In article ,
Bob Engelhardt wrote:
Bob Engelhardt wrote:
I just made a 2-MOTE (MOT electromagnet) brake ...
I checked the phasing of the MOTE's & they were indeed out-of-phase,
meaning that there was some flux shorting through the brake's structure.
I put them in phase and tried bending some test pieces. There was a
distinct improvement: it will now bend 20 ga (.040 +-), where it would
only do .030 before.
That's good news and bad news. The good news is obvious, the bad news
is now I'm uncertain what to do. Before, I had ruled out using MOTE's -
now, I don't know. MOTE's would be easier and faster than winding my
own electromagnet in the style of the Magnabend, but just barely strong
enough.
The other thing I did was look into doubling up the windings by adding a
winding from another MOT. The problem is that other windings either
don't fit at all, or just fit. The just-fit ones are really too tight
to use without damaging them. This is 30 ga wire.
The MagnaBend 650E (24" wide) draws 4 amps at 220/240 volts, call it 230
volts AC, which is 4*230= 920 watts. One can simplify the winding task
by using a stepdown transformer, if one can come by such a big
transformer cheaply enough.
I suppose that a MOT or a pair of MOTs would do, a task similar to
making a spot welder from MOTs.
The general approach would be to remove the MOT secondary winding, and
replace it with a new secondary having a reasonable number turns and
feeding a fullwave rectifier which in turn feeds a big filter capacitor
and the MagnaBend flux coil. The main limit on how few turns the
secondary can be is that the voltage has to be high enough the one does
not have excessive losses due to the forward voltage drop in the
rectifiers, at least 12 volts or so.
So there - we found a way to use MOTs.
We also need to find out the needed number of ampere turns. Probably
the easiest approach (aside from being told the answer by the inventors)
is to figure out the ampere-turn product to just saturate a mild (1018)
steel magnetic path of the dimensions given in the users guide.
Length is 630 mm. Cross section is EI, apparently (to my eye) with
standard transformer-lamination proportions. The center leg is 30mm
wide. The magnetic pathlength (computed from the cross-section drawing)
is 160mm. Needed are the magnetic properties of 1018 steel, which is
known. Anyway, this is enough to figure out the maximum possible
ampere-turns needed.
But it will probably turn out that keeping the coil from melting
prevents us from going quite that far. The MagnaBend duty cycle is 30%
maximum, and the built-in thermal cutout trips at 70 C.
The surface area (not including the clamping bar) is 2(98+45)*630=
286+630 mm^2= 279.3 square inches, call it 280 square inches, so the
power loading is 920/280=3.28 watts per square inch. The rule of thumb
for air cooling is 0.008 watts per square inch per degree centigrade
http://www.vias.org/eltransformers/l...mers_03_11.htm
l, we get (0.008)(105-25)(280)= 179 watts can be dissipated in
continuous duty, or one fifth of the actual dissipation. The flaw in
the calculation is that the transformer rule-of-thumb assumes a
spherical object, a reasonable approximation for a transformer, but the
Magnabend is instead long and thin. Some empiricism may be in order.
Potting the winding in situ may also be useful in getting the heat out
of the copper and into the steel, thus reducing the core temperature.
But we need to insulate the coil from the iron to withstand something
like 1000 Volts AC, for safety.
Is there a power tranformer and/or motor designer in the house?
Joe Gwinn