This is an old question , but maybe this info might be useful to someone some day.
454565 Z565 transformer measurements taken today on an inductance analyzer out of curiosity...
Arbitrarily chose 20 Hz 1 Vrms because I had to chose something. Inductance varies with level, typically increasing with applied voltage, at least in range system would provide (10 V rms max., as long as it doesn't object to the impedance, which was not a problem with this transformer). I did not have time to collect data at different voltages. I also am not sure at the moment which are the plate leads and which are the screen/UL leads, so I may have listed them in a strange order.
Also, they were purchased used from an amp designer who extended the leads with add'l wire and heat shrink tubing over the splices...the colors made 'sense' but I think they were not exactly the same...it's been a while since I looked beyond the splices...I think using brown & brown/white was a mismatch but identifiable when I looked up a schematic. Only measured one as I didn't expect there to be a significant difference.
PVC insulation, late 60's IIRC (it's back in a box in a box at the moment).
black-yellow 1.01 ohm DC 910 mH (20 Hz 1 Vrms)
yellow-orange 0.45 ohm DC 69.5 mH (20 Hz 1 Vrms)
black-orange 0.74 ohm DC 480.8 mH (20 Hz 1 Vrms)
red-blue 166 ohms DC 46.2 H (20 Hz 1 Vrms)
red-blue/white 191 ohms DC 46.5 H (20 Hz 1 Vrms)
blue-blue/white 358 ohms DC 133.6 H (20 Hz 1 Vrms)
red-brown 42.1 ohms DC 5.5 H (20 Hz 1 Vrms)
red-brown/white 48.7 ohms DC 5.65 H (20 Hz 1 Vrms)
brown-brown/white 91 ohms 16.345 H (20 Hz 1 Vrms)
Leakage inductance: measuring blue-blue/white then shorting combinations of secondary taps (I didn't know which were of most interest but knew they would produce different results...just not their significance)
blue-blue/white: 133.5 H with secondary not shorted.
b-b/w with bk & org shorted: 195 mH
b-b/w with bk & yel shorted: 111 mH
b-b/w with org & yel shorted: 780 mH
I think therefore that the blue-blue/white measurement is the full primary (plate-plate) and black-yel is the full secondary.
These are just for reference if someone wants to know what to compare to. Don't put too much weight on the inductance measurements because the winding excitation current at 1 Vrms is probably very different than at most actual operating conditions.
What I think you may find to be a useful takeaway is that the way leakage inductance is measured is to short one winding while measuring another. That also is a quick way to tell if there is a persistently shorted winding failure - if a winding is shorted, it's reflected to the other winding, preventing the 'production' of normal winding inductance...only leakage inductance is then measurable. I call it a 'production' problem, not a measurement problem...the inductance is simply suppressed by the reflected short circuit. You can also get very approximate turns ratio from the square root of the inductance ratio of two windings (because inductance is proportional to impedance)...but frustratingly, I never get the same ratios when comparing ac voltage applied to a winding and measuring the other.
The meaningful measurement of iron core inductance is full of variables - I learned this talking to an engineer at a transformer company, so I accept that if I don't know the right conditions that result in 'gospel' measurements, I better disclaim them as 'for reference only'.
I also measured the primary with a (2-wire) DMM before discovering the inductance analyzer had a DCR mode, and the two measurements were close enough to be happy. For example, blue-blue/white was 360.0 ohms on the DMM, and 358 on the other instrument.
I kind of expected (without much basis for argument) to measure a higher primary inductance. Comparing inductance ratio 133.5 H/0.91 H, I get 146.7. I thought this was full primary vs full secondary and if that is on the order of 7.8k:16, the Z ratio should be 487.5.
So I think this indicates either 1 Vrms is insufficient to produce enough flux to produce full primary inductance, or I misidentified the windings by assuming what the largest winding inductance measurement on each side actually identified. Or this is a weird transformer - did I read somewhere it's bifilar-wound like McIntosh transformers?
So I think my DC resistance measurements are credible, and probably the leakage inductance.