On Jan 16, 11:57*pm, "Wild_Bill" wrote:
I just saw the force listed in the eBay auction shown in this thread as 6
tons.
The retarded seller doesn't even clearly state what the size of the machine
is.. one might assume that it's 48" from the model number, but I didn't look
it up.. and the machines are from a metricized country, so the model number
may not be relative to the size in inches.

--
WB
.........

It's in sillimeters, a Model 2500=2.5 meters. Big enough for anything
I'm likely to dream up. Smaller the model number, the shorter the
bed. And "Tonnes" is likely metric tons, 2200 lbs, thereabouts. So
probably closer to 5000 lbs if it's factory listed at 4 tonnes.
Wonder if you could rig a magneticly operated punch off that? Would
be kind of neat to do pop rivet holes on the same unit.

Stan

Rick wrote:

More detailed manual and schematics:
http://www.aaybee.com.au/MagnaBend%20UserManual.pdf
That is an interesting de-mag ckt. :-)
...lew...

On Jan 15, 9:25*am, "Wild_Bill" wrote:
The steel in transformers isn't particularly hard.. it bends freely and cuts
quite easily with tin snips.
A quality hacksaw/bandsaw blade shouldn't have any difficulty cutting thru
the welds.

A fiberglas reinforced cutoff wheel will make easy work of cutting thru the
weld of microwave xfmers, or most any welds for common steels.

The entire weld doesn't need to be cut away to separate the E and I
sections, just cutting from the outer surface to the junction of the E-I
sections will do.

Cutting thru multiple layers of steel (where the welds aren't located) can
present problems if the layers start to flex or shift, which could cause
pinching or binding of a saw blade.

--
WB
.........

"Denis G." wrote in message

...

Nice work and good ideas! *Did you need anything special to cut the
MOTs in half? *I haven't tried it, but I thought that they were high
silicon steel metal and quite hard.

I remember when I took apart transformers as a kid, I think that they
were bolted together and the lamination slid apart quite easily. We
used to smash them with hammers on the sidewalk and they would
shatter. They probably don't make them like that anymore, so maybe
the steel is nothing special nowadays and they don't make them like
they once did.

Jon Danniken wrote:
....
On a side note, I played with the little transformer I posted pics of
earlier, using it as a lifting magnet. ...
I turned the power on, and it lifted all 30 pounds, ...

That was my big fun for the day.

Well, I just had some big fun of my own. Based on your report, I had
pretty much dismissed the idea of MOTs being converted to any kind of
serious magnet. But, what the hell, I was curious about the MOTs that I
had converted, so I did some testing.

Mine is a MOT that I had cut in half & used the secondary. The cut
surface was bandsaw flat, but not machined. The core is 2+ x 4+ & I put
a 3/4" plate across the poles. I applied 120v, full wave rectified.

I started out about where you had been: 20-30 lbs. No problem, so I
started adding more. Well, I overflowed the bucket that I was putting
weight in (my MOT collection), switched to a milk crate and filled that
with all the MOTs, plus all the lead I had. Beefing up the lifting
system as I went. I got to 207+ pounds and had to stop 'cause the milk
crate was full and near its breaking point, and the lift system was also
at its limit.

So, I'm going to bring in my engine hoist, scrounge up some more weight
and see just what this baby will do. Not because I think that it might
actually be strong enough for a magnetic brake (it won't be), but "I
gots to know" (Dirty Harry).

Bob

Bob Engelhardt wrote:
Jon Danniken wrote:
...
On a side note, I played with the little transformer I posted pics of
earlier, using it as a lifting magnet. ...
I turned the power on, and it lifted all 30 pounds, ...

That was my big fun for the day.

Well, I just had some big fun of my own. Based on your report, I had
pretty much dismissed the idea of MOTs being converted to any kind of
serious magnet. But, what the hell, I was curious about the MOTs that I
had converted, so I did some testing.

Mine is a MOT that I had cut in half & used the secondary. The cut
surface was bandsaw flat, but not machined. The core is 2+ x 4+ & I put
a 3/4" plate across the poles. I applied 120v, full wave rectified.

I started out about where you had been: 20-30 lbs. No problem, so I
started adding more. Well, I overflowed the bucket that I was putting
weight in (my MOT collection), switched to a milk crate and filled that
with all the MOTs, plus all the lead I had. Beefing up the lifting
system as I went. I got to 207+ pounds and had to stop 'cause the milk
crate was full and near its breaking point, and the lift system was also
at its limit.

So, I'm going to bring in my engine hoist, scrounge up some more weight
and see just what this baby will do. Not because I think that it might
actually be strong enough for a magnetic brake (it won't be), but "I
gots to know" (Dirty Harry).

Bob

COOL!

Thanks Bob. I anxiously await the next chapter.

--Winston

"Bob Engelhardt" wrote in message
...

Well, I just had some big fun of my own. Based on your report, I had
pretty much dismissed the idea of MOTs being converted to any kind of
serious magnet. But, what the hell, I was curious about the MOTs that I
had converted, so I did some testing.

