Prune wrote:

I'm poor. So I rebuilt the core of a 16 lbs microwave oven transformer and
now it's very quiet -- outside the chassis. I mounted it in the 3u power
supply chassis, and the chassis resonates like crazy, at least 30 dB more
noise than the transformer alone. Rubber/springs/foam didn't help much. I
spent quite some time building a pneumatic vibration isolator with a pump
and everything. Imagine my surprise when it turned out most of the
vibration wasn't transmitted mechanically...the hum was being induced
electromagnetically into the chassis! I almost cried! Repositioning
didn't make much difference (more of the effect seems to come from the side
with the primary winding, but orienting that side away from the vertical
walls means having the core horizontal, which makes the hum worse; standing
the transformer on its narrowest side doesn't fit in the chassis). All my
testing was done with no load as the salt water resistor boiled too quickly
and I took it apart. Would it get better/worse/same when it's loaded? And
what can I do to deal with the issue, in terms of shielding or anything
else? I swear I'm going to go crazy trying to figure this out.

Why not re-mount the transformer on a wooden chassis. I've built a few projects
this way, and as long as you're careful with regards to heat, it works just
fine. This solution would be much less prone to vibration. Use 3/4" plywood,
sand it, and then make sure to coat it completely with shellac, or some similar
synthetic product to make sure the wood dosen't absorb moisture.

Another thing to consider, is that it's possible the transformer is noisy under
load, and when it's hooked up in the chassis, it's also drawing current, and
that's what the problem is. I have a number of transformers which are dead
silent under no load conditions, and emit a noticible buzz when loaded.

Prune wrote:

Patrick Turner wrote in
:

This usually reduced the magnetizing current because the
primary inductance rises since the U of the assembled core
is higher, compared to the butted and welded cores.

So then the primary now definitely has too few turns?
I had to take the welds off as they made too much noise. What's the point
of insulating each lamination with shellac if you're going to short them by
welding? There were actually five or six deep lines of welding right
across the thing.

Without seeing this darn SOB transformer, I cannot kmow exactly what you have.

Six lines of welds eh?
It must have taken ages to get it apart, and re-assemble it, with the
the lams all re-varnished.
Using some *new* E&I material would be *so* much better.

I said I suspected the original bobbin had too few turns.
I did ask you to find out what the Bmax was, since I suspected it was high,
about 1.2T,
maybe 1.5T, and then if more turns are added to the P the B goes down,
and the tranny will be quieter.

You are not aware of what all the engineering parameters are for the
transformer you have.

And its your ignorance which has led to a regretable situation where the
preformance of what you are building is unsatisfactory.
Don't worry, I went through all this years ago, but I then learnt
there were no short-cuts, and that if I wanted silent running high power
trannies, I had to wind them myself, or
pay through the nose to some specialist winder who hated doing
what I told him.
I had to learn, learn, learn!!!!!
This meant spending lotsa time in tech libraries, copying texts,
applying the theory in my well equipped workshop, and measuring every darn
thing I did.



In the other post you said that adding more turns is a problem. For the
filter I wound my own choke and it works fine, so I think I can handle it
Of course I'll keep the right direction. There's plenty of space as
originally some was taken up by the magnetic shunts (which I threw out).

I want to ask about power factor correction. I've seen around the web that
some people using MOTs add PFC capacitors across the primaries, usually at
least 20 uF.

I see no point in placing 20 uF across the primary.
Could be dangerous, and illegal, and I have never seen any need to do it.

However, these are unmodified MOTs. In my case, with the
shunts removed and core reassembled, is this necessary? The transformer
does tend to draw a good deal of current when UNloaded (thin jumper wires I
used during some testing heat up), so I'm guessing the PF is still bad. I
don't have a scope so I don't know how to determine a proper value. Or
should I just forget about it?

The amount of current flow with no load should be no greater than 8%
of the mains input current at 110% of fully loaded condition.
The wave form of the idle current should have no more than
5% harmonic distortion relative to the existing mains wave form.

You need to measure and examine these things, and get right out of the
"I guess its right" attitude.




You need to get some 1mm thick mild steel sheet which is dirt cheap.

What is this usually used for?

Making 1,001 things and any sheet metal working fabricating plant should have
lotsa stuff in stock,
maybe plenty in their rubbish bins.

I ask as the hardware store people probably
wouldn't know what mild steel was if I asked.

