LTR caps

["Paul Mathus" <pmathus-at-learningco-dot-com> (by way of Terry Fritz <twftesla-at-uswest-dot-net>)]

     Terry, et al,
     
     I am a little confused about your recommendations for LTR cap sizes.  
     
     you say below 1.5 times the resonant size, but both the cap sizes you 
     are using below and your excellent MMCcalc program give values closer 
     to 2.5 times the resonant size.
     
     Thanks,
     
     Paul


______________________________ Reply Separator
_________________________________
Subject: RE: performance/tuning question
Author:  "Tesla list" <tesla-at-pupman-dot-com> at INTERNET
Date:    7/26/00 8:13 PM


Original poster: Terry Fritz <twftesla-at-uswest-dot-net>
     
     <snip>
     
The next is point about 1.5 times the resonant size (static gap) where the 
NST list still able to charge the cap to the full NST rated peak voltage. 
This is the "LTR" type coil we often speak of.  The trick here is to charge 
a large cap to a specified voltage and take advantage of the increased 
energy.  With a rotary gap, one can really tune the inductive kick effects 
and use really large primary caps to get the full VA rating of the 
transformer into the coil's primary.  Unfortunately, this tuning is a bit 
tricky but tunable multi-string MMCs have mostly solved this.  My 15/60 LTR 
coil uses a 24nF cap and my small 9/30 uses 27nF.  The resonant size caps 
would be only 10.7nF and 8.84nF.  The 27nF cap stores three times the 
energy and fires at 120BPS.  The 8.84 cap would have to fire at 360BPS for 
the same power through put.  However, 360BPS does note "ring" well with the 
60Hz line voltage so the throughput is worse due to erratic gap firing.
LTR coils were literally born from line frequency timing theory so they are 
truly optimal in this respect.  LTR coils do not over voltage the NST and 
they only over current them a "little" (50%) which NSTs seem to take in 
stride.  If the spark gap on an LTR sync coil fails, the voltage actually 
drops about 30%...  It is interesting to note that the sync gap 
configuration was "discovered" by computer modeling before the real 
hardware was demostrated and "real" testing proved the computer's 
prediction.  No "seat of the pants" stuff there! :-))
     <snip>