Mine is a MOT that I had cut in half & used the secondary. The cut
surface was bandsaw flat, but not machined. The core is 2+ x 4+ & I put
a 3/4" plate across the poles. I applied 120v, full wave rectified.

I started out about where you had been: 20-30 lbs. No problem, so I
started adding more. Well, I overflowed the bucket that I was putting
weight in (my MOT collection), switched to a milk crate and filled that
with all the MOTs, plus all the lead I had. Beefing up the lifting
system as I went. I got to 207+ pounds and had to stop 'cause the milk
crate was full and near its breaking point, and the lift system was also
at its limit.

So, I'm going to bring in my engine hoist, scrounge up some more weight
and see just what this baby will do. Not because I think that it might
actually be strong enough for a magnetic brake (it won't be), but "I
gots to know" (Dirty Harry).

Bob

What AC current are you seeing when you do this? Is the winding getting
hot? How ling did you leave it powered on? Since you are putting DC
thru the MOT, it isn't acting as an inductor and the secondary resistance is
all that is limiting the current.
If there isn't any significant heating you could try putting 240VAC into the
rectifiers for increased mag flux and holding power. Check for heating
again.
Art

Bob Engelhardt wrote:

[snip] Mine is a MOT that I had cut in half & used the secondary. The cut
surface was bandsaw flat, but not machined. The core is 2+ x 4+ & I
put a 3/4" plate across the poles. I applied 120v, full wave
rectified.

Holy mackerel, 2x4 cross section on a core is a big sucker, sounds like it
must have been out of a commercial oven?

I started out about where you had been: 20-30 lbs. No problem, so I
started adding more. Well, I overflowed the bucket that I was putting
weight in (my MOT collection), switched to a milk crate and filled
that with all the MOTs, plus all the lead I had. Beefing up the
lifting system as I went. I got to 207+ pounds and had to stop
'cause the milk crate was full and near its breaking point, and the
lift system was also at its limit.

I'm impressed, and anxious to see what you can get up to. I'm also
encouraged that you were able to do this with the secondary winding.

Jon

This continues to amaze me. I had tested with 207 lbs, then set up a
better test rig and added 85 lbs more (292 total) - no problem, 52 more,
to 345 - no problem, then 393 (48 more) - still holding!! But at 425,
my engine hoist bent! It was homemade, by the that-looks-about-right
method, so perhaps to be expected.

Back to the drawing board, as they say. Then inspiration: use 2 equal
pole pieces & just load one. I.e., I would be measuring 1/2 the pull.
So I got another 3/4" bar and used each of the 2 bars to cover 1/2 of
the pole area (the long way) and fastened the load to one of them.

I backed off the load, some, to 369 and it _just_ lifted it before
letting go. I had _finally_ found the limit. That means that this
little MOT electromagnet is capable of holding 740 lbs! WHOA! Double WHOA!

The coil is 3-3/4", i.e., in a magnetic brake it would take 3-3/4" of
clamp space. And the magnetic pull rate would be 740 divided by that,
or about 200 lbs/in. Compared to the 250 lbs/in of the MagnaBend. Now,
if it isn't saturating at 120v & could be run higher, with more flux,
and more pull, we could be in the DIY magnetic brake business.

Some tidbits: it's not easy finding a 400 lb load - I had 2 milk crates
FULL of MOTs, lead ingots, exercise weights, chunky steel stock, and a
lathe tailstock. It's also not easy lifting that much - I was using a
come-along & doing one click at a time.

At the 345 level, I tried something. I backed off the coil voltage (I
was using a variac) until it released the load. It was at the 30/100
mark. 36 volts if I hadn't wired in the boost section of the variac. I
don't know if this means that the core runs saturated at 120v, or what.
But I intend to find out. It would be much better if it didn't have
to run at 120v.

An aw-**** moment: when the magnet did let go, a piece of lift chain
smashed into the coil & cut through a couple of turns. It still works,
but probably not for long. But I have 2 other MOT electromagnets, and
8-10 untouched MOTs.

Stay tuned,
Bob

Artemus wrote:
What AC current are you seeing when you do this?

I was thinking it was an amp, but I just looked at FunWithMOTs & the 1
amp is a secondary in air (no core) with ac on it. I'll have to measure it.

Is the winding getting hot?

Warm, but not too hot to touch.

How ling did you leave it powered on?

A minute or two each trial. Duty cycle would definitely be less than
100%. (The MagnaBend has a duty cycle of 30%.)

Since you are putting DC thru the MOT, it isn't acting as an inductor and
the secondary resistance is all that is limiting the current.

Well, there is definitely a lot of ripple, so that portion of the
current would be inductor affected.

If there isn't any significant heating you could try putting 240VAC into the
rectifiers for increased mag flux and holding power. Check for heating

Yes. First, I intend to determine the saturation voltage and pull at
that point. Then heating, by change-of-resistance method.

