Ct-100
 

I picked up a CT-100 earlier this year from a local collector, and after finishing off a bunch of other projects over the past few months, I've now turned to this wonderful RCA.

This set has seen a bit of work over the years. Many of the original paper caps have already been replaced, sometime between the 1960s and 1980s, based on the appearance of the replacement capacitors. The main power supply electrolytics were replaced with nice GE FP electrolytics that are still doing fine. The selenium rectifiers were replaced with diodes, and the ballast has been removed and replaced with fixed power resistors mounted in the area where the seleniums were.

Cosmetically, the set is not bad, with a cabinet that could use refinishing, but with the wood in pretty good shape except perhaps for a couple of the feet. Unfortunately the main front panel knobs are gone, so I'll be eager to find a possible source for replacements.

Everything else is there, and the good news is that the original 15GP22 tests good on all three guns.

I recapped a number of the already recapped paper caps (the replacement caps were a mixture of paper and polymer dielectric caps). I haven't done a full recap yet, but tried to cover the sweep circuits and video output stages.

Powering up briefly, I got sound immediately, as well as anemic high voltage (12 kV) and some semblance of a dim picture (bad sync, quite dark, but some video information making it to the screen).

I then recapped the HV cage, hoping that would bring the HV up to normal, but it didn't. I replaced some resistors in the focus section, putting original values in place of various jury-rigged components, and now notice that if the focus rectifier is connected, I get no HV. With the focus rectifier disconnected, I get 10 kV. There may be a (possibly brand new) short in the vertical dynamic convergence transformer, but we'll get to that later. For now, I'm leaving the focus rectifier disconnected.

Unfortunately, the flyback gets hot. I have a nice horizontal drive waverform on the horizontal output tube, and about -40V bias on the HO grid. Running the set briefly and then feeling the warmth on the flyback seems to indicate clearly that the source of the heat is the HV outer winding on the flyback, and most likely not the primary windings. It heats up pretty quickly, so I'm running much less than a minute. No nasty damage, but it is clearly getting too hot too fast. I gather this indicates a shorted turn in the HV winding; any other diagnosis possible? That would certainly be consistent with my anemic HV. The heating is the same whether the HV rectifier is connector or not, so that seems to rule out something loading down the HV as the cause of the heating.

Moyer does not have a replacement flyback available. Any junker chassis out there for a possible replacement?

Otherwise, I may try some surgery on the flyback. However, I don't want to do that prematurely. The original *looks* very nice and clean, and whatever surgery I do will irreversibly change that, and also has a fairly low chance of success. I've saved one flyback on a different set in the past by removing some outer turns on the HV winding, where the short happened to be. No guarantee by any means that would be the case here.

Open to advice on how to proceed.

P.S.: Moderator, I suppose this really belongs in the COLOR section! :) I'm so used to posting in the black and white section that I forgot about that...

Sounds like you need one of John's transformers, shoot him an email. RE low HV: disconnect the shunt regulator (6BD4), see if HV pops back up. The HV pot may be bad, do some futzing with the shunt circuitry, I wouldn't be surprised if you locate the fault. I've never seen a CT-100 flyback go bad, they seem to be better made than the later stuff. The outer winding getting hot tells me it's trying to put out the voltage, but it's getting lost to ground somewhere between there and the CRT.

I was probably editing while you posted -- I added one comment above mentioning that the flyback heating occurs whether the 3A3 is connected or not, so it seems to rule out something loading the HV as a cause. I had also tried pulling the anode cap off the 6BD4 earlier, and that didn't affect the low HV.

I'm encouraged to hear that CT-100 flybacks rarely burn up, so perhaps that means the likelihood of someone having a good one on a junker chassis is better than zero.

I most likely will need one of John's transformers, but I figure I don't really need it until I get the flyback situation sorted out.

It's probably just the convergence transformer, put one of John's in before you do any more testing. Otherwise you'll be chasing your tail, multiple faults and so forth.

With the focus rectifier disconnected, isn't the vertical dynamic convergence transformer completely out of the flyback circuitry?

True Tom, but the 1X2B is still in the circuit, which drives the focus circuit, which is why Nick is so adamant about replacing the convergence transformer. One never knows for sure what is wrong until it's fixed of course...

Pete

Tom, I just read your latest post. Yes, I agree that removing the focus rectifier should remove a convergence transformer short from the flyback; I did not see a 1X2B reference earlier in the thread.

Hi Pete,

Thanks for chiming in -- I know both of you have a lot of good experience with these sets.

Anyone know of a spare CT-100 flyback floating around out there? According to Sams, a Thordarson FLY-100 is an option. The RCA part number is 79012.

Tom, sorry, the FLY100 is NOT a replacement, just another Saqms data error. I acquired a FLY100, and it is some B&W flyback. Grrrr.

I would agree that with the plate cap of the HV and focus rectifiers removed (and the tubes pulled) if the flyback HV winding still heats up rapidly, it is most likely a shorted turn.

I don't have a spare chassis, but I do have a universal coil winding machine and have wound several flybacks with varying degrees of success and failure. I unwound a bad CT100 flyback to get the turns count recipe and wound a replica on the old frame and core. Installed in my CT100, it lasted about 2 minutes, then the primary winding failed open somewhere.

I wound a replacement flyback for my 15" Motorola which worked and produced good HV and sweep, but the HV pulse and all the auxiliary pulses supplied by the flyback have a big negative going ring right in the middle of the pulse, which is upsetting the circuits which use the pulses (AGC, burst gate, color killer, color bandpass amp etc.) So the internal LC ratios are not quite right, even though I followed the recipe as closely as possible.

So winding replacements is an art I have not mastered. While I would not be eager to wind a complete flyback, I could be coaxed into wind a replacement HV winding. But first, do some searching for a junker chassis. I bet there are some out there...

John,

Thanks for the straight scoop on the FLY-100. When I called Moyer, the person I spoke with also indicated that his cross reference did not support the idea that a FLY-100 would be a suitable substitute for the flyback in a CT-100. In any case, he didn't have a FLY-100, so at least I didn't end up wasting money on that.

Great to hear that you've tried rewinding a CT-100 flyback. I also know of a coil winding machine I can borrow. The one here is a pretty simple affair, with just a variable speed motor (with foot pedal control), a turns counter, and a drill chuck. Is yours a little more sophisticated, with perhaps something to control the wire winding pattern?

I also have the perhaps misguided notion that it might be possible to remove most of the high voltage winding from another random flyback (keeping it almost entirely intact), and mechanically mount that on my CT-100 flyback after removing the original HV winding from it. Naturally I'd have to pick a donor flyback with an inside diameter on the HV winding that is slightly larger than the CT-100 primary, so it would fit. Could easily end up with tuning problems like you encountered, however.

I won't go after either of those just yet, since I agree it is worth trying to get the word out to see if someone may have a spare available.

In the mean time, however, I might look into some fine surgery on the existing flyback, on the outside chance that the short is near the outer diameter of the HV coil. Unwinding a relatively small number of turns from the outside might remove the short but leave enough for sufficient HV.

Tom,

I think attempting to unwind some of the HV coil is worth a try. If you can expose all of the HV winding, you might then be able to localize the shorted turn by seeing where it gets hot.

And the transplant idea might work, but it would be tricky to find one that fits mechanically ad has the right number of turns.

My coil winding machine was mad by the George Stevens Co, and is a very sophisticated device. It has interchangable cams and gears to set up the coil width and the ziz-zag winding pattern (also known as a universal type winding). Here is a photo of the machine winding a HV coil, like what you need.

That's a very nice winding machine. I always wanted something like that for rewinding power transformers, but I could never justify the cost or space requirement, for the few times I'd actually use it.

If you could remove the HV winding from the flyback ,could you then use the resulting flyback for all of the other windings except the HV winding and also use an independent external HV supply for the CRT anode? I suppose it may require some additional winding with L/R/C components to tune the flyback to match the original characteristics?

Dave

This George Stevens Co winding machine is designed for universal (flyback and peaking coil) type windings, and does not do layer type windings used in power and audio transformers.

On newer roundys (like CTC-16 era) I've heard of folks having successfully removed the fly's HV winding and getting their HV from a solid state HV trippler the input of which being connected to the HO Tube's top cap....This may also be a solution for your CT-100.

Two options to further confirm your flyback is at fault are to compare it's winding resistances to those listed in the Sam's folder, and to get a vintage flyback/yoke ringer (aka a flyback tester) and see how yours tests.

Good luck fixing it.

Dave,

I've seen it done. When I got my Philco 48-2500 projection TV, a prior owner (an engineer) had unhooked the original HV supply, and built his own little HV supply with an 807 tube and a nice big RF coil for the HV. I hastily removed all that and put the set back to original, only to discover there was a good reason why he had done that -- the original HV supplies in those sets had too high of source impedance, resulting in fluctuating HV with scene brightness, which in turn affected focus stability.

So yes, I think that could work. I'll probably try that as a last resort if necessary. Hopefully I can either get a replacement flyback, or modify/repair the existing one in some manner that works OK. So one way or another, we'll get this problem solved! Glad to see lots of ideas here from you guys.

Tom C.,

The tripler idea sounds very interesting! On the Philco set mentioned above, I eventually replaced the original poorly-designed vacuum tube voltage tripler with a solid state tripler to improve the HV supply, so I have a little experience working with a tripler module. That would be a very nice fix that would fit into a small space somewhere and not look too crazy. If the drive right off the primary is high enough, then the original regulator, etc., should still be usable. It seems quite possible that this could work. I notice that the focus supply runs off a tap *below* the 6CD6 drive connection, and generates 4.5 kV with a simple half-wave rectifier. All I need is for the top of the primary to generate 6.7 kV to get enough to triple to 20 kV.

This might be the next thing I try after seeing if some minor surgery can remove the shorted turn. Easier than trying to rewind the HV winding or retrofit an HV winding from another flyback. I suppose I could even "borrow" the tripler out of my Philco to do a feasibility test, and then order another if the result looks viable.

I did check all the resistances, and they are quite close to the values on the Sams schematic. If it's a shorted turn, it seems to be isolated to just one turn or a small number of turns, having no appreciable effect on the total DC resistance of the winding.

Tom

Comparing winding resistance will likely not reveal a shorted turn, as the short will only remove a small number of turns resistance from a large resistance total.

Do you have access to a "ringer" tester? They are pretty good at spotting shorted turns...

Have you checked the value of the cathode resistor on the 6CD6 yet, or read for cathode current? All this talk about serious mods so soon, and we're not even certain the flyback is the problem yet! Slow down, take a deep breath, do some trouble shooting. In the end if it ends up being the flyback, at least you did your homework first. All I'm saying is I definitely rule out every other possibility before jumping straight to unobtanium parts like flybacks, especially in a chassis like the CTC-2 where they rarely fail.

There are several controls on the chassis which impact HV production:

Drive varies the waveform going into the grid of the output tube, if it's low in amplitude the negative bias of the tube will not be right and tend to cause more current to be drawn heating things up. R175 (Sams) on my set had drifted causing incorrect operation of the drive control, I couldn't get rid of the drive line till I got the correct resistor in there.

Horizontal may be off frequency (wave coil), which will cause bad sync and low HV production- 2 of your original symptoms. The waveform in the oscillator itself may not be the right shape (equal peaks), the cure for both of these is to do the horizontal setup procedure in Sams.

HV pot may have drifted, if so the shunt may be dragging down HV even if everything else is working correctly.

You have a known bad convergence transformer- stop thinking about the fly till you cure that problem.

Is the CRT boot clean? If not, it will make a sizzling sound and even if it's not it could still have a path to ground somewhere that's draining the high potential away from the CRT. I went through 3 different ones before I found one my set 'liked'.

Or any one of a million other things on a very long list, but the point is you have to start somewhere before assuming the worst. My starting point is always to make sure the horizontal oscillator is in perfect health and on the right frequency FIRST. All other circuits in the chassis depend on horizontal to tell them when to do their thing: burst keying, color killer, convergence, ect. All other circuits are secondary to horizontal being correct.

My AGC pot was also bad, leading to a lengthy and unnecessary foray into CTC-2 IF alignment. Much was learned, including the important lesson of LOOKING FOR SIMPLE FIXES FIRST! :yes:

Slow down, maybe even walk away for a few days and think about it. Put John's convergence transformer in it, then retest. If it's still jacked up, go through the horizontal circuit with a fine toothed comb. Report to us with your results, the last thing we want is for your fly to be bad.

N2IXK,

I have a Ringer, but it works poorly. I've been on the lookout to buy a better one, but that's one acquisition I still need to make. The one I have has a sticky meter movement, and I have not been able to improve it. That ends up making the readings very unreliable. Would be nice to have a good Ringer right now, just to make doubly sure that the problem isn't somewhere else and the flyback really shows the expected lossy behavior in a Ringer.

Miniman,

I'm actually reasonably experienced at TV restoration, so quite a few of the important things you point out are indeed things I already paid attention to, or have reasons to believe are not relevant for the particular way I'm going after the problem. Nonetheless, I'm glad to see you ask all the right questions here, since I don't want to leave out something critical and make an unnecessary mistake.

The horizontal frequency is correct, because I can see just enough video stuff on the screen to tell that it actually locks weakly to a TV signal on the antenna input. Nonetheless, I got out the frequency counter to make sure, and yes, it's ~15.75 kHz. My ear had also confirmed from the start that the pitch was "about right."

HO grid bias is -46 volts, and the peak-to-peak drive waveform is 150 V on the grid (cf. 120 V P-P in Sams), adjustable with the drive control.

HO cathode current is 193 mA (cf. 167 mA in Sams), which seems reasonable given the improper power dissipation in the flyback itself, although if any of these numbers raise a red flag for any of you, don't hesitate to speak up.

The real smoking gun on suspecting the problem is a shorted turn in the HV winding of the flyback transformer is that it is getting quite hot in about 1 minute of operation, with neither the high voltage rectifier (3A3) or focus rectifier (1X2) connected. The only way there can be a load on the high voltage winding is if the load is within the coil itself - a shorted turn or something else very lossy. The heat really does seem to be generated in the HV secondary winding; it is much hotter to the touch than the primary windings. My finger may not be a 100% reliable temperature sensor, however, so I've checked it many times and I seem to be coming to a consistent conclusion on that.

I have the convergence transformer out of circuit for all of these measurements (by disconnecting the focus rectifier), so even though it looks quite likely that it is bad, it can wait until the HV problem is cured. It is definitely not having any effect on the flyback right now.

I have the CRT out of circuit most of the time, but have connected it a few times (as well as the 3A3 HV rectifier) and can see a very dim, out-of-focus raster on the screen, and it does not substantially load down the already weak HV (because the CRT anode current is almost zero under these circumstances). HV is quite low, around 10-13 kV -- a little lower in that range yesterday, a little higher today. No evidence of any current in the 6BD4 regulator tube (which is how it should be with the HV too low). Disconnecting or connecting the regulator tube has no effect (which indicates it is working correctly, or at least doing no harm).

I think I'm properly covering all the things that could account for improper flyback function, but if you see something I missed, I'll be more than happy to go after it. I, too, am not eager to prematurely modify the flyback in an irreversible manner.

One could always suspect the yoke, but if it's bad, it shouldn't be causing my HV secondary on the flyback to get hot. I have a test yoke I can connect, and perhaps I'll do that next just to make sure (in case my finger temp measurements are misleading me as to what's the source of the heat in the flyback).

On the other hand, there are lots of other circuits in the set that have nothing to do with the horizontal sweep, and my preference is to solve this problem before going after all those. I'm confident we'll get this set fixed in a step-by-step manner, and for me, horizontal sweep and HV are priority #1 pretty much in any set I work on. So indeed I agree with you that the horizontal oscillator needs to be functioning properly as step #1, and that was indeed done here.

When I get to the color circuitry, that's where I've got the most learning to do. I've restored dozens and dozens of black and white sets, so I've got a decent grasp of all of the circuitry and functions in black and white sets. But color is something new, so that part of this experience is going to be educational!

The B&K 1076 analyst is a good flyback checker. Another good way is with a WO-91B RCA scope with the mod from RCA bulletin. The mod is adding a capacitor to get a sweep output. The pattern during test is a good ringing waveform if the flyback is good.

I spent some time last night and this morning trying to check the flyback and yoke in additional ways.

I have a Sencore SS137 Sweep Circuit Analyzer (picture attached) that I had not used before. I spent a little time making repairs to it (new electrolytics, fixed a broken wire) and confirmed that everything on it appears to be working properly. It has many functions, of which two are useful here. One is that it has a substitute horizontal yoke coil. Connecting this substitute yoke in place of the yoke in the set is supposed to tell you whether there is a problem with the set's yoke. Connecting the sub yoke, I find the HV is slightly higher (14-15 kV), but no where near where it should be, so my conclusion is that the yoke in the set is probably OK. It also measures the current in the substitute yoke and displays whether it thinks the flyback (and overall horizontal sweep system) is OK. Although the measurement doesn't seem very sophisticated, it does claim the flyback is bad (provided the surrounding circuitry is functioning normally). The meter has lines for various deflection angles of black and white sets (above line is good, below bad), and a single line for "color" sets, requiring the highest amount of measured signal. My result was that the meter came up to the level for a 50 degree B&W tube, which is way below what is deemed correct for a "color" tube. Not sure I trust this one-size-fits-all test, given that the 15GP22 is probably a bit unique, and its sweep system may not have enough in common with what this tester thinks is appropriate for "color" to give a trustworthy result.

Fant's post above reminded me that people sometimes do a ringdown on a scope to check a flyback. Although I have never done that before, it occurred to me that although I don't trust the meter on my Ringer, it does put out pulses. I captured the ringdown on a scope (see attached photo). What do you guys think about the waveform?

The manual for the Ringer says 10 cycles is the boundary between good and bad, but I don't know how small a cycle needs to be before it is no longer counted. Perhaps one of you knows.

I have no idea where the threshold is set for the ring test, but I would judge that to be somewhat less than 10 rings.

I used the 1076 with 100% success back in the day. Even with its simple neon indicator, it never once gave a false "bad" reading or showed a bad unit "good".

I would try the ring test on several known good (B/W is ok) flyback transformers to get an idea what the display should look like... I agree that that seems to fall short of the 10 ring threshold, but what does another flyback do in your set-up? Perhaps the scope input impedance is lower than that in the tester circuit, causing more rapid damping?

jr

Tom-

I am not too far away from you, and I have several devices that include flyback-testing ability if I remember right. They are a B&K 1077, a Sencore VA48, and a Sencore VA62. The catch is that I have not yet used any of them to do flyback tests, so there would be some learning to do in the process. I am happy to lend you any of these instruments and/or to help out if I can. I also have a CT-100 that I restored a couple of years ago, and I did a lot of work on its horizontal circuits as well. (In fact, it is still not 100% either, but I had put it on hold while other tasks came up.) You can probably find its discussion here with a search of my name.

Hi Chris,

It sounds like your situation is not so much different than mine -- various equipment that should do the job, but nothing that you've used often enough to really have confidence in its accuracy. I also have two B&K 1077 units that I could repair and try.

All,

I did a bunch of ringdown tests on various flybacks to see how they compare to what I'm seeing on the CT-100 flyback.

Here are the attached pictures in order:

1. CT-100 flyback, this time isolated from the circuitry (all connections removed, including yoke) - looks like 5 or 6 decent cycles only

2. Stancor A8134 NOS flyback, no connections

3. RCA 104309 NOS flyback with big coil, for 110 deg CRT (probably black and white), no connections

4. Traid D305 NOS flyback, for color set, no connections

5. Traid D311 NOS flyback, for color set, no connections

6. Traid D313 NOS flyback, for color set, no connections

Quite an obvious difference in terms of seeing a lot of decent ringing cycles on all the NOS yokes.

So that all looks nice and conclusive, but wait until you see the next post (I can't upload any more pictures in this one).

Now a little more data to make things confusing...

I happen to have an Emerson 637 in the shop. It's a 10" B/W set from 1949 and it works very nicely. Took the chassis out and tried the ringdown test on this set with a "known good" flyback, and look what I get!

1. Emerson 637 ringdown, with flyback in circuit (yoke, etc., connected)

2. Emerson 637, with yoke disconnected (removing other connections had little effect, also tried pulling 1B3 to remove filament load, but didn't matter)

3. Merit HVO-5 NOS, very similar to what's in the Emerson. No connections.

OK, so the Emerson flyback doesn't look very good in a ringdown test, but it works in the set! And it compares badly with an NOS flyback that looks very similar and might even be the right replacement (it's the fairly common early style flyback used in the late 1940s with a more metallic-looking core, not ferrite or laminated).

HV in the Emerson is 9 kV, which is just what it should be.

Not sure what to make of that!

And a couple more for a sanity check.

1. Crosley 9-407, with all circuits connected. Similar early style flyback as Emerson 637 above (the other one which rings down with very few cycles, even though the set works well).

2. GE 805 Locomotive, with all circuits connected. More cycles on this one. This set also works fine.

Still not sure how to interpret the Crosley and Emerson results. Maybe the early style metallic looking core is lossier than more modern ferrite cores? Then why does the Merit HVO-5 NOS flyback not show the same loss? It looks very similar in construction.

Reading some of the details in the Ringer manual, it says that the usual threshold is 25% of the amplitude of the first cycle. So the bar is pretty high.

Tom, when testing with the flyback in circuit, if you leave the rectifier tube in the socket, the filament circuit looks like a shorted turn. Take one of your out of circuit flys and put a single shorted turn of wire around one leg of the core and compare the results.

John,

Here are some pictures to show exactly what you're referring to. The first is a NOS FLY-107 flyback with no circuit connections. The second is the same flyback, but with the filament winding for the HV rectifier shorted directly. The third is with an actual 1B3 filament across the filament winding.

These all show the expected behavior. A dead shorted turn has a huge effect, and a turn shorted with a small, but finite resistance, has a noticeable, but smaller effect.

The CT-100 flyback cases I've shown always had both the HV rectifier (3A3) and the focus rectifier (1X2) unplugged, since I was aware that this could be an issue. In the case of the post above with 6 pictures, the CT-100 flyback was tested with no connections whatsoever (no yoke, no filaments, no pulse windings connected to anything).

The one case I want to go back and double check is the Emerson 637 picture above, where I mention the yoke is disconnected and other connections didn't matter. I definitely tested it with the 1B3 removed, and also tested with most other connections removed (including yoke), but I want to make doubly sure I did it with all of the above disconnected simultaneously.

The Crosley 9-407 case above had everything connected, including HV rectifier tube filament.

So far, all tests are consistent with the CT-100 flyback having a shorted turn. The only confusing cases that are causing me to hesitate are the ringdowns on the Emerson 637 set, where I did not see the expected result for a working flyback.

The CT-100 ringdown isn't as fast as the direct shorted case posted here, but I think that makes sense. Shorting the filament winding has very low resistance, since the filament winding is thick wire. An unwanted short inside the HV winding (such as what I think I have on the CT-100 flyback) will have higher resistance both due to much thinner wire, and possibly because the short itself has some resistance (dirty/burned contact point).

This has turned into quite a sidetrack, but I'm actually glad to develop a good technique for testing flybacks, with enough experience on many different flybacks to have decent confidence in the results. These ringdown scope traces are probably a more thorough method of testing than simply relying on what some particular piece of equipment says in a "good / bad" test.

I think you are right, the resistance of the HV coil wire ends up being in series with the shorted turn and thus reducing the effect of the shorted turn on the ringing response.

One other thing you can try.. remove the solder from the terminal which connects to the plate cap of the HO and remove the flyback wire which goes to the HV winding and leaving the flyback wire going to the primary attached to the terminal. Run the ring test again, and you should see a normal response, proving the shorted turn. Of course, doing this does have some risk, one could break either of the two wires in such a way that they could not be repaired. But as the flyback is most likely NG, you don't have too much to loose.

Hi John,

I can do that this evening when I get home from work, but I don't understand why disconnecting this wire will change anything. If there is a shorted turn in the HV winding, it will still have the same effect, whether the bottom end of the HV winding happens to be connected to the primary or not. Top end, of course, is open circuit with the 3A3 removed. A completely isolated shorted turn will still overdamp the ringdown, just like a shorted filament winding, wouldn't you agree?

Um.. not sure i follow correctly. But you're saying leave the HV winding ('tire') electrically floating but still physically in place(?).
Would not a shorted turn in the 'tire' still load the primary whether or not it's electrically connected? Seems like you'd have to physically remove the tire to do the test.
Or maybe i'm misreading whut U sed. :o

Yeah, my bad... what was I thinking? Bit I am pretty much convinced you have a shorted turn.

I retested the Emerson 637 set, which appeared to show too few cycles for a known-good flyback when I tested yesterday (and posted on page 2). This time I really had all connections removed, and it looks decent.

I think previously I had noticed that removing the 1B3 in that set had only a minor impact on the ringdown and then plugged it back in. However, after removing all other connections, it may have been having a bigger effect, being the only remaining load. So the previous picture with "no yoke" still had the 1B3 filament load. Got to be careful to make sure all connections are removed for a final test of a flyback.

So I think that removes all doubt about this particular method for testing flybacks. If you get something like 10 decent cycles of ringing or more, the flyback is OK. If you don't, it has high loss (low Q) from a shorted turn.

Looks like this will be a good technique for checking flybacks going forward.

Decided to unwind the HV winding of the flyback and see how many turns would need to be removed to find and remove the short. I was hoping it would be a short in the outer layer or two of the coil.

Got started by removing the black hard shell that coats the outer diameter of the HV coil. The material is quite different than most flybacks -- quite hard and durable. It was difficult to nibble at it with a diagonal wire cutter. First picture below shows the first nibbles out of the shell. Second picture shows the whole thing removed, which I also used a Dremel tool to assist. Once I get a few big nibbles out, then I was able to crack the whole thing off easily.

I left the Ringer and scope attached the entire time while I unwound the coil, so I could see immediately when the short was gone. After unwinding a few tens of layers, I could tell the unwinding was getting more difficult, and one spot in particular looked well cooked. Getting closer to the short!

Eventually got the short, and I also unwound one more layer of well cooked wire with the thought that a damaged winding layer would be likely to fail later. Still there are quite a few cooked layers present, but the short is gone. Alas, I had to remove quite a bit of the coil (perhaps something like 15%). Will it have enough HV back in the set? We'll find out.

The new ringdown is shown in the 3rd attached picture, and looks convincingly good. The last picture is the remaining HV coil, obviously smaller than when we started. I'll reattach the 3A3 anode lead and reinstall it in the set.

Dang, this is one of the more fascinating of troubleshooting mini sagas.

:)

Good Work! It will be interesting to see how much HV the coil can produce now.

jr

If it is not good enough you could always take the removed wire dip it in lacquer a few times to build up it's insulation and try to rewind that wire back on to the fly...