Transformer winding direction

[isw]

I recently came across a YouTube video where a guy rewound a microwave
transformer to make a spot welder.

In talking about it, he stated -- and repeated -- that it was very
important for the secondary winding to have the same "sense" as the
primary -- that is, both windings had to go around the core in the same
direction.

I know that matters if the transformer is handling very asymmetric
waveforms such as in a flyback configuration, but I have never heard
that it matters for plain old 60 Hz. sinusoids.

Does it? Or was the guy just confused?

Isaac

[Phil Allison]

** A spot welding transformer is almost the direct opposite of a microwave oven tranny - big step-down instead of step-up and the lowest possible leakage reactance instead of heaps.

To help with the latter, it could easily be important to wind with the same sense. Also winding multiple, identical secondaries and connecting them in parallel is the way to go plus using other techniques like interleaving and use of strip conductors for the secondary.

Getting 3V at 200Amps is non trivial.


... Phil

[N_Cook]

Even using it in reverse, swapping primary and secondary, you will not
get the low volts and high amps for welding.
Did u-tubber just use the core and existing primary and rewind the
secondary secondary with 1 to 2mm diameter wire? should be obvious in
the vide, those sorts of dimensions

[Miguel Giménez]

May be he is winding an autotransformer, although it's not very safe for
welding :) .

--
Saludos
Miguel Giménez

[whit3rd]

On Thursday, February 12, 2015 at 9:59:58 PM UTC-8, isw wrote:
> I recently came across a YouTube video where a guy rewound a microwave
> transformer to make a spot welder.
>
> In talking about it, he stated -- and repeated -- that it was very
> important for the secondary winding to have the same "sense" as the
> primary -- that is, both windings had to go around the core in the same
> direction.

I think he was confused. Look at a transformer-style soldering gun,
the hairpin secondary doesn't have ANY helical orientation at all,
and it transforms down to high current at low volts just fine.
Ditto for feedthrough AC current meter transformers.

Output windings on such high-ratio transformers don't couple capacitively
to the primary (the low resistance and inductance completely swamp
tiny capacitive currents), and it's hard to imagine any importance of flux
coupling defects at low (50 to 60 Hz) frequency with soft-iron cores.

[Phil Allison]

whit3rd wrote:

> I think he was confused. Look at a transformer-style soldering gun,
> the hairpin secondary doesn't have ANY helical orientation at all,

** Huh ?

See hi-res pic of soldering gun tranny.

http://en.wikipedia.org/wiki/Soldering_gun

Note use of copper strip and close coupling of the secondary.

> Output windings on such high-ratio transformers don't couple capacitively
> to the primary

** Do they ever ?

> and it's hard to imagine any importance of flux
> coupling defects at low (50 to 60 Hz) frequency with soft-iron cores.

** Mick Faraday used soft iron in his toroidal job, but it was judged no much good. Silicon steel laminations have been the norm for over a century.

Also, many power transformers have separate windings on adjacent limbs of a U or C core. Invariably, there are primary and secondary coils wound on each limb that are later coupled in series or parallel.

Never seen the primary on one and the secondary the other - cos that results in a tranny with very poor regulation due to high leakage reactance.



.... Phil

[mike]

Problems I discovered with these welders are:
600W ain't nearly enough power to weld anything substantial.

Very thin stuff like battery tabs require very accurate energy
delivery. The difference between no weld and blasting a hole
thru everything is a small pressure difference holding the weldment
together.

If you don't turn it on/off at zero crossings, the saturation
state of the core can make a great difference in the next weld.

I gave up trying to weld with a MOT.

CD welders deliver a known energy to the weld and are very much
less dependent on contact resistance.

[isw]

In article <mbkard$1b6$1...@dont-email.me>, N_Cook <div...@tcp.co.uk>
wrote:

That's what he was doing.

Isaac

[isw]

In article <mbmj31$a7a$1...@dont-email.me>, mike <ham...@netzero.net>
wrote:

> On 2/12/2015 10:30 PM, Phil Allison wrote:
> > isw wrote:
> >
> >> I recently came across a YouTube video where a guy rewound a microwave
> >> transformer to make a spot welder.
> >>
> >> In talking about it, he stated -- and repeated -- that it was very
> >> important for the secondary winding to have the same "sense" as the
> >> primary -- that is, both windings had to go around the core in the same
> >> direction.
> >>
> >> I know that matters if the transformer is handling very asymmetric
> >> waveforms such as in a flyback configuration, but I have never heard
> >> that it matters for plain old 60 Hz. sinusoids.
> >>
> >> Does it? Or was the guy just confused?
> >>
> >
> > ** A spot welding transformer is almost the direct opposite of a
> > microwave oven tranny - big step-down instead of step-up and the lowest
> > possible leakage reactance instead of heaps.
> >
> > To help with the latter, it could easily be important to wind with the same
> > sense. Also winding multiple, identical secondaries and connecting them in
> > parallel is the way to go plus using other techniques like interleaving and
> > use of strip conductors for the secondary.
> >
> > Getting 3V at 200Amps is non trivial.
> >
> >
> > ... Phil
> >
> Problems I discovered with these welders are:
> 600W ain't nearly enough power to weld anything substantial.

Mine easly pops the 15 Amp. breaker if I set it for too long a pulse.

> Very thin stuff like battery tabs require very accurate energy
> delivery. The difference between no weld and blasting a hole
> thru everything is a small pressure difference holding the weldment
> together.

Mine does a fine job on things like coathanger wire and the stainless
steel strips from windshield wipers.

> If you don't turn it on/off at zero crossings, the saturation
> state of the core can make a great difference in the next weld.

NP; just use a good SSR for switching.

Isaac

[mike]

Yep, if your objective is to weld your breaker, you're on the right track.

>
>> Very thin stuff like battery tabs require very accurate energy
>> delivery. The difference between no weld and blasting a hole
>> thru everything is a small pressure difference holding the weldment
>> together.
>
> Mine does a fine job on things like coathanger wire and the stainless
> steel strips from windshield wipers.

So does mine on things that aren't damaged by overheating.
Just hit it until it glows red.
Battery tabs aren't so forgiving.
I spring loaded the tips separately to give some repeatability.
I got about 90% good welds. When it's a 10-cell battery pack,
90% ain't nearly good enough.

>
>> If you don't turn it on/off at zero crossings, the saturation
>> state of the core can make a great difference in the next weld.
>
> NP; just use a good SSR for switching.

Agree, but I expect most DIYers don't.
I set the weld time by an integral number of cycles of 60 Hz.
Took about 6 cycles to weld a battery tab.

Still claim that a controlled energy dump is far superior.
It's relatively insensitive to contact resistance.
My CD welder is rated for 7V peak across .001 ohms.
It's over before the cell case even gets warm.
>
> Isaac
>

[Phil Allison]

mike wrote:

>
> Problems I discovered with these welders are:
> 600W ain't nearly enough power to weld anything substantial.
>
> Very thin stuff like battery tabs require very accurate energy
> delivery. The difference between no weld and blasting a hole
> thru everything is a small pressure difference holding the weldment
> together.
>
> If you don't turn it on/off at zero crossings, the saturation
> state of the core can make a great difference in the next weld.
>

** Zero crossing switching is the worst possible for creating large in-rush surges and hence fully magnetising the core of an AC supply transformer.

