[Phil_S]

Well, this certainly gives one an appreciation for the sublime. It's a good
thing wire is cheap ;-}

I chose 22AWG for the secondary and barely fit 152 +/- turns, with taps at
76 and 104. I was thinking this would result in 8K to 4-8-16. That was the
plan.

I hand wound the secondary. I found that it is very difficult to keep track
of turns. I did, however, accurately figure that I could get 38 turns in a
layer. In practice, due to my lack of manual skill, I ended up with about
35 turns per layer and wound 4 layers. At first, I thought I had badly lost
track. I did loose it somewhere, but it wasn't all that bad. It is also
hard to do this neatly.

I saved the fish paper for the primary and put it back. Actually, it was
probably from the secondary, but I used it to cover the primary. I used
green high temp polyester (?) tape (stuff meant for this sort of thing) to
wrap outer layer of the secondary. It all barely fit on the tongue of the E
lams. I chose to leave the existing gap, which mic'd at about 0.6mm.

I think the most significant error was my estimate of primary turns. I
believe now that primary turns are approximately 3800 and the primary wire
gauge is 39AWG. I bought a surprisingly nice digital caliper on eBay for $8
and its measure says the wire is #39. The number of turns makes sense with
the DCR of the winding.

Assuming that I did wind 150 or so secondary turns -- I think I did -- these
are the test results from the 3 taps:

10225:16 ohms, turns ratio = ~25
9544:8 ohms, turns ratio = ~35
10000:4 ohms (exactly, oddly enough), turns ratio = 50

The 16z and 4z taps both suggest primary turns are around 3800; the 8z tap
must be in the "wrong" place, as it suggests primary turns are around 3500
and the primary z is "low".

I have no idea if I will ever use this OT. It will go on the bone pile for
some future project, or not.

It was educational. I had some fun. I learned more than I bargained for.

Many thanks to Patrick and Flipper for informing me along the way.

Regards,
Phil