21CT55 #2628 restoration
 

I'm finally beginning the restoration of my RCA 21CT55. I had intended to begin it about three years ago after I restored some vintage test equipment. Then I ran across Phil's excellent restoration web site. Then I found the Early Television Foundation site and realized my set was kinda rare. I had always thought it was special, but I had no history of early color TV development to know how few were made and that it was based on the CT100 CTC-2. Then I found Videokarma and learned how much I did NOT know about restoring old TVs. The depth and breadth of the knowledge of the members is a treasure. SteveK sent me a copy of the 21CT55 RCA Service Clinic document that is very helpful. I then decided that instead of starting with the 21CT55, I would warm-up with early B/W sets that I was extremely happy to find were actually available. I studied all of Bob Andersen's YouTube restoration videos and learned a lot. So I hunted, bought, and restored a RCA 721TS, a Dumont RA-103, and a RCA 630TS. I mixed those restorations with a Zenith 6S321 Stars-and-Stripes radio that I have used since the 60's, a 1931 Philco Model 90, 1928 Atwater Kent Model 49, and a 1924 Atwater Kent Model 20 radio. As I learned, I picked up additional test equipment along the way. Here is a picture of my restorations arranged in my living room with my 21CT55 waiting patiently. And a few pictures of the 21CT55. Its time has come. :yes:

Dave

Heh, looks like the HV cap in your blew at some point. Interesting way to make the repair, with them all arranged in a circle like that laced with dope.

Can't wait to see the "Glorious Lollipop Color"....(grin)

I was college kid at the time (1974) and could not afford the right part. It functioned ok, but had a lot of corona, hence the slathered on corona dope. I hope to find a better replacement, but I have not started searching yet.

Dave

The original gray 30KV doorknob on my CT55 also went bad. I found a suitable 40KV doorknob on ebay. Mounting was slightly different, but was easily modified to work like the original. I also see that the discharging assembly that grounds the doorknob has been removed. I installed a low RPM cooling fan on the HV cage to keep the flyback cool.

Here is a link to the restoration pages of my CT55 on my web site.

http://antiquetvguy.com/Web%20Pages/...CA-21CT55.html

Good luck and welcome to the CT55 club!

I bought this set from the service manager at Electronic Specialty Company in Charleston, WV in 1974. My dad worked there as the accountant and I worked there a couple of summers during college. The service manager told my dad that the set was originally purchased by a doctor at West Virginia University through Electronic Specialty Co. They maintained it over the years and when it blew a flyback after the CRT had been replaced the service manager got hold of it when they bought a new set. He replaced the flyback, but had not completed the rewiring. The selenium rectifiers had been replaced with diodes before I got it. I bought it for the $75 from the service manager who had lost interest in the project. I found the 30KV capacitor shorted, so the first thing I did was fix that and a couple bad filter caps. That got the set working but it did not see much use. The CRT is a 21AXP22A dated 1957. My CR-70 shows its got emissions in the good range, so I am hoping I will not see purple in the neck when I get HV to it.

It also has a few of the white peaking coils that Nick found were bad in his, so I expect to be replacing mine.

Dave

Thanks for the link Bob! All advice is very welcome. I have the shorting bar along with others bits and pieces. When I got the set it had a chipped brightness knob, was missing the fine tuning knob, the volume knob, and the channel number insert. Here is a picture of the knobs I have, and another of a new brightness knob, a knob that I think is the correct volume knob, a fine tuning knob that I hope to modify to work, and an original "spare" peaking coil the service manager said I would probably need.

Dave

Bob, thanks for the ebay tip for a capacitor. Would a 2700pF 40KV work ok as a replacement for the 2500pF 30KV?

Dave

I've done some cleaning to the chassis. I used naval jelly (thanks bandersen) for the cleaning and to treat the rusted areas. The chassis "before" picture is above, here are some "after" pictures. Also is a picture of HV capacitors that I will replace with the 40KV 2700pF door knob capacitor to replace the mess. I have not yet figured out how to clean the HV cage, any suggestions? I compared my flyback with the two Bob shows in his restoration page: http://antiquetvguy.com/Web%20Pages/...CA-21CT55.html
I think mine is the same replacement Bob is using.

