My Predicta crapped out
 

Howdy All,

Well, tragedy struck last night. We were watching "What's My Line" on the Predicta Holiday and the picture was a little unstable. No adjustment helped. It was just jiggling, it didn't seem to be the horizontal lock. I have noticed a little occasional jiggle over the last few months, now that I think about it. The jiggling got worse and then *poof* we lost picture. I can't say if we lost sound because we have been running the sound through the stereo and I was on my feet and hitting the power off as fast as I could. This morning I tried turning it on. It has a picture, disturbed as it was last night. I am guessing if I left it on it would go dark again. Sounds like something is getting warm. The set smells fine, no hot smell.

Some background: I "restored" this set, finishing about 4.5 years ago. "restored" is in quotes since it was my first and only TV that I have worked on. "Got it working with the help of this forum" might be a better description. It has been recapped, new tube sockets installed on the PCB, new couplates, new diodes, replaced the fusible resistor. It has been working fine for years with hundreds of hours on it. I am hoping for some help with troubleshooting, I am not an expert and my theory is weak.

Many thanks in advance!

You may have encountered a random failure of some part you did not replace in the last 4 years, you did real well for a Predicta. :thmbsp:

Anyone could assume the worst-case: flyback arc-short is possible as I have witnessed color sets behave this way prior to losing HV for good, which yours didn't, fortunately! But shorted turns (AKA internal spot-welds between adjacent windings) never fix themselves.

Your HV came back because whatever part was failing decided to heal itself on a cold start. Get some freeze spray after all the parts around the horizontal oscillator. I am assuming H - jitters caused by H-hold or H-range pots being dirty or having tin-whiskers. I hope its that simple for you.

There are some Predicta-savvy techs here, but I avoid THOSE Philco products myself.

Most all us Philco-techs would agree: OEM Philco caps/resistors/pots are all SUSpect until confirmed otherwise:sigh:!

You are enjoying the true vintage Television experience of having your set die mid program, just like it was back in the day. :D

Reading your description I wonder if it has a bad tube in the Horizontal section, oscillator or output. You might tap around on them and see what happens.

Eric H., you reminded about one of my TV repair customers in the mid 1960s. You will remember that at the end of every show Red Skelton would look into the camera and say, "Good night and may God bless." My customer was watching that show and the instant after Red uttered those famous words, the screen went black and a puff of smoke came out of the back of the TV. Flyback!

If it repeats that you probably want to have meters on the grids (G1 and G2) of the h output, and possibly meter the HV.

I few weeks ago I had something similar with a CTC-25. I powered back on and it was fine for a bit, then a few sparks flew inside the output and as the HV faded the damper began to redplate...Next power up it was dead and the output wouldn't light in my tester....The output base cracked and it had filled with air (hadn't had enough time to get white getter yet).... Surprisingly the damper was fine and with a new output it was happy as a clam.

I've got 3.5 of the Predicta tandem/penthouse model on my bench right now. One of the 4 is necked and I'm going to be hard pressed to make 3 very nice cabinets out of the 4. Three chassis have life so far so once I work out the kinks I should have a keeper and 2 flips. Not sure what to do with the last one, build an extension cable for the HV, try to make a color set out of it, keep it for parts, or sell it cheap?

Pulled the chassis out and ran the set again. All seemed normal. I wiggled tubes and aside for one in the IF that needs the socket cleaned, no problems. The picture disturbance was the same. After about 10 minutes the screen went dark. No hot smell, noise or any sort of drama, just poof. The flyback was cool and no evil smells.

The TV and the work bench are in different houses so I will have to grab some meters and come over and check voltages. I do have an HV probe, not a great one but it will give me an idea. I will do as you recommend Tom. Will be next week before I am back here, I live a nomadic life and only spend 2-3 days a week where the set is.

