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

[Manfred Mornhinweg]

Bill,

> There is a better way if you have a dual-channel scope. Use both
> channels in ADD mode and set one of them to INVERT and DO NOT connect
> the probe ground to anything. That way both probes are "floating" and
> either one can be used for hot, ground or anything else.

Yes, that's a valid technique, but it has important limits. One is the 
voltage rating of the probes, another is that of the scope inputs, and a 
very important one is the headroom of the vertical amplifiers. If you 
will be using two 100:1 probes rated for 1kV, to measure the drain 
waveform of a line-connected amplifier, that should be fine. But if you 
want to measure the gate waveform, which is only a few volts, these few 
differential volts are mounted atop several hundred volts of common mode 
voltage into each scope channel. If you set the scope to a gain that 
allows you to see the few RF volts with a reasonable height on the 
display, most likely the vertical amplifier stages that are before the 
summing point will totally saturate from the high common mode voltage, 
and you will see nothing more than a blurry mess, if anything at all.

So, it's a technique one should be aware of, but it's not usable in all 
cases. You would still need some of the other techniques, to measure the 
smaller signals in such an amplifier.

A good technique is to use a small, but well insulated RF transformer 
between the scope probe and the circuit. It can be wound on a small 
ferrite toroid. It needs a series coupling capacitor, and for higher 
voltages it also needs a voltage divider in front of it. It all can be 
built into a special, homemade, galvanically insulated, magnetically 
coupled RF scope probe.
This won't let you see DC, of course, but it shows you the RF signal 
nicely. The DC component can be measured separately with a common plain 
multimeter. Almost all present-day multimeters are rated for direct use 
at line voltage, at least on home circuits.

Anyway most people I know just float the scope and connect it directly 
to the circuit under test, trying to remember to unplug the device under 
test before touching the scope! We even did that at the job. If somebody 
forgets to unplug, the GFI will come down as soon as the scope is 
touched, forcing a trip to the electrical panel to switch it back on. 
GFIs are fast enough that the shock is barely noticeable. I know from 
experience, because I am somewhat forgetful, and on my past job I often 
had to service high power motor controllers that were directly line 
connected. The funny thing is that the panel was down the corridor, past 
another 6 office doors that usually were open, and my colleagues knew 
why I was trotting down to the panel making a sheepish face!

Manfred

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