Another Westinghouse H840CK15

[old_tv_nut]

Dumb question: why can't you use the internal 3.58 oscillator to do the 90 degree adjustment?

[Electronic M]

Originally Posted by old_tv_nut

Dumb question: why can't you use the internal 3.58 oscillator to do the 90 degree adjustment?

Pretty sure Westinghouse didn't use a crystal referenced oscillator In that model so unless you are synched to a color burst it will probably drift off frequency.

[trinescope]

Not a dumb question, because this particular set does not use a 3.58MHz crystal controlled oscillator like pretty much everything else. It's a bit more like a horizontal sweep oscillator AFC circuit controlling a reactance tube based off of the color burst and it actually has a control labeled "color hold". Guess that's why the alignment instructions call for an external crystal oscillator.

[old_tv_nut]

I knew it used an LC oscillator, but if you first adjust the oscillator and have it locked properly to an incoming signal, then it should be just fine for adjusting the 90 degree transformer.

By the way, I don't know why the designers used a tuned transformer to get a 90 degree phase shift. Better to use the transformer to get the impedance match needed (as they do), but not for the phase shift. The R-Y demod drive goes directly to the burst phase detector, so R-Y drive phase is stable at 90 degrees to the burst. Why not then use a simple RC circuit to get the additional 90 degrees for B-Y drive? It would not need tuning to obtain the correct and much more stable angle.

[trinescope]

I have the winding machine here now. I did a test wind of one layer, 333 turns of 41AWG wire and it turned out very well. Now I'm waiting for the Kapton tape to get here. The machine did the 333 turns in well under a minute. I'm going to try the winding data I calculated for the Westinghouse part since it uses only 41AWG wire and see how it measures as far as winding resistance. It should not take a whole lot of time to wind the transformer with this machine.

If all goes well I will order some 43AWG wire and try to do the CT100 winding data as well. 43AWG wire is more expensive than the thicker, more usual gauges.

[etype2]

[trinescope]

I measured the resistance of the 333 turn layer a bit later last night and it was 175 ohms. This calculates to 1050 ohms for the 2000 turns which is close enough to the original 1000 ohms.

[ohohyodafarted]

Post a photo of the coil winding machine.

[trinescope]

This is the same machine I have

https://youtu.be/7PXXoA3897M

[trinescope]

I tried winding the replacement transformer, but my first two attempts were failures. The first one wound ok, but two of the free wire ends broke off and could not be retrieved. The second time I slowed down the winding speed and got most of the way through but started having problems with wire breakage because the wire was snagging on mold flashing on the edge of the wire spool. I did end up finishing the part and got the wire leads attached. One of the wires broke because the coil was too much of a tight fit to the lamination stack, so that one also became scrap. Now I need another spool of wire to continue. I did another test wind of the primary only and it seems to be ok, but I will need more wire to do the secondary. I'm new to this and am learning what works and what doesn't as far as securing the free wire ends and getting an even layer of wire to lay down. I also will make the bobbin a bit shorter than the core window so it hopefully will fit after all the windings have been laid down. I'll get this right eventually.

[trinescope]

John Folsom was kind enough to give me a copy of the winding data for the replacement transformers that he had made available. Good news is my own calculations were close enough to have resulted in a working part if my first two attempts had succeeded. No 43AWG wire was used, so the winding resistances should be close to the Westinghouse part. Now I just need to wait for the new spool of wire to get here and start practicing with the machine again. I tried two more test winds of the primary which appeared to have turned out OK, but that one is easy since it's only a few layers of wire. The problems start with the secondary simply due to the number of layers of wire and insulating tape needed. Just a bit of unevenness from a wrinkle or air bubble seems to get magnified when winding the new layer over it. I probably won't get it 100% perfect, but I think I can get it close, at least enough to recreate the transformer. Apart from a few wax capacitors, the transformer is the last hurdle I have at this time to getting this set running again barring any unforeseen problems.

[etype2]

[trinescope]

I did get the transformer working, at least for about a half hour before I heard the sound of insulation breakdown. I did at least get to see the results of the IF and chroma bandpass alignment I had performed. Much better detail and color quality than before. Back to the drawing board on that transformer, though. I was trying to do the convergence procedure when the transformer went bad on me, so I didn't get the chance to take any pictures.

I need to find something for winding layer insulation that has a bit of grip to it so the windings don't move around so much. Kapton is good for insulating the inside and outside of the windings as well as between primary and secondary, but is too slick to hold the individual layers. Double sided tape is too thick for this purpose as it will result in a coil which will not fit in the core window. I used fiberglass resin in a glass jar under vacuum for impregnating, it seemed to work but is too thick, also expensive since I had to throw away over 90% of the material afterwards. I could try something like good old Minwax polyurethane and hope it will cure all the way through. At least the unused portion can be saved. This has been an interesting, although a bit costly learning experience. I would like to try doing some audio output transformers later on down the road with what I have learned so far.

[etype2]

That winding machine looks expensive.

[trinescope]

Yes or no depending on how you look at it. It sells for just under $600. It looks to do a pretty good job of laying down an even layer of wire. It's certainly not as good as a much more expensive machine, but for small production runs it will do fine. Materials and construction technique are just as important as getting an even layer of wire on the bobbin. The main weaknesses are the tensioning system and the wire guides, but those could be improved with a bit of ingenuity and work. The basic machine itself is pretty solid otherwise.