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

DRSSTC-3 VCO drive test (long)



Original poster: Steve Ward <steve.ward@xxxxxxxxx>

Hey Guys,

Im not doing anything "new", but rather something new for me.  I made
a nice little VCO box that can plug into my coils where the feedback
normally plugs in.  I did a brief test and was pleasantly surprised by
a few things.

I set my spark length to 25" with my little DRSSTC-3.  Now looking at
my recent records on this coil, i needed about 330W at 120bps, 100uS
burst length, 400VDC buss, and 450Apk in the primary to get 25".  Now
compare that to the VCO at the same settings (100uS, 100bps, 400VDC
in) i find i only need 300W... not bad :).

Also, tuning the VCO for best spark was really easy (im tuning it on
the fly!).  I have my single turn pot along with a 12t trimmer, as
well as several dip switches and timing capacitors so i can really
dial the VCO in on what range i want to operate in.  The single turn
pot (1k) will sweep the frequency from any range of 6khz to 30khz
depending on how i set the other "stuff".  Anyway, i could vary the
drive frequency by +/- 5khz in the "sweet spot" at full power.
Detuning the coil too much at full power had no ill effects aside from
a reduction in primary current and spark length.  But the VCO in its
current state is limited to about 100uS on this setup.  The fact that
the streamer detunes the coil is causing a bit of an issue.  My idea
for a solution is a ramp generator synched with the interrupter to
slowly down-shift the frequency during the burst.  I wish to control
the amount of frequency shift as well.  I believe the VCO drive will
be just perfect then.

Things get a little better.  Tested it with heavy ground strikes and
the system seems to respond very well to this.  The primary current
actually goes DOWN (most likely due to the slight detuning and very
low secondary Q at that point).  I like :-).  Any there were no real
hints that the IGBTs were hard switching.  I could see switching noise
on the primary current waveform, but they were pretty much by the zero
crossings from what i could tell.

Some other thoughts about any possible risks of hard switching in a
VCO driven system...  It seems to me that when the DRSSTC performs
best is when the drive frequency is the same as the primary tuned
frequency, which is tuned to one of the poles (right now the upper
pole seems to be the winner!  In fact this coil runs upper pole with
feedback as well).  The fact that just a slight mis tuning of any
"component" lessens the spark length considerably says to me that if
its making big sparks, all is well.  So to get to my idea... there
must be very little hard switching when the coil is operating at its
best, because being in tune implies a situation of being very close to
a ZCS condition.  Anyone have comments on that or some reason why im
wrong?

Just as a side note.. i believe what was killing the CM300s in my big
DRSSTC was the devestating hard switching of 1200+ amps when the
secondary feedback would go "bad" and lock to 3X Fo.  Thats the worst
kind of hard switching possible i think ;-).  Now, a VCO system can't
possibly have a situation like this, where everything is fine until
say a cycle or 2 before the end of the burst when suddenly it starts
switching at random.  And if the VCO is out of tune, the primary
current never builds to super high levels anyway.

Well, thats all for now, its time to get working on those VCO mod's to
get this thing really tweaked up ;-).

Seeya,

Steve