The Set: Pete Deksnis's Site about the CT-100

Restoring a Vintage Color Television Set

THE SET

restoration log

9-4-99. Well, some stuff has happened. This is the beginning of the Labor Day weekend and I plan to spend some quality time with THE SET. Safely running (meaning with typical operational B+ voltages) the rf-to-sound chain without the variable supply is my next goal in the process of restoring the keyed AGC and horizontal sync separator. Only the bias supply section of the variable supply will still be used. Incidentally, by cranking the bias up to minus eight to nine volts lowers 275-volt bus current to about 130 mA at a 250 volt level, a more reasonable burden for a 39-year old Heathkit PS-4 to assume (so last evening after work and a walk on the Ocean Grove boardwalk I 'burned in' the set for a good 45 minutes, enjoying the aural functionality of my technological artifact).

Bad news has surfaced that I've really known all along but didn't want to confront. Gee Doc, I've had these nightmares....

Here it is: the 15GP22 tricolor 3-gun picture tube made in January 1954 is a dud! Why the neck gets warm (as noted earlier) when 6.3 Vac is applied to the filaments is one big "?". Is there a shunt of some sort or a heater in the base for some reason? Doesn't make any sense to me. Maybe by brain is faking me out. Either way, there's no color, no glow, no secondary photon emission, no electrons, no nothing but the warmth at the neck-end of this (I think) Lancaster, PA-made, marvelously inefficient electron-to-light converter. Shat!

Plan B. Before beginning this restoration last June, I wanted to be reasonably certain that any effort would accomplish something positive if, as suspected, the CRT had become inoperable in the 33 years it had languished, powerless. My final justification for undertaking the task became plan B -- feed the red, green, and blue video signals developed by THE SET for 15GP22 control grids into a suitable interface that, in turn, would drive a run-of-the-mill 14-inch VGA monitor. [rest of paragraph updated 9-18-99] There are a zillion of them sitting around waiting to be junked because their former users just bought a K6-2 or -3 with 64 meg of RAM and a 17-inch monitor for about double what just a 14-inch monitor cost in 1994. I measured the height of the display area of the 15GP22. It is 8-1/4 inches tall and 11-1/4 inches at the widest. The 14-inch monitor I'm using to write this is 8-1/4 inches tall and 11-inches wide. Pretty close match. Only the monitor need be a reproduction. Everything but the picture tube will be cooking away in the cabinet. It'll be as close as possible to a fully functional, original CT-100 as is now possible.

I may not need plan B. The primary purpose of this site is to present reference information about this rare progenitor artifact of the twentieth century. It is also to communicate with the very few others in the world who have the hardware and so can share the experience. Thus far, there are three [9-18-99: it's now four][9-27-99: still taking names! It's now five] other collector/restorers with whom I have communicated. Guess what everybody needs.

In 1978, I learned of a company who could have rebuilt my CRT, but I was busy and passed up that opportunity. Last week, a collector/restorer found a small stash of replacement electron-gun assemblies made decades ago when consumers still watched old color sets. If the four of us have CRT's rebuilt there will still be a healthy number of guns remaining for the future. One collector has begun the process. I keep my fingers crossed. More to come.

9-7-99 Quick update: Spoke to the CRT rebuilder today and discovered we're not out of the woods yet. There doesn't seem to be anyone who has experience rebuilding a 15GP22. The stash of electron guns is for the 21-inch round tubes of 1955 through to about 1968. 21CYP22A, for example. Wrong gun for the 15GP22, which has a fourth element, an internal, static-convergence electrode (pin 13). That's in addition to the usual focus, acceleration, and control grids. Will hear more from the rebuilder after he studies the 15GP22 now enroute to him. More when I know more... [9-27-99 still no new info...]

9-13-99 The puzzle continues. The CRT rebuilder is consulting with someone who has rebuilt 15GP22s. The gun supplier insists there is a "new improved" gun available for rebuilding 15GP22s. Hope so. Also, I disassembled my own 15GP22. Under the mu metal shield there was a Handle with Care and X-ray Warning label. It included a serial number (6255) stamped in black ink over a machine-printed s/n in red ink (IK 1273). While checking the CRT for faceplate cracks, I found the 6255 s/n hand etched inside the tube, clearly visible through the thick glass. There's gotta be an interesting story behind that. [10-3-99 For more info, click on 'the 15GP22' in the 'simplified theory' menu.] If I get a s/n from at least one other collector, I'll start a separate page of 15GP22's known to have survived.

10-1-99 Steve sent some set-up and convergence data from a 1953 RCA document he acquired. There's a little tricolor crt history included (the 1950 version had only 117,000 dot trios; a color filter in front of the phosphor screen was need to enhance the weak red phosphor). It probably answers my question about the serial number visible inside the tube. "In order to obtain precise alignment, which is absoultely necessary, between the holes in the mask and the phosphor dots, the mask and phosphor dot screen are mounted together in an assembly. This assembly is then placed in the tube and held in proper resationship to the electron guns." Upon more recent inspection of the crt, I noticed the s/n was written twice. Once on the phosphor dot plate and once, you guessed it, on the shadow mask.

We still are waiting to hear whether or not the original gun can be rebuilt or the gun for the 21-inch round version (21AXP22, for example, which are available), can be modified.


10-18-99 Just received word from Steve that the existing vintage guns cannot be modified to rebuild a 15GP22. He has developed an alternative process, however. There is a considerable up-front tooling cost, but once that's met, guns for the 15GP22 can be manufactured on a custom basis. All it takes is one rich collector or a dozen or more guys like me. The out of pocket expense for the first 15 rebuilds will be less than what a RARE NIB 15GP22 costs. So in that sense, it may be a cost-effective process.

11-13-99 There is currently an effort underway spearheaded by Bruce to develop a protocol for rebuilding the electron gun assembly of 15GP22 tricolor kinescopes. That's where the action is right now.


1-19-00 Had a twenty-minute telecon with Bruce last week. He has his 15GP22 at the rebuilder. Also, so the rebuilder can have hands-on access to the genuine article, Steve has contributed a 15GP22 gun assembly he previously had removed from a dud. Apparently, the big challenge is not remanufacturing the gun assembly (in itself, a relatively modest task for experienced rebuilders).

The critical process seems to be defined by what may be the only television CRT ever manufactured with a 20-pin base. Although the exact process is unclear (to me, at least) at this point, it seems a tool that can reseal a 20-pin gun assembly in the glass neck of a CRT is unheard of today. Perhaps only RCA, who manufactured relatively few of these tubes within a narrow window of time half a century ago, had the manufacturing capacity. Although we know from an existing 15HP22 (the CBS-manufactured equivalent of the 15GP22) that it too had the 20-pin base (which confirms a Sams parts list that cites the 15HP22 as a replacement for the RCA 15GP22). Maybe somewhere in an old warehouse, there's a tool for remanufactuing 20-pin CRTs. What the hey: if it were easy, it wouldn't be any fun...

Incidentally, not all 20 pins were used. Pin 13 is connected to the modulated 10 kV convergence anode voltage. Three unused pins on either side of pin 13 provide isolation for the rest of the tube pins and functions.

1-21-00 THANK YOU for all your comments. I try to respond to all e-mail received by this site. But to those to whom I am not able to reply, please be assured that your time and efforts are fully appreciated. Thanks again.


2-4-00 Received some GOOD NEWS from Bruce yesterday. It seems there is a company that can make the tool we need to have new 20-pin bases built for the 15GP22 tricolor picture tube. The part must of course be manufactured from drawings, but such drawing do not exist, and so must be generated. Thus far, we have not been able to locate mechanical drawings of the 15GP22 that would allow a draftsperson/designer to do the job. However, and although we would prefer to find original RCA drawings, our contact at the gun rebuilder will generate the drawings using, as a model, the gun Steve had removed from the neck of a damaged/dud 15GP22.

Here's what all the fuss and trouble is about. It's a glass disk through which all the wires from the outside world feed into the vacuum world inside a tube. In the parlance of tube manufacturing, it is the stem. Here you see the stem of a 9-pin miniature tube such as the 6AN8. A tool that makes the stem for (based on a 20-pin pattern) the 16-pin 15GP22 does not now exist. Or, more correctly, does not yet exist.

Once a stem is manufactured using the new tool, its leads will be welded to the rebuilt gun, and the assembly will be shipped to the crt rebuilder. Then the stem/rebuilt-gun assembly will be attached to the glass neck of the soon-to-be glowing-again 15GP22. Thanks Bruce, for all the footwork and phone calls. You've taken us all one big step closer...

2-9-00 Received more information from Bruce yesterday, and I've heard from a previoulsy unknown CT-100 collector from Canada who is sending some early '50s drawings of the 15GP22. Access to mechanical drawings of the 15GP22 is perhaps the final hurdle that we face.

From Bruce:
Today I talked to a Frederic's Glass Company technician in the stem-making department. It was quite consoling to hear her tell me that they make stems for all kinds of situations, and that this sounded "right up their alley."

They were okay about the idea of using an exact specimen to copy. She says
they have some old machinery that was used by RCA , and that it is often just a matter of dialing in numbers to reproduce an item. Hard numbers are needed if this machine is the one that can form the button stem we need. Original drawings of the 15GP22 will tell us those numbers.

2-13-00 As you might imagine, others have attempted to rebuild a 15GP22 tricolor picture tube. Here's an interesting approach taken years ago. First investigated last September by Steve, it is now deemed an unacceptable solution. We believe for authenticity and functionality that the RCA design
(with its isolated pin-13) must be retained.

Recently, Bruce obtained a never-installed sample using this rebuild technique. The collector who owns it is unsure of its origin. It had been packaged in a receptacle containing inert gas.

This is a photograph taken in the first week of February, 2000. The solid element on the right is the convergence electrode (G4), which is what makes any 15-inch color crt from 1954 unique, rare, and as it turns out, such a pain in the proverbial butt. Only this crt and the CBS-Hytron version, the 15HP22, used electrostatic convergence and so used this element.

This is an 8-bit grey scale image with high compression for quick loading, but it's still fairly detailed.
Netscape users can right-click on the picture and select View Image to examine a X2 image,

Well, I got the gun today. Studied it well and yes it's a 15GP22, but it is welded onto a 14AH button. This was the base for the 21AXP22 tube. This would be too risky to try. The 15G needs 13 leads, and you know about G4/pin13. On this one they put the 11,000V right between Blue cathode and Red screen, and oops, +#@$&O.sht

3-14-00. It's time to open another restoration log page, so to close this longest and over-the-millennium log, here's a quick and brief update. We keep adding to the group. I spend more time on this site than hands-on work on my CT-100. The quest for a rebuild protocol for the 15GP22 continues. The next restoration log page begins at this
15 March 2000 link.

[Return]