Re: Differential voltage probes Safety Issues

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

Original poster: "Terry Fritz" <teslalist-at-qwest-dot-net>

Hi Richard,

At 01:57 PM 7/19/2003 -0400, you wrote:
>
>Terry,
>
>I've given some thought lately to safety issues regarding overvoltage of
>the front end of the differential voltage probe.  I disagree that a "fuse"
>is going to do much to protect the differential probe and the scope (and
>operator) attached to it.  Once an arc occurs and plasma forms there's no
>way to control the destruction of probe and scope.  I believe a new design
>is in order.

The probe will hopefully be rated for +-1500Vpeak or 1000VACrms.  That 
would be well in the range of solid state coil work.  Hopefully the CMMR 
and all will be good so we can pick out fine signals riding on these high 
voltage common mode signals which is what the probe is designed to do.  At 
~1800 volts, the MOVs are going to clamp the voltage and begin stressing 
the fuses.  Assuming solid state coil's power supplies are in the <5000 
volt range.  The forward divider resistors alone will hold the voltage. I 
will test things out to be sure there is a wide safety margin.  The real 
threat here is an unfused MOT at about 2000VAC.  However, the fuses will 
have 2000-1800 volts across both of them which is well within their ability 
to break.

The next threat would be a say 20000V 10J cap discharge.  That is easily 
within the MOV's ability to handle and then clamp down.  A 30mA NST will 
overheat the MOVs eventually, but they will go short.  So really, the probe 
has far more margin than it needs and far more overload protection than 
typical.  Even if it does get through the fuses, MOVs, ICs...  It then 
meets up with 5 V TVS's which can hold off an NST literally forever...

Of course, if one hooks it to a 14400Vac 500 amp power line, it will 
explode and also isolate the scope :o)))


>A fiber optic system should be placed between differential probe and scope.
>Continue to design the front end of the probe as is, but without fuses.
>The differential output of the probe should drive a fast fiber optic op
>amp.  The fiber optic leads are to a second fiber optic receiver op amp 
>and the
>scope.  There are fast fiber optic cables and receiver op amps available.
>An example is the SA 5212 at 150 mhz.

At that point, one might as well go to standard fiber probes in the 
differential mode.  We are talking of an entirely different application now.

What we are trying to do is bring the abilities of the TEK 5205 style 
differential probe to a far more affordable level for "us".  Two P5205's 
cost $2070.  Thus, people here are resorting to floating scopes or blowing 
channels trying to use "ordinary" probes...  We're just trying to make a 
little device that allows SSTC measurements to be made easily and far more 
safely than floating scopes or "trying" to do differential two channel 
measurements with 300V probes...  Tek has a dirty little secret about the 
5205.  If you put too much voltage in them, they break.  They are serviced 
as a full unit for about $650.  In our case, I hope to make them hard to 
break and easy to fix if a mistake does occur.  The big thing is getting 
the cost down to where there is no problem getting on if you need one.  I 
hope the cost will be about $50.

Cheers,

         Terry


>You should recognize the technology.  You pioneered fiber optic Tesla coil
>instruments.  For a few dollars more you can design and build a much safer
>instrument for both scope and users.  Much cheaper in the long run than
>scope destruction or operator injury.
>
>RWW