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

Re: [TCML] pole pig protection



Hi Gary,

Unfortunately, an NST or distribution transformer HV winding doesn't behave as a lumped inductance to incoming high voltage transients. As with other distributed LC networks, a HV winding exhibits multiple resonances when frequency-swept by a signal generator. The outermost portions of the winding appear much more inductive than the interior portions at high frequencies. So, instead of the transient voltage stress being evenly applied across the entire winding (as in a lumped inductor), the voltage distribution is much higher across the outermost turns and winding layers.

If you apply a voltage step function to the output of a large ideal lumped inductor, the resulting current will simply gradually increase over time. However, if you do the same test using a real multilayer HV winding, you'll see the current immediately jumps during the step transition due to the combination of winding inductances and interwinding capacitances. The instantaneous voltage distribution will actually be governed more by the distribution of capacitances within the winding than by the inductances.

Bert

Lau, Gary wrote:
It's true that the choke's impedance increases with frequency, but
the same is true for the HV winding.  Their impedances BOTH scale
identically with frequency, so the division ratio is a constant,
regardless of frequency.

I don't know the specific numbers for a pole pig, but for a 15/60
NST, the secondary leakage inductance is in that ballpark.  I'm
struggling at the moment to locate the specific numbers, but was in
the thousands of Henries, and this value sat well with the numerous
simulations that Terry and I did.  What I can't resolve just now is
the several thousand Hy figure, vs. a value of 663 Hy, which would
achieve mains-resonance with a 10.8nF tank cap (the 60 Hz
mains-resonant value cap for a 15/60 NST).

I had assumed that a pig's HV winding inductance would be at least
that of the NST's.  As my familiarity with pigs is not great, I can't
say this with conviction, and any corrections are welcome.  But I do
feel confident that its inductance is orders of magnitude greater
than a choke, and that any voltage division between an air-core choke
and the iron-core HV winding would be insignificant.

Regards, Gary Lau MA, USA

-----Original Message----- From: tesla-bounces@xxxxxxxxxx
[mailto:tesla-bounces@xxxxxxxxxx] On Behalf Of Jim Calvin Sent:
Wednesday, September 10, 2008 10:13 AM To: 'Tesla Coil Mailing
List' Subject: RE: [TCML] pole pig protection

Gary, The division ratio will not be the same at all frequencies.
The RF choke's impedance increases with frequency (hence a low pass
action) also I am fairly certain the inductance of the HV winding
is not 1000 henries.  In fact, the impedance at the HV winding
should be the impedance at the low voltage input times the turn
ratio (1/65).  This of course is going to be the source impedance
plus the ballast impedance.

-----Original Message----- From: tesla-bounces@xxxxxxxxxx
[mailto:tesla-bounces@xxxxxxxxxx] On Behalf Of Lau, Gary Sent:
Wednesday, September 10, 2008 9:51 AM To: Tesla Coil Mailing List Subject: RE: [TCML] pole pig protection

Correct - a choke in series with the HV winding does create a
voltage divider.  But -

1) The division ratio is constant for all frequencies, 60 Hz as
well as the tank frequency.  Filters usually imply that the
attenuation is frequency-dependant.  We want a filter that
attenuates high frequencies more-so than low frequencies - a
low-pass filter.

2) Let's say that the HV winding L1 has an inductance of 1000
Henries.  If we put a 10 mH inductor L2 in series with this, the
voltage seen by the HV winding is L1/(L1+L2): 1000 / (1000+.01) =
0.99999 of the unfiltered value.  No difference.

Regards, Gary Lau MA, USA

-----Original Message----- From: tesla-bounces@xxxxxxxxxx
[mailto:tesla-bounces@xxxxxxxxxx] On Behalf Of Jim Calvin Sent:
Wednesday, September 10, 2008 8:53 AM To: 'Tesla Coil Mailing
List' Subject: RE: [TCML] pole pig protection

As a point of fact I have destroyed a pole transformer last year
with this very same coil.  So they are not indestructible.

Gary, you are right, an inductor by itself is not a filter, but
in series with the HV winding of the transformer forms a voltage
divider with an
input
impedance of greater than 5Kohm (at 90Khz).  I have not put a
scope on it with my HV probe so I can not say for certain it is
helping.

Can you email me this filter DC.  This is exactly what I am
looking for. My email is jcalvin@xxxxxxxxxxxxxxx

-----Original Message----- From: tesla-bounces@xxxxxxxxxx
[mailto:tesla-bounces@xxxxxxxxxx] On Behalf Of DC Cox Sent:
Tuesday, September 09, 2008 11:52 PM To: Tesla Coil Mailing List Subject: Re: [TCML] pole pig protection

I disagree with some of this.

If there were zero oscillations then we would not have to use
Terry
filters
even on NST circuits as the spark gap would be shorting out the
oscillations
there also.  If you throw a scope on the circuit you can see some
 oscillations coming back from the tank circuit directly toward
the xmfr
---
yes, even thru the shorted spark gap (conduction resistance is constantly changing and not a fixed value so it is not always at
zero value).

and yes, one should employ a simple air core inductor and also a
resistive filter (12 pcs 600 Ohms in parallel, 200 watts each)
even on pole xmfrs.  again, if it is scope you can see the non-60
Hz oscillations.

pole xmfrs are testing usually for a few single shots at 125 kV
BIL (basic impulse level) using a Marx generator, however
lightning oscillations tend to be short-lived and certainly not
continuous as with a LC tank circuit.  I know many experimenters
who do not use a simple filter but these may be the same guys who
drive around without auto insurance!

the filter is simple and will save your $300 pole xmfr, plus the
$250 it cost to ship it if in another state.  The filter costs
approx $30  --- $300/$30 risk/reward ratio.