Dave

Looks like the photo of the fine tuning knob in the picture (if that is what it is) is wrong. That fine tuning knob is probably for a ctc5. You need a fine tuning knob like a CT100 uses. IF has a great big hole in the center (about 1.25" dia)

RE: HV cage. If you have an electroplating shop that does tin plate, I would take it there. Zinc can also be done but it doesn't look as good as tin plate.

Otherwise a simple paint job with aluminum spray paint looks OK.

Yes, I have not been able to find one, so I bought this one in hopes of fabricating a stronger central part and attaching it to the outer part of this one. The channel changing knob will fit within the outer grip area of this fine tuning knob.

Dave

Mine also looks like his replacement, but I'm unsure if it's just that the wax melted off Bob's donut or there actually was a different one. I always figured it was original.

I was judging from the wider looking HV donut on his replacement flyback and its brightly colored lead insulators. Or maybe the old one discolored its leads when it died. Here is a shot of some numbers on one end of my flyback.

All of the original chassis mount caps tested open, leaky, or way off value. The "modern" (1977-ish) replacements that I had tacked in, way back then, all reformed nicely. I am not going to trust them, so I have started re-stuffing the old electrolytic caps with 105C caps. Here are some pictures of a few of them being re-stuffed. I have all but one of the chassis mount cans re-stuffed and I have started wiring them in. The 2700 pF door knob capacitor arrived and I think it will fit in the HV cage well.

Nice job on the re-stuffing!

I also use the same method, removing the can from the chassis and pealing back the rolled edge of the can to remove the tab ring. I much prefer this method over the method that cuts through the aluminum can above the base of the can. It's a lot more work using this method, but this method retains the original can in one piece, pretty much un-molested. And when re-installed on the chassis, it is hard to detect that it was re-stuffed.

On my first attempts I was using a small pair of wire cutters to pry the edge up. It was pretty easy that way, but left ugly marks on the outside of the wider ring at the bottom of the can. With the right small jewelers flat blade screw driver I can force up the edge and then wedge it in, applying pressure along the circumference and then pry up a very short bit at a time. It gives a nice result but is tiring on the hands.

Dave

I use an old Stanley scratch awl with a wooden handle. I have it ground and curved the tip in such a way that I can work it under the rolled edge and use it to "roll" the edge open once I have it started. Then after I have the tab ring removed. I use a piece of water pipe just the right size, clamped in a vise, as a curved anvil. I take a small ball peen hammer to flatten the unrolled aluminum edge against the pipe anvil. You need to be careful with the awl so you don't slip and run the point into your hand when working the rolled edge open.

Re-stuffed the last electrolytic after the parts came in. This one is the 1000uF 3V non-polarized cap. Here are some pictures of before, cover un-crimped, the parts, back together, and back in the set. I did this for the other little electrolytics except for three that I did not have the original caps to re-stuff. I wet the paper cover before I uncrimp the end to try to keep the paper from crumbling.

My selenium rectifiers had been replaced by the time I bought the set. They had been replaced with the 5A 600PIV silicon rectifiers as shown in the first picture. The original(?) soldered-in-fuses were blown and snap on fuse holders with replacement fuses were used. This mess is kinda hidden by the rectifier cover. The plug-in dropping power resistors still measured ok, but I decided to replaced them with chassis mount power resistors. I also wanted to add the recommended 7.5 ohm dropping resistor to match the drop from the selenium rectifiers. And I wanted to put in chassis mount fuse holders. So I rewired it all. Here are pictures before and after all the changes.