The potting compound I use sometimes is molten roof pitch, heated to
about 180C

Since I didn't have shellac, I used nail polish to assemble the core (I got
the idea from this guy: http://geek.scorpiorising.ca/articles.html). I'm
not sure it can withstand that temperature.

plain old polyurethane varnish is OK for coating lams.

There is also some shellacked
paper around the windings (originally there, I didn't add it), and I think
it will carbonize at that temperature and will become conductive.

It probably is *not* shellac, but special high temp rated varnish.



I see some low ionic content silicone based potting compounds around the
web, but they are expensive. I wonder if regular hardware store silicone
can be used.

No.



What about the copper or brass bands sometimes seen parallel to the
windings but outside the core (as in Figure 13.4 at
http://sound.westhost.com/xfmr2.htm)? Maybe this together with a few more
primary turns will take care of the problem and I won't need potting.

I doubt it.

Try it out, then you'll know.



BTW, speaking of sound.westhost.com, I notice the author of
http://sound.westhost.com/vi.htm is ... Phil Allison. So I guess not all
he writes is flamebait...

Not all PA; Rod Elliot is the main author I thought.
Not much tube stuff there....

Anyway, I'm sorry if anyone else was offended by
my reply to him. Obviously, just like when one is cursing, literal meaning
is not intended -- just insult, not injury.

What's a news group without a few feathers flying occasionally?

Patrick Turner.

"Patrick Turner"

I see no point in placing 20 uF across the primary.
Could be dangerous, and illegal, and I have never seen any need to do it.


** Agreed - a totally stupid idea that just makes the PF worse.


I ( Prune) notice the author of
http://sound.westhost.com/vi.htm is ... Phil Allison. So I guess not all
he writes is flamebait...

Not all PA; Rod Elliot is the main author I thought.


See also:

http://sound.westhost.com/absw.htm

http://sound.westhost.com/project66.htm

http://sound.westhost.com/project86.htm

http://sound.westhost.com/project67.htm






........... Phil

Phil Allison wrote:

"Patrick Turner"

Lord Valve wrote:

Prune wrote:

Multiple myeloma is too good for you.
I hope your kids get it.

That's a ****ty ****ing thing to say, asshole. Stuff like that
is the reason I don't respond to Allison. You're no better
than he is. **** off.

Lord Valve
American

But I thought you said this news group is populated by
a mob of no hopers, so WTF are you doing here mr valve?

Mr Prune is giving PA something he may not want, but needs,
like we all do sometimes, something laxative......

I really shouldn't have said what I just did, but I canna help it....

Patrick Turner.


** The Turneroid has posted outright death treats on this NG - makes
merely wishing cancer on someone and their kids look tame.

I suppose being a common criminal means he really can't help it.

............ Phil

Ha.

Patrick Turner.

Phil Allison wrote:

"Patrick Turner"

Lord Valve wrote:

Prune wrote:

Multiple myeloma is too good for you.
I hope your kids get it.

That's a ****ty ****ing thing to say, asshole. Stuff like that
is the reason I don't respond to Allison. You're no better
than he is. **** off.

Lord Valve
American

But I thought you said this news group is populated by
a mob of no hopers, so WTF are you doing here mr valve?

Mr Prune is giving PA something he may not want, but needs,
like we all do sometimes, something laxative......

I really shouldn't have said what I just did, but I canna help it....

Patrick Turner.


** The Turneroid has posted outright death treats on this NG - makes
merely wishing cancer on someone and their kids look tame.

I suppose being a common criminal means he really can't help it.

............ Phil

Ha.

Patrick Turner.

I said I suspected the original bobbin had too few turns.
I did ask you to find out what the Bmax was, since I suspected it was
high, about 1.2T,

I'm not sure. I look at the Bmax formula he
http://www.du.edu/~jcalvert/tech/transfor.htm
However, I don't understand the definition of K. I'm also unsure about
the number of turns. The primary seems to be about 110 turns, +/-10.
The core format is wasteless, 5.45 x 11.1 x 13.3 cm. Bolt holes through
the four corners, smaller ones in the middles of the long sides.

applying the theory in my well equipped workshop

I'm a grad student renting a room There's a workshop here but not
well equipped. The money for that is at least a year away...

I see no point in placing 20 uF across the primary.

Power factor correction is very common for reactive loads. Many
industrial motors have them. In a (unmodified) MOT current is quite out
of phase with voltage. That's what the capacitor corrects, and many
people using scavenged MOTs do that (tesla coils and other HV projects
out there). However, I don't know how the values used change when I
modified the thing by ditching the magnetic shunts and interleaving the
laminations. If I had a scope I could just do trial and error, but I
don't. If you know any other way I can measure PF please tell me. I
don't think this has anything to do with your comment about harmonic
distortion.

