Re: OLTC Update

["Tesla list" <tesla-at-pupman-dot-com>]

Original poster: "Bert Hickman by way of Terry Fritz <twftesla-at-qwest-dot-net>" <bert.hickman-at-aquila-dot-net>

Terry,

Congratulations - the system is really beginning to look good! Regarding
current monitoring - how about monitoring current through just one of
the tank capacitor leads via a smaller current transformer - should be
the same (just smaller) as the total primary current. 

It'll be interesting to see how the system behaves with a secondary in
place and how much quenching flexibility you've actually got. You should
be able to quench at every other current zero crossing (limited by the
HexFRED diodes). However, this is great, since you should be able to
tweak the coupling so that there is always an even number of "half
cycles" to complete secondary ringup. This implies that you may be able
to hit "magical" coupling coefficients that are not easily achieved in
disruptive systems - 0.28 or even 0.60(!) assuming the geometries
permit. 

As previously mentioned, a possible downside is that, through
transformer action, some of the secondary's energy will be extracted
during secondary ringdown, resulting in partial recharging of the tank
caps through the internal HexFRED diodes. But even this shouldn't be too
much of a problem since the tank caps will only recharge when the
induced tank EMF is greater than the voltage stored on the tank caps.
Since the induced primary voltage will be constrained by the
secondary-to-primary turns ratio (i.e., with the diode in the circuit
energy transfer is no longer resonant between secondary and primary),
the actual amount of power lost in this fashion should be comparatively
small. And, unlike a spark gap, the energy is not actually lost, but is
"recycled" back to the tank cap instead of being burned up.  

Great job, Terry - dang, you're getting me excited now...!

Best regards,

-- Bert --
-- 
Bert Hickman
Stoneridge Engineering
"Electromagically" Shrunken Coins!
http://www.teslamania-dot-com
  

Tesla list wrote:
> 
> Original poster: "Terry Fritz" <twftesla-at-qwest-dot-net>
> 
> Hi All,
> 
> I got the OLTC's primary circuit running off low voltage DC supplies tonight.
> 
>         http://hot-streamer-dot-com/temp/OLTC08-18-01.jpg
> 
>         http://hot-streamer-dot-com/temp/OLTC08-18-02.jpg
> 
> Hard to find the thing under all the leads and test equipment...  Here are
> some close ups of the good parts:
> 
>         http://hot-streamer-dot-com/temp/OLTC08-18-03.jpg
> 
>         http://hot-streamer-dot-com/temp/OLTC08-18-04.jpg
> 
> It will look much simpler when cleaned up ;-))
> 
> So no more of those goofy MicroSim charts...  Here is a real waveform from
> the primary:
> 
>         http://hot-streamer-dot-com/temp/OLTC08-18-05.gif
> 
> I was able to do a very important stability test where the IGBT controller
> circuits were actually all running but with a bad input waveform.  This
> checks to see if the coil would do anything "bad" if the control
> electronics go nuts:
> 
>         http://hot-streamer-dot-com/temp/OLTC08-18-06.gif
> 
> No problem at all :-))
> 
> Here is the primary voltage and drive signal at about 50 volts input (all I
> can reasonably do right now):
> 
>         http://hot-streamer-dot-com/temp/OLTC08-18-07.gif
> 
> I will have to figure out the loss and divide it by ten for the 500 volt
> case where the coil will actually run.  The losses of IGBTs or more or less
> proportional to current, but the power is proportional to current squared.
> So as you increase the current (drive voltage), efficiency of the coil
> increases.  So right now, the all important primary loss looks fine.
> 
> There is one odd thing.  The coil's Fo frequency seems to be 38kHz.  That
> is higher than I was thinking by about 6kHz.  I will have to check to be
> sure all the IGBTs are coming on, but it may just be due to the
> difficulties in measuring (and guessing) sub micro-henry inductances.  When
> you dealing with 500nH inductors, you just never know how they are going to
> turn out ;-))  Higher frequency is very nice since it reduces the size of
> the secondary.  However, current increases too and I am a bit concerned
> about that.  In any case, it looks like we have some room to play :-))
> 
> The diagram of the IGBTS, caps, and control stuff is here (sorry about the
> very rough sketch ;-))
> 
>         http://hot-streamer-dot-com/temp/OLTC08-18-09.gif
> 
> Don't let this seem too complicated.  Barely 10$ worth of odd parts there
> and you could hold them all in your hand.  If you take out the resistors
> and divide the basic circuit by ten.  It only has two "important" parts.
> The IGBT and the primary cap.
> 
> Here is something fun!  Ever try to quench a gap with all the power still
> in the primary ;-))
> 
>         http://hot-streamer-dot-com/temp/OLTC08-18-08.gif
> 
> No doubt about it.  Even though the anti-parallel diodes are still in
> circuit, Turning off the IGBTs stops the whole show right quick!  The
> trailing "screeeeem!" is the primary wondering where to dump the energy.  I
> have no idea where it goes...  Probably would not want to do that at full
> power without a secondary :-))
> 
> Oh!  It's noisy!!  Forget that "silent" thing!"  The caps squeal like pigs!
>  I have not figured out the current yet, but the low frequency and high
> current vibrates the caps like little speakers.  It sounds like, and is
> about as loud, as a florescent light ballast that has gone bad.  I wonder
> if mechanical vibration of the caps is a problem...  Probably no big deal
> but I guess solid state coil's that depend on high currents or going to
> have noisy caps.  I was only running at 1/10 voltage tonight.  I imagine it
> gets really loud at 550 VDC =:O
> 
> I could kick myself for one thing...  I have no way of getting a current
> monitor in the primary circuit :-((  I have like six Pearson current
> monitors here (one is a 50,000 amp!) and no way to insert one in the
> circuit.  I guest that is why they make the split clamp one types too (and
> charge 5X for them!  They know when they gotcha!)
> 
> So I decided to stop and think at this point tonight rather than keep
> fiddling with it till I blow it up :-))  But all seems to be going fine.
> 
> Cheers,
> 
>         Terry
> 
>