Switching on at a voltage peak, while under load, is the best option.

A triac switch will always go off near zero current, which will be nowhere near zero volts.

... Phil

[whit3rd]

On Friday, February 13, 2015 at 4:01:02 PM UTC-8, Phil Allison wrote:
> whit3rd wrote:
>
> > I think he was confused. Look at a transformer-style soldering gun,
> > the hairpin secondary doesn't have ANY helical orientation at all,
>
> ** Huh ?
>
> See hi-res pic of soldering gun tranny.
>
> http://en.wikipedia.org/wiki/Soldering_gun
>
> Note use of copper strip and close coupling of the secondary.

That's NOT a hairpin secondary! Look instead at this

<https://plus.google.com/photos/118343199678883181101/albums/6115815766613132033/6115815770882958370?banner=pwa&pid=6115815770882958370&oid=118343199678883181101>

Even if the transformer manufacturer works on good flux coupling, the external
loop of the iron's tip will leak lots of flux; you should expect every CRT in the vicinity
to shimmy while you hold down the trigger. There's no particular reason to care, in this
case.

[Phil Allison]

whit3rd wrote:

> On Friday, February 13, 2015 at 4:01:02 PM UTC-8, Phil Allison wrote:
> > whit3rd wrote:
> >
> > > I think he was confused. Look at a transformer-style soldering gun,
> > > the hairpin secondary doesn't have ANY helical orientation at all,
> >
> > ** Huh ?
> >
> > See hi-res pic of soldering gun tranny.
> >
> > http://en.wikipedia.org/wiki/Soldering_gun
> >
> > Note use of copper strip and close coupling of the secondary.
>
> That's NOT a hairpin secondary! Look instead at this
>
> <https://plus.google.com/photos/118343199678883181101/albums/6115815766613132033/6115815770882958370?banner=pwa&pid=6115815770882958370&oid=118343199678883181101>
>

** So it's a one turn secondary on a toroidal core.

Magnetic coupling will be poor, but still good enough for a soldering tool.

> Even if the transformer manufacturer works on good flux coupling, the external
> loop of the iron's tip will leak lots of flux;

** Any external flux will come from the iron core, not the loop.

Removing the soldering tip will not eliminate external flux or even reduce it much.



... Phil

[mike]

Well, I don't recall mentioning volts at all.
Yes, triacs switch off at the zero crossing of the current.
Not much you can do about that.

I question not turning it on at zero volts???
Please explain the large in-rush current when there's zero volts
on the transformer primary?

[Phil Allison]

mike wrote:


> >> If you don't turn it on/off at zero crossings, the saturation
> >> state of the core can make a great difference in the next weld.
> >>
> >
> > ** Zero crossing switching is the worst possible for creating large in-rush surges and hence fully magnetising the core of an AC supply transformer.
> >
> > Switching on at a voltage peak, while under load, is the best option.
> >
> > A triac switch will always go off near zero current, which will be nowhere near zero volts.
> >
> >

> Well, I don't recall mentioning volts at all.

** You did, cos "zero crossing" means at zero voltage crossings.

> Yes, triacs switch off at the zero crossing of the current.
> Not much you can do about that.

** Then why did you say it was needed ??

> I question not turning it on at zero volts???

** Really ? Did you try Googling the topic?

> Please explain the large in-rush current when there's zero volts
> on the transformer primary?

** The surge current peak comes a little after the first voltage peak.

Consider that it takes half the time for the applied voltage wave to first average zero if you switch at a peak - half a cycle later instead of a whole cycle - minimising the low frequency component of the wave.


... Phil

[mike]

if you say so.
As long as you stay out of saturation, the main component of the
primary current is due to the shorted secondary.
I think I'll give my SCR the benefit of switching on when there's zero
voltage.

[whit3rd]

On Sunday, February 15, 2015 at 1:02:31 AM UTC-8, mike wrote:

> As long as you stay out of saturation, the main component of the
> primary current is due to the shorted secondary.
> I think I'll give my SCR the benefit of switching on when there's zero
> voltage.

Not really the best idea; it is then possible that the previous turn-off
happened after a half-cycle (+), and if this ON state starts with a half-cycle (+)
as well, that's two half-cycles of the same polarity. The likelihood
of saturation is very high. Turn-on at peak V is a strategy that
minimizes saturation risk.

[Phil Allison]

mike wrote:

> >
> if you say so.


** Wot a smug prick you are.

> As long as you stay out of saturation,

** No chance - you are all wet right now.

> the main component of the
> primary current is due to the shorted secondary.

** The secondary is not shorted.

> I think I'll give my SCR the benefit of switching on when there's zero
> voltage.

** More fool you.


... Phil

[mike]

Well, I'll repeat what I wrote earlier in the thread:

I set the weld time by an integral number of cycles of 60 Hz.

The key word there is "integral" as in complete as in full as in
you don't get two half cycles of the same polarity.

Designing your welder to avoid saturation is a

[mike]

On 2/15/2015 3:46 AM, Phil Allison wrote:
> mike wrote:
>
>>>
>> if you say so.
>
>
> ** Wot a smug prick you are.

Seems to be your opinion of everyone.

>
>
>> As long as you stay out of saturation,
>
>
> ** No chance - you are all wet right now.
>
>
>
>> the main component of the
>> primary current is due to the shorted secondary.
>
> ** The secondary is not shorted.

Once again, your signature condescending tone declares that
the other guy is always wrong.
Once again, you nitpick instead of attempting to understand.
Communication is difficult enough, even when you try.
When you try NOT to understand, you're just being you.
I'm sure you win a lot of arguments when the other guy
just gives up trying to influence your thinking.

My presence in this thread is to help others get the most
out of their MOT welder. I gave up trying to influence you
long ago. Just trying to give some balance to the view.

>
>
>> I think I'll give my SCR the benefit of switching on when there's zero
>> voltage.
>
>
> ** More fool you.

There it is again.
>
>
> ... Phil
>

[jurb...@gmail.com]

>"Once again, your signature condescending tone declares that
>the other guy is always wrong. "

That's not fair. I have seen Phil be nice to people a dozen times.

In the last thirty years.

Go ahead and hate the MF but respect him, he knows WTF he is doing and I know enough to know he knows WTF he is doing.

[mike]

One of the issues with people who know everything is that they often
don't bother to understand the question before pontificating on the CORRECT
answer.
So you get a snarky response on their way to solve the problem that they
inferred based on their experience.
They have no trouble calling you names if you disagree with their
interpretation.
Since they know everything, anything you say is WRONG.
And it's no use trying to support your position, cuz they ain't listenin'.

Communication is a two-way process. The objective of the mentor is
to use terminology that the newbie can understand. Sometimes
trying to state the exactly perfectly technically correct description
obfuscates the key issue. The nitpickers jump on that to tell you
that you're wrong...and stupid.
It's not about education.
It's not about being right.
It's about telling the world that the other guy is WRONG!!
Throw in a few of your favorite pet names for good measure.

I like to have technical discussions. That requires listening
on both sides and supporting the argument with logic.
Calling me stupid doesn't help anybody.

[Chris Jones]

Although he is rude, Phil is usually right, and this time is no exception.

I suggest you get a hall-effect current transducer connected to a DSO so
that you can measure inrush current, and try out switching the
transformer on at both the peak of the mains voltage, and also at the
zero-crossing, and look at the current waveforms. When you run out of
working triacs you could also google it.

The key to understanding this is to realise that the magnetic flux is
proportional to the time-integral of the applied voltage, and that in
continuous operation the flux is normally close to zero when the voltage
is close to maximum, and the flux is close to maximum when the voltage
is close to zero. Have a look at the first link here:

www.te.com/commerce/DocumentDelivery/DDEController?Action=srchrtrv&DocNm=13C3206_AppNote&DocType=CS&DocLang=EN
http://sound.westhost.com/articles/inrush.htm
http://en.wikipedia.org/wiki/Inrush_current#Transformers

Chris

[Phil Allison]

Chris Jones wrote:

>
> Although he is rude, Phil is usually right, and this time is no exception.
>
> I suggest you get a hall-effect current transducer connected to a DSO so
> that you can measure inrush current, and try out switching the
> transformer on at both the peak of the mains voltage, and also at the
> zero-crossing, and look at the current waveforms. When you run out of
> working triacs you could also google it.
>
> The key to understanding this is to realise that the magnetic flux is
> proportional to the time-integral of the applied voltage, and that in
> continuous operation the flux is normally close to zero when the voltage
> is close to maximum, and the flux is close to maximum when the voltage
>

> http://sound.westhost.com/articles/inrush.htm


** FYI:

I had a large input to the writing of that article - Rod is a friend and colleague, we talk and email regularly.

The most surprising thing is how the in-rush surge of an unloaded transformer consists of brief pulses all with the same polarity - it's DC current.

Also, if you power a 240V tranny from 120V, surges are eliminated.



.... Phil

[mike]

been there done that
AEMC MR461 current probe.
TEK TDS540 scope.
I designed my first production forward converter ~40 years ago.

so
> that you can measure inrush current, and try out switching the
> transformer on at both the peak of the mains voltage, and also at the
> zero-crossing,

with the secondary heavily resistively loaded,

and look at the current waveforms. When you run out of
> working triacs you could also google it.
>
> The key to understanding this is to realise that the magnetic flux is
> proportional to the time-integral of the applied voltage, and that in
> continuous operation the flux is normally close to zero when the voltage
> is close to maximum, and the flux is close to maximum when the voltage
> is close to zero. Have a look at the first link here:

Are we talking about inductors or transformers with load resistors
that cause a steady state primary current 2X their design rating?

>
> www.te.com/commerce/DocumentDelivery/DDEController?Action=srchrtrv&DocNm=13C3206_AppNote&DocType=CS&DocLang=EN
>
> http://sound.westhost.com/articles/inrush.htm
> http://en.wikipedia.org/wiki/Inrush_current#Transformers
>
> Chris
>

Thanks for the links. I like to learn new stuff.
I'm still trying to get my head around why the graphs in the first
link are reversed in time, but if I stand on my head, it looks
like the drive signal is optimized to maximize inrush current.
I don't have any argument with that. You can certainly manage
the drive so the core saturates.

My attempts were to arrange the drive signal to MITIGATE inrush
current.

The key point is in the wikipedia link:
"Worst case inrush happens when the primary winding is connected at an
instant around the zero-crossing of the primary voltage, (which for a
pure inductance would be the current maximum in the AC cycle) and if the
polarity of the voltage half cycle has the same polarity as the remnance
in the iron core has. (The magnetic remanence was left high from a
preceding half cycle)."
end quote

If you always turn off the current at the current zero crossing with
a positive voltage slope, then always turn on the next weld pulse
at zero voltage on the positive voltage slope, doesn't that leave
you in a remanence position to avoid saturation at the next turn on?
If not, why not?

The SSR is gonna turn off near zero current. About all I can control
is the slope of the voltage sinewave when I give the command.
To turn it on it's far easier to sense the zero crossing of the line
voltage than the peak. Isn't a major portion of the primary current
in phase with the primary voltage due to the resistive secondary load?
Isn't it the leakage inductance that causes the phase shift?

Under the control conditions described above where we control both
ends of the waveform to manage remanence and have a very low value
resistive load, how much would I gain by waiting for the peak line
voltage at turn on?

I'd go look, but it's stored behind a bunch of junk in the garage.

As I recall, I didn't make many measurements without load. But
with the secondary (almost) shorted in the weld mode, I don't
remember any horrible input current transients.
I do know that synchronization with the line made a major improvement
in the repeatability of the welds.


I'm up for some education.

My thinking was that, if not for saturation, the SCR would be less
stressed if I turned it on at zero voltage when the primary current was
zero. And from the unpowered state, the voltage and current can't
be anything but zero.
And that, if I could arrange the resting place on the B-H curve
from the previous pulse such that the first half-cycle wouldn't
saturate the core, that's the best I could do easily.
Measurements didn't show any horrible first cycle inrush.
Welds got more repeatable.

Let me say the same thing in different words.
If the load is linear resistive, the transformer current and voltage
will be approximately in phase. If the SCR shuts off at zero
current, the voltage will also be near zero volts (plus whatever the
leakage inductance allows).
Case 1, you start the next pulse in a nanosecond.
Isn't the initial current still pretty near zero?
Isn't the point on the B-H curve still about the same?
Case 2, you start the next pulse next week at the zero crossing
of the input voltage headed in the same direction.
What's the initial current?
What's the initial point on the B-H curve?
How is restarting it synchronously significantly different from
just leaving it running?

I'm not disputing the articles you posted.
I'm not saying anything about the general (worst) case.
I'm suggesting that this is how you engineer a spot welder using a MOT.

Where did my thinking go wrong?

[mike]

Why is that surprising?
In the steady state, it is traversing a very nonlinear B-H loop.
If you restart it from a place different from where you left it, you drive
it "off center". To regain steady state, you have to apply a
DC, component.

[Phil Allison]

mike wrote:


> >
> >> http://sound.westhost.com/articles/inrush.htm
> >
> >
> > ** FYI:
> >
> > I had a large input to the writing of that article - Rod is a friend and colleague, we talk and email regularly.
> >
> > The most surprising thing is how the in-rush surge of an unloaded transformer consists of brief pulses all with the same polarity - it's DC current.
> >
> Why is that surprising?

** Most people are *very* surprised to find this out.

Just like YOU were surprised that zero switching produces maximum surges in transformers.

It's counter intuitive in both cases.



.... Phil

[Phil Allison]

mike wrote:

>
> Thanks for the links. I like to learn new stuff.
> I'm still trying to get my head around why the graphs in the first
> link are reversed in time,

** Figures 3 & 4 showing scope screens have been published up side down so need rotating by 180 degrees.

> but if I stand on my head, it looks
> like the drive signal is optimized to maximize inrush current.

** By simply switching the AC supply on at zero volts.

> I don't have any argument with that. You can certainly manage
> the drive so the core saturates.

** By simply switching the AC supply on at zero volts.

> My attempts were to arrange the drive signal to MITIGATE inrush
> current.

** Bollocks.

> If you always turn off the current at the current zero crossing with
> a positive voltage slope, then always turn on the next weld pulse
> at zero voltage on the positive voltage slope, doesn't that leave
> you in a remanence position to avoid saturation at the next turn on?

** Nope.

Read the link.

> To turn it on it's far easier to sense the zero crossing of the line
> voltage than the peak.

** Then allow a 4mS delay before firing the triac.

> Under the control conditions described above where we control both
> ends of the waveform to manage remanence and have a very low value
> resistive load, how much would I gain by waiting for the peak line
> voltage at turn on?

** Try it and see.

Most of the transformers I see are under heavy load at switch on, charging hefty filter caps. Only makes the combined switch on surge worse, compared to no load.



.... Phil

[Chris Jones]

If there is unnecessarily large inrush current that is due to core
saturation late in the first half-cycle, then adding a resistive load on
the secondary won't fix that.

> and look at the current waveforms. When you run out of
>> working triacs you could also google it.
>>
>> The key to understanding this is to realise that the magnetic flux is
>> proportional to the time-integral of the applied voltage, and that in
>> continuous operation the flux is normally close to zero when the voltage
>> is close to maximum, and the flux is close to maximum when the voltage
>> is close to zero. Have a look at the first link here:
>
> Are we talking about inductors or transformers with load resistors
> that cause a steady state primary current 2X their design rating?

It doesn't much matter whether we are talking about transformers or
ungapped inductors, in that the transformer with its secondary
unconnected can still draw a lot of inrush current, and adding a load on
the secondary won't fix that.

>>
>> www.te.com/commerce/DocumentDelivery/DDEController?Action=srchrtrv&DocNm=13C3206_AppNote&DocType=CS&DocLang=EN
>>
>>
>> http://sound.westhost.com/articles/inrush.htm
>> http://en.wikipedia.org/wiki/Inrush_current#Transformers
>>
>> Chris
>>
> Thanks for the links. I like to learn new stuff.
> I'm still trying to get my head around why the graphs in the first
> link are reversed in time, but if I stand on my head, it looks
> like the drive signal is optimized to maximize inrush current.

Yes I don't know why they put the scope plot backwards. It adds
unnecessary confusion, though at least they do mention it in the text. I
can only assume they lacked the ability to flip it easily in their
chosen method of document preparation.

> I don't have any argument with that. You can certainly manage
> the drive so the core saturates.
>
> My attempts were to arrange the drive signal to MITIGATE inrush
> current.

I know, and I was just suggesting that turning on when the mains voltage
goes through zero is not the best way to do that. Turning on at the
zero-crossing of the mains voltage is good if your load is capacitive,
e.g. the input of a SMPS.

>
> The key point is in the wikipedia link:
> "Worst case inrush happens when the primary winding is connected at an
> instant around the zero-crossing of the primary voltage, (which for a
> pure inductance would be the current maximum in the AC cycle) and if the
> polarity of the voltage half cycle has the same polarity as the remnance
> in the iron core has. (The magnetic remanence was left high from a
> preceding half cycle)."
> end quote
>
> If you always turn off the current at the current zero crossing with
> a positive voltage slope, then always turn on the next weld pulse
> at zero voltage on the positive voltage slope, doesn't that leave
> you in a remanence position to avoid saturation at the next turn on?
> If not, why not?

I think the remanence is clouding the issue. It is a relevant effect but
even without it, there are good and bad times to switch on the
transformer primary, and it would be better to consider remanence only
after the basic situation with a soft-magnetic core is thoroughly
understood. You might be able to use the remaining flux in the
switched-off transformer to choose the least worst of the two zero
crossings to switch it on at, but even then, I think you would do better
to switch on at a different time. Why not seriously try it out with a
current transducer and DSO, (and a vastly over-sized triac or even
better a pair of big SCRs, just in case!). It would be nice to see the
plots, and it is one way to end an argument.

>
> The SSR is gonna turn off near zero current. About all I can control
> is the slope of the voltage sinewave when I give the command.
> To turn it on it's far easier to sense the zero crossing of the line
> voltage than the peak.

The mains frequency (or period) is accurate and stable enough, and
microcontrollers or even 555 timers are cheap enough that as Phil
mentioned, you can figure out the time of the voltage peak from the zero
crossing.

> Isn't a major portion of the primary current
> in phase with the primary voltage due to the resistive secondary load?
> Isn't it the leakage inductance that causes the phase shift?

That sounds reasonable, but if the core saturates then that's the least
of your worries. The primary current is not necessarily a good way to
determine the core flux density, as you can make the primary current be
whatever you want by choosing the secondary current.

>
> Under the control conditions described above where we control both
> ends of the waveform to manage remanence and have a very low value
> resistive load, how much would I gain by waiting for the peak line
> voltage at turn on?

Time for an experiment. It depends a lot on the transformer design. I
have read that toroidal transformers produce more problematic saturation
effects than E-I types, and if the core was nominally run at less than
half of its saturation flux density then there will be no problem. Due
to the more uniform geometry I think they can run toroidal transformers
close to saturation in normal operation.

>
> I'd go look, but it's stored behind a bunch of junk in the garage.
>
> As I recall, I didn't make many measurements without load. But
> with the secondary (almost) shorted in the weld mode, I don't
> remember any horrible input current transients.

I wonder whether they even bother to switch the transformer on at the
right time in a microwave oven. The current that your transformer draws
in steady state might be so high (due to the secondary current) that
saturation doesn't make it all that much worse, but given the choice (or
given a bigger transformer than your circuit breaker likes), you might
as well make it optimum if that is just a matter of inserting a small delay.

I have an arc welder that sometimes trips the breaker if I turn it on at
the wrong time (with a mechanical switch), and it would be nice if it
didn't.

> I do know that synchronization with the line made a major improvement
> in the repeatability of the welds.
>
>
> I'm up for some education.
>
> My thinking was that, if not for saturation, the SCR would be less
> stressed if I turned it on at zero voltage when the primary current was
> zero. And from the unpowered state, the voltage and current can't
> be anything but zero.

The risk of saturation occurs well after the zero crossing when the
voltage is turned on. If the load is not capacitive (not a big SMPS)
then the SCR won't mind if you turn it on when there is voltage across
it, and later on in the cycle it will be much happier.

> And that, if I could arrange the resting place on the B-H curve
> from the previous pulse such that the first half-cycle wouldn't
> saturate the core, that's the best I could do easily.
> Measurements didn't show any horrible first cycle inrush.
> Welds got more repeatable.
>
> Let me say the same thing in different words.
> If the load is linear resistive, the transformer current and voltage
> will be approximately in phase. If the SCR shuts off at zero
> current, the voltage will also be near zero volts (plus whatever the
> leakage inductance allows).

And the magnetizing current too.

> Case 1, you start the next pulse in a nanosecond.

This is not the same thing as I was discussing. If the transformer is in
steady-state operation and if you were able to turn off the primary at a
zero-crossing of the mains voltage, that is a time when there is maximum
flux in the core. If you instantly switch it back on again, sure this
will be pretty much a continuation of steady-state operation.

If instead you turn it off for an integer number of mains cycles that
adds up to a few seconds, the core flux will not be the same when you go
to switch it back on again.

> Isn't the initial current still pretty near zero?

Perhaps but that doesn't really matter as regards the risk of saturation.

> Isn't the point on the B-H curve still about the same?

Yes, if you only switch it off for nanoseconds. No, if you wait a few
seconds until you have repositioned your parts for the next weld.

> Case 2, you start the next pulse next week at the zero crossing
> of the input voltage headed in the same direction.
> What's the initial current?
> What's the initial point on the B-H curve?
> How is restarting it synchronously significantly different from
> just leaving it running?

The flux in the core is different.

Put a big inductor (maybe a car ignition coil) across a 12Vrms AC supply
and make sure you are holding the terminals a nanosecond after you
disconnect the supply, at the instant when the AC supply is at zero
voltage (and the inductor is carrying maximum current).

Then give it to me and I will hold the terminals a week later.

I think you will notice the difference. The state of the flux in the
core matters.

Note that in the case of a transformer, it is possible that some value
of secondary current could result in the primary current being zero (or
any other chosen value) at the zero-crossing of the mains voltage. That
is not relevant to my point, which relates to the flux density in the
core, which won't be affected much by the secondary current if a
low-impedance supply is driving the primary winding.

>
> I'm not disputing the articles you posted.
> I'm not saying anything about the general (worst) case.
> I'm suggesting that this is how you engineer a spot welder using a MOT.
>
> Where did my thinking go wrong?
>

It may just mean that you need a larger rating for your fuse or circuit
breaker and more expensive triac or SCRs than you could otherwise get
away with.

When I tried spot welding, I was never able to get enough current from a
MOT-sized transformer with a few turns on the secondary. I could sort of
weld things if I applied very light pressure so that the workpieces made
poor contact with each other and the resistance was high enough for the
(insufficient) current to heat them, but this wasn't really satisfactory
because getting the force and contact resistance just right was not
reliable.

If I made a transformer big enough to weld thick workpieces with proper
contact pressure, it might cause excessive drop in my mains supply,
and/or trip breakers.

I think the best option for me is a series-parallel array of Maxwell
boostcaps. This would eliminate the requirement for a large mains supply
capability. The 3000F ones are rated for 1900 Amps each, so about 5 in
parallel would probably supply about enough current for any normal sheet
metalwork up to a couple of millimetres thick which seems to require
close to 10kA capability. (Aluminium welding requires several times more
current so I won't try that.) Most of the references that I have seen
tend to suggest that the weld itself requires somewhere in the region of
1.5 Volts, but the electrodes etc. will have some resistive drop so I
think at least 2 banks of boostcaps in series will be desirable. Due to
the capacitors holding more energy than the total that you would want
for one weld, it would be necessary to find a way to switch them off,
and it would also be very useful to be able to adjust the current by PWM
during the weld. Therefore a lot of MOSFETs would be required. It seems
that the best current rating per dollar occurs for individual MOSFETs
rated at about 100A, so about 100 of these in parallel would be required
for 10kA. I think a totem-pole style half-bridge topology might work,
using the output cables as an inductor to smooth the output current. A
multi-phase PWM arrangement with multiple output inductors could make
better use of the current rating of the caps. It would be an interesting
project but I don't have time to do it yet. I am somewhat concerned
about what would happen in the event of one failed MOSFET, and I would
like to think of a way to mitigate that. Perhaps the bondwire or package
pin would be an adequate fuse.

Chris

[whit3rd]

On Sunday, February 15, 2015 at 9:08:03 PM UTC-8, mike wrote:

> >>>> the main component of the
> >>>> primary current is due to the shorted secondary.
> >>>
> >>> ** The secondary is not shorted.

> My thinking was that, if not for saturation, the SCR would be less
> stressed if I turned it on at zero voltage when the primary current was
> zero. And from the unpowered state, the voltage and current can't
> be anything but zero.

So, this is a concern that the SCR's dI/dt rating may be exceeded
if peak voltage is present at the turnon time (the specific
microsecond of time, because SCRs turn on faster than
an AC period). That's a valid concern, with two solutions:
(1) use an IGBT instead of SCR (yeah, I know, it's a big deal),
or (2) use a transformer with enough stray inductance that
the worst-case current risetime is tolerable.
One hopes that the transformer is appropriate to this kind of
frequent-switching use, and fulfills requirement (2). You
really HAVE to hope, the microsecond-scale inductance of
the transformer cannot be easily measured (the core isn't
fully magnetized that quickly, so 60 Hz measurement tells
an inappropriate value). Flux coupling to the secondary
will also be poor for that short time.

A 'typical' SCR (TYN640, in stock at DigiKey, needs dI/dt under
50A per microsecond...) doesn't need much inductance to keep its
turnon conditions satisfied.

> And that, if I could arrange the resting place on the B-H curve
> from the previous pulse such that the first half-cycle wouldn't

> saturate the core...

Again, this depends on the transformer design and material. Remember Phil's
comment that a 240V transformer on 120V excitation wouldn't saturate-
because a 240V transformer has an oversize core for 120V excitation.
The key concept is that the core remanent field also might be zero,
and a first half-cycle starting at zero magnetization is more stressful than
that same half-cycle starting at the inverse remanent field (which is what
subsequent cycles of AC excitation provides). Starting at peak V
is a safe bet if the remanent field is negligible, like when
the gadget has been powered down for a while.

[mike]

This has been much more of a discussion than the "argument" you typically
see in these newsgroups. I appreciate your use of logic instead of
the usual name-calling. A discussion doesn't need to end with a loser.

I had it in my head that remenence was the primary source of the inrush
problem and all you'd need to do is prevent the first peak from
driving the core into the saturation. If you left the core in the
optimal state given your intent to
turn it on at zero voltage crossing, you'd be done.
I'll certainly examine it more carefully next time I use a transformer.

I dug out the documentation for my welder. Turns out that I couldn't
delay the turn-on if I wanted. All the solid state relays I had
were/are zero voltage turn on units. Best I can do is control the slope
of the AC input at turn on and turn off.

The relay is rated at 600A non-repetitive for the duration of a typical
weld. I was more concerned about drawing 40 amps out of a 15 amp breaker
thru a daisy chain of connections thru outlets along the way from the
breaker to the test bench in the back room. I never popped the breaker,
but I'm sure I wasn't doing it any favors.
I took it out to the garage and hooked it up to the dryer plug
when I wanted to use it.

Along the way, I picked up a 125 joule CD welder off ebay for $20 shipped.
Solved all the battery tab welding problems without stressing the
electrical system.
The MOT is in the pile of superseded projects.