Bob

Bob Engelhardt wrote:

(Very intriguing MOT as electromagnet stuff)

Stay tuned,

If you are not at saturation yet, you could
stack a couple secondaries and run them in
series or parallel for double flux.

--Winston

On 2011-01-17, Bob Engelhardt wrote:
Rick wrote:
More detailed manual and schematics:
http://www.aaybee.com.au/MagnaBend%20UserManual.pdf

Good stuff! It makes me want one even more! Lots of data one could use
for building one, too.

The most significant spec is the 3 tons ("tonnes") of clamping force on
the 24" brake. That's 250 lbs/running inch. I'm skeptical that a
reworked MOT is going to come anywhere near that.

I suspect that the magnet is a continuous stack of E-cores the
full active width of the brake -- all in a single coil. If you had a
bunch of MOT cores, you would have areas where the grip was a lot weaker
than you need.

Enjoy,
DoN.

--
Remove oil spill source from e-mail
Email: | Voice (all times): (703) 938-4564
(too) near Washington D.C. | http://www.d-and-d.com/dnichols/DoN.html
--- Black Holes are where God is dividing by zero ---

DoN. Nichols wrote:
I suspect that the magnet is a continuous stack of E-cores the
full active width of the brake -- all in a single coil. If you had a
bunch of MOT cores, you would have areas where the grip was a lot weaker
than you need.

It's true that the field would be uneven. My test magnet, for instance,
has coil width of 3.75" & a core width of 2.25". 40% of the length
would be field-less, more or less. However, if the average field/grip
is the same, the 4 x 5/8" clamping bar should be beefy enough to provide
grip in the field-less areas. In other words, if the total clamping
force is the same, the clamping bar should spread it out evenly (enough).

Bob

On Mon, 17 Jan 2011 13:22:29 -0500, Bob Engelhardt
wrote:

Jon Danniken wrote:
...
On a side note, I played with the little transformer I posted pics of
earlier, using it as a lifting magnet. ...
I turned the power on, and it lifted all 30 pounds, ...

That was my big fun for the day.

Well, I just had some big fun of my own. Based on your report, I had
pretty much dismissed the idea of MOTs being converted to any kind of
serious magnet. But, what the hell, I was curious about the MOTs that I
had converted, so I did some testing.

Mine is a MOT that I had cut in half & used the secondary. The cut
surface was bandsaw flat, but not machined. The core is 2+ x 4+ & I put
a 3/4" plate across the poles. I applied 120v, full wave rectified.

I started out about where you had been: 20-30 lbs. No problem, so I
started adding more. Well, I overflowed the bucket that I was putting
weight in (my MOT collection), switched to a milk crate and filled that
with all the MOTs, plus all the lead I had. Beefing up the lifting
system as I went. I got to 207+ pounds and had to stop 'cause the milk
crate was full and near its breaking point, and the lift system was also
at its limit.

So, I'm going to bring in my engine hoist, scrounge up some more weight
and see just what this baby will do. Not because I think that it might
actually be strong enough for a magnetic brake (it won't be), but "I
gots to know" (Dirty Harry).

Bob

It would also be interesting to see how much force you get when a
non-ferromagnetic gap of 1/16" or so (e.g., .062 aluminum) is
introduced.

If the faces are shaped they will focus the forces.

I used to use a research magnet that had cone shaped with small flats -
an when we turned it on, all of our pocket things, watches, rings, etc
were in another room. A ring would get red hot near the poles.

A company I worked with had one that was the diameter of 48" and
had a focus cone down to 8" diameter. It was to energize wafers :-)

Martin

On 1/17/2011 9:41 PM, Bob Engelhardt wrote:
DoN. Nichols wrote:
I suspect that the magnet is a continuous stack of E-cores the
full active width of the brake -- all in a single coil. If you had a
bunch of MOT cores, you would have areas where the grip was a lot weaker
than you need.

It's true that the field would be uneven. My test magnet, for instance,
has coil width of 3.75" & a core width of 2.25". 40% of the length would
be field-less, more or less. However, if the average field/grip is the
same, the 4 x 5/8" clamping bar should be beefy enough to provide grip
in the field-less areas. In other words, if the total clamping force is
the same, the clamping bar should spread it out evenly (enough).

Bob

Testing for saturation: I added a small secondary, so I could get an
induced voltage to measure. Then I measured this induced voltage vs.
the applied voltage in 10v steps up to 120v, applied. My thinking is
that there will be a linear relationship until the core saturates and
then the induced voltage will level off.

That didn't happen so far. Before I rig to test up to 240v, I want to
confirm that my thinking is correct about output voltage at saturation.
Anybody know?

Thanks,
Bob

BTW - in the microwave oven, this coil generates 2000v, so 240v won't be
a problem (current might be).

Bob Engelhardt wrote:
Testing for saturation: I added a small secondary, so I could get an
induced voltage to measure. Then I measured this induced voltage vs. the
applied voltage in 10v steps up to 120v, applied. My thinking is that
there will be a linear relationship until the core saturates and then
the induced voltage will level off.

That didn't happen so far. Before I rig to test up to 240v, I want to
confirm that my thinking is correct about output voltage at saturation.
Anybody know?

Thanks,
Bob

BTW - in the microwave oven, this coil generates 2000v, so 240v won't be
a problem (current might be).

The inductance of that coil fell significantly when you lopped
off the 'I' core, yes?

Can you place a small value sense resistor in series with the
'common' line to your coil? The way I misunderstand magnetic
core saturation is that we expect a 'knee' where current begins
to increase in a nonlinear fashion in relation to applied
voltage. You could set your 'scope up as an X-Y display to
show this nonlinearity very clearly.

--Winston

On Tue, 18 Jan 2011 21:47:49 -0500, Bob Engelhardt
wrote:

Testing for saturation: I added a small secondary, so I could get an
induced voltage to measure. Then I measured this induced voltage vs.
the applied voltage in 10v steps up to 120v, applied. My thinking is
that there will be a linear relationship until the core saturates and
then the induced voltage will level off.

That didn't happen so far. Before I rig to test up to 240v, I want to
confirm that my thinking is correct about output voltage at saturation.
Anybody know?

Thanks,
Bob

BTW - in the microwave oven, this coil generates 2000v, so 240v won't be
a problem (current might be).

I thought you had a bridge rectifier in there. Are you exciting it
with AC or with rectified AC which is DC with ripple?

On Jan 18, 9:47*pm, Bob Engelhardt wrote:
Testing for saturation: I added a small secondary, so I could get an
induced voltage to measure. *Then I measured this induced voltage vs.
the applied voltage in 10v steps up to 120v, applied. *My thinking is
that there will be a linear relationship until the core saturates and
then the induced voltage will level off.

That didn't happen so far. *Before I rig to test up to 240v, I want to
confirm that my thinking is correct about output voltage at saturation.
* Anybody know?

Thanks,
Bob

BTW - in the microwave oven, this coil generates 2000v, so 240v won't be
a problem (current might be).

If I recall correctly when one designs a transformer, you calculate
what the ET product is. That is the voltage times time.

E{avg}= 4 f N A B{peak} is the basic formula

So rearranging and eliminating constants.

ET = N A B ( peak )

Where T is the inverse of f. And N, the number of turns, and A the
cross sectional area are constant for a transformer.

So ET ~ to B ( peak )

So the flux ( B ) depends on the voltage and frequency. And using the
same line frequency and the voltage at 240 volts RMS ( much less than
the usual 2000 volts on the secondary, you should not have to worry
about saturation. ( this assumes you are using AC voltage.).

Dan

Winston wrote:
The inductance of that coil fell significantly when you lopped
off the 'I' core, yes?

Yes and no, I suppose G. 'Though I did lop it off, I replaced it with
a shorting bar, so it shouldn't be drastically different.

Can you place a small value sense resistor in series with the
'common' line to your coil? The way I misunderstand magnetic
core saturation is that we expect a 'knee' where current begins
to increase in a nonlinear fashion in relation to applied
voltage. You could set your 'scope up as an X-Y display to
show this nonlinearity very clearly.

Funny you should mention that. I was going to do that and had dragged
my scope out, only to find it busted! Dang - those things are expensive
to fix!

Bob

Don Foreman wrote:
I thought you had a bridge rectifier in there. Are you exciting it
with AC or with rectified AC which is DC with ripple?

Oops - you're right, I was going to switch to AC for the test & kinda'
skipped that part. Still, there's enough ripple to give me output & the
DC component helps with getting to saturation.

Bob

On Tue, 18 Jan 2011 21:47:49 -0500, Bob Engelhardt
wrote:

Testing for saturation: I added a small secondary, so I could get an
induced voltage to measure. Then I measured this induced voltage vs.
the applied voltage in 10v steps up to 120v, applied. My thinking is
that there will be a linear relationship until the core saturates and
then the induced voltage will level off.

That didn't happen so far. Before I rig to test up to 240v, I want to
confirm that my thinking is correct about output voltage at saturation.
Anybody know?

Thanks,
Bob

BTW - in the microwave oven, this coil generates 2000v, so 240v won't be
a problem (current might be).


Basically correct but with most ferrous materials onset of
saturation is gradual and not sharply defined.

Jim

wrote:
If I recall correctly when one designs a transformer, you calculate
what the ET product is. ...
snip
So the flux ( B ) depends on the voltage and frequency. And using the
same line frequency and the voltage at 240 volts RMS ( much less than
the usual 2000 volts on the secondary, you should not have to worry
about saturation. ( this assumes you are using AC voltage.).

Thanks, that helps!

I've been thinking about those 2000 volts. If that's the normal voltage
on this coil, it's probably going to take close to that to saturate the
core. I mean, wouldn't they design for the core to be close to
saturation, to minimize the core size needed?

My interest in saturation is 2 fold: I don't want to run beyond
saturation 'cause of the extra heat, and at saturation is where the
maximum pull will be.

Now, about ACC. To use as an electromagnet, I have full-wave
rectification, unfiltered. Although the coil's inductance will do some
smoothing (I wish my scope was working). My intuition is that the DC
component of the current will be determined by the coil resistance and
that will produce flux proportional to the number of coil turns. So,
the question is whether the DC current will saturate it before 240v. Or
be too high for the coil's wire guage.

But the bottom line is that I'm going to be using 240v max (full wave)
and as long as it isn't saturated then, I'll be getting the maximum pull
available.

Thanks,
Bob

Bob Engelhardt wrote:
Winston wrote:
The inductance of that coil fell significantly when you lopped
off the 'I' core, yes?

Yes and no, I suppose G. 'Though I did lop it off, I replaced it with
a shorting bar, so it shouldn't be drastically different.

Now's the time to break out your $0.99 inductance bridge.

I haven't tested this but it looks reasonable:

http://www.aronnelson.com/gallery/main.php/v/BinOfBrett/Inductance_meter.jpg.html

http://www.aronnelson.com/gallery/main.php/v/BinOfBrett/Inductance_method.jpg.html

Can you place a small value sense resistor in series with the
'common' line to your coil? The way I misunderstand magnetic
core saturation is that we expect a 'knee' where current begins
to increase in a nonlinear fashion in relation to applied
voltage. You could set your 'scope up as an X-Y display to
show this nonlinearity very clearly.

Funny you should mention that. I was going to do that and had dragged my
scope out, only to find it busted! Dang - those things are expensive to
fix!

Not if it is old enough.

Got schematics?

--Winston -- Borrow an 'X' to repair an 'X',
where 'X' = anything.

On Wed, 19 Jan 2011 09:39:14 -0500, Bob Engelhardt
wrote:

wrote:
If I recall correctly when one designs a transformer, you calculate
what the ET product is. ...
snip
So the flux ( B ) depends on the voltage and frequency. And using the
same line frequency and the voltage at 240 volts RMS ( much less than
the usual 2000 volts on the secondary, you should not have to worry
about saturation. ( this assumes you are using AC voltage.).

Thanks, that helps!

I've been thinking about those 2000 volts. If that's the normal voltage
on this coil, it's probably going to take close to that to saturate the
core. I mean, wouldn't they design for the core to be close to
saturation, to minimize the core size needed?

My interest in saturation is 2 fold: I don't want to run beyond
saturation 'cause of the extra heat, and at saturation is where the
maximum pull will be.

Now, about ACC. To use as an electromagnet, I have full-wave
rectification, unfiltered. Although the coil's inductance will do some
smoothing (I wish my scope was working). My intuition is that the DC
component of the current will be determined by the coil resistance and
that will produce flux proportional to the number of coil turns. So,
the question is whether the DC current will saturate it before 240v. Or
be too high for the coil's wire guage.

But the bottom line is that I'm going to be using 240v max (full wave)
and as long as it isn't saturated then, I'll be getting the maximum pull
available.

Thanks,
Bob

You're right, DC current will be determined by coil resistance and
applied voltage, and flux will be proportional to current and # of
turns.

Focus on keeping the current to a level that won't burn out the
winding. Don't worry about saturation in a DC electromagnet. Having
it saturate is not a problem. Saturation is a big problem in a
transformer, but not in an electromagnet.

It's a bit surprising that, for given core cross section and winding
window area, the max flux possible with a copper winding is
independent of wire size. Voltage and current change, of course, but
Bmax doesn't for given current density in the copper.

The attraction force you get will be B^2*A / 2*muzero where B is flux
density, A is area, muzero is 4*pi*10E-7 newton/amp^2. Flux density
will depend on the total magnetic circuit including whatever is being
attracted.

Bob Engelhardt wrote:

Winston wrote:
The inductance of that coil fell significantly when you lopped
off the 'I' core, yes?

Yes and no, I suppose G. 'Though I did lop it off, I replaced it with
a shorting bar, so it shouldn't be drastically different.

Can you place a small value sense resistor in series with the
'common' line to your coil? The way I misunderstand magnetic
core saturation is that we expect a 'knee' where current begins
to increase in a nonlinear fashion in relation to applied
voltage. You could set your 'scope up as an X-Y display to
show this nonlinearity very clearly.

Funny you should mention that. I was going to do that and had dragged
my scope out, only to find it busted! Dang - those things are expensive
to fix!


What kind of scope? Brand? Model?


--
You can't fix stupid. You can't even put a band-aid on it, because it's
Teflon coated.

Bob Engelhardt wrote:

wrote:
If I recall correctly when one designs a transformer, you calculate
what the ET product is. ...
snip
So the flux ( B ) depends on the voltage and frequency. And using the
same line frequency and the voltage at 240 volts RMS ( much less than
the usual 2000 volts on the secondary, you should not have to worry
about saturation. ( this assumes you are using AC voltage.).

Thanks, that helps!

I've been thinking about those 2000 volts. If that's the normal voltage
on this coil, it's probably going to take close to that to saturate the
core. I mean, wouldn't they design for the core to be close to
saturation, to minimize the core size needed?

My interest in saturation is 2 fold: I don't want to run beyond
saturation 'cause of the extra heat, and at saturation is where the
maximum pull will be.

Now, about ACC. To use as an electromagnet, I have full-wave
rectification, unfiltered. Although the coil's inductance will do some
smoothing (I wish my scope was working). My intuition is that the DC
component of the current will be determined by the coil resistance and
that will produce flux proportional to the number of coil turns. So,
the question is whether the DC current will saturate it before 240v. Or
be too high for the coil's wire guage.

But the bottom line is that I'm going to be using 240v max (full wave)
and as long as it isn't saturated then, I'll be getting the maximum pull
available.


Are you removing the secondary windings?


--
You can't fix stupid. You can't even put a band-aid on it, because it's
Teflon coated.

On Jan 19, 11:39*am, Don Foreman
wrote:



You're right, DC current will be determined by coil resistance and
applied voltage, and flux will be proportional to current and # of
turns. *


Minor correction. The flux will be proportional to the current and #
of turns up to where the core is saturated. Or at least that is how I
remember it.

Dan

Winston wrote:
Bob Engelhardt wrote:
... dragged my
scope out, only to find it busted! Dang - those things are expensive to
fix!

...
Got schematics?

No. Not the skill, either. The idea of a 10,000v power supply is
another deterrent G.

Bob

Michael A. Terrell wrote:
What kind of scope? Brand? Model?

Analog - Tektronix 465B

Bob

On Wed, 19 Jan 2011 15:06:08 -0500, "Michael A. Terrell"
wrote:


Bob Engelhardt wrote:

wrote:
If I recall correctly when one designs a transformer, you calculate
what the ET product is. ...
snip
So the flux ( B ) depends on the voltage and frequency. And using the
same line frequency and the voltage at 240 volts RMS ( much less than
the usual 2000 volts on the secondary, you should not have to worry
about saturation. ( this assumes you are using AC voltage.).

Thanks, that helps!

I've been thinking about those 2000 volts. If that's the normal voltage
on this coil, it's probably going to take close to that to saturate the
core. I mean, wouldn't they design for the core to be close to
saturation, to minimize the core size needed?



Only at line frequency.

With DC exitation the same flux density will be reached when the
DC reaches the same level as the normal line freqency magnetising
current (i.e the no load current). With DC, the voltage needed to
reach this current will typically be about 5% of the rated line
frequency voltage provided the magnetic circuit is closed (no air
gap)



My interest in saturation is 2 fold: I don't want to run beyond
saturation 'cause of the extra heat, and at saturation is where the
maximum pull will be.

Now, about ACC. To use as an electromagnet, I have full-wave
rectification, unfiltered. Although the coil's inductance will do some
smoothing (I wish my scope was working). My intuition is that the DC
component of the current will be determined by the coil resistance and
that will produce flux proportional to the number of coil turns. So,
the question is whether the DC current will saturate it before 240v. Or
be too high for the coil's wire guage.

Because the rated line frequency full load current is very much
larger than the no load magnetising current, DC excitation will
saturate the core long before rated full load current is reached.

This applies when the magnetic circuit is as fully closed as in
the original transformer. If it is only partially closed with
perhaps inferior iron or small airgap, proportionally
more current will be needed.

But the bottom line is that I'm going to be using 240v max (full wave)
and as long as it isn't saturated then, I'll be getting the maximum pull
available.

240V should be ample.

Jim

Don Foreman wrote:
...
Focus on keeping the current to a level that won't burn out the
winding. Don't worry about saturation in a DC electromagnet. Having
it saturate is not a problem. ...

I have kinda' backed into that conclusion - given that it's not
saturated when it's outputting 2000v, and seeing the heating that I'm
getting so far (120v).

It's a bit surprising that, for given core cross section and winding
window area, the max flux possible with a copper winding is
independent of wire size. Voltage and current change, of course, but
Bmax doesn't for given current density in the copper.

It makes sense from the fact that flux is a function of amp-turns. 1
amp through 100 turns or 100 amps through 1 turn. In my MOT
electromagnet it's a lot easier to put higher voltage & less current
through 2000 or so turns of secondary than small voltage & large current
through 120 turns of primary.

The attraction force you get will be B^2*A / 2*muzero where B is flux
density, A is area, muzero is 4*pi*10E-7 newton/amp^2. Flux density
will depend on the total magnetic circuit including whatever is being
attracted.

I suppose before I had started my experiments I could have used that to
calculate force. I would still have verified it, but it would have
given me a better idea of where to start.

Thanks,
Bob

Michael A. Terrell wrote:
Are you removing the secondary windings?

Actually, I'm removing the primary and exciting the secondary.

Bob

Today's results:

I abandoned my search for saturation. I will not be applying the kind
of voltage to the coil that will get it anywhere near saturation. Also,
with the kind of force that I've measured already, I don't need to saturate.

What the limiting factor will be is heat. Exciting the MOT's secondary
causes a lot more current that when it's used in the microwave. I
gathered some data on heating: with 160v AC applied to the bridge, it
drew 1.8A initially and dropped to 1.2A after running for 4 minutes and
heating up. The average coil temperature had risen 210F, to 280F!
(There was some bubbling sound from the coil potting material!) This is
24ga wire+-.

I let it cool for 13 minutes (to give a 30% duty cycle) and ran through
a couple of iterations of 3 minutes on, 7 minutes off. The peak
temperature kept climbing, reaching a max of 240F rise to 310F.

At this point I decided that this test is not realistic. The 30% duty
cycle would likely give the same average temperature as an operating
brake, but the long on period was giving exaggerated max temperatures.
In use, the full on time would only be seconds and 1/2 power on a few
seconds more. I am going to build a rig that will do 30% duty cycle
with short on times (10sec maybe). And maybe do 10% duty cycle also.

Let me add that the temperatures that I was seeing were not alarming me.
I had looked into transformer temperature ratings and found some
interesting stuff. First, their ratings are not so much limits above
which they shouldn't operated, rather they are values which can be
expected to not be exceeded when the transformer is operating at
capacity (80, 115, & 150 degree C rise). It's more a rating of
transformer efficiency (hot is less efficient). And, independently, the
real limiting parameter is the winding insulation temperature limit.
Typically 220C (428F). I was not anywhere near this insulation limit. See
http://www.copper.org/applications/e...fficiency.html

Damn, I'm having more fun than a boy in a sandbox,
Bob

wrote:
....
With DC exitation the same flux density will be reached when the
DC reaches the same level as the normal line freqency magnetising
current (i.e the no load current). With DC, the voltage needed to
reach this current will typically be about 5% of the rated line
frequency voltage provided the magnetic circuit is closed (no air
gap)

So my 2000v coil would saturate the core with 100v DC applied. Yet I am
applying much more than that, without saturation, I think. 'Must be the
ripple in my voltage.

Because the rated line frequency full load current is very much
larger than the no load magnetising current, DC excitation will
saturate the core long before rated full load current is reached.

This applies when the magnetic circuit is as fully closed as in
the original transformer. If it is only partially closed with
perhaps inferior iron or small airgap, proportionally
more current will be needed.

I'm sure that there is SOME air gap - the shorting bar is just sitting
there (under considerable magnetic force).

....
240V should be ample.

It's looking like heat will limit it to less than that.

Thanks for the good stuff.

Bob

Bob Engelhardt wrote:

Michael A. Terrell wrote:
What kind of scope? Brand? Model?

Analog - Tektronix 465B


Damn! I had a lot of spare parts, but all I have left is a power
transformer.


--
You can't fix stupid. You can't even put a band-aid on it, because it's
Teflon coated.

wrote:

On Wed, 19 Jan 2011 15:06:08 -0500, "Michael A. Terrell"
wrote:


Bob Engelhardt wrote:

wrote:
If I recall correctly when one designs a transformer, you calculate
what the ET product is. ...
snip
So the flux ( B ) depends on the voltage and frequency. And using the
same line frequency and the voltage at 240 volts RMS ( much less than
the usual 2000 volts on the secondary, you should not have to worry
about saturation. ( this assumes you are using AC voltage.).

Thanks, that helps!

I've been thinking about those 2000 volts. If that's the normal voltage
on this coil, it's probably going to take close to that to saturate the
core. I mean, wouldn't they design for the core to be close to
saturation, to minimize the core size needed?


Only at line frequency.

With DC exitation the same flux density will be reached when the
DC reaches the same level as the normal line freqency magnetising
current (i.e the no load current). With DC, the voltage needed to
reach this current will typically be about 5% of the rated line
frequency voltage provided the magnetic circuit is closed (no air
gap)

My interest in saturation is 2 fold: I don't want to run beyond
saturation 'cause of the extra heat, and at saturation is where the
maximum pull will be.

Now, about ACC. To use as an electromagnet, I have full-wave
rectification, unfiltered. Although the coil's inductance will do some
smoothing (I wish my scope was working). My intuition is that the DC
component of the current will be determined by the coil resistance and
that will produce flux proportional to the number of coil turns. So,
the question is whether the DC current will saturate it before 240v. Or
be too high for the coil's wire guage.

Because the rated line frequency full load current is very much
larger than the no load magnetising current, DC excitation will
saturate the core long before rated full load current is reached.

This applies when the magnetic circuit is as fully closed as in
the original transformer. If it is only partially closed with
perhaps inferior iron or small airgap, proportionally
more current will be needed.

But the bottom line is that I'm going to be using 240v max (full wave)
and as long as it isn't saturated then, I'll be getting the maximum pull
available.

240V should be ample.

Jim


None of the text you quote was written by me.


--
You can't fix stupid. You can't even put a band-aid on it, because it's
Teflon coated.

On Wed, 19 Jan 2011 19:41:33 -0500, Bob Engelhardt
wrote:

Today's results:

I abandoned my search for saturation. I will not be applying the kind
of voltage to the coil that will get it anywhere near saturation. Also,
with the kind of force that I've measured already, I don't need to saturate.

What the limiting factor will be is heat. Exciting the MOT's secondary
causes a lot more current that when it's used in the microwave. I
gathered some data on heating: with 160v AC applied to the bridge, it
drew 1.8A initially and dropped to 1.2A after running for 4 minutes and
heating up. The average coil temperature had risen 210F, to 280F!
(There was some bubbling sound from the coil potting material!) This is
24ga wire+-.

I let it cool for 13 minutes (to give a 30% duty cycle) and ran through
a couple of iterations of 3 minutes on, 7 minutes off. The peak
temperature kept climbing, reaching a max of 240F rise to 310F.

At this point I decided that this test is not realistic. The 30% duty
cycle would likely give the same average temperature as an operating
brake, but the long on period was giving exaggerated max temperatures.
In use, the full on time would only be seconds and 1/2 power on a few
seconds more. I am going to build a rig that will do 30% duty cycle
with short on times (10sec maybe). And maybe do 10% duty cycle also.

Let me add that the temperatures that I was seeing were not alarming me.
I had looked into transformer temperature ratings and found some
interesting stuff. First, their ratings are not so much limits above
which they shouldn't operated, rather they are values which can be
expected to not be exceeded when the transformer is operating at
capacity (80, 115, & 150 degree C rise). It's more a rating of
transformer efficiency (hot is less efficient). And, independently, the
real limiting parameter is the winding insulation temperature limit.
Typically 220C (428F). I was not anywhere near this insulation limit. See
http://www.copper.org/applications/e...fficiency.html

Damn, I'm having more fun than a boy in a sandbox,

Yeah, and a whole lot fewer cats hang around the shop than a litter
box.

We're having fun with you (albeit vicariously) too, little Bobby.
Carry on.

--
Live in the sunshine, swim the sea, drink the wild air...
-- Ralph Waldo Emerson

On Wed, 19 Jan 2011 13:54:15 -0800 (PST), "
wrote:

On Jan 19, 11:39*am, Don Foreman
wrote:



You're right, DC current will be determined by coil resistance and
applied voltage, and flux will be proportional to current and # of
turns. *


Minor correction. The flux will be proportional to the current and #
of turns up to where the core is saturated. Or at least that is how I
remember it.

Dan

Right. Thanks for that correction. It isn't really proportional in
the nonlinear region below saturation either, but it works as a first
approximation.

Winston wrote:
Bob Engelhardt wrote:
Winston wrote:
The inductance of that coil fell significantly when you lopped
off the 'I' core, yes?

Yes and no, I suppose G. 'Though I did lop it off, I replaced it
with a shorting bar, so it shouldn't be drastically different.

Now's the time to break out your $0.99 inductance bridge.

I haven't tested this but it looks reasonable:

http://www.aronnelson.com/gallery/main.php/v/BinOfBrett/Inductance_meter.jpg.html

http://www.aronnelson.com/gallery/main.php/v/BinOfBrett/Inductance_method.jpg.html


You can also measure inductances of this size using your PC and tone
generating software. Wire up the inductor with a capacitor, apply the tone,
and look for the resonant frequency.

I did this a few years ago with a MOT, testing the difference in inductance
by using different sized paper shims between the "E" and the "I"
laminations.

There's a webpage describing this somewhere out there, but of course I can't
find it right now. I think I made the mistake of saving it instead of
adding it to my bookmarks.

Jon

Jon Danniken wrote:

Winston wrote:
Bob Engelhardt wrote:
Winston wrote:
The inductance of that coil fell significantly when you lopped
off the 'I' core, yes?

Yes and no, I suppose G. 'Though I did lop it off, I replaced it
with a shorting bar, so it shouldn't be drastically different.

Now's the time to break out your $0.99 inductance bridge.

I haven't tested this but it looks reasonable:

http://www.aronnelson.com/gallery/main.php/v/BinOfBrett/Inductance_meter.jpg.html

http://www.aronnelson.com/gallery/main.php/v/BinOfBrett/Inductance_method.jpg.html

You can also measure inductances of this size using your PC and tone
generating software. Wire up the inductor with a capacitor, apply the tone,
and look for the resonant frequency.

I did this a few years ago with a MOT, testing the difference in inductance
by using different sized paper shims between the "E" and the "I"
laminations.

There's a webpage describing this somewhere out there, but of course I can't
find it right now. I think I made the mistake of saving it instead of
adding it to my bookmarks.


It's simple: Connect a known value capacitor across the coil. Put a
1K ohm resistor in series with the generator to the hot side and connect
the ground to the other end of the coil. Then use a scope or AC
voltmeter to look for resonance across the L/C pair.


--
You can't fix stupid. You can't even put a band-aid on it, because it's
Teflon coated.