: --Previous Message--
: Hi Bert,
: A time ago I posted a few messages/questions about magnifiers.
: Now the magnifier is already build.It looks very cool and Iīve designed
: it with magsim,inca,optmag and mandk.
: The driver has a diameter of 31,7" and is 40" high.
: There are around 250turns of 12gauge. The last 15 turns are spaced.
: The driver construction was very difficult,because the wooden varnished
: main frame.(there are no nails,screws;only wooden sticks and wood glue)
: The turns lay on a frame,consists of 22 (1,18" wide) arcrylic rods.
: The complete driver is packed in three layers (each 0,0315")clear PET
: (mylar?) plastics.Iīve made a isolation test with this plastic,and I
: couldnīt see those carbon tracks like Bert Pool.
: The primary is a flat coil with 2,36" wide copper ribbon.
: Distance prim.sec is only 0,4-0,5" because the wanted highest
: A coupling test at lowest prim. position gives k12=0,41
: The teritary has a diameter of 12" and has a turn length of 62"
: Topload C3: main toroid 59"x12,2"
: lower toroid 29"x6,7"
: Iīve made a few low power tests with
: around 3-4kVA and 10kV with static gap.
: Results: Around 16" stremers from breakoutpoint C3 and than
: racing sparks on L2 from top turns to the lowest turns.
: I called they racing sparks,but i donīt know what it is.
: They arenīt very bright.
: Iīve than arced the transmissionline to a grounded metal rod.
: If I have a distance to the rod less than 8" the racing sparks
: disappear.And the transmissionline arecd to ground.
: The problem is the daylight,so i couldīnt see where they comes from.
: They donīt arc to the primary,but between turns and plastic downwards,
: one time upwardsfrom the lower turns to the middle.
: Iīve made photos which I can post you,but because the daylight you
: couldnīt see a spark/streamer.
: Iīve a photo postet to a coiler,which has uploaded it to his internet
: The sphere on this photo is only for measurements and is now displaced
: (now 33"x5" toroid)
: best regards
: Hi Alex,
: Congratulations on first light - that's a very nice looking system! As
: you're discovering, insulation is one of the most challenging aspects of
: when trying to achieve tight magnifier driver coupling. Just about
: everyone who has built a magnifier has struggled with similar problems
: (usually L1:L2 flashovers). Sparking between the secondary and primary
: would be very hot, bright, and white. Since your sparks aren't bright,
: they do sound very much like racing sparks.
: As you are probably aware, the detailed mechanism underlying racing sparks
: is still not very well understood. In classical two-coil systems, racing
: sparks appear when the coupling is too high relative to the intrinsic
: dielectric strength of the secondary. The usual fixes are to reduce
: coupling or to reduce the topload breakout voltage (using a breakout
: point) to reduce overall voltage stress seen by the secondary. Changing
: the secondary's insulation system or increasing its length may also help,
: but these changes are not usually recommended since the other fixes are so
: much easier. Following are some questions, some thoughts, and some things
: you might try.
: 1. What type of wire/insulation are you using on L2 and what is the
: physical length of the closewound and spacewound sections of L2?
: 2. Is the space between L1 and L2 tightly filled by the rolled Mylar? BTW,
: PET is indeed the same thing as Mylar.
: 3. Is the Mylar tightly rolled on top of L2 or is there an air gap
: Some thoughts and some things you may wish to try: You appear to be
: exceeding the transient dielectric strength of L2's dielectric system.
: Adding a material with a relatively high dielectric constant (Mylar -
: k~3.3) in close proximity to your secondary winding can cause excessive
: electrical stress within any intervening air gap(s). If the resulting
: E-field is sufficient to produce corona in the gap then, in addition to
: possible corona damage of nearby polymers, you may encounter electrostatic
: effects due to corona charging of nearby dielectric surfaces. Nearby solid
: dielectrics that are "sprayed" by corona will develop localized
: regions of stranded surface charge. The better the dielectric material,
: the higher the localized potential and the longer these charged regions
: During normal operation, subsequent voltage reversals on the secondary can
: further increase E-field stresses between the secondary and these nearby
: charged regions to the point where a surface spark can develop along the
: surface of the dielectric. This is sometimes called a guided spark. For a
: given voltage stress guided sparks propagating along a dielectric surface
: can be 3-5X longer than the distance in free air... and sometimes
: considerably longer. The surface spark is merely discharging excess
: stranded charge between regions of the dielectric, or between the
: dielectric and the winding. However, although they are initially
: electrostatic discharges, they may initiate followthrough RF discharges
: between portions of the secondary winding. Since these are lower energy
: sparks, they are relatively dim but they can cause progressive damage to
: your secondary and Mylar insulation.
: If you can prevent L1:L2 flashovers, the natural E=field grading across
: the single layer L2 winding and the E-field shielding from L2's toroid may
: be sufficient to prevent overvolting of the coil without the need for
: supplemental insulation. I suspect the root of your particular problem is
: the close presence of the Mylar insulation layer over L2. Unfortunately,
: it may be difficult to eliminate the insulation layer with your existing
: driver design.
: Some things you might want to try: 1. Try running the system in the dark
: to see if you are developing any corona on/across L2, especially
: underneath the Mylar. Also look for any consistent points where discharges
: seem to repeat and any possible anomalies.
: 2. Increase the gap between the primary and secondary and eliminate the
: Mylar layer. Elevate L1 versus L2 (if necessary) to increase coupling to
: the desired level
: 3. If you need to provide supplemental L1:L2 insulation, try using layers
: of LDPE film or thin LDPE sheet material. LDPE's lower dielectric constant
: (k ~ 2.3) should also help to reduce E-field stressses within any air gap.
: 4. If changing the insulation system still doesn't fix the problem your
: compact primary design may, in fact, be introducing excessive transient
: stresses on the lower portion of the secondary (particularly during
: ringup). Most other successful magnifier designs have used helical
: primaries which may tend to more evenly bathe a larger portion of the
: secondary with primary flux. There has recently been some discussion of
: this (only educated speculation at this time) on the Pupman Tesla List.
: There is no clear evidence of this as yet. However, it would seem to make
: intuitive sense...
: Good luck and best regards,