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

120 bps vs. 240 bps comparison tests

All,

In previous work, I found that 120bps sync operation produced longer
sparks than higher non-sync break rates using the same power
input.  I wasn't sure at the time if this was due more to greater
transformer losses, or rather to non-linear spark growth with 
increased bps.  What was needed was to measure the voltages
on the caps, and I did this in these new tests.

Since I generally used 300 to 800 bps for the non-sync tests, there
was also some question as to whether lower break-rates around
240bps might be better.  There was also the question of whether
or not 240 bps sync operation might confer some special advantage.
Generally, I've always thought that sync gaps lose most of their
advantages at higher break rates.

In these tests, I ran my TC at both 120 and 240 bps to compare the
results, and also to measure the cap firing voltages, Joules, etc.  I'm
powering the TC using two 7.2kV, 1.5kVA potential xfrmers in series,
but I only turned up the power part way to limit the spark to 42".  I've
posted the full specs for my coil numerous times, so I didn't include
them here again.  I used the 4" by 17" corregated aluminum toroid,
with the "measuring wire" set at 42" away.  The 42" sparks are max
spark lengths with occasional hits to the measuring wire.

Test 1.    The 42" spark TC was run at 120 bps, at 550 watts.  A
HV probe and o'scope was used to measure the cap firing voltage
which was found to be 22.9kVAC.  This was higher than I expected,
so some resonant charging must be occuring with the 0.0147uF cap
and ~25mH ballast I'm using.  I also disconnected the transformer 
output from the TC and measured the voltage at the same variac
setting to compare the voltages and see how much resonant build
up was occuring.  This test showed 17.2kVAC, so there was a
substantial resonant voltage build up.  This degree of voltage build
up may agree with data given in "Principles of Radio Communication" 
by J.H. Morecroft, 1921.  I calculate about 3.85J*120bps = 463 "cap"
watts.  This indicates an 84% efficiency in the charging circuit.  

Test 2.   I increased the break rate to 240 bps.  I had to adjust the
sync gap phase carefully to obtain two equal sized bangs, if fact,
they are not yet equal.  One bang fires at 17.2kV, and the other
fires at 21kV.  I'll work some more on this and try to even it out,
but I don't think it will affect the results very much.  The sound of
the TC is higher pitched as expected, and has an irritating sound
that reminds me of Richard Hull's small thyratron switched maggy.
I figured the Joules of each bang and added them together and 
multiplied by 120 bps to obtain the cap watts for the system.  This
came to 650 cap watts, to produce a 42" spark.  The wattmeter 
read 800 watts.

Here's a chart that summarizes the results:

                      cap                            spark   
                     watts    watt     amp     length   power     charge
BPS   Joules   calc    meter   meter   inches   factor %    eff. % 

120     3.85      463      550     2.6        42         88           84
240     5.41      650      800     3.8        42         88           81

Note:  I suspect there may be about a 6% error in my scope voltage
readings, but this shouldn't have much effect on the relative
measurements between the 120 and 240 bps.  But it would affect
the charging efficiency figures, joules, etc., but again the relative
efficiencies wouldn't change much.  

Conclusions:   I am getting good power throughput at 240 bps, but
this is not translating into long sparks.  Whether one looks at Joules,
cap watts, wallplug watts, VA, etc., the higher break rate is using
up almost 50% more power to produce the same spark length.  The
sparks at 240 bps appear brighter, fuller and more fan-like, and less
bolt-like than the 120 bps sparks.  This is the same effect I saw at
higher break rates with non-sync gaps.  I did not try different toroid
sizes, but I did try different toroid sizes in the past when using the
higher break rate non-sync gaps, and it didn't help the situation. 
I'll work some more on trying to even out the bang sizes, and verify
the absolute firing voltages.  It would appear from these results, that
a large bang size contributes much more greatly to spark length 
than higher break rates, for a given power input.  It would seem to 
me that both small and large TC's might benefit greatly by using
large caps and 120 bps sync gaps, unless the spark growth behaves
differently at high powers.  I doubt it would behave differently, but I'm 
not in a position to test this (I have no room to build a large coil).  

Comments welcomed.

John Freau