I wanted to test the power supply rewiring and cap re-stuffing. The power cord was intermittent, the chassis mounted cheater socket removed, and the cheater cord cut and soldered in. So I removed the kludge and replaced the chassis mounted cheater socket. Its now back to factory configuration. I removed all of the fuses expect for the main 4.5A 400VAC supply fuse. I brought the power up on a variac with a dim-bulb while measuring the 400VDC supply rail across the 80uF 450VDC cap. At about 300VDC I saw the dim-bulb flicker brighter, heard a "pfffft" sound from under the chassis and smelled smoke! :sigh:

I cut power and turned the chassis upside down. After not finding anything really wrong, I brought the power back up and saw the source of the smoke. The phenolic insulator of one of the voltage doubler capacitors had carbonized and was heating/burning/smoking intermitantly. The voltage on that capacitors' case is at 200VDC, so there is 200VDC across the phenloic insulator. I guess that's why I removed that cap from the set so many years ago. So I removed the re-stuffed caps' ground wire from the can ground lug to remove the 200VDC across the carbonized insulator and then wired the re-stuffed caps' ground directly to the other voltage doubler capacitor. Here are some pictures. Its all better now. :D

Now to start the long process of replacing all the paper caps and out-of-tolerance resistors....

Yep, carbon tracking can cause a lot of heartache. I once had a Philco 17MT80 roundie with intermittent reds, which I traced back to a carbon track going to ground from the plate of the red output tube. I was able to grind out the carbon from the PCB with a Dremel, then fill it in with epoxy. Whole PCB would have been a loss if I hadn't been able to fix it.

I've been working on measuring and replacing out-of-spec resistors and all the paper capacitors. I found that the vertical blocking oscillator transformer secondary is open. I melted it out of its tar hoping to find a simple loose connection. No such luck, the open is deep inside somewhere. I was able to find a replacement from oldradioparts.com. Here is a picture of the old and NOS part. It has a -2 where the original has a -1, but its resistance measurements match the Sams. I also have found that some of my sets resistor values match the Sams values where they differ from the RCA schematic values. And also some in my set differ from both schematics. Here is a picture of the HO screen resistor which Sams and RCA (R269) list as 11K but my set has a 10K. Its soldering looked original, but I could be wrong. I guess they are production run changes. Is it best to start with my sets original values and see if I need to change from there?

I have also found all but one of my sets white peaking coils are open. I got some replacements from oldradioparts.com but had to get some from digikey. Can modern coils of the same inductance be used as peaking coils? I think the old ones as Pi-wound, no clue on the modern ones.

11k sounds picky - 10 k should be good.

New peaking coils should be OK unless they have so much stray C that they self-resonate in the video band. Try them, they probably work.

Meh, people told me not to use the ferrite cored ones from Digikey, but they work just fine in my CTC-2 and 2B.

Cool, thanks for the information guys. I'm eager to fire it up, but I still have to rework the HV cage. Any good way to remove corona dope?

Dave

I replaced all the bad peaking coils and then compared all of the wiring and component values against the schematics. Then I installed the F104 300mA fuse ... and ... another carbon tracked part let out some smoke. Here is a picture of the burst amp transformer with a carbon track. I used a dremel tool to remove the carbon and trim that area away from the case and re-installed it.

I then turned my attention to the HV cage that needed to be completely rebuilt. I decided to remove it from the chassis to try to clean out the goo left from the original doorknob cap meltdown. Here is a picture of the mostly empty cage. I noticed my flyback had an opening right under it, but the cage did not. So I added an opening for better cooling. Then decided to add a fan. Here is a picture after the cage cutout. Here is a picture of the rebuilt cage with all new HV doorknobs. And picture from the bottom of the rebuilt chassis.

Next I installed all the tubes except for the V and H output tubes as did what voltage checks I could without them. I found that my audio output transformer primary was open. I will have to search for a proper replacement, but in the mean time I am using a 120V/12.3CT power transformer. Yeah, I know, power transformers are not designed for this, but its only temporary and it does make noise you could call sounds.

Before I could test the rebuilt V and H sections, I needed to replace the remains of the yoke cover. Here is a picture of the yoke with whats left of the original plastic yoke cover. And another picture of a Home Depot paint can cover taking its place.

Then I installed the V and H output tubes and actually got good, adjustable HV. I found that the H lin. inductor core was frozen in place and had to remove it to use lacquer thinner carefully dripped into the core to free it. I re-installed it and went through the horizontal adjustments. I monitored the current through the F101 450mA fuse and got a minimum of about 210mA. So that means what is next is.. try the chassis with the 30yr dormant 21AXP22A..

Here is a picture of the first light after 30yrs. After a quick V lin/size adjustment I got full deflection. Then I drove it with a VA62 channel 3 crosshatch pattern as shown in the next picture. After wiggling the not-glowing 6BQ7 tuner tube I got a signal. It only had chroma noise in the shape of the test pattern. After wiggling the 2nd video tube I got luminance and the crosshatch in this last picture.

I actually said, "Wow!" when I read the latest post...

Great news, she lives again!
Now let's see some color damit! lol

I have not done any convergence yet but here is a picture of the VA62 multiburst video response test pattern. And lastly a black and white picture of some video. Color does not yet work, and IIRC, that was the main problem I had intermitantly 30yrs ago.

I checked and found the 3.58MHz oscillator is not working. I tried the one other crystal I have and that made no difference. I guess I will start changing what parts I can in that circuit. Is it right that I can disconnect the reactance tube plate coupling capacitor (just to isolate and simplify the circuit) while I get the oscillator working?

Yeah, kvflyer and miniman, I am really amazed and thankful the old CRT still works. It tested good on the CR70, but until I saw the screen light and no blue neck glow, I was holding my breath.

I can't wait to see color too, but that's for another day, soon hopefully.

Dave

Great progress!

Keep up the good work!

After swapping tubes, disconnecting the oscillator components, testing them, finding them ok, testing the caps inside the reactance transformer, and finding them ok I examined the crystal socket and thought it looked a bit corroded. I cleaned its socket and tried the oscillator adjustment procedure again. This time the oscillator started :banana:. So I adjusted it to the 5Vpp, and ran out of time for tonight. So I guess I will not need the replacement crystals and oscillator output transformer I ordered today. But its nice to have spares. I ordered an audio output transformer that I think may work even if its not a listed replacement. Its from an amp that also uses a 6AQ5 for its output tube.

Before I did the color alignment I wanted to try to lower the flyback drive current as much as I could without messing up the width, linearity, or adding visible drive lines. I tried making the HOT drive DC voltage more negative by increasing the grid to ground resistance, but too much adds drive lines. I found the HOT screen voltage was high, so I changed the 10K screen resistor to 11K and ended up adding 2.2K to that and just barely have enough width with the width switch set to max. I only got it down to around 200 mA without getting visible problems. I also messed with the 7.5 ohm silicon diode dropping resistor. I ended up with a 4 ohm dropping resistor to get right at 400VDC with my AC input voltage. To make checking voltages and currents easier later, I added a terminal strip on the transformer side of the selenium rectifier tower with low ohm resistors in series with the supply voltage rails so I can check voltage and current on the four rails. I also added a terminal strip to the back of the HV cage with a jumper to insert a current meter in the HOT cathode and a 10 ohm resistor in series with the HV regulator cathode to measure its current. With the chassis back in the set I did another horizontal and vertical alignment and ended up with 25KV HV, 205-200 mA HOT current, full vertical and full horizontal deflection with decent linearity. With the color oscillator now running I can see that the color is not locked.

