Quote from: OilBarren on April 05, 2008, 09:36:39 PM

SURELY EVEN YOU MUST BE TAKEN ABACK BY A PRIMARY POWER FACTOR OF 0.26 WITH A LOAD POWER FACTOR OF 1? DEFINATELY NOT NORMAL TRANSFORMER BEHAVIOR.

Thane

I think it can be explained by the action of the additional secondaries in there.

According to your measurements, the PF is 0.2 even with no load - note that the transformer is consuming 25W at no load, which is also "definitely not normal transformer behaviour" unless you have a shorted winding in there - which I think is actually what you have, isn't it? That other winding is responsible for the internal load that both consumes all that power and also reflects such a low power factor to the input.

PB

OilBarren:
Could I ask what country and city you live in,
I noticed a link to Chatham, is that Chatham ON. ?

Or is that link relating to JustMe?

Quote from: OilBarren on April 05, 2008, 09:36:39 PM
...
THE PROBLEM IS THE PHYSICAL SIZE OF OUR PRIMARY CORE IS 5MM SO HIGHER CURRENTS (HIGH ENOUGH TO USE THE Pf METER DON'T WORK WELL AND WE HAVE 200 FEET WOUND ON OUR PRIMARY NOW BUT WE SHOULD HAVE ABOUT 1000 TO GET THE IMPEDANCE DOWN TO A POINT WHERE WE CAN INCREASE THE VOLTAGE WITHOUT MELTING OUR PRIMARY WIRE....
Thane

But the results you just posted used an input current as high as 3.5A - so that shows that even with the existing primary it's possible to get it up that high where solid, reliable measurements can be made. That's far away from the 0.01A that you're using for the other measurements.

If your 177% (or even 700%) at 0.01A is really true (but can't be confirmed because the meter isn't good enough), and we already know that at 3.5A it's 38%, there should be a current between those two extremes which delivers 100%+ and is also at a measurable resolution and measurable PF. This will give you the proof you need.

How about taking a whole series of measurements at different input currents, starting at 0.01A and going on up as high as you can, eg to end up at 3.5A. You'll then have a complete picture of both power factor and efficiency for the whole range of input conditions. By starting at the low end first you'll guard against melting the winding before the experiment is done. And if measurement errors distort the results at the low end, you should be able to draw a curve and extrapolate down there from the more reliable results at the higher currents.

PB

Quote from: AbbaRue on April 06, 2008, 02:42:50 AM
OilBarren:
Could I ask what country and city you live in,
I noticed a link to Chatham, is that Chatham ON. ?

Or is that link relating to JustMe?

He lives in Almonte, just outside of Ottawa, ON, Canada.  It's strange how many Canadians there are in this forum, and that a lot of us seem to live in the same urban area of under 2 million people.  I wonder if Alsetalokin was from around here, too?

Hey Thane, and all:

You might be interested to know that I have been doing some experiments and math around how flux travels from the primary to the secondary in a transformer over time.

By driving a standard EI isolation transformer well above its design frequency (around 30KHz), I have been able to predict and experimentally verify several of the anomalies reported in this thread, including:

- For resistive loads, output voltage proportional to impedance (constant current), over a wide range of R from 0 through 1KOhm or so.  This is very odd, and confirms Thane's assertion that he got less power from smaller loads.

- Output power factor 1, input power factor close to 0.

- Output about 90 degrees out of phase with input.  Input is a square wave, output is nearly triangular.

- As the load varies, there is a wide range through which drawing more from the output reduces the input power.

Cheers,

Mr. Entropy