I have the filter circuit that I use if anyone wishes to contact
me
off-list
I will email it to them.

better safe than sorry and out a few hundred bucks.

Dr. Resonance




On Tue, Sep 9, 2008 at 6:59 PM, miles waldron <mileswaldron@xxxxxxxxxxx>wrote:

These comments are in response to Gary Lau, who writes on this
thread.

Gary is a very nice person. The following opinions are mine,
and for the record I agree completely with everything Gary has
said. Someone please
let
me know if I have missed the point entirely, and if that is the
case
then
I
am sorry.

Terry filter protection for a pole pig is pointless. A pole pig
is
designed
to be struck by lightning over and over again during its
lifetime. Have
you
ever been hit by lightning? As equipment goes, a pig is
typically used
to
avoid transformer limitations. You can do just about whatever
you want
with
a pole pig, and it will still be there (not hindered in any
way) after
the
explosion.

Miles Waldron




-----Original Message----- From: tesla-bounces@xxxxxxxxxx
[mailto:tesla-bounces@xxxxxxxxxx] On
Behalf
Of Lau, Gary Sent: Tuesday, September 09, 2008 6:25 PM To:
'Tesla Coil Mailing List' Subject: RE: [TCML] pole pig
protection


Opinions vary on this topic.  Here is mine.

First off, in contrast to NST's and MOT's, pole pigs are
extremely
rugged
and well designed.  They are designed to be undamaged by
lightening strikes. While it is not unheard-of, damage to pole
pigs is a very rare thing.

As to protecting the pig from RF - I assume that your TC
circuit has the main gap in parallel with the pig HV secondary.
Please consider that
the
ONLY time there is RF oscillating in the TC primary or
secondary, the
spark
gap is conducting.  If the spark gap is conducting, it is
essentially shorting out the pig's HV secondary, so there's not
a lot of RF to
protect
against.

Now, it's not entirely true that the gap is conducting
continuously
during
each bang.  Each time the current through the gap passes
through zero at the resonant frequency, the gap arc
extinguishes briefly for a few(?) nanoseconds.  During this
brief time the primary rings with its self-capacitance, in the
tens of MHz region.  The initial magnitude of
this
oscillation is Vbang, and since it is in series with the tank
cap, also initially at Vbang, the combination of the two in
series (what the HV winding sees) is 2Vbang.  This, IMHO, is
the thing that needs to be filtered, at least for NST's and
less sturdy transformers.  And this is easily and effectively
attenuated by the R-C filter in the Terry filter.

Simply adding a choke in series with the pig's HV terminal does
not a filter make.  The pig's secondary may be viewed as a many
thousand HENRY
inductor.
Adding a few mH in series with it accomplishes nothing.  Low
pass
filters
are constructed of typically two or more elements - L-C, R-C.
L-C-R,
etc.
Just adding a choke inductor does not achieve any filtering.
You might
as
well string together cloves of garlic or St Christopher
medallions.

I apologize for being so wordy and blunt, but the use of chokes
to "protect" transformers is a practice that was once done,
just because it "seemed"
at
first glance like a good idea, and persists despite strong
logic not to. It has since been shown through rigorous circuit
analysis, simulation, and measurement, to be ineffective at
best and counterproductive when done
with
bypass caps.

Regards, Gary Lau MA, USA

-----Original Message----- From: tesla-bounces@xxxxxxxxxx
[mailto:tesla-bounces@xxxxxxxxxx] On Behalf Of Jim Calvin Sent: Tuesday, September 09, 2008 3:07 PM To: 'Tesla Coil
Mailing List' Subject: [TCML] pole pig protection


I have a 25KVA pole pig I am using for my large tesla coil
(the coil only uses about 5-6KVA).  I am using a hand wound
RF choke to protect the RF from the coil from getting back
into the pig.  It is wound with 16 gauge wire onto a 8 inch
sonatube.  I measured the inductance with RLC meter to be
10mH.  My pole pig has two HV leads with one grounded and the
other connected through this choke to my coil. I have not used the coil in awhile (been a busy summer). I was going to
start using it again but I wanted to make sure this is
sufficient to protect my piggy!  Any advice would be helful.
_______________________________________________ Tesla mailing
list Tesla@xxxxxxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla

_______________________________________________ Tesla mailing
list Tesla@xxxxxxxxxxxxxx http://www.pupman.com/mailman/listinfo/tesla

_______________________________________________ Tesla mailing
list Tesla@xxxxxxxxxxxxxx http://www.pupman.com/mailman/listinfo/tesla


_______________________________________________ Tesla mailing
list Tesla@xxxxxxxxxxxxxx http://www.pupman.com/mailman/listinfo/tesla
_______________________________________________ Tesla mailing list Tesla@xxxxxxxxxxxxxx http://www.pupman.com/mailman/listinfo/tesla


_______________________________________________ Tesla mailing list Tesla@xxxxxxxxxxxxxx http://www.pupman.com/mailman/listinfo/tesla
_______________________________________________ Tesla mailing list Tesla@xxxxxxxxxxxxxx http://www.pupman.com/mailman/listinfo/tesla



_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla