I just won this off Ebay, a mint condition WR-64B for only $50. All original tubes, chassis you could eat off of. But something's wrong with it, the rainbow pattern isn't right. It starts off red and has lots of blue, no greens. I've tried the alignment procedure in the manual, but nothing so far has helped. Crosshatch doesn't look the greatest either, maybe more caps need replaced. Anyone with one of these share their experience?

Not having a schematic or other resource, I would bridge the electrolytic with a good one.
I have seen some realy strange symptoms and behaviors because of a bad power supply filters in test equipment. It might explain the varied or multiple symptoms.
If it looks messed with in any way your best bet is if someone has a schematic. If not you could carefuly shotgun all the paper caps. Good luck

I don't think it's a power supply issue, there's only around 1.5v ripple and the lytic is cool as a cucumber. This was a later production unit, I'm guessing '67 at the oldest based on the tubes in it. But I suppose I could give it a try, can't hurt anything.

http://bama.edebris.com/manuals/rca/wr64b/

manual & schematic.

Looks like a lot of stuff dependent on that 189K osc....

If too many symptoms, yes check the ps & voltages first.

Caps don't have to be hot to be bad.

I would also check that the 3.563795 mhz crystal is on frequency.

The 189K osc most likely is related to placement of the lines, dots, and spacing. The 3.58 of course is the color burst osc. So it will give you correct colors, and stable colors, not changing or barber poling. If you have evidence of both the bars, or dots being wrong, and some type of color problems, then you have one common problem, or two or more problems. Be neat if you could post pictures..... It could even be off channel frequency, you using RF on something with fine tuning?

It's not the RF section, this unit has a blank audio function to tell you if it's tuned in right or not. I played with the tuning slug, and it's definitely right on ch 3 where it should be. Posting pictures wouldn't do a lot of good, it really has to be video to see what it's doing but I'll try and describe what I'm seeing.

In dot mode, the dots dance around a bit as if there's a power supply problem. They never stay right where they should, like they do with my modern silicon powered dot/bar gen. In crosshatch mode the horizontal lines look good, but the vertical ones are shimmering and slightly wavy.

Bars actually look decent, just not the right colors. According to the front of the machine it's supposed to start out orange, but instead I have red. At the far right of the screen where green should be I instead have blues, so it sort of seems like one or more of the timing circuits aren't at the right frequency or something. I don't have a counter at home, but I do at work. Maybe I should take it in and do some fiddling?

I still have a color bar generator for sale, $25 plus freight.

It's not the RF section, this unit has a blank audio function to tell you if it's tuned in right or not. I played with the tuning slug, and it's definitely right on ch 3 where it should be. Posting pictures wouldn't do a lot of good, it really has to be video to see what it's doing but I'll try and describe what I'm seeing.

In dot mode, the dots dance around a bit as if there's a power supply problem. They never stay right where they should, like they do with my modern silicon powered dot/bar gen. In crosshatch mode the horizontal lines look good, but the vertical ones are shimmering and slightly wavy.

Bars actually look decent, just not the right colors. According to the front of the machine it's supposed to start out orange, but instead I have red. At the far right of the screen where green should be I instead have blues, so it sort of seems like one or more of the timing circuits aren't at the right frequency or something. I don't have a counter at home, but I do at work. Maybe I should take it in and do some fiddling?

Not sure about the WR-64B, but in most color circuits, the phase of the signal determines the hue. Old Motorola color sets would display the color bars incorrectly - the problem was a uuf (pF) cap that caused a phase shift when warm. The gripe would be most noticeable when watching a football game- the grass field would shift from green to blue.

If it is the phase, look for out of tolerance parts - most notably resistors and capacitors, then check the slugs in the coils with a bright light - cracked slugs were a problem in early GE Color sets - the ferrite was inferior and would crack, shifting the color phase slightly, giving you either orange greens or color bars drifting through the picture. Years later, RCA had a problem with slugs that lost some permanence (their ability to behave magnetically) and you'd get garbled audio from the phase shift in the discriminator circuit. Rock the quad coil slug slightly, and the sound would come back to normal.

I'm also not sure of the tube complement -but if one or more tubes were supposed to have a shield, make sure they now have the shield. Also, put the chassis back in the cabinet of the unit to test - the case might affect it slightly. I have an Eico sig gen that will shift frequency slightly with a pressing action around the top handle.

I've never owned an older RCA color bar-dot generator, but I have an RCA WR-504A of Dad's that is rock solid (solid state) for going on 39+ years. Great little unit, used a mercury battery for operation, we converted it to NiCd.

Cheers,

You say at one point in your description some of the lines are wavy, for this I would go to the power supply. You say it was recapped once, well look for a bad cap. Someone recapped this unit and chose the wrong caps, or got bad caps and now you need to do it again....

Does your good generator have gated bars? Post pix of the gated bars from the good one and the new one.

It is pretty hard to get wrong colors from one of these, because if the crystal is at 3579545-15750, there is automatically one cycle of color from left to right. The first burst of subcarrier acts as the color burst, but actually it's changing phase slightly during the burst since the frequency is deliberately different. Then the subsequent bars are each shifted in phase.

One problem you might look for is poor horizontal sync causing the burst gate in your receiver to be offset. Try rocking the horizontal hold of the receiver and see if the hue changes. If so, look for a problem in the H sync waveform out of the generator.

Edit: the 189 KHz is supposed to generate the H sync pulses (when divided by 12) and the gating for the color bars. - So, if this is way off, it might cause some trouble too.

Wayne,

I think you're on to something!

I bypassed the RF section of the generator, and ran composite to my monitor. The monitor has the advantage of telling me when it's receiving an NTSC signal, so I should know when it's on frequency. Trouble is, it never really gets to the correct frequency. So it looks like I need help troubleshooting the divide by 12 part of this thing, but the trouble is I don't have a reliable way to measure the frequency. I do have a meter with a HZ function, but it loads the circuit too much throwing it off. Right now I'm running off the assumption that the main 189khz gen is good, because it's crystal controlled. So I'm thinking if I can figure out what's up with the horizontal sync section I might be able to get this right.

It looks like an off-frequency signal when viewed on the monitor, a lot of diagonal lines across the screen but I can't get them to line up no matter what I do.

Do you have a scope? You could just check the period of the horizontal rate and divide 1 by the period to get frequency. Accuracy might not be perfect, but would tell you if you are off by more than a few percent.

Edit - or maybe h sync is just missing

Actually, it was more bad caps- maroon drops!

I had a CTC-9 full of them that worked flawlessly, so I assumed they were at least as good as orange drops but I guess I was wrong. Or probably they were fine as coupling caps in the video path of a receiver, but they sure upset the sensitive timing circuits in this thing. There was slight DC leakage from the 31.5 and 15.75 sections, which was sporadically throwing off everything down river. You could hear it randomly changing frequency, and that made any sort of scope alignment impossible. Never mind the pictures of scope traces in the manual are ambiguous at best...

In the end I fashioned a low capacity probe for my DMM, because trying to read frequency directly was loading the circuits and changing the frequency. With a .001 cap and 100k resistor in series with the DMM probe, I was finally able to get accurate readings and went through the alignment that way. Next time I hooked it up, I saw that beautiful rainbow of color! :thmbsp:

Seems pretty stable now, so on the shelf it goes till I need to use it on a set.

http://www.videokarma.org/attachment.php?attachmentid=181586&d=1391141953

Beauteous!