RE: HV instrumentation for coil work, other

["Tesla list" <tesla-at-pupman-dot-com>]

Original poster: "Mccauley, Daniel H" <daniel.h.mccauley-at-lmco-dot-com> 

Greg,

It all depends on what you consider high voltage.
If its less than 60kV, you can easily get away with a nice high
bandwidth voltage divider.
Commercially speaking, they are relatively inexpensive and work very
well.

I've have also built a high bandwidth voltage / capacitive divider which
is good for up to 60kV and has a flat bandwidth of about 50MHz.

http://www.easternvoltageresearch-dot-com/hv_divider.htm

This is great if you are looking for transients directly on the high
voltage.  However, if you are looking for very small transients within
your circuitry thats floating on the high voltage (which i think you
mean), then the HV divider may not be the best option.

As a high voltage engineer (transmitters / radar) we frequency have a
need to perform measurements like you describe.  Commercially, there
really isn't much available (other than big HV dividers stuff which may
not be able to discern small transients floating on high voltage)  The
best thing to do is to float the entire measurement device.  You can
float an old battery operated scope (don't try with a nice new one), or
the big oscope companies also sell adapter kits to use with their scopes
for just this purpose.  We have one from tektronix that interfaces with
our battery operated tek scopes.  Its basically an insulated box, and
there are long insulated extenders etc... which allow the user to
operate the scope from a long distance.

Dan




 >Hi All,
 >
 >I seem to remember that Terry [and perhaps others] had developed
 >isolated
 >test equipment that could perform measurements on coils while floating
at
 >HV and perhaps relay the data to ground via fiber-optics or some such.
 >
 >We have a need at work to measure transient waveforms in circuits that
 >float at high potentials.  I've been trying to find a commercial vendor

 >that offers a *compact*, bare-bones transient digitizer module that can
be
 >easily clipped into equipment under test, and downloads data to ground
 >through FO cables.  (A Fluke scopemeter is far too large; hoping more
for
 >a single high-density SMT PC board, measuring about 1" x 2".)
 >
 >Anyone know of a suitable product or homebrew design?  It's important
 >for
 >the floating digitizer portion to be as small as possible, and we'd
like
 >to get at least 50MHz [20nS] resolution.
 >
 >
 >Thanks,
 >Greg Leyh
 >