She's lookin' mighty purrty.:yes:

As Someone who knows NOTHING about the Electrical stuff...You GO , Kid...Save another 621 TS....We'll congratulllate you when the stuff separates out...

Fired the set up for the first time and got a surprise. C128B (one of the main filter caps) blew up in a cloud of foul smelling smoke.
**THIS IS IMPORTANT TO ANY ONE WHO WILL READ THIS AND START A RESTORATION OF THEIR OWN**
The Rider schematic has a misprint. It incorrectly states the value of C128B.
http://farm8.staticflickr.com/7008/6...1e2c9a5be7.jpg
Why is this important? Because C128B is supposed to be rated at 450v not 150v and I'd like to save you the :mad: look your wife will give you when you smell up the house.
After replacing the offending capacitor I fired the set up again. No explosions :), however I didn't hear the horizontal oscillator so I investigated in that area. After swapping the 6SN7 out (it was one of the few non RCA branded NOS tubes) with another tube in the set I got the characteristic sound of the oscillator.
I took some voltage measurements and every thing was in the neighborhood so I installed the crt fired the set up and was greeted with.
http://farm8.staticflickr.com/7001/6...bc8e8f6d7c.jpg
No video or sound just yet but it seems like a good time to quite for the evening.

I would restore this thing to CURRENT standards...it's too valuable to restore, IMHO, to 1946 standards...

That is the money shot of any vintage set as it tells you the electronic unobtainium is in good shape.

Great job so far!

And don't worry your not the only one whom has popped a freshly installed lytic. Heck I've done it twice now. LOL

I've done it once before also, but that was entirely my fault. Doing it because the schematic was incorrect is a bugger :rolleyes:. I'm sure it wont be the last time I run into that in my life and as I get better I'll be able to spot those mistakes before I make them. Had I analyzed the schematic and thought about where that capacitor was going/what it was doing instead of just going on faith I probably would have known it to be incorrect.

On multi-section cans I tend to check with the labeling of the can and the existant wiring to make sure of my work.

The first time I made that sort mistake I blew two lytics in a row because I could not find a way to get the chassis loose (the mounting nuts were recessed out of reach of a wrench), and a rubber AC mains line had disintegrated and shorted to one of the above chassis lytics.
The second time was a TV that I'd miss wired the replacements for the selenium doubler rectifiers on.....I was powering it up on a variac, went to have dinner, and started hearing a regular popping sound as the two sets of 5 22uF caps I had put in parallel to make up a replacement for each doubler cap started firing their guts out one by one.

Well I have run into my first stumper. I'm having a problem with the vertical. There's a short video of the problem in the link below (the set is on it's side btw).
http://www.youtube.com/watch?v=WXtdZeuZpO8
I've gone through the resistors in the vertical section and they all read within 20% of spec. I've also swapped the 6SN7 out. The problem kind of looks like the oscillator is running too slowly but the dual "rolling" lines have me stumped. Any one got any ideas?

What does it look like with a regular picture?

Looks like the vertical is not locked, and it's got a sine wave rolling through it. Scope the sync section, see if you can catch the sine wave. Could be a heater/cathode leak or short.

I haven't gotten that far yet. There's no output from the tuner/if stages. I am working backwards from the crt. The picture above was injected at the grid of the video amp. If you're asking for the same reason I wished I had a regular picture, to see if there is two complete pictures on the screen, then the answer is yes. One of the patterns available on my video injector is three horizontal lines. When I select this setting I get six lines on screen, with the "rolling" hum bars separating them. This is what makes me think that my oscillator is running too slowly.

If you adjust the vert hold control, will it lock to the hum bar? Or does the bar still "roll" irrespective of the v hold setting?

It's tough to explain, when I adjust the vert hold it will start to roll the way sets normally do when you are out of vertical lock. Then I can get it to "lock" on the horizontal bar but it still rolls slowly in the way that you can see in the video. One thing I noticed is that the voltage on the grid of the vertical output (pin 4) moves up and down 5-6 volts in tune with the rolling bars.

Sounds like a heavy power supply ripple is getting into the vert circuit. Possibly one of the filters is miswired(?).

I'll buy that for a $. I'll give a third look at every filter cap and see if I can't find an offending miss-wire by yours truly.

The video reminds me of when my 1953 Zenith had it's vertical working at double frequency.

Have you made sure all resistors in the vertical are in tolerance? Those can really louse things up if they have drifted too far off value.

I have gone through the resistors in the vertical. I even drew the circuit out as I worked through it. That of course is no guarantee that I didn't miss something, but they all tested within 20%.

Checked and rechecked the filters for a miswire but came up empty. I've gone through the resistors from the sync tube through the vertical output tube and they are all where they need to be. This one's a real :headscrat.

I recently acquired a B&K 1077 so I can show what a regular picture looks like now. I had prevously thought that the vertical was running too slowly but you can see now that is not the case. The horizontal white bands do not roll with this signal, instead they are steady right where you see them.

http://farm8.staticflickr.com/7004/6...93536124b3.jpg

Any thoughts?