>
>>
>> The SSR is gonna turn off near zero current. About all I can control
>> is the slope of the voltage sinewave when I give the command.
>> To turn it on it's far easier to sense the zero crossing of the line
>> voltage than the peak.
> The mains frequency (or period) is accurate and stable enough, and
> microcontrollers or even 555 timers are cheap enough that as Phil
> mentioned, you can figure out the time of the voltage peak from the zero
> crossing.

I used a GAL20V8 as the controller. Wouldn't have been hard to delay
the control, but the relay would have waited for zero volts anyway.

that was my assumption. I'm already running the transformer way past its
rated steady state load capability.

but given the choice (or
> given a bigger transformer than your circuit breaker likes), you might
> as well make it optimum if that is just a matter of inserting a small
> delay.

Yep, I learned something. I'll take a look at that next time I do
something with a transformer.

Would be interesting to learn the trajectory of the magnetic characteristics
when you turn off the input with the output (almost) shorted.

>
>> Isn't the initial current still pretty near zero?
> Perhaps but that doesn't really matter as regards the risk of saturation.
>
>> Isn't the point on the B-H curve still about the same?
> Yes, if you only switch it off for nanoseconds. No, if you wait a few
> seconds until you have repositioned your parts for the next weld.
>
>> Case 2, you start the next pulse next week at the zero crossing
>> of the input voltage headed in the same direction.
>> What's the initial current?
>> What's the initial point on the B-H curve?
>> How is restarting it synchronously significantly different from
>> just leaving it running?
> The flux in the core is different.
>
> Put a big inductor (maybe a car ignition coil) across a 12Vrms AC supply
> and make sure you are holding the terminals a nanosecond after you
> disconnect the supply, at the instant when the AC supply is at zero
> voltage (and the inductor is carrying maximum current).
>
> Then give it to me and I will hold the terminals a week later.
>
> I think you will notice the difference. The state of the flux in the
> core matters.

I think that's a red herring. I can't argue the result, but don't think
it's relevant. Short the output and repeat the experiment.

>
> Note that in the case of a transformer, it is possible that some value
> of secondary current could result in the primary current being zero (or
> any other chosen value) at the zero-crossing of the mains voltage.

If the secondary is loaded heavily with a resistor, why wouldn't the
primary voltage
be pretty much in phase with the primary current? The leakage inductance
should be a small part of the equation?

That
> is not relevant to my point, which relates to the flux density in the
> core, which won't be affected much by the secondary current if a
> low-impedance supply is driving the primary winding.

OK, but the (almost) shorted output should have a major effect
on what happens when you stop supplying primary current and where
the core ends up on it's B-H curve.

I appreciate the principle. But, how much does it matter?
If you have an unloaded transformer and see a 40 amp input surge,
you're likely to be concerned. If your load causes a 40 amp
input current and you do what you can to minimize the surge current,
how much does it really matter? Unfortunately, I have no easy way
to measure it given my SSR characteristics.

>
>>
>> I'm not disputing the articles you posted.
>> I'm not saying anything about the general (worst) case.
>> I'm suggesting that this is how you engineer a spot welder using a MOT.
>>
>> Where did my thinking go wrong?
>>
> It may just mean that you need a larger rating for your fuse or circuit
> breaker and more expensive triac or SCRs than you could otherwise get
> away with.
>
> When I tried spot welding, I was never able to get enough current from a
> MOT-sized transformer with a few turns on the secondary. I could sort of
> weld things if I applied very light pressure so that the workpieces made
> poor contact with each other and the resistance was high enough for the
> (insufficient) current to heat them, but this wasn't really satisfactory
> because getting the force and contact resistance just right was not
> reliable.
>

That was my problem. Slight changes in contact pressure made huge
differences in the energy contributing to the weld.

Here's what I did to improve MOT welds in order of
decreasing benefit.
Control the turn on/off phase of the input voltage and control
the number of integral cycles of 60Hz. When your weld takes
six cycles, being off by one is significant.
Spring load the contacts for pressure repeatability.
Cutting a slot lengthwise down the battery tab forces
more of the current thru the weld points decreasing
sensitivity to contact resistance somewhat.

With all that, I still had trouble constructing a whole
battery pack without blowing at least one weld.
The same fixture with a CD welder gives almost 100% good
welds.

http://i.imgur.com/yd8c0rf.jpg
http://i.imgur.com/er1BqSb.jpg
http://i.imgur.com/OeQZdWH.jpg

> If I made a transformer big enough to weld thick workpieces with proper
> contact pressure, it might cause excessive drop in my mains supply,
> and/or trip breakers.

It's all about time. For battery tabs, you need very localized heating
and a very short weld time.
I have a commercial spot welder. You just push the button and wait several
seconds for the metal to turn red. Works great for angle iron,
but would kill a battery instantly.

>
> I think the best option for me is a series-parallel array of Maxwell
> boostcaps. This would eliminate the requirement for a large mains supply
> capability. The 3000F ones are rated for 1900 Amps each, so about 5 in
> parallel would probably supply about enough current for any normal sheet
> metalwork up to a couple of millimetres thick which seems to require
> close to 10kA capability. (Aluminium welding requires several times more
> current so I won't try that.) Most of the references that I have seen
> tend to suggest that the weld itself requires somewhere in the region of
> 1.5 Volts, but the electrodes etc. will have some resistive drop so I
> think at least 2 banks of boostcaps in series will be desirable. Due to
> the capacitors holding more energy than the total that you would want
> for one weld, it would be necessary to find a way to switch them off,

Just lower the initial voltage on the caps. E = C*V^2/2.
You could also disconnect some of the caps. Or maybe not fire
all the FETs...although that may be easier said than done.
I had great difficulty measuring output current because the huge
magnetic fields generated coupled into
everything.

> and it would also be very useful to be able to adjust the current by PWM
> during the weld.

See the link below

Therefore a lot of MOSFETs would be required. It seems
> that the best current rating per dollar occurs for individual MOSFETs
> rated at about 100A, so about 100 of these in parallel would be required
> for 10kA. I think a totem-pole style half-bridge topology might work,
> using the output cables as an inductor to smooth the output current. A
> multi-phase PWM arrangement with multiple output inductors could make
> better use of the current rating of the caps. It would be an interesting
> project but I don't have time to do it yet. I am somewhat concerned
> about what would happen in the event of one failed MOSFET, and I would
> like to think of a way to mitigate that. Perhaps the bondwire or package
> pin would be an adequate fuse.

Each FET has to be able to handle all the current it can deliver with
the output shorted at the weld. Limit the charge current to the caps
and a short won't cause much problem.
I wonder if you couldn't run a wire from each FET thru a common toroid
before the junction point to help equalize the peak currents??
As for failures, anything that would "blow" under short conditions would
seriously undermine your ability to deliver current to the "short"
at the weld.
>
> Chris
>




The advantage of CD is that you get a defined energy pulse.
The energy delivered to the weld point is relatively insensitive
to the resistance of the contact. Far less so than with a MOT.

I've watched a lot of youtube videos of successful welds using
racks of low voltage caps and arrays of FETs.

My Unitek uses about 400V discharged into a pulse transformer.
That V^2 factor for energy storage adds up rapidly.
The magic is in the pulse transformer. It's smaller than my
fist and claims to put 7KA into .001 ohms.
Would be interesting to learn how to build one of those.
From this:
http://i.imgur.com/ZeZerGx.png
it appears that they might pass current backwards thru the transformer
to "reset" it.

