[Prev][Next][Index][Thread]

Re: MOT saturation tests



Hi Bill, all,

> Original Poster: "Bill the arcstarter" <arcstarter-at-hotmail-dot-com> 
> 
> Coilers,
>  I finally got around to performing my microwave oven transformer 
> saturation tests.  The goal was to characterize the primary winding in 
> terms of current - ie - see just how nonlinear this reactance is as a 
> function of current.
> 
> Setup:
> * One MOT (out of a 120vac 60Hz oven)
> * This MOT contains about 122 turns of wire on the primary.
> * No secondary windings (hacked off the core)
> * Shunts undisturbed (left in place).
> * The resistance of the primary winding is 0.32 ohms +-0.04 ohms.
> *
> * Two generic non RMS digital volt/ammeters (Alfa electronics)
> *
> * Variac.
> *
> * Fire extinguisher (never hurts...)
> 
> Procedure:
> Apply various values of 60 Hz voltage.
> Record drawn current (as sensed by an inline DVM ammeter).
> 
> Results:
> V=volts, I=amps, Z=complex ohms
>  V   I     Z(approx'd as Z=V/I) (Neglecting the 0.32 ohm primary)
>  50  0.26  192.
>  60  0.31  194.
>  70  0.38  184.
>  80  0.50  100.
>  90  0.73  123.
> 100  1.16   86.2
> 110  2.14   51.4
> 120  3.86   31.1
> 130  5.90   22.0
> 
> Clearly this core is starting to do "something odd" even for low 
> currents near 0.5 amps.  An ideal linear (non-saturating) inductance 
> would show no change in Z as the applied voltage was changed.  Perhaps 
> this is the action of the shunts?

All cores of any type change their induction factor (permeability) 
with magnetization. As domains in the core line up, it is clear there 
are fewer left to do so. This is mapped as a function of magnetizing 
force and known as incremental permeability. Initial permeability is 
the figure one starts with for an unmagnetized core. The shunts will
be having some influence but not a huge amount  at low currents as 
they have an airgap in series with them. As magnetization increases, 
the shunts come progressively into play. 

    The importance of shunts in transformer cores is thus: if you 
apply progressively greater loading on the secondary core leg/s (load 
the secondary windings more and more), the opposing flux generated by 
the loaded secondary progressively opposes more and more primary flux 
from building up in the secondary core legs and instead funnels it 
into the shunts. The shunts in effect limit the amount of loading you 
can place on the primary winding (and by proxy, the secondary).

> Interestingly enough - my LCR DVM indicated an inductance of 0.065 H, or 
> Z=24.5 complex ohms, which is surprisingly close to the measured value 
> at around 120-130 volts!  I wonder how it 'knew' that?
> 
> I'm not sure what conclusions to draw from this data, only that using 
> MOTs as current limiters will most definitely introduce some odd 
> nonlinearities as a function of the current, especially at either higher 
> currents or voltages.  (We've used MOTs as current limiters on the 240 v 
> side of our pig.)

Very true. The effects can be seen rather dramatically on o'scope 
current waveforms.

> Possible future directions:
> *I suppose I should have measured things at voltages <50V...
> *Maybe I'll knock the shunts out tomorrow and redo the measurements.  
> *Maybe I'll fire up my Tek Type 543B oscilloboatanchoroscope...
> *Running the tests at voltages over 130 volts might be amusing!
> *Make measurements using a MOT w/ the secondary shorted-out.
> 
> Comments invited.
> -Bill the arcstarter

If you knock the shunts out, you are left with a core which may not 
be large enough to prevent saturation from the primary magnetizing 
force in the case of a uWave transformer. If it is still big enough,
the secondary can supply whatever you care to suck from it.

Malcolm