If I have to drag it back to the house with the work bench I will but I sure hope not. Not only will I have to bring the set but also the RF modulator and all that jazz. I have been meaning to get some sort of test jug so I could just bring the chassis. I also have a spare chassis I would like to restore, then when this sort of thing happens I can just slide the other chassis and take the other to the bench.

Any other simple things to check?

Does it have high voltage after it goes dark?

Unknown so far. Bench and tools in different places. Will bring some meters and probes for next week and find out.

If you put the tubes on a tester let them sit for a while and see if they change readings and also at this time check for shorts.

When the set goes out are the filiments still lit?

My stab at it

Thermistor going open once it warms.

Sand resistor goes open when warm.

Weak solder joint in a area that has higher current or high heat.

The way you described the flickering I wonder if you're losing horizontal oscillation, and in turn losing horizontal output, HV and B+boost.

I would start it back up and see if the filiments go out, and if they stay lit move on to the horizontal section.

I assume loss of pix means "went black". Almost always thats loss of HV.
Easy first. There are 4 tubes that will cause it. Hoz osc, hoz output,
damper, & HV rect. I have never seen a HV rect go int. so leave that out.
With the set running lightly tap the tubes & see what happens.
This is usually better than a tube tester for this type fault. You can
also do it on a checker to smoke out an int tube but having the TV raster
works BEST !

Freeze mist was mentioned also. A real life saver sometimes. Best way to
use it is spray it on a Q-tip & touch it to suspect parts one at a time.
It will almost always be a cap.
BUT keep in mind if you chill down some caps to much they will drift
out of tolerance & act up but still be good.

good luck
73 Zeno:smoke:
LFOD !

Thanks for the help!

I am at work for a 72 hour shift but should be back to the set on Friday. I will bring meters and an HV probe. I will try the tube tapping, I did wiggle them. The filaments do remain lit when the screen goes dark. Sand resistor has been replaced when the set was restored. The thermistor is new but you never know so I will check it.

I think there has been a progressive deterioration. It started with some screen jiggle a few months ago. The first time it went dark we had run it for 20 mins or so. Last time I tried it took 10 minutes or so. I have plenty of spare tubes so when I go there this week hopefully I will find the fault or at least gather some more clues.

Finally where the set it. Now when I power up I get nothing.

Swapped out V8, V10, V11, V12, V13. No change. The horizontal output tube looks like it might be red-plating a little but I am not sure. It is just a dim orange.

Voltage at the rectifiers is a little low, 247 instead of 290. I couldn't find my HV probe before I came over to where the set is. How did I misplace something 12" long with a cord on it? I dunno. Anyway, I measured the flyback resistances and all are exactly as they were when I restored the set.

What do do next? I guess start trying to track down the low B+ voltage. I think I have a test socket I can put on the horizontal output tube and do some looking at the voltages there.

Voltage is seriously low. Look at power supply first.
Next check the voltages on the hoz output tube.
DO NOT measure the plate cap !
The G-1 voltage is most important. It needs to be very
( see manual) negative. If not look to the Hoz osc stage.

73 Zeno:smoke:
LFOD !

the voltage on the plate cap CAN be measured, if IF it is unhooked from the HOT and there are NO horizontal pulses, it should be aprox the same that is going into the damper, the danger is that ANY horizontal pulses or feedback from the FBT can toast your meter, but if horizontal not running, and its unhooked, it's usually safe.

True. I tend to assume OP hasnt had 40+ yrs in the entrails of a TV. So
I try to keep it simple & safe, one step at a time. In a perfect world
the OP, TV & me would be all in one place with a great parts stock &
the best of equipment. Without that its a challenge for all of us to help.

73 Zeno:smoke:
LFOD !

The HO tube beginning to red plate proves there's voltage on it. With the cap disconnected, the screen grid will get cooked in short order.

For the very short duration that the damper heats up and gives power through the FBT to the plate cap, long enough to be measured is what I meant, there is really no risk to the screen grid or its resistor, it's not like you are going to leave it on any longer to take the voltage measurements after it first comes up, 30-50 seconds or so, I have done it many time, to confirm that power IS getting through the damper and FBT to the HOT OK.

