TSSP: List Archives

From: "Terrell W. Fritz"
Date: Sat, 12 Oct 2002 17:49:58 -0600
Subject: Re: [TSSP] Top Voltage

Hi All,

At 11:27 PM 10/12/2002 +0100, you wrote:
>Bart wrote:
>> I have one of Terry's current sensors, so I can
>> run the base lead through and set the scope to
>> capture first breakout
>
>Got anything to calibrate it with?  At least we don't
>demand too much from the scope, just a peak reading to
>better than 5%.

Wha!  Ya think I sent Bart a piece of junk or something :o))))   Accuracy is
+1% -0%.  He may want to check his "scope" against a battery, meter and all,
but no worries about the Pearson probe ;-))  I did check it before I sent it
too ;-))

>
>> When testing for "break-out" conditions, the gap arc voltage
>> is the adjustment tool. The variacs allow us to ramp up to
>> what ever is required to arc the gap, but the gap arc voltage
>> sets the topvolts breakout potential.
>
>Ok, I see.  Perhaps not so easy then to find the point
>where it is only just beginning to break out.  Hopefully
>the procedure for taking these measurements won't be too
>hard, so that we can collect quite a few. 

It seems this voltage seems to increase with time.  It is like the first
breakout blows off dust or charges the surrounding area requiring more and more
voltage to get breakout.  After awhile, one starts to loose confidence that
this is an "accurate" test.  Things just seem to move on you too much...  I
have all solid state stuff here.  I imagine with a spark gap and all, things
get really squirly...

>
>It would also be very helpful to measure the secondary
>Fres (with primary gap open) and the two mode frequencies
>(with primary gap closed).  Not everyone is going to be able
>to do this, and in those cases there will be some doubt that
>we modeled the tuning correctly.  
>
>Without frequency measurements, there could easily be 
>another 5% error in converting the base current reading to 
>a peak topvolts.  With frequency measurements the extra
>error drops to around 1%, or the accuracy of the frequency
>measurements, whichever is the worst.

With top breakout voltage varying by like 200% maybe it's all relative ;-)))

>
>I'm wondering if we should restrict to coils setup outdoors.
>Perhaps the wrong time of year for most of us!  The thing is,
>any doubt about the E-field around the top of the coil adds
>directly to the error in the resulting breakout voltage.  The
>presence of the ceiling is a big unknown quantity.  Likely    
>not a problem for table top coils, or coils operated in a big
>room.  The trouble is, even if a nearby wall or ceiling doesn't
>add much in the way of extra C, it could nevertheless make
>some redistribution of the topload charge, which is critical
>to determining the location and size of the highest surface
>field.  For example, a nearby wall might shift the charge to
>one side of the topload by 30%, but only make 10% change to
>the total charge, so we would see a 5% error in Fres but 
>have a 30% error in breakout voltage.
>

Probably a good thing to look into!  Would a fan blowing air around change
things too???  I'll try and find a fan tonight.  Seems like we need to look at
breakout statistically as a percent over 100 firings or some messy thing like
that.  But I feel that the "variable" is just not under our control or
understanding at the moment...

>Are we likely to still be able to find several coils with
>this outdoor requirement?  Perhaps we should only trust
>indoor measurements if the coil in question will breakout
>equally readily in any direction from the toroid, indicating
>that at least cylindrical symmetry of the E-field has been
>obtained.

I think I can get outside easily.  Maybe time to order rack mount kits for my
toys :-))  I'll see what I can do here.

>
>Wonder how many candidate systems that leaves within our
>reach?  We want to get accurate figures, but going to far
>will leave too few systems, so we won't get a clear
>picture.  

I think we should try and eliminate variables.  We seem to have some wild
unknown out there...  If we could pin it down, we could work it...  It must be
known.  Maybe Marco can ask around his high voltage lab or play with some
stuff.  Something basic is going on here I don't think we understand.

Cheers,

        Terry

>--
>Paul Nicholson,
>--


Maintainer Paul Nicholson, paul@abelian.demon.co.uk.