Tripped over this overview today:

http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=12&ved=0CFcQFjAL&url=http%3A%2F%2Fwww.amadamiyachi.com%2Fservlet%2Fservlet.FileDownload%3FretURL%3D%252Fapex%252Feducationalresources_fundamentals%26file%3D01530000000Jybm&ei=-lXiVL3lBMi5oQTb4YDYDQ&usg=AFQjCNFc_ush2PjqSdItUaNVEJsM3Wmh7g&sig2=ZJclqkBCdA3AOHUL_l1eew&bvm=bv.85970519,d.cGU&cad=rja

I've given up trying to weld thin aluminum.
With wire feed and Argon, I can lay a bead on THICK aluminum.
I experimented briefly with TIG. Better, but would take practice
that exceeds my attention span and my Argon budget.

Welco 52 rod with a torch makes a very strong bond
on stuff that can take the heat. There's very small margin
between a successful bond and a puddle on the floor.
I learned the hard way that most stuff that looks like aluminum
is really potmetal and has the same melting temperature as the Welco 52.

[Phil Allison]

whit3rd wrote:

> Remember Phil's
> comment that a 240V transformer on 120V excitation wouldn't saturate-
> because a 240V transformer has an oversize core for 120V excitation.

** Switching at a zero crossing creates a short term, low frequency component in the AC wave - a component approximately equal to the same AC voltage at half frequency.

Most AC supply transformers operate right on their low frequency limit - but if you only apply half voltage then that limit is now at half the original frequency - so the switch on condition is tolerated.

IOW, a given saturation condition is proportional to V / f.

I often see this in action when testing US model amplifiers meant for a 120V at 60Hz supply. A step-down device only adjusts the voltage leaving the frequency at 50Hz. The resulting magnetising current is the SAME as if one had applied 144VC to the transformer instead of 120V so Imag is way higher.

Some transformers are OK with this and others run very hot PLUS it has a significant effect on the VA rating of a suitable step down transformer.



... Phil

[Chris Jones]

Ok.

My point was just that the state of the field in the core would be
different just a nanosecond after you turn off the primary current vs a
week later in your examplesl, hence the consequences of switching it
back on would be different.

>>
>> Note that in the case of a transformer, it is possible that some value
>> of secondary current could result in the primary current being zero (or
>> any other chosen value) at the zero-crossing of the mains voltage.
> If the secondary is loaded heavily with a resistor, why wouldn't the
> primary voltage
> be pretty much in phase with the primary current? The leakage inductance
> should be a small part of the equation?

It quite possibly would, but I was just saying that the primary current
by itself isn't a good way to determine what the flux in the core is.
Anyway I am pretty sure you get it.

> That
>> is not relevant to my point, which relates to the flux density in the
>> core, which won't be affected much by the secondary current if a
>> low-impedance supply is driving the primary winding.
>
> OK, but the (almost) shorted output should have a major effect
> on what happens when you stop supplying primary current and where
> the core ends up on it's B-H curve.

I guess it might in that the SSR will turn off when the primary current
is zero which might be a different time instant, and therefore a
diferent point on the voltage waveform from what it would have been
without the secondary current. Still, some cores might go to nearly zero
flux density, regardless of when they are turned off, if they don't have
much remanence.

>
> I appreciate the principle. But, how much does it matter?
> If you have an unloaded transformer and see a 40 amp input surge,
> you're likely to be concerned. If your load causes a 40 amp
> input current and you do what you can to minimize the surge current,
> how much does it really matter? Unfortunately, I have no easy way
> to measure it given my SSR characteristics.

Ok.

To get the energy in these capacitors low enough for one weld, the
voltage would be tiny and I think it may not deliver enough current (but
would deliver it for too long) in that case. Also I would like the
possibility of current waveforms other than exponentially decaying.

> You could also disconnect some of the caps.

I was hoping to stay within the current rating of the caps which is
1900A each, so I would want several in parallel to get enough welding
current. The caps are capable of developing nearly 10kA each according
to the datasheet but that supposedly damages them and they are somewhat
expensive.

> Or maybe not fire
> all the FETs...although that may be easier said than done.

Again, I was intending to stay within the current rating of the FETs so
I would want all of them on to share the current as evenly as I can,
even if I have to do something else to reduce the current. I could
reduce the current by adding some cable resistance if there is not
already enough. I would still want to have the FETs there as a way of
switching it off at the end of the weld. Maybe PWMing it to regulate the
current would be worthwhile, but it probably means twice as many FETs.
Still I might need catch diodes if I didn't put the second lot of FETs
so maybe that is no extra problem.

> I had great difficulty measuring output current because the huge
> magnetic fields generated coupled into
> everything.

Yes, that sounds difficult. I guess you could use that as the way of
measuring the current, e.g. a Rogowski coil or whatever it is called.

>
>> and it would also be very useful to be able to adjust the current by PWM
>> during the weld.
> See the link below
>> Therefore a lot of MOSFETs would be required. It seems
>> that the best current rating per dollar occurs for individual MOSFETs
>> rated at about 100A, so about 100 of these in parallel would be required
>> for 10kA. I think a totem-pole style half-bridge topology might work,
>> using the output cables as an inductor to smooth the output current. A
>> multi-phase PWM arrangement with multiple output inductors could make
>> better use of the current rating of the caps. It would be an interesting
>> project but I don't have time to do it yet. I am somewhat concerned
>> about what would happen in the event of one failed MOSFET, and I would
>> like to think of a way to mitigate that. Perhaps the bondwire or package
>> pin would be an adequate fuse.
> Each FET has to be able to handle all the current it can deliver with
> the output shorted at the weld. Limit the charge current to the caps
> and a short won't cause much problem.

I think it could be a problem... Hehe these caps are pretty amazing, one
of them will easily melt a nail well after the charging current is
disconnected. Then another nail, and another. 3000 Farads is more like a
small capacity battery (but with less resistance than a battery).

> I wonder if you couldn't run a wire from each FET thru a common toroid
> before the junction point to help equalize the peak currents??

I was more thinking putting the wire from each FET stage through its own
toroid, and sensing the currents in each wire and regulating it with
individual PWM for each stage. That sounds like too many parts so maybe
just having separate longish wires from each FET stage would be enough
sharing resistance even with common PWM for all stages. Anyway I won't
build this for ages so I will have forgotten by then.

> As for failures, anything that would "blow" under short conditions would
> seriously undermine your ability to deliver current to the "short"
> at the weld.

Well I was not intending to exceed the ratings of the FETs, so if there
are 100 FETs, each with wires that fuse at say 200A, but normally each
conducts 100A, then if one FET fails short and starts carrying a few kA
it will be good if its wires fuse cleanly rather than keeping conducting
and starting a fire. As long as I don't exceed each FET's ratings it
would not be a problem for each FET to have a fuse action at some higher
current. I am reminded of the photos of inside a Tesla's battery pack
that seems to have a lot of cells in parallel, each cell with an
individual thin fuse wire in series with it.