Correct you are amigo. :yes:

The low B+ could be from the horizonal output tube red plating and loading the B+ down. I would re-measure the voltage at the rectifiers with the horizontal output tube removed. I'm assuming this is a series string set with 2 rectifiers in a voltage doubler circuit?


Any chance you can post up the schematic?

Sorry for the delay in replying. It has been a crazy week and now I am at work for a 72 hour shift.

I will figure out a way to get the chassis on its side to take voltage measurements. I could set up a jumper on the output of the HOT to the flyback. Then I could use the test socket.

Here is the schematic:

http://www.earlytelevision.org/pdf/p...sams_439-1.pdf

Thanks for the help. I know that slow progress can be annoying to those helping. At best I have one day a week I can work on the set.

You really need to hunt up your HV probe to see if you are really getting the 14-15kv or not, and check for the 435 boost. It looks like point C can be unhooked, at L14 point 20, If so, and you DO have HV and boost, unhooking that and grounding it will turn the CRT full on, giving a full raster, IF you have full scan that is, this will rule out any video problems.

I will do another look. It has to be around here somewhere! But I did do the CFL lightbulb test and it showed nothing. I forgot to mention that.

I am at work but will look at the schematic to understand the test you suggest.

It all really depends on if you have HV or not, if you don't, your HOT may be overloading the bias as someone mentioned, this could be from a problem of the FBT side or more likely a failure of the horizontal oscillator, the Sams shows that the voltage on the g1 of the HOT is -45v, this should start positive, then go negative as the horizontal oscillator starts up, if it does not go to -45v, it's a good sign the the oscillator has failed, suspect the couplet K6, if it's not replaced if the oscillator is not starting.
Example low or no voltage pin 6 on V10.

Oh, if the bulb doesn't light you won't have HV. It also indicates that the HO tube isn't switching any current to speak of.
Most probably the horizontal oscillator isn't running or is way off frequency.
It could very well be that the oscillator signal isn't make it to the HO tube.

I know this guy and he does excellent work. I think he still makes replacement couplet kits.
https://tvrestorerguy.com/

Thanks notime. I will be back at the set in a couple of days. The couplates have been replaced with ones made by banderson when I did the set 4-5 years ago. There might be one original underneath the chassis, I think I was told that it never went bad.

I ordered a little sains-smart 'scope that will be easier to bring over to where the set is. I'm still hoping to avoid moving it back to where my bench is. I will get some voltage readings this week. Since the FBT tests the same as when I restored it and the HOT is red plating a little my guess is as you have suggested that the horizontal osc isn't running. I will take another dig looking for my HV probe to confirm the lack of HV.

It does sounds like the horizontal oscillator is not running. Another part to check would be the resistor 15K 1 watt (R62) on the horizontal oscillator to make sure it hasn't drifted out of spec. Kind of a long shot, but it wouldn't hurt.

If the HO tube is red plating, it's in self-destruct mode. What you could do is this: leave the plate cap disconnected, and clip the wire feeding the screen grid (pin 4). Then you can troubleshoot without cooking the screen grid. Have you checked the Horz Freq coil (L22) for continuity?

OK, I am here with the set and trying to test a few things.

I made a little jumper so I could use my test socket on V11, the HOT, 12DQ6A. Pin 4 has 105V instead of 175. Pin 5 is just barely negative at -.15 instead of -45. So I think we can say the horizontal multi-vibrator isn't running. I did actually get a small Sains-Smart oscilloscope to look at the horizontal multi-vibrator wave form but without it running at all it seemed pointless to try it.

Looking at the power supply side of things, the junction of the rectifier diodes (M1 and M2) starts at 150V and drops to 126 as the set warms up. The output of M1 is 245V instead of 290. So I decided to see what the current passing into the set is. The schematic shows it as 270mA. Once the set is warmed up the output current reaches 350mA. So I guess the power supply is OK and something is dragging the voltages down.

