I was replacing a bad power transformer with another one of very close specs from a group of spares I keep on hand that I bought from Gary at PTOTP.
Each used transformer was tagged with voltages and curent ratings by Gary.

I was double checking and I was testing the voltages at each winding and checking the ohms for any possible inter-winding shorts and/or leakage to the case or other windings.

One puzzle I see is that even though the transformers seems to be good ....and has no shorts anywhere.

And even though there is NOT ANY ohms leakage from any winging to the case... I DO IN FACT measure AC voltages of some weird amounts from each winding to the case when the primary is under power.
Why is that?

As an example, I have a 350-0-350 HV winding that measures 8-vAC from the CT to case and 130-VAC from either end of the HV winding to the case!
Why?
Why should there be ANY ac voltage reference to the case?

I think I also measure about 30 - 60 vac from the filament windings to the case.

I don't see why the case should be a reference because wouldn't that cause leakage to the chassis and other windings?

BTW.
Gary labeled each HV winding on these "pulled" transformers with the appropriate current ratings ie 40ma or 70ma etc.

My question is ...how can one tell for certain the proper rating of a winding of an unknown transformer?

Is it purely a function of accurately measuring the wire gauge?

Peter:
Maybe the use of an 11 MegOhm Voltmeter is causing you to read the capacitive leakage, rather than the resistive. I suggest you load your meter down with a oh, say, a 10K resistor, and see if the readings look a little better. When we got these high Z meters, we saw all sorts of things we never saw on the Triplett 260 20K/Volt jobs,
Lewis

:I was replacing a bad power transformer with another one of very close specs from a group of spares I keep on hand that I bought from Gary at PTOTP.
:Each used transformer was tagged with voltages and curent ratings by Gary.
:
:I was double checking and I was testing the voltages at each winding and checking the ohms for any possible inter-winding shorts and/or leakage to the case or other windings.
:
:One puzzle I see is that even though the transformers seems to be good ....and has no shorts anywhere.
:
: And even though there is NOT ANY ohms leakage from any winging to the case... I DO IN FACT measure AC voltages of some weird amounts from each winding to the case when the primary is under power.
:Why is that?
:
:As an example, I have a 350-0-350 HV winding that measures 8-vAC from the CT to case and 130-VAC from either end of the HV winding to the case!
:Why?
:Why should there be ANY ac voltage reference to the case?
:
:I think I also measure about 30 - 60 vac from the filament windings to the case.
:
:I don't see why the case should be a reference because wouldn't that cause leakage to the chassis and other windings?
:
:
:BTW.
:Gary labeled each HV winding on these "pulled" transformers with the appropriate current ratings ie 40ma or 70ma etc.
:
:My question is ...how can one tell for certain the proper rating of a winding of an unknown transformer?
:
:Is it purely a function of accurately measuring the wire gauge?
:
:
:
:
:
:

As Lewis suggests, such voltage measurements are very common due to stray capacitive coupling. Many AC radios have such voltage readings.
Doug

:Peter:
:Maybe the use of an 11 MegOhm Voltmeter is causing you to read the capacitive leakage, rather than the resistive. I suggest you load your meter down with a oh, say, a 10K resistor, and see if the readings look a little better. When we got these high Z meters, we saw all sorts of things we never saw on the Triplett 260 20K/Volt jobs,
:Lewis
:
:

Thanks.. that's probably it.
I'll use my little old $10 analog VOM and compare results.

Maybe Gary inferred the xfmr ratings from the sets he pulled them out of?

A transformer's total VA rating is determined by a combination of the winding-wire sizes and the mass of iron in the core - the amount of iron, by itself, is a pretty good indication of total VA rating.

I don't know how you can estimate the rating of an individual winding of a mult-winding xfmr.
Doug

:
:BTW.
:Gary labeled each HV winding on these "pulled" transformers with the appropriate current ratings ie 40ma or 70ma etc.
:
:My question is ...how can one tell for certain the proper rating of a winding of an unknown transformer?
:
:Is it purely a function of accurately measuring the wire gauge?
:
:
:
:
:
:

magnetic field induces voltage on surrounding metal case. your hi-z meter reads that.

:Maybe Gary inferred the xfmr ratings from the sets he pulled them out of?
:
:A transformer's total VA rating is determined by a combination of the winding-wire sizes and the mass of iron in the core - the amount of iron, by itself, is a pretty good indication of total VA rating.
:
:I don't know how you can estimate the rating of an individual winding of a mult-winding xfmr.
:Doug
:
::
::BTW.
::Gary labeled each HV winding on these "pulled" transformers with the appropriate current ratings ie 40ma or 70ma etc.
::
::My question is ...how can one tell for certain the proper rating of a winding of an unknown transformer?
::
::Is it purely a function of accurately measuring the wire gauge?
::
::
::
::
::
::

.....Also due to capacitive coupling.

I use my digital meter for high impedance work, but when I want really accurate readings in low impedance work, I trust my Simpson 260. I was using a digital voltmeter on my car whilst adjusting the voltage regulator, and I couldn't get the voltage to go below about 10 or 12, even though the ammeter and the lights all looked normal. As some of you know, the generating system in a car is regulated by a device that switches it on and off very rapidly (in the case of a mechanical regulator, from 150-250 times a second; with electronic regulators, perhaps more often). When I wired up my Simpson, I was able to adjust for a steady 7.4 volts. When I connected the digital meter back to the system, it still read excessive voltage....because it was reading the spikes and not the average.

T.

At Delta, we used Simpson 260s up until we got Fluke 87 DMMs. Still, there were times when we couldn't get the true RMS Flukes to get a correct reading, so we had to go back to the good ol' 260s. A lot of things, like windshield heat controllers use pulse width for the temperature control, and a square wave will confuse the dickens out of a DMM.
Lewis

:.....Also due to capacitive coupling.
:
:I use my digital meter for high impedance work, but when I want really accurate readings in low impedance work, I trust my Simpson 260. I was using a digital voltmeter on my car whilst adjusting the voltage regulator, and I couldn't get the voltage to go below about 10 or 12, even though the ammeter and the lights all looked normal. As some of you know, the generating system in a car is regulated by a device that switches it on and off very rapidly (in the case of a mechanical regulator, from 150-250 times a second; with electronic regulators, perhaps more often). When I wired up my Simpson, I was able to adjust for a steady 7.4 volts. When I connected the digital meter back to the system, it still read excessive voltage....because it was reading the spikes and not the average.
:
:T.