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

Re: Ballasting question



Original poster: "R.E.Burnett by way of Terry Fritz <twftesla-at-qwest-dot-net>" <R.E.Burnett-at-newcastle.ac.uk>


Hi Michael, Chuck, all,

I see the functions of the ballast as follows:

1. It limits the current flow at the instant when the spark gap fires and
   shorts out the HV supply.  Without some ballast inductance,  the arc
   would reflect an almost dead short across the line and trip the supply
   breaker.  The ballast inductor limits the rate at which this short 
   circuit current can rise during the brief conduction of the spark gap.

2. It defines the capacitor charging current.  The inductance in series
   with the supply defines the rate of rise of the current after each
   firing of the gap.  Without any ballast,  the charging current would
   be infinite immediatley after each discharging of the capacitor.  Not
   healthy for either the transformer or capacitor ;-)

   The inclusion of a ballast inductor (or resistor) defines the charging
   profile for the tank capacitor in between bangs.

3. The ballast limits the fault current to a safe value if a short circuit
   occurs on the HV side for any reason.

The ballast can be either inductive or resistive,  and can be placed at
either the low voltage (primary) side of the transformer or at the
secondary.  The operation of resistive ballast is fairly simple.  It
results in the classic RC exponential charging characteristic.

I posted a step-by-step description of how I percieve inductive ballasting 
to work the other day.  The fundamental difference that I see is this:

A resistive ballast wastes energy as heat in order to limit the
current.  Whereas,  an inductive ballast utilises this energy to charge
the capacitor to a higher voltage. (No energy is wasted with the 
inductor.)

I agree with Chuck that the ballast resonates with the tank capacitor at a
particular frequency,  but I personally think we should get away from the
fascination with resonance at the supply frequency.  (50Hz or 60Hz)

I particularly dislike the "Xl cancels Xc" theory for one reason.  This
only applies in the STEADY STATE and at ONE FREQUENCY !  Yes,  if you
apply a pure 50Hz signal with the cap across an NST and wait a while,  you
will get terrific voltages and currents if there is no spark gap.  But,
this occurs due to resonance over MANY cycles,  and is not what generally
happens in normal TC operation.

Just because Xl = - Xc,  does not imply that the charging current is
always limited by stray resistance alone.  This is a resonant circuit in
which the steady state current will be high,  but we typically do not let
the resonant charging circuit ring for more than half a cycle at most,
before depleting the cap of all its stored energy !  In my opinion
resonance at the line frequency is a moot point since it never has chance
to really get going ???

I feel that the resonant frequency of the charging circuit should be
chosen to complement the firing rate of the spark gap,  not the supply
frequency.  i.e.  The ballast and tank capacitor should be chosen so that
the capacitor is charged to peak voltage in the time between firings of
the spark gap.  This has been the aim of the work I presented a couple of
weeks back.

Chuck,  the behaviour of the circuit can be modelled with differential
equations.  It reduces to a series resonant circuit,  in which the
capacitor voltage is periodically reset when the spark gap fires.
Unfortunately, most circuit equations are "geared-up" for analysing
resonant circuits in the steady state,  rather than this transient or
"time-domain" analysis.

I use Microsim for most of my analysis work because it does all the
Laplace transform stuff for me.  (I'm an EE not a mathematician.)
Microsim is really great for getting a feel for what is going on with
resonant charging and sync rotaries.

There is some info on ballasting and resonant charging on my web page too:

www.staff.ncl.ac.uk/r.e.burnett/ballast.html

I'm glad that there are more discussions about ballasting and res charging
lately,  as I feel that the charging circuit is pretty important for good
TC performance overall.
							Cheers,

							-Richie,
On Tue, 26 Jun 2001, Tesla list wrote:

> Original poster: "Charles Hobson by way of Terry Fritz
<twftesla-at-qwest-dot-net>" <charles.a.hobson-at-btinternet-dot-com>
> 
> ----- Original Message -----
> From: Tesla list <tesla-at-pupman-dot-com>
> To: <tesla-at-pupman-dot-com>
> Sent: Monday, June 25, 2001 11:15 PM
> Subject: Ballasting question
> 
> 
> > Original poster: "Michael O. Poley by way of Terry Fritz
> <twftesla-at-qwest-dot-net>" <mpoley-at-mindspring-dot-com>
> >
> > Okay, since I got such a fast response on the other question, let me ask
> > this...
> >
> > Can someone explain the theory and practice of "ballasting"?  I'm familiar
> > enough with the physics behind the Tesla coil, but the term has been
> > bandied about a lot and I haven't seen anything explaining what exactly
> > ballasting is nor when it should be done, and when it is acceptable to
> > overlook it.
> >
> > Mike
> >
> > Michael O. Poley
> > mpoley-at-mindspring-dot-com
> 
> Hi Mike,
> 
> I hope you get some good clear answers to your question. I am looking
> forward to them also. My understanding for ballast is limited relative to
> Tesla Coil application, but as I see it, the ballast in needed for systems
> using "pole pigs", transformers where if the secondary is shorted, "like
> with a spark gap" , the primary see the short also. Hence, the minimum
> inductance needed in series with the primary appears to be a function of the
> transformer's maximum VA rating. (You experienced guys please correct me on
> this if I am wrong)
> 
> But that's not where it seems to stop. Between bangs the transformer is
> looking at the Tesla Coil's primary tank circuit capacitor Cp. Thus, there
> appears to be an LC series circuit comprising Cp and L where L = ballast
> inductance x N^2 (pole pig turns ratio). Please again you experienced guys,
> jump in and correct me if this is wrong. If I am right in this, then the
> ballast inductance can be increased from its minimum value to where it will
> resonate with Cp at 50/60Hz. Again the transformer primary circuit will
> behave as a short circuit at resonance. So--- some value between the minimum
> inductance and the resonating inductance needs to be worked out. I think
> inductances greater than the resonant value can also be considered.(Please
> again you experts, help!!) There seems again here to be computer programs
> around to work this all out, but I would like to see and understand the
> algorithms used.
> 
> I hope this is helpful here or has stirred up better and more correct
> answers to your question.
> 
> Chuck
> 
> 
> 
>