It probably is *not* shellac, but special high temp rated varnish.

Most varnish doesn't dissolve in alcohol or smell like shellac when you
burn it.

Prune wrote:

I said I suspected the original bobbin had too few turns.
I did ask you to find out what the Bmax was, since I suspected it was
high, about 1.2T,

I'm not sure. I look at the Bmax formula he
http://www.du.edu/~jcalvert/tech/transfor.htm
However, I don't understand the definition of K. I'm also unsure about
the number of turns. The primary seems to be about 110 turns, +/-10.
The core format is wasteless, 5.45 x 11.1 x 13.3 cm. Bolt holes through
the four corners, smaller ones in the middles of the long sides.

The core has 44mm tongue width, ( 1.75" ) and has a 52mm stack.

The core would have a rating of 240 VA.

If you can thread 10 turns of enamel wire around the bobbin with the core in
place,
and turn on the transformer with no load, then you can measure the voltage
across 10 turns.
say you get 5 vrms. Then you have 2 turns per volt.
Then if the mains is 115v, you know there are 230 primary turns.

Then since you know the the core sectional area,
the voltage applied, the turns, and the frequency, you can work out the B
at which the transformer is running.

This is very basic engineering that you should study.



applying the theory in my well equipped workshop

I'm a grad student renting a room There's a workshop here but not
well equipped. The money for that is at least a year away...

Hmm, so why not save your pennies for something cool?



I see no point in placing 20 uF across the primary.

Power factor correction is very common for reactive loads. Many
industrial motors have them.

But not in amplifier power supplies.

Seen the schematic of an electric motor?
Know all about how it works?
Know all the relationships between R, L and C involved?
Know all the phase relationships of the voltages and currents?

When you find out, talk to me about 20 uF across the mains.

In a (unmodified) MOT current is quite out
of phase with voltage. That's what the capacitor corrects, and many
people using scavenged MOTs do that (tesla coils and other HV projects
out there). However, I don't know how the values used change when I
modified the thing by ditching the magnetic shunts and interleaving the
laminations. If I had a scope I could just do trial and error, but I
don't. If you know any other way I can measure PF please tell me. I
don't think this has anything to do with your comment about harmonic
distortion.

If you have no CRO, then you cannot easily examine any electronic
behaviours,
so its no use me explaining deeply what you need to know.


It probably is *not* shellac, but special high temp rated varnish.

Most varnish doesn't dissolve in alcohol or smell like shellac when you
burn it.

As I said, I have not seen the item.

But all the commercially produced transformers I have seen have varnish
insuation
which is immune to action by alcohols, and which
is a far better substance to use for the job compared to shellac.

Shellac has its uses, but not as an impregnant varnish in power
transformers.

Patrick Turner.

It's about 1 turn per volt. Rechecking my measurements and taking into the
account the thickness of the in-between-lamination varnish, the equation
gives me 1.7 T, which is obviously very high. According to
http://www.turneraudio.com.au/htmlwe...powertrans.htm, grain
oriented silicon steel saturates at 1.6 T. So should I add 10 more turns
to the primary?

Prune wrote:

It's about 1 turn per volt. Rechecking my measurements and taking into the
account the thickness of the in-between-lamination varnish, the equation
gives me 1.7 T, which is obviously very high. According to
http://www.turneraudio.com.au/htmlwe...powertrans.htm, grain
oriented silicon steel saturates at 1.6 T. So should I add 10 more turns
to the primary?

It might saturate at about 1.6T, but any tranny is in trouble
even when B is 1.3T, and
I routinely use 0.9T or less, so I have to use twice the weight of everyone
else.
My trannies used for my 300 watt tube amps have to handle 550 watts even at
idle,
yet after 3 hrs the t rise is less than 10C.
The idle current is less than 20 mA. The GOSS E&I wasteless core material
made here in Oz is a 125mm stack of 50mm.
It has a U of 17,000 when assembled.
These trannies are inaudible at a 600mm away on a quiet evening,
even though they are inside a steel box and not potted.

Such are the benefits of a bit of weight.
No bloomin bean counters are employed at this address.

