Re: Soft transformer turn on without a variac

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

Original poster: "Malcolm Watts" <m.j.watts-at-massey.ac.nz> 

Hi Dave,

On 30 Oct 2003, at 6:26, Tesla list wrote:

 > Original poster: davep <davep-at-quik-dot-com>
 >
 > Speaking as an EE, with some transformer test experience:
 > My understanding/observation:
 > 	Basic Physics:
 > 	It takes a (brief: milliseconds) period of time for
 > 	the mag field to 'penetrate' the core material.
 > 	Thus when initially powered up any transformer LOOKS
 > 	LIKE an air core coil: Very Low Impedance.  The effect
 > 	is not _noticed_ on small transformers as the
 > 	surge is small.  Rule of Thumb for large transformers
 > 	is: 'worst case' about 10x normal full load current.
 >   	The surge is independent of the _actual_ load on the
 > 	transformer.  (Until mag field can build up in the
 > 	core the load is 'invisible' to/from the primary.)

Observation suggests that a high initial peak current in the primary
is due to an unfavourable switch closure at the peak of the mains
cycle coupled with a remnant core bias in the same sense causing near
saturation over the first few cycles. The fact that it does not
always happen means either that the switch-on wrt phase is
inconsistent (i.e. the switch could be closed at either a zero
voltage crossing or a mains peak to give the extremes), the
transformer core is not always unfavourably biased wrt the mains
polarity at switch-on or a combination of both (statistically
favourable).

      It would be instructive to run a test where power is applied a
number of times with identically polarized mains peak but switched
off at both the same and oppositely polarized peaks. This would
establish where the truth lies. I have a lot of difficulty accepting
the mS-range penetration hypothesis. Perhaps someone would perform
this expt for us all. If no-one does, I would be happy to do it
albeit with some time delay due to other committments.

 > The magnitude of the surge also varies with the (accidental)
 > phase of the closing of the main switch.  If closed at or near
 > a voltage peak, the surge current is maximum: 'full' voltage
 > across a (temporarily) 'air core' coil.
 >
 > I've never tried a zero crossing switch: solid state devices
 > around Tesla/10s of amps environments need to be well done
 > to work.  They should help.  I tend to think in terms of
 > resistors in series, with bypass relays, or whatever.
 >
 > In general transformer coil materials tend to be chosen to
 > have low magnetic remnance (tendency to store a field):
 > cuts hysteresis losses...

Agreed, but it is easy to check that the remnance in an ungapped
typically laminated core is not insignificant. I have demonstrated
this many times using a lab demo transformer which has a C-I core
with the I-section removable. Sometimes it sticks like glue, other
times it doesn't.

Malcolm

 > 	(Curious Coincidence:
 > 	I first got a 'sharp education' in this effect powering
 > 	up unloaded, mid sized, variacs...  8)>>  Later
 > 	observed and measured with mid size power supplies.)
 > --
 > 	best
 > 	dwp
 >
 > ...the net of a million lies...
 > 	Vernor Vinge
 > There are Many Web Sites which Say Many Things.
 > 	-me
 > ...for he has read everything, and written nothing...
 > 	A J Raffles
 >
 >
 >