Re: Coil Efficiency (and true wattmeter)

From: 	David E. Sharpe[SMTP:sccr4us-at-erols-dot-com]
Sent: 	Thursday, June 26, 1997 12:45 AM
To: 	Tesla List
Subject: 	Re: Coil Efficiency (and true wattmeter)

Tesla List wrote:
> From:   FutureT-at-aol-dot-com[SMTP:FutureT-at-aol-dot-com]
> Sent:   Tuesday, June 24, 1997 2:39 PM
> To:     tesla-at-pupman-dot-com
> Subject:        Re: Coil Efficiency (and true wattmeter)
> In a message dated 97-06-21 07:13:23 EDT, you write:
> << Hi all,
> >          I think we could be on the verge of making some real
> > progress in nailing down the merits of actual coil parameters. There
> > is much to be done in the wall socket -> coil efficiency stakes. But
> > if we look at the raw E x BPS vs spark length figures, we can start
> >to see where the coils themselves might be improved. I think there is
> > now a most fruitful area of research open to us. We might at last
> > have a platform from which to examine L/C ratios and the like. It
>  >seems obvious now that when using current efficiency measurements we
> > are floundering around in the dark (with an arc to light the way :)
> > Malcolm
>   >>
> Malcolm, All,
> I agree.  Can you suggest a simple way to measure the capacitor voltage?
<SNIP>   The max power input for the circuit is 1300 watts,
but maybe this can be "upped" using appropriate shunts.  I haven't built
one yet, but I don't really think there's any choice, unless someone can
suggest something simpler of comparable accuracy.

Maybe Dave will add some more info here about his unit.

John Freau
Chip, John, Malcolm, ALL

The wattmeter that I have previously posted can be upgraded
easily to at least 10,000 W by simply increasing the Vmultiplier
resistor from 100k, 1/2W to 220K, 1W (115 to 230VAC input),
and changing the Rshunt from 4.6" of #14 AWG THHN copper wire
(1.00 milli-ohm) to 3.8" #8 AWG THHN wire (250 micro-ohm).
The 0 and span of op-amp is adjusted for 0-10V = 0-10kW.  Full
scale up to +rail of op-amp (about 14kW) is possible with
this arrangement.  Higher current capability is possible,
Rshunt should provide 15 millivolts FS at IFS. The above 250 micro-ohm
shunt would be suitable to currents up to 60A.  Large shunt wire should
use mechanical lugs, and sense wires tapped off lug to optocoupler
as a standard Kelvin voltage sense connection.

The 741 opamp can be replaced by a MAXIM MAX427CPA instrumentation
grade precision op amp (cost less than $5.00, Digikey, Jamesco,
MCM other in USA). This upgrade will improve accuracy to significantly
better than 1% across full span, and pin-outs are identical (direct
one-for-one swap).

Several indicating meter options are available, I like the Yokogawa
hybrid digital/baragraph analog 51 segment panel meter.  Even though
pricey (about $200 through outlets for 0-10VDC input), can be scaled
at no cost to engineering units (0-10.00KW), is specifically designed
for switchgear service, and is a very professional indicator.

The only concern I have with this design is there is minimal circuit
board clearance through the optocoupler (0.1 and 0.3 inches).  Fusing
of the V multiplier with 250V fast blow fuses (1/16A) is probably
a prudent thing to do.  The optocoupler and opamp should be carefully
shielded from air currents, and should be allowed 5-10 minutes warm-up
(actually thermal equilibration) for max accuracy.  Info for the
op-amp is available for download at the MAXIM WWW website.  Several
multichannel optoisolators with up to 5kV isolation have recently
become available from several suppliers (Hewlett-Packard) which have
significantly wider clearances than above mentioned.  I can not
infer performance of these HV devices as an analog pass-thru device
as the NEC 4 channel optocoupler part is used.

I will keep list informed as a large unit is built and tested.