If you add 10% more turns, perhaps it may make some difference;
no harm in trying it if there is room to place the turns, and room to place
enough insulation between the existing windings and the core, to stop the mains
voltage
ever arcing to where it might.
Mains trannies are usually tested to withstand maybe 3,000v applied between the
primary
and other things nearby.

If you have one turn per volt, and there are already 120 turns there,
then adding 10 turns will reduce the Bmax only about 8%, which
probably won't be nearly enough, and 40 turns
extra might be more like it.

Then you'd find the sec voltage will be too low.

You just may not be able to win here.

Patrick Turner.

Patrick, would it help if I also add turns to the secondary?

Prune wrote:

Patrick, would it help if I also add turns to the secondary?

Only if it is done to comply with safety rules for insulation
and isolation.

You are playing with high voltages, and they are potentially deadly.

If you asked me to perfom the necessary alterations to make such a
transformer
work properly as you wish, I would refuse, because I don't provide
bandaid professional
services, and i would leave myself open to being sued if the tranny
goes up in flames or you get a big shock.
Maybe you can sue yourself, or God, if something goes "BLAT".


It seems like you are trying to make a silk purse from a sows ear,
and I really don't like your chances of success at addressing all
matters
of concern raised.

Sorry to seem to throw a wet blanket over a hot idea,
but caution is uppermost in my mind.

Patrick Turner.

"Prune"

Patrick, would it help if I also add turns to the secondary?



** No one has yet pointed out the simplest answer of all - just increase
the frequency of the primary voltage ;-)





................ Phil

Prune wrote:

I've seen power transformers with a ribbon of copper wrapped
around the outside of the lams. Concentric with the windings.
That is, the ribbon goes along the windings in the same direction
as the wires in the windings go under the transformer bells, but
this ribbon then goes *OUTSIDE* the lams between the bells.

I'm not sure what the purpose of this is, but it might
shield the outside world from stray magnetic fields from
the transformer.

** No one has yet pointed out the simplest answer of all - just
increase
the frequency of the primary voltage ;-)

I tried that once with my headphone amp. I fed 100 Hz from my power amp
through a step-up transformer, adjusting volume to get 115 V RMS under the
load. It worked. Of course, I couldn't hear any difference

You are playing with high voltages, and they are potentially deadly.

A hundredth of what the Tesla coil guys are playing with. Power-wise, too:
some of their power supplies have as much as ten MOTs feeding the coil.

If you asked me to perfom the necessary alterations to make such a
transformer work properly as you wish, I would refuse

I didn't ask that.

I have already added extra insulation (mica, which has very high breakdown
voltage) and there's enough space for more windings and insulation (the
space previously taken up by the magnetic shunts). Everything is also in a
grounded case, with fuses, temperature controlled cooling, capacitor
discharging, etc.

All I'm asking is, this: if I add say 10% more turns to each winding, it
seems that Bmax wouldn't change any more than if I just added 10% to one
winding. Is this right? So adding turns to the secondary after I've added
turns to the primary only helps to increase ouptut voltage, not decrease
Bmax (because the formula only has numbers for one winding), no?

"Prune"

You are playing with high voltages, and they are potentially deadly.

A hundredth of what the Tesla coil guys are playing with. Power-wise,
too:
some of their power supplies have as much as ten MOTs feeding the coil.



** Standard lunatic's excuse - " others are much crazier than me
........ "




............... Phil

All I'm asking is, this: if I add say 10% more turns to each winding, it
seems that Bmax wouldn't change any more than if I just added 10% to one
winding. Is this right? So adding turns to the secondary after I've added
turns to the primary only helps to increase ouptut voltage, not decrease
Bmax (because the formula only has numbers for one winding), no?


Why not? After all, one good turn deserves another . . . .



;-(

Prune wrote:

You are playing with high voltages, and they are potentially deadly.

A hundredth of what the Tesla coil guys are playing with. Power-wise, too:
some of their power supplies have as much as ten MOTs feeding the coil.

If you asked me to perfom the necessary alterations to make such a
transformer work properly as you wish, I would refuse

I didn't ask that.

I have already added extra insulation (mica, which has very high breakdown
voltage) and there's enough space for more windings and insulation (the
space previously taken up by the magnetic shunts). Everything is also in a
grounded case, with fuses, temperature controlled cooling, capacitor
discharging, etc.

OK, so you realise that there is a danger in what you are up to.

Probably no need to use mica, its brittle, and why not some nomex,
or mylar polyester sheet?
About 0.2 mm thick would be OK, although
a mm between P&S wight comply better with your countores safety standards
codes.
And you ought to read the codes if you have not already done so.




