[keith49vj3]

All I'm working on a 38-62 which I believe is the same as the 37-62 (schematic here) and I discovered the oscillator transformer (part number 32-2330 [#8]) is open.  What are my options here?  Do I attempt a rewind, try to find a used one, or are there generic replacements available?
Thanks in advance,
Keith

[Chas]

If it is a solenoid, single layer wind, the schematic seems to say so. Rewind, takes patience an exact count of turns and exact wire size or next larger, even size. For example if it is #39 wire, 38 would work.. A generic replacement may have tracking problems...

GL

Chas

[keith49vj3]

Chas, this wire is very thin and has a woven insulation (litz?).  Do I need the same type of wire or can I use magnet wire of a similar diameter?  What's the secret to unwinding, do I soak it in solvent first, use heat or just try unwinding?
Thanks,
Keith

[OZ4]

I rewound one for a Philco 45 IF transformer and used plain enameled magnet wire of the same size. It had a wood core. I looked for about three hours trying to find where it was open with no luck.
I am sure someone know why they used that strange woven wire.

I slid the old coil off whole after a little cleaning of the wood dowel. I carefully measured the distance from the end before removing the old one for placement of the new one.

I used masking tape ( the green stuff) to make a form for winding the new coil. After I had enough thickness of tape I used an xacto knife to cut out a section where the new coil is to sit on the core.

Now I cold not measure the inductance of the old one but since it looked just like the other coil on the same wooden core, I wound the new one to match It's inductance (at this point what have I got to lose.) I used my capacitor checker which can measure inductance to match them. I didn't try and match the number of turns.

I used my cordless drill to wind the wire. It took three tries to get something that might work. The first time I had about three times too much inductance. I just kept removing wire until I had a match.

I used a little super glue to hold it's shape and baking soda to cure it. It wasn't real pretty but it worked.

It actually tuned.

I will try and post a picture of it later.

[Radioroslyn]

I found they the post '36 sets the rf and osc coils are different from the older models in that they are smaller and have a white coating that is like cement. This makes them difficult to dissemble and repair w/o destruction.
I'd go with a universal replacement as it has a slug and can be used in many application.
Now if you like experimenting I can send you a couple of 1/2" dia slug tuned forms and you can roll your own. Takes abt 80-100t of 36-38ga wire.

[Chas]

The textile covering is braided silk. The silk and the enamel coating provide a spacing to reduce turn to turn capacitance. The enamel on modern magnet wire is thick so no problem.. The windings can be loosened by soaking the coil in alcohol or mineral spirits both of which will soften even damar hardened bees wax or paraffin. You did not state it it is solenoid or a universal wind. Count the turns carfully.

Scramble wind will work to replace a universal coil if an old winder cannot be found. However, as suggested a slug tuned oscillator coil may work. Just be sure it will resonate using the tuning condenser in the radio.

Place a want ad in the for sale page. for a good oscillator coil too.

Chas

[keith49vj3]

I'm not sure yet how it's wound.  I'l unwind it tomorrow if I can find the time.  If I understand this correctly it's critical to count the turns.  If I get this right I'll measure the inductance and post the results.
Keith

[keith49vj3]

I soaked the transformer overnight in xylene and it removed all the wax that was holding the wire together.  I had no problems with the wires being stuck together. The wire had at least 10 breaks.  What would cause this? the breaks were all on the same side of the coil as if someone cut it with a saw.   I counted the turns and came up with 88 +/- two  turns.  I wound a new one with 100 turns of #36 wire.  I figure better long than short.  I will test it soon and remove windings if necessary.  I will report back  with the final results but it may be a few days.
I'll keep you posted.
Keith

[keith49vj3]

Well, I turned on my radio and surprisingly it plays well. With the dial set at 1200KC I'm picking up a station that broadcasts at 940KC. Please correct me if I'm wrong but I believe I've got too many turns on the coil (I added an extra 12 turns when I wound it). Tomorrow I'll start removing the extra windings to see if I can get the dial in the right ballpark. Is there a way to calculate how many turns to remove or should I do this empirically?
Thanks,
Keith

[Radioroslyn]

Yup too many turns.

[keith49vj3]

Well, it looks like the magic number is 87 turns using #36 wire. This job was actually pretty easy. I post pics soon.
Keith

[morzh]

Well,

a long solenoid inductance depends on:

1. Proportional to the Number of turns (square of it), so N^2
2. Proportional to the cross-section of the bobbin (considering wire diameter being very small relative to it), and
3. Inversely proportional to the Length of the winding.

So, you need to observe both geometry of the winding (the exact size of the cylinder formed by the cross-section and the length) and the number of turns.

The diameter of the wire is only important, if:

1. The gauge is too large and so the winding cannot fit in the same length as before.
2. Current rating of the wire is too small (it is not too large in the first place) so you need a larger gauge.
Note that the absolute DC resistance by itself is not very important, in most cases anyway.

The Litz (it is not the cloth insulation, but multiple strands forming the wire, every single strand being its own enameled insulated wire) vs solid is important at higher frequencies, but not necessarily very important: it will make the coil less lossy due to diminishing the skin-effect, but it might be not that much of a loss in the first place in tube radios. Resonant tanks are more sensitive to it as they need especially high Q-factor and could use lower losses, but again, most times solid wire will work fine.

[Radioroslyn]

<Well, it looks like the magic number is 87 turns using #36 wire. This job was actually pretty easy. I post pics soon.

Glad to hear you got it working! Sometimes the tricky part is to get the tracking ( low end and high end where there suppose to be) You didn't mention if you used a coil form w/slug tuning? The slug helps to get the low in the proper place and using the trimmer on the tuning cap to set the high end.

[keith49vj3]

All, I rewound the coil on the original form and the tracking is pretty good.  The old and new winding were 0.25" wide.
As promised here are some pics.
Keith

[OZ4]

And here's my handy work.