The couplates, including K6 have been replaced. I have subbed another 6CG7. I routed the sound through the set. I can get some sound with the volume cranked and the local/distance switch set to distance. However, it is weak and distorted.

@old_coot88 I just tried to check L22. There are three tabs visible from the top. Between 2 of them I get essentially 0 resistance. Between either of those two tabs and the third, I get infinite resistance. I though I had a drawing/picture that showed the traces on the back of the PCB so I could see how it is wired into the circuit. I can't find it. I do have a parts set I could dig out and see how it measures there. You may be on to something!

Thanks for the help!

@old_coot88 I think you nailed it. I dug up one of my parts chassis and the inductor reads as it should, 139 out of 140 ohms. So the one on my set is indeed bad.

So, the next question is: Are there other issues that could have caused the inductor to go bad that I should check for? Or is it just a luck of the draw thing?

lt's generally a luck o the draw thing unless there was an overcurrent condition that made the coil go open. Check that 15K 1 watt plate resistor (R62) that's in series with L22. Does it look charred by any chance?

No, the resistor looks fine and it is a metal oxide resistor that I put in. It is probably a 2 watt, IIRC.

Now, would you please tell me where I put the spare chassis that I already canibalized parts from? If I can't find it I can pull it off the one I found but I would rather leave that intact. Then I have to go and disconnect the restored chassis, undo all 28 or whatever wires plus the grounds on both PCB's, pull the parts on both, replace it on the restored chassis, then reconnect all 28 wires. No problem! Why do people say these sets are a pain in the arse?

Ha-Ha!
I try to warn them when I can.

Mad: to avoid all that riggamaroll, is there any chance you could cut the bad coil off just above the lugs (Dremel cutting wheel maybe), leaving the lugs undisturbed, and then piggyback the replacement onto the lugs? just a thought.

I believe the oscillator will run if you short out the coil. That's part of the setup instructions to set the horizonal frequency control. Might not have a stable sync, but at least it should restore the HV.

@banderson Just went and shorted out the coil. We have HV again!

:banana:

Oddly enough the the signal was pretty stable. But there is light on the screen!

@old_coot88 Maybe I could just piggy back. I will look and see if there is enough room. I found my other parts chassis and L22 is open on it. So it seems that this part likes to fail, 2 out of 3 that I have are dead. So I will see if I might be able to repair one of the dead ones first, maybe it is at the end of the coil or only a few turns down. If not I think I will start looking for another as I hate to have no spare for a part that obviously likes to fail. I would assume that there is no realistic way to make or use a modern part there?

Thanks very much for the help!

Sams doesn't give the inductance value, but chances are any two-lug, slug tuned H osc.coil would work if you have a stray one floating around. Just jump it in with a pair of clip leads for trial.

Some sets you can get around removing the board and unwiring it. On the 17" sets and the 21" console with the power transformer you can unscrew the vertical transformer and board grounds, lift the board 0.5-1" and snake an iron underneath (IIRC there are small holes under the board that facilitate access.
The tandems can also be opened like that.I just recapped 2 tandem main boards like that... I've never built a ship in a bottle but I'd imagine it is probably a similar experience.

Like O,C-88 said, “chances are any two-lug, slug tuned H osc.coil would work”.

https://talonelectronics.com/shop/item.aspx?itemid=9543

this is a NOS replacement for my big mama Magnavox, who knows, it may work, if it does not, you are not losing much for that price!

Great news :) The coil isn't a crucial part of the multivibrator oscillator. I believe the RC combo of C43 (the 390pF mica cap) and the horizontal rang/hold controls set the free running frequency.

The parallel combo of that coil and 3900 pF cap are set to resonate at 15,750 HZ and help to keep the oscillator from drifting.

By golly you're right. On closer examination of the circuit, the LC combo is just there for stability enhancement.