>
> The advantage of CD is that you get a defined energy pulse.
> The energy delivered to the weld point is relatively insensitive
> to the resistance of the contact. Far less so than with a MOT.
>
> I've watched a lot of youtube videos of successful welds using
> racks of low voltage caps and arrays of FETs.
>
> My Unitek uses about 400V discharged into a pulse transformer.
> That V^2 factor for energy storage adds up rapidly.
> The magic is in the pulse transformer. It's smaller than my
> fist and claims to put 7KA into .001 ohms.
> Would be interesting to learn how to build one of those.
> From this:
> http://i.imgur.com/ZeZerGx.png
> it appears that they might pass current backwards thru the transformer
> to "reset" it.

Interesting, but I hope I don't need to build one, partly becuase it
restricts pulse length etc.

[mike]

On 2/17/2015 5:03 AM, Chris Jones wrote:


>> You could also disconnect some of the caps.
> I was hoping to stay within the current rating of the caps which is
> 1900A each, so I would want several in parallel to get enough welding
> current. The caps are capable of developing nearly 10kA each according
> to the datasheet but that supposedly damages them and they are somewhat
> expensive.

The inductance in the path limits the rate of rise.
The resistance in the path limits the peak current.
Depending on your construction methods and means to equalize
current sharing, you might find that removing some caps
doesn't much affect the stress on the others.
Doesn't take many of those extra milliohms in wires and
connections to put a serious dent in how much current you can deliver.

>
>> Or maybe not fire
>> all the FETs...although that may be easier said than done.
> Again, I was intending to stay within the current rating of the FETs so
> I would want all of them on to share the current as evenly as I can,
> even if I have to do something else to reduce the current. I could
> reduce the current by adding some cable resistance if there is not
> already enough. I would still want to have the FETs there as a way of
> switching it off at the end of the weld. Maybe PWMing it to regulate the
> current would be worthwhile, but it probably means twice as many FETs.
> Still I might need catch diodes if I didn't put the second lot of FETs
> so maybe that is no extra problem.

It's all about your objectives. For me, I decided that I needed to weld
battery tabs. I have stick and wire feed and Acetylene and Miller
handheld spot welders for big stuff with long weld times.
Building something more universal would be a lot of effort with little
benefit.

>
>> I had great difficulty measuring output current because the huge
>> magnetic fields generated coupled into
>> everything.
> Yes, that sounds difficult. I guess you could use that as the way of
> measuring the current, e.g. a Rogowski coil or whatever it is called.
>>
>>> and it would also be very useful to be able to adjust the current by PWM
>>> during the weld.

Closed-loop PWM sounds great in theory. But, for short weld times, you
need a wide
bandwidth control loop. But you want to filter out all those
transients. And there are lots of internet discussions about
blowing up all their FETSs and being unable to adequately protect them.
You're gonna have a lot of really big parts in a design where resistance
and inductance are your enemy. Something as simple as a foot-long
buss bar connecting the some might not behave quite like you'd expect
by looking at the schematic. Thick conductors aren't much help if
the skin effect confines the current near the surface.
Would be interesting to see a simulation
of the current distribution when you short the array of cap busses with
a weld. I don't think I have any tools that model skin effect.
Might be interesting to weave some "litz-type" wire out of smaller
conductors.

>> See the link below
>>> Therefore a lot of MOSFETs would be required. It seems
>>> that the best current rating per dollar occurs for individual MOSFETs
>>> rated at about 100A, so about 100 of these in parallel would be required
>>> for 10kA. I think a totem-pole style half-bridge topology might work,
>>> using the output cables as an inductor to smooth the output current. A
>>> multi-phase PWM arrangement with multiple output inductors could make
>>> better use of the current rating of the caps. It would be an interesting
>>> project but I don't have time to do it yet. I am somewhat concerned
>>> about what would happen in the event of one failed MOSFET, and I would
>>> like to think of a way to mitigate that. Perhaps the bondwire or package
>>> pin would be an adequate fuse.
>> Each FET has to be able to handle all the current it can deliver with
>> the output shorted at the weld. Limit the charge current to the caps
>> and a short won't cause much problem.
> I think it could be a problem... Hehe these caps are pretty amazing, one
> of them will easily melt a nail well after the charging current is
> disconnected. Then another nail, and another. 3000 Farads is more like a
> small capacity battery (but with less resistance than a battery).

If you have 100 parallel paths, you should be able to do some good with
fusible links. Just watch those milliohms. Doesn't take too many to
ruin your ability to deliver that current.

Yep. That wasn't part of my plan because I want the shortest reasonable
pulse width. You can get your weld fixture up close and personal to the
secondary of the transformer. All your drive circuitry gets easier
by the transformer ratio. One SCR is far easier to manage than 100
FETs. It's easy to get seduced into making something far more complex
than is required, "just because you can." ;-)

My thinking was that, if I have so many joules to deliver,
if I deliver them fast,
more goes into the weld and less is conducted away by the weldment.
In that scenario, you should be able to control the cap voltage
over some range and range-switch by disconnecting some caps.

Twice, I welded tabs onto a 2032 coin cell for a laptop.
Both times, the open circuit voltage dropped from 3.2V to 3.0V.
Even a short, single pulse did some damage.

I have one of those handheld spot welders that welds "nails" onto
car bodies for pulling dents. Might be interesting to examine the
characteristics of that transformer. I expect its inductance precludes
use as a short pulse transformer, but might be interesting as a replacement
for the MOT.

[Clifford Heath]

On 13/02/15 16:59, isw wrote:
> I recently came across a YouTube video where a guy rewound a microwave
> transformer to make a spot welder.
>
> In talking about it, he stated -- and repeated -- that it was very
> important for the secondary winding to have the same "sense" as the
> primary -- that is, both windings had to go around the core in the same
> direction.
>
> I know that matters if the transformer is handling very asymmetric
> waveforms such as in a flyback configuration, but I have never heard
> that it matters for plain old 60 Hz. sinusoids.
>
> Does it? Or was the guy just confused?

I think he's confused.

I have a professionally-made spot welder, designed for the days when
orthodontists needed to weld stainless-steel brackets for people's
teeth. It uses a variac and bridge to charge an electrolytic capacitor,
which is dumped via a contacter into the primary of a transformer with a
secondary having a half-dozen fat turns (think, flattened water-pipe).
The secondary has a fat braid running to the contacts.

The contacts are brought together by pressing a foot pedal, and are
spring-loaded with a screw adjuster. When the preset force is reached, a
microswitch closes and activates the contacter.

To use it, adjust the variac to set the pulse energy, and adjust the
activation force, place your work between the contacts and press the
foot pedal. You get repeatable energy and force with both hands free to
position the work.

It is fairly easy to remove an MOT secondary since they tend to use an
EI core that's not interleaved. Grind away the weld line, remove the I
section, pull off the secondary and insert your new one made from a few
turns of fat copper. Weld the I-section back on the transformer. Not
sure if you need to remove the magnetic shunts... anyone know?

The rest is easy, but you'd use an adjustable HV supply not a variac
these days.

Clifford Heath.