[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [TCML] BIG solid state




From: Steve Ward<steve.ward@xxxxxxxxx>

Hi Greg,

I wasn't aware of this article about your 120L coil, I really like the use
of a car engine as a power source, how well did that work for you?  Im also
curious to extract more information about how you changed the tank Z and
obtained such a large improvement in efficiency.  Did you also increase the
charging voltage for the capacitors by some factor of sqrt(3) to make up the
bang energy from cutting C to 1/3?  Or i guess you were making bigger sparks
with less energy (as input power went down), but still maybe had to bump up
the voltage somewhat?

The problems driving the Zpri and Csec changes are described in better detail here: http://www.lightninglab.org/Projects/120L50K/120L50K.html

Essentially, Zpri had to be raised since the rotary gap resistance turned out to be much higher than I had originally calculated, so the rotary gap originally wasted considerable power. Doubling the primary turns increased the total primary circuit inductance by about three [including the fixed stray inductances] so the # of caps had to drop from 13 to 4.5. To compensate Epri, we reconfigured the hv xfmrs from delta-delta to delta-star but with a less aggressive primary tap setting, yielding a new Epri that's slightly lower. One other important change was the max rotary gap speed. Break rates above 240BPS seemed to provide little extra performance, so we lowered the max break rate from 360BPS to 240BPS.


And furthermore, what do you think the real impact was of increasing the
toroid?  Do you think the benefits were mostly from dropping the resonant
frequency?  Or do you think the lower secondary impedance was better suited
for driving power into streamers?  I personally am questioning whether or
not i really need an 8 foot diameter toroid, or if i could get away with ~6
foot diameter (that way i can use 20' lengths of aluminum tube to make a
multi-element ring toroid).  I had often looked at some of your big coil
projects to try and make some decision on whether or not the monster toroid
was required for making monster sparks.  If it means i have to put in an
extra 10% power with a smaller toroid, it might be worth it.


For years I had resisted the notion that a larger toroid improves spark length, since Conservation of Energy clearly dictates that a larger toroid must necessarily lower the output voltage. However, live arc msmts on Electrum in 1998 revealed the strong presence of large, fast transients, possibly indicating an arc growth mechanism analogous to dart leaders. In this light, a local, low impedance reservoir of charge might be very important indeed. In the end, the larger toroid did in fact improve the arc performance.


There really weren't any big technological hurdles in the building of this
DRSSTC, though I have been working extensively with this topology for the
last 6-7 years or so, this was merely the next step up in size (the details
of the electronics are pretty well understood by now).  With no specific
project schedule, i pondered the quad primary drive for about the last 2
years, which in my opinion is the only "new" thing going on here, and even
that seemed too easy once it was actually implemented.  De-bugging took just
a few hours.

The main thing preventing it from happening for so long was really just a
lack of man-power.  The scale of things was just beyond what i could do by
myself in the garage.  Getting people together to work on projects is always
harder than it should be, especially when there is no deadline involved.

Steve


Coil logistics seem to scale roughly with the volume of the secondary; perhaps someone should come up with a scaling law for that. -GL




_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla