No, I don't want to buy one, I want to make one.
You know the sets, the ones that wipe out the power supply every time. As we know there are several different causes, not which the least is a CRT, and in these parts we see alot of bad 32", and surprisingly fewer
36".Anyway there are a few other things and I would like some input as to what such a list would be. Now I really do prefer empirical troubleshooting, but I have to take into consideration time, as well as buying 2SC4834s by the truckload. Also everytime those short they might take those caps with them, and there a few different styles of them, while not expensive what worries me is running out of the right type at the wrong time.
On those occasions that it's not an intermittent short in the CRT, but the CRT is pulling too much current I think an upgraded HOT would solve the borderline ones, I mean the ones that take two weeks. As the person who makes such desicions at work, we will be using 2SC5150s for outputs in all Sonys that use the 2SC5148. I have not yet decided on the ones using other HOTs, but will definitely go to the highest transistor in the "family", just as I use a 2SA1302 for a 2SA1301. If a "1303" were available I'd use that (I think there is a 1303 but it's not the same family). Our shop will also not be buying 2SD1885s anymore, all 1887s from now on.
So far the "kit" will have at a minimum the HOT, the driver coupling cap and I haven't decided on what else. The "top" tuning cap might be a good choice, so if anyone hass seen a certain value of these going bad I like to know it. Coils as well, perhaps a certain coil is found to be bad in an insidious way such as when hot it would arc, or even open up a crack with thermal expansion (or whenever it damwell pleases).
One time between the other tech and me, we had replaced EVERY component in the horizontal circuit and the ($$$) Sony tech line told us about the CRTs. Some kind of internal leakage path.
Of course I can think of two things that were overlooked, the 503Khz crystal and it's caps. I don't think I'd want to go with a jungle IC every time. I would throw a crystal in though due to the low cost. I believe that today's technicians need to figure the cost of the parts relatively when it is difficult or impossible to determine which is the culprit.
Now how to isolate that pesky CRT leakage condition. Even without an HSTAT adjustment there is still an HSTAT element in a bunch of these tubes, so without making a jig socket to ground the terminal it's not so practical to run them with the CRT socket off. That can easily be solved, an even quicker way might be to simply interrupt the filament source, but then is that test conclusive ?
Now comes the idea of test playing it with the anode cap removed. It would certainly be conclusive. At least we can be pretty sure that it's not the CRT if it blows again. Then what if it doesn't ? Run it with the anode on and the filament off ? I'm sure the HSTAT will cause somewhat of a load so everythings a bit warmer. But then what to consider a limit ?
You know I don't normally ask questions of this type, asking for what might be called a aggregate of case histories, but I have to consider economics. It is getting to be too much pickup/delivery, having the sets at the shop, processing even the minimal paperwork for an inhouse warranty, etc. We can actually come out AHEAD charging the same amount, and maybe even less for these repairs!
When there is a situation where none of the parts are bad right now, a technician cannot determine the problem no matter what his skill level. I'm sure many of us have looked at the schematic and ordered parts for at least some of these jobs going by whay takes the highest voltage or current, with the requisite focus on caps.
All are invited to bring in their ideas, and I will of course report the results. I think others may benefit as well, you might have a sonofabitch on the bench, or even under the bench right now and say "hmmmmmm" after reading someone else's post.
It's hard to say just how prolific those leaky CRTs are, there's one in now I did and it failed on the bench. I saw it and I did not see interrupted H scan, you know it, it looks like the medical sign or something. Instead it was just a very short round spot that was quite out of focus. No bright raster with retrace or anything like that. That can lead me to believe it's not the 503Khz rock, then remember that failure mode. I have described it before but will include it here. The oscillator stops when the countown chain has the HOT in the ON state, current ramps up too high. The field collapses in the HDT and the HOT turns off with a terrible slew rate, but dissipation is not the problem. After a much longer rampup of current the "flyback" (remember the reactive scan theory) is REALLY flying back. No shutdown circuit could catch it, it's simply too fast. Any last vestiges of oscillation would be completely damped.
You might not see the oscillations of the H yoke circuit, since the HOT is an effective short even before it shorts out.
To actually cure these elusive problems, those that won't go away, I know of two times where the crystal was 100% at fault. Umm, three. And I mean for sure, those who know me in here know that when I say for sure, I am sure, there are no unknowns. I'll say in about 20 units where It may have been the rock, but I am NOT 100% sure. Those three cases, I AM 100% sure (one was a Sharp, one Sony and one RCA CTC169).
I'm basically looking for the bit more expensive parts known to fail, and the cheap parts that were hard to isolate as a problem.
Anything is appreciated and with any luck people will be helping each other as well as me. I will reciprocate.
Thanks in advance.
JURB