All I'm asking is, this: if I add say 10% more turns to each winding, it
seems that Bmax wouldn't change any more than if I just added 10% to one
winding.

If you add 10% more turns to the primary, the Bmax will reduce by 10%.

The Bmax will be the same for P and S windings at all times.



Is this right? So adding turns to the secondary after I've added
turns to the primary only helps to increase ouptut voltage, not decrease
Bmax (because the formula only has numbers for one winding), no?

If you add 10% of turns to each of the P and the S windings, the voltage ratio
will stay the same, since the turn ratio is the same.

Patrick Turner.

I wonder, is the saturation issue made any worse or not due to the fact
that the transformer is only loaded at the peaks of the waveforms when it
is charging the filter capacitors through the rectifiers?

Prune wrote:

I wonder, is the saturation issue made any worse or not due to the fact
that the transformer is only loaded at the peaks of the waveforms when it
is charging the filter capacitors through the rectifiers?

The saturation issue is not made any worse by the loading of the transformer.

Saturation is a voltage issue.

The noise one gets from a transformer used in a linear cap input power supply

is usually far worse than when the load is a pure resistance.
Its because the current in the diodes is a series of pulses over a small % of
the whole cycle.

So when you have say +1,000v at 100 mA of DC power from the input cap,
it means you have 100 watts of power.

This output power will always be slightly less than the input power to the
diodes, due to losses in diodes.

The power going into a cap off the diodes is in the form of current pulses
much higher than the steady DC current output from the cap.
But the pulses don't last long, maybe only 15% of the sine wave cycle.

And voltage x current x time for the input energy must be slight;ly greater
than the
voltage x current x time for the output energy.

You need to study physics, and conservation of energy, its a very basic
concept.

Basically, the cap input off the diodes is like a bath being filled up by a
bucket at a time,
but there is a steady flow of water out the plug hole.

Darn it, If I catch that critter who let the water out of my bath again,
I'll......

Patrick Turner.

Dude, that is what I said. Diodes are only conducting at the peaks of the
waveform, recharging the filter capacitors. I know how linear AC-DC PSUs
work.

What I don't understand is what the problems that saturation creates are.
Is the secondary waveform distorted by saturation? Is output voltage
changed? If not, what's the problem?

For example, at http://www.allaboutcircuits.com/vol_2/chpt_9/11.html it
says that the secondary waveform will be distorted. So it seems to me that
if the core is saturating, output voltage can be lower than the turns ratio
would suggest.

Therefore, if by adding 10% more turns to the primary I bring my MOT away
from saturation, output voltage will decrease not quite 10%.

What's wrong with this reasoning?

"Prune"

What I don't understand is what the problems that saturation creates are.
Is the secondary waveform distorted by saturation?


** Can become very distorted.

Is output voltage changed?

** Yep - it gets reduced and distorted when the primary winding is
supplying lots of current just magnetising the core.

The *effective* primary voltage is the AC supply voltage MINUS any voltage
lost in the primary resistance by magnetising current. Eg. If the
magnetising current is 2 amps and the primary has 5 ohms resistance then 10
volts from the AC supply voltage is lost this way.

If not, what's the problem?


** The major problem is excess heat in the primary winding - especially
off load.



................ Phil

Prune wrote:

Dude, that is what I said. Diodes are only conducting at the peaks of the
waveform, recharging the filter capacitors. I know how linear AC-DC PSUs
work.

What I don't understand is what the problems that saturation creates are.
Is the secondary waveform distorted by saturation? Is output voltage
changed? If not, what's the problem?

The saturation causes noise and heat.

For part of the input wave cycle, there are high current peaks
in the copper because the inductance reduces to zero when saturation occurs.

Try reading all about magnetics and transformers at you local tech library for
a couple of days.

Patrick Turner.

"Prune"
For example, at http://www.allaboutcircuits.com/vol_2/chpt_9/11.html it
says that the secondary waveform will be distorted. So it seems to me
that
if the core is saturating, output voltage can be lower than the turns
ratio
would suggest.

Therefore, if by adding 10% more turns to the primary I bring my MOT away
from saturation, output voltage will decrease not quite 10%.

What's wrong with this reasoning?


** It depends if you are looking at the loaded or unloaded secondary
voltage - for unloaded you are right.

A transformer with a high magnetising current ( ie saturating heavily at
idle) has a BETTER secondary voltage regulation factor that one which has no
saturation at idle. It is common with small transformers, like those in
wall warts, to have the same (rms ) primary current on and off load - or
even to have LESS when on load.