I bolted the chassis into the cabinet and tipped it on its side so that I could get at the underside of the chassis and see the picture. I started through the color AFC alignment procedure but I could not get a zero beat pattern with the reactance tube input grounded. I put my Heathkit frequency counter on the color oscillator and found it running at 3.581391 MHz, pic attached. After scoping a few points I found a bad connection on the reactance transformer supply input. I guess it had been cut at some time and poorly soldered back. This is probably the reason I could occasionally get the color to lock 30+ years ago. With that fixed and some fiddling with the fine tuning I got the color to lock, but I still could not get a zero beat pattern when adjusting the reactance transformer. I could reduce the number of out-of-sync color bars, but there was not enough range to reach zero beat. I put the frequency counter back on the color oscillator and found that when it locked the frequency counter read 3.579445 MHz, pic attached. With the reactance transformer adjusted for the minimum number of color bars, the counter read 3.579653 MHz. So the reactance transformer would let me adjust the frequency down, but not down far enough to reach zero beat. I had one old crystal and five NOS I just bought, so I tried them and got these frequencies:

RCA crystal dated 11/56 3.579653
Crystek #1 3.579651
Crystek #2 3.579644
Crystek #3 3.579625
Crystek #4 3.579647
Crystek #5 3.579632
COYSON dated 3/75 3.579614

I put the Crystek #5 crystal back in, since it was about the middle of the set of frequencies, and verified it read the same after putting it back in. Then I added a 1pF ceramic cap to across the crystal and saw it lower the frequency. I ended up with 3pF across the crystal to get a frequency lower than the locked frequency with the reactance transformer adjusted for its lowest frequency. Now I could adjust the frequency up to get a zero beat of the color bars with the reactance input grounded. I did the rest of the color AFC alignment and ended up with color that remained locked with acceptable fine tuning.

The matrix alignment had me confused at first when looking at the R-Y scope pattern from a color bar generator. I realized the RCA color bar generator color sequence is not the same as my Sencore VA62 color bar sequence. Here is a list of the correspondence:
Pulse after burst---RCA--------------------Sencore VA62
--------1-----------yellow-orange----------green-blue
--------2-----------orange------------------cyan
--------3-----------red----------------------blue-green
--------4-----------blue-red----------------green
--------5-----------magenta----------------yellow-orange
--------6-----------blue---------------------orange
--------7-----------green-blue--------------red
--------8-----------cyan---------------------blue-red
--------9-----------blue-green--------------magenta
--------10----------green-------------------blue
(Sorry for the -'s, I don't know how to set up columns)
I was able to do the matrix alignment after translating from the RCA to the VA62 color sequence numbers, but it was confusing for me.

I dug out my old degaussing coil and ran it over the set. Here is a pic of the purity with all of the neutralizing magnets pulled out all the way. I took a shot at the convergence and found that even with restoration there is still a lot of interaction between the controls. My set has a CTC3 convergence chassis that matches the RCA service manual schematic, while my CTC2B chassis has what look to be running changes that more closely matches the Riders schematic. I may try modifying the CTC3 chassis to match the Riders to see if that improves convergence. Anyway, here are some pictures of the set with my attempt at the alignments. So far I have avoided doing anything inside the tuner, but the noise I see in the picture may motivate me to see if there are some drifted resistors as wiseguy found in his. The pictures on this set brings back memories of the color I saw 30+ yrs ago on this set and how amazed I was when I compared its colors to the 70's sets. I need to get a DVD of the Wizard of Oz and see how my sets compares to the CT100 pictures. :D

It's looking good.

Are you going to get around to the IF alignment check?

Another 21CT55 Operational
 

Hey Dave, its looking good. Less than 18 weeks working the restoration so far is quite fast considering you went silicon rectifiers and big electrolitic cap stuffing. Of course you replaced the white RCA peaking coils and a number of small under-chassis yellow caps. Was this a less than mandatory complete re-cap or did it come this way? I really respect your logical, systematic approach to each problem you encountered and shared the process with the Members. I hope you get the Wizard of Oz for subjective evaluation of your 21CT55 picture quality and suggest also getting the " Digital Video Evaluation" DVD's test patterns for critical setup and objective PQ evaluation. Looking forward for your full-screen width, no room light, full computer monitor posted JPGs. I know what the 21CT55's are capable of so lotsa luck finishing this one off...........Tom