Predicta vertical integrator replacement

[AmigaMan]

Hello all.

I'm finally getting back to working on my Predicta Holiday. Got a nice and bright CRT, but no vertical lock, and weak horizontal.

I've only done the basic recap so far, so I haven't yet looked into the resistors and couplates. They're next, after a long period of waveform tracing and value checking.

My question now is about the vertical integrators used in these sets. It's a part of the couplate K3, and the integrator itself is indicated in the Sams as a resistor with a capacitor plate across it. I've seen these before in other sets, most recently in my Zenith porthole. In that case I had an NOS spare, so I didn't need to dig into it too far.

So how might I be able to build an equivalent with discrete components? I've read about others making them out of a series string of 3 or 4 resistors with capacitors to ground between them, and I think that would work fine as long as the filtration covers the correct range with minimal gaps.

The values for the resistor and capacitor elements in the original integrator are clearly stated, and they are 90 kΩ and 4000 pF respectively. I can easily divide out the resistance, but capacitance I'd assume would be tricky since the parallel array of capacitors is separated by resistors.

So have any of you made a good K3 replacement, or vertical integrator replacement in general?

[Phil Nelson]

Yes, the vertical integrator couplates are not hard to build. For that component you can use a 90K resistor with a cap on either end, or two 47K resistors in series, with three caps. As long as the caps are pretty close to the total value, it should work.

Phil Nelson
Phil's Old Radios
http://antiqueradio.org/index.html

[DaveWM]

if the horz lock is weak I would not suspect the vert integrator, but rather look around the sycn separator tube. scope the plate of that and make sure its clean.

[snelson903]

thats interesting you are talking about that ,the same issue has showed up in that 24mc32 zenith iv been putting back together k1 k2 both are crumbling apart . i was going to try to make up something or try to hunt for two replacements , but i can't find any info about the values of them. resistance / capa. i dont no if my sams photofact is comeplete ,or are they just giving a part number and thats all .

[Don Lindsly]

The dark red Eric Pacs are more reliable than the beige Sprague flat packs. The later Erie Pacs are not insulated and intended to be repaired.

What is the part number? There are diagrams by part number in the parts book.

[snelson903]

K 1 -the zenith part # for vertical integrator is 87-4 / aerovox pa-765 / centralab pc-408 K-2 the zenith part # for vertical feedback is 87-5 / aerovox pa-765 / centralab pc- 408 if i could get aliitle more description of the values i could probley make something up ,i could vary gently try to measure the orginal resistance / cap. but i dont really trust it ,not with the outer shell crumbling like that. its seem the other ceramic cap's that are made the same way are all crumbling and are crack up too, they the brown ones with yellow painted # on them a / b 3 wire type all others are like new condition.

[josephdaniel]

You may find this helpful for rebuilding the integrators in zenith sets http://boxcarcabin.com/integrator.html

[snelson903]

Originally Posted by josephdaniel

You may find this helpful for rebuilding the integrators in zenith sets http://boxcarcabin.com/integrator.html

yeah that looks like what i need. thanks wear did you find that at ( boxcarcabin .com ) never hurd of some of these websites

[josephdaniel]

I had to replace one on one Of my sets and I asked about it and someone replayed with that link and I used 1kv disk caps

[snelson903]

that resistance value 33k - c=680 pf is that the overall finished value, or is that each componet value. like when its finished total reading will be 66k c=2040pf. i went down and measured the old one for k1 it came out 51k so they must be each seperate value. i went ahead took a measurement of k2 it was 166k that one must have drifted way off value.

[josephdaniel]

I believe that The finished value of the whole thing should come out to 66k and 2040pf but I could be wrong since I have not even tested that chassis yet.