Remember that primary saturation current is approximately 90 degrees out of
phase with the AC supply PLUS has peaks in value near each ZERO crossing
of the supply. The larger the AC peak voltage of the supply the larger the
resulting current peaks at ZERO volts - but they are not proportional,
the relationship becomes exponential when saturation set in.

Secondary load current will be either sine wave (with resistive loads) and
in phase with the supply OR consist of current pulses that peak in time with
each maxima when rectifiers and filter caps are used. In both cases, the
secondary load *reduces* the effective peak value of the AC supply voltage
through resistive losses in the primary - hence reducing magnetising
current and saturation.


It ain't so simple !!!





................. Phil

Prune wrote:

For example, at http://www.allaboutcircuits.com/vol_2/chpt_9/11.html it
says that the secondary waveform will be distorted. So it seems to me that
if the core is saturating, output voltage can be lower than the turns ratio
would suggest.

Therefore, if by adding 10% more turns to the primary I bring my MOT away
from saturation, output voltage will decrease not quite 10%.

What's wrong with this reasoning?

Your reasoning is clouded by your complete lack of experience
at measuring the behaviours of transformers and various turns per volt ratios
and their effct on the current.

The above URL article does not explain it very well to lead the novice to any
understanding as to what
really is seen and heard in the real world with a real transformer.

There are some things you can't learn from books, or from a screen;
you need to measure and observe, and if you don't have a CRO,
you can't see, and if you don't look, you won't know.

Patrick Turner.

Thanks for clearing that up. The primary resistance I have is a small
fraction of an ohm, so I'm probably loosing only a couple of volts in the
primary at my expected load, not enough to improve the situation much. I
expect the transformer to be operating at 2/3 to 3/4 load 99% of the time,
so the heat issue is not too great, as I also have a fan cooling the thing
with heat sensor I'll thermal-paste to the core. I figure I'll only need
to add enough primary turns to reduce the leakage enough so that the
chassis quiets down. Is there a problem of a saturating transformer in
terms of interference to other equipment on the same mains circuit?

NP

What the, posts don't seem to be nesting any deeper here... Must be
something wrong with Xnews... In case it's not clear, grandparent post was
intended as reply to Phil Allison's post.

"Prune"

What the, posts don't seem to be nesting any deeper here... Must be
something wrong with Xnews... In case it's not clear, grandparent post
was
intended as reply to Phil Allison's post.



** So NP = ????


BTW Snipping someone's entire post and replying with cryptic code is
*****ing* bad mannered.







.......... Phil

"Phil Allison" wrote in news:2vtek7F2p788jU1@uni-
berlin.de:

** So NP = ????

Sorry about that. NP is used to abbreviate No Post on some forums, meaning
no body text. I made the mistake of not putting it in the title (NP).

"Phil Allison" said:

** It depends if you are looking at the loaded or unloaded secondary
voltage - for unloaded you are right.

A transformer with a high magnetising current ( ie saturating heavily at
idle) has a BETTER secondary voltage regulation factor that one which has no
saturation at idle. It is common with small transformers, like those in
wall warts, to have the same (rms ) primary current on and off load - or
even to have LESS when on load.

Remember that primary saturation current is approximately 90 degrees out of
phase with the AC supply PLUS has peaks in value near each ZERO crossing
of the supply. The larger the AC peak voltage of the supply the larger the
resulting current peaks at ZERO volts - but they are not proportional,
the relationship becomes exponential when saturation set in.

Secondary load current will be either sine wave (with resistive loads) and
in phase with the supply OR consist of current pulses that peak in time with
each maxima when rectifiers and filter caps are used. In both cases, the
secondary load *reduces* the effective peak value of the AC supply voltage
through resistive losses in the primary - hence reducing magnetising
current and saturation.


It ain't so simple !!!

Nice post, Phil. Keep this up!

--
Sander de Waal
" SOA of a KT88? Sufficient. "

Is there any simple way to determine the winding direction of the
secondary? I can't do it by visual inspection.

Prune said:

Is there any simple way to determine the winding direction of the
secondary? I can't do it by visual inspection.

Yes! Put some DC on the secondary, and measure whether you get plus or
minus at the primary ends.......oh wait............

:-)

--
Sander de Waal
" SOA of a KT88? Sufficient. "