I know you have probably checked this but going by just the picture it looks like a power supply cap has let go.

It sure looks like a huge power supply ripple. The bars not rolling would be the case if the B&K's vert sweep is locked to the powerline (whereas a TV station's vert is offset slightly, which would cause the bars to roll). Possibly one of the replacement filters has gone open.

If you've got an extra 'lytic cap of equivalent rating as the main filters, you can jumper each one and see if the ripple disappears.

Or alternately if you have an analog AC voltmeter, you can measure ripple. Just put a cap of .1mf or so in series with the meter's hot lead (to block the DC component). I'm not sure i would trust a digital meter in that application.

I did try adding a cap to each of the electrolytics except for the 1,000mfd cap as I do not have a replacement for it.
How would I measure ripple with an analog meter? I have an analog VTVM and a .1mfd cap. That is to say what would I look for?

First i think it'd be a good idea to see perzackly what's connected between the center tap of the power xfmr and ground. Some of those filters' negative sides are insulated from ground and are connected to the CT. Just for the heck of it, take a resistance reading from that CT to ground. How much resistance does it show? If by chance it's inadvertantly shorted to ground, there would be some hairy ripple and hum bar effects.
In that pic of the power supply area that you posted earlier, it doesn't show where the center tap goes to (kinda fades into the twilight zone). Is it possible you could post a pic showing more of that area?

I've got some readings on the electrolytics and a better picture of the schematic.
http://farm8.staticflickr.com/7157/6...577fae063e.jpg
There are 5 multi stage electrolytic capacitors in the set. Towards the top of the picture, just above, and just to the right of the vertical centering control you can see two sections of C116. The negative on this cap reads shorted to ground which according to the schematic is correct. The other four caps are isolated from ground. Coming right off the center tap of the power transformer is the negative leads of C127, C128, and C131. Further along the line you have C132. Rather than give it away, how would I read the schematic and figure out what resistance reading those four caps should have to ground?

Hmm.. so the power xfmr CT is the negative voltage source*, supplying 110V negative for stuff elsewhere in the chassis. And the main filter for it is the 40mf.cap (square symbol). Can't read it too good, but i b'lieve it's C128A). The positive side of this cap should go to ground. Is this wired correctly? If it's not, the negative supply would have a large ripple on it. So check this first, and then check for ripple. And verify that there's 110V negative on it (+/- 10% or so).

To check resistance of the -110V source to ground, just look for a DC path to the nearest ground symbol in the schematic (hint: it goes thru the ion trap coils), and add up the values of the resistances in that path. And see if the sum agrees with the ohms reading you get on the meter (+/- 10% or so).
------------------------------------------
*This technique is called "back biasing". You'll find it used in old equipment quite often, particularly in 30s radios. The CT of the power xfmr winding, instead of going directly to ground, goes through a chain of resistors or a large tapped resistor to form a voltage divider, supplying negative voltages for biasing other tubes in the set. A Google under "Back bias voltage" should turn up some good info on it.

Such an easy answer to my question. Seems silly that I had to ask, but now I know. All of the resistance readings on the negative sides of the capacitors read correctly.

This is wired correctly, and there is -110v present.
I believe that I am checking for ripple, but perhaps I have the test wrong. I have put my meter on an AC 4v scale. I've connected the ground lead to chassis ground and the probe through a .1mfd cap to negative on C128. Is my test correct, and am I seeing ripple on the meter?
http://www.youtube.com/watch?v=b5BBxQlqp1o

Yes your hookup is right. It's showing just a little over 2VAC of ripple. That shouldn't be anywhere near enough to cause the huge hum bar effect you're seeing. This pretty well eliminates the negative supply as the source of the problem.

So check for ripple on the positive supply, first on the 'upstream' side of the spkr field* coming out of the rectifier (there'll be a large ripple here) and then on the 'downstream' side (top of C127A). Then at the top of C116B. The ripple should smooth out to just a volt or two and get progressively smoother the further downstream you go, such as at C116A.
-------------------------------------
*The speaker field serves double duty as a filter choke, a common practice in old equipment.

Yep, this is exactly how the set is behaving. Progressively less AC as I move further downstream. By the time I get to C116A the ripple is less than a volt. A real :headscrat for sure.
I thought outside the box today and considered that there was outside interference so I unplugged my LED work light transformer and every thing else on my bench to see if the problem was external to no avail.

Have you considered an H-K short in the CRT?

By golly it looks like power supply ripple has been pretty decisively ruled out. I went back and reread the thread and looked more closely at the pics. Post#43 shows the raster as having nice straight sides (no hourglassing), indicating no appreciable ripple in the B+ feeding the H output stage (good). And post#58 shows the CRT displaying strong video which would rule out a heater-cathode short in the CRT (which would kill the video).

So.. it kinda looks like the hum is arising somewhere in the video chain 'upstream' of the CRT. And since the sync is affected too, the hum must be occuring upstream of the sync take off point. (Just guessing, but the sync may be picked off at the vid. output which was common practice in the old stuff.) It could possibly be nothing more than a H-K short in a tube. Sometimes tapping the tubes will show up an intermittent that won't show up on a tester.
If it ain't a tube, maybe you could post a pic showing the video amp and sync areas of the schematic.

Occam's Razor: you've been making cap upgrades which means you've been messing with the wiring, find the wiring fault and repair it.

Finding the wiring fault can be fun and challenging if you make it so. :)Also helps a Noob hone his skills. Sorta like in the old days an electronics instructor would deliberately introduce a fault into a piece of gear.

Tell me about it. When I went through AE 'A' school in Pensacola, I was forever finding faults with their trainers. Then again, they were as old as most of our sets...

Check your work young student! Good advice for every restoration and I should have done it sooner. I traced every wire on every connection that I had my hands on and three hours later I'm vindicated but sad to report that I have not found a single wire/connection out of place.
I tried a few taps on each tube and no changes there. Maybe I can retest the tubes for HK shorts.

It looks like the sync is taken off of the plate of the 2nd video amplifier. The video at this point carries the sync pulses with it to the sync amplifier which would magnify these signals. The sync separator would remove the sync pulses from the video (or does the term "separate" come from separating the horizontal pulses from the vertical?) and pass them on to the vertical and horizontal output sections. If this is correct and the problem is originating in the video section, is there a way to isolate every thing upstream from the 6SN7 sync amp/separator to quickly confirm this? Would removing the video amplifier tube achieve this?
I could be completely off base on my thoughts above but just trying to think it out as much as I can on my own so I can learn.

http://farm8.staticflickr.com/7027/6...6ab1ee59_z.jpg

http://farm8.staticflickr.com/7025/6...5bb984c5_z.jpg

You can isolate everything from before the video amp by lifting the grid connection (V105 pin 1), then injecting composite video right into the video amp on pin 1. If the problem goes away, the issue is in the tuner/IF. If you remove the sync tube, the picture won't lock at all and I think it would be difficult to tell if there problem were still there or not.

One last thought.
If the problem is coming from the video section would it still manifest itself with no video signal present?

Yes it would, assuming the raster has still got that heavy white bar in it. That bar's still present, right? If it is, there's another trick you could use that doesn't involve lifting any connections or injecting a signal (the bar is the "signal"). Get that trusty .1mf cap again* and hook one end to ground, and the other end to a clip lead. Starting with G1 of the CRT (this set injects video via the grid instead of the cathode), and connect the cap to it. This shorts the signal and should produce a blank or near-blank raster. Then take the clip lead and progressively work back, shorting the plate of the 2nd video stage, then the grid, then the plate of the 1st video, then the grid, and on back through to the plate of the 6H6 and then the cathode. If at any point the bar remains unaffected, you'll know it's originating in the stage you just left, 'downstream' of the shorting cap.

Be sure and discharge the cap to ground after taking it off the plate of a tube!
-------------------------------------
*If the single cap is not killing the bar sufficiently, and if you have more than one of those .1 caps, put as many as you can in parallel to get higher capacitance for better shorting effect.

When you say blank raster, do you mean just a normal raster minus my white bar problem?
Shorting (through a cap) G1 to ground makes no change in raster. .1 didn't do it so I moved to .25 which didn't do it so I jumped all the way up to 10mfd which also had no change, aside from the fact that it was a bit more dramatic to discharge. Assuming I didn't do this test wrong would that put the problem in the CRT?
Fearing I did the test wrong I injected my pattern generator on the video amp to ensure that when I shorted G1 to ground that the video went away, which it indeed did.

Da-yum. You're sayin' that even with the CRT's G1 clamped to ground with a 10mf cap, the white bar remains unchanged? That means the bar is definitely not originating in the video chain (see how one quick test can eliminate or confirm a whole signal chain?:)). Does the bar roll if you turn the vert hold control or not? I'm gonna take a SWAG that the bar is locked to the vert sweep, in which case it will not roll. And another guess that the bar is actually appearing on the CRT's G2.. and that since G2 has boost voltage on it, something's allowing vertical pulse to get into the boost source.

Try this- do the ripple test thingy again with the .1cap, on the CRT's G2. I'll betcha there's some signifigant ripple there, causing the white bar.

Of course i could be completely wrong. :worried:

The CRT appears to have a partial aquadag coating on it. Is that coating securely grounded? An ungrounded or poorly grounded 'dag can cause some really weird symptoms mimicking sync and AGC problems (a little factoid worth bookmarking).

Oldcoot88: "Is that coating securely grounded? An ungrounded or poorly grounded 'dag can cause some really weird symptoms mimicking sync and AGC problems (a little factoid worth bookmarking)."

Good one! A spring arcing in a curb-found 1961 Philco BW, left the room, and the CRT "blowed up". Made a hell of a mess out the back. Cant be too careful grounding.

Dang.:eek: First time i heard of that happening (other than those SS color Zeniths with the 4-legged Safety Cap which would fail, letting HV soar to over 50KV, cutting the neck off the picture tube).