is
> through the two high impedance probes. Remember, the scope chassis isn't
> connected to anything. 
> 
> I just went out to the shack and measured my Tek 10x probes. They are 9 megs
> from tip to center conductor of the BNC. As I recall the resistor is
> paralleled by a few pF. Wouldn't hurt a fly at 60 Hz. 

Yes. I agree that that would be safe in practice, even if it won't 
satisfy safety regulations.

The problem is that it might still not work well, due to saturation of 
the channels! When you measure around in an amp that is tied to the AC 
line, with the two-channel method, and the scope's body completely 
floated except for those "few pF" of stray capacitance, the body of the 
scope will float more toward the amp's potential or more toward ground. 
depending on the impedance ratio of the probes and the stray 
capacitance. As long as the stray capacitance of the whole 
battery/inverter/scope setup is really low, the scope body should be 
pulled reasonably close to the potential of the amp, and the scope 
should work fine. But as soon as you touch a scope knob, you are adding 
your body's stray capacitance, pulling the scope body closer to ground, 
farther from the amp under test, possibly enough to cause saturation of 
the channels.

It might work, but it's not a sure bet. It depends on the stray 
capacitances.

> Or you could get a 40 MHz Fluke ScopeMeter for several thousand dollars.
> That works too.  :-)

Yes, and I love that solution - as long as someone else pays for it! One 
of the good sides of that solution is that the body of that scope is 
fully insulated, so that you don't need to use the two-probe method. You 
can go ahead, clip the ground lead of one probe to the amp's line-tied 
"ground", and happily press all the buttons you want on the scopemeter, 
safely and with good measurement performance - as long as you properly 
set up the scopemeter. Like all sampling scopes, it has a tendency to 
hide important features of the signal, unless you know what you are 
looking for, and set it up for that.

> I grant you the inverter approach is still dangerous, but not to an
> experienced tech. The problem with all the above approaches is there is
> still significant capacitance between windings of the isolation transformer
> or the scope's power transformer and you are still lightly coupled to the AC
> line.

That's true.

> With the inverter approach, you have a truly "floating" scope. A few
> stray pF maybe, but that's about it. 

Yes, could even be 100pF, but that's still lower than the capacitance in 
a transformer.

But even if we use nothing else than the internal transformer of the 
scope for insulation, and cut its ground line, thus ending up with maybe 
0.01uF from the scope to ground, this is NO PROBLEM for the measurement, 
if we use the single probe method! The 0.01uF of the transformer plus 
stray plus RF filters will appear in parallel to one of the rectifier 
diodes in the line-connected amp, and at 50Hz will do essentially nothing.

That's why the simplistic method of opening the scope grounding lead, 
and then measuring away happily in a libe circuit, works well, as long 
as we remember to either unplug the circuit being tested BEFORE touching 
the scope, or else if we use well insulated electrician's rubber gloves! 
I have done that too, and it works well, it doesn't even tickle, but 
this is NOT a recommendation to do it, specially for people who don't 
know the fine points about voluntary yet accidental electrocution.

> Essentially a larger version of the Fluke ScopeMeter.  :-)

Take a normal scope, an inverter, a battery, thickly wrap it all in 
insulating tape, completely covering everything, including all the knobs 
of the scope - there you have your scopemeter equivalent! Or rather, 
ALMOST. Because with your common scope the grounds of all test probes 
are shorted together. With the real scopemeter instead they are isolated 
form each other! That adds a lot of flexibility to the scopemeter! I 
love it - but not its price!

Manfred

========================
Visit my hobby homepage!
http://ludens.cl
========================