5 ms·

A Tour of the Tektronix Ceramic CRT "Catacombs"

77 points by dangle1 3y ago

CamperBob2 3y ago
And for extra geek cred, get your calendars here! https://vintagetek.org/2024-calendar/ https://vintagetek.org/2024-calendar/
aurizon 3y ago
I recall some radars used those large round screens, long persistence phosphors. Often these were mechanically rotated in time with the antenna, with slide contact rings. Later electronic rotation superseded that.
- dtgriscom 3y ago
  Tektronix was famous for actual storage screens, where once a signal was drawn across the CRT it would persist until the screen was reset. Very handy for capturing a single event and viewing it at your leisure. https://en.wikipedia.org/wiki/Storage_tube https://en.wikipedia.org/wiki/Storage_tube
  - ddingus 3y ago
    Those computers were fun to use! The one I used had a joystick built right into the keyboard, and it ran a CNC programming environment. Data I/O was paper tape via serial and data cassette. Ours did not have the floppy or fast refresh graphics options. 2048x2048 vector space! The big CRT could resolve most of that when well tuned. The work process was: Load editor from tape cartridge Type it in, or read text from paper tape. I had punched a fair number of shortcut G-codes and I would use them to get the bones of a new program started. Write finished program text (g-code) to paper tape. That could and should include comments. (We put a setup sheet on the master tape, for example) This was a master. Make another stripped down one, and this is production. Prep a tape for reading by sticking the beginning of program end into the reader and put the reader into handshake mode. Read it back to validate against program in RAM. Or... Load backplotter/ simulator. Prep program tape for reading. Run the simulator, plotter. If it was a master tape, it would know the tooling setup and just draw all the operations to the CRT. And because it was a storage tube, those all would display and one could see what metal gets cut out easy enough. This worked fairly well! Once it was done, the production tape went into a can and was moved to a library. That was a whole wall of cabinets filled with little plastic canisters containing paper tape programs for everything. 1Mhz 6500 CPU. And like I said, the Tek terminals were just fun to use.
    - adql 3y ago
      And g-code is still in use till this day
      - ddingus 3y ago
        Indeed. G-code is a very well specified protocol. It mostly just works.
    - aurizon 3y ago
      Nice stuff, I was a Tech-Tel officer in the 60's(RCAF) and solid state was slowly penetrating the MIL-space, due to the huge waits for MIL part validation and approval - for want of a nail - a shoe was lost.... I can recall 10K 1/2W resistors in their own package with their own frame of a 16MM filmed x-ray enclosed as repair back stock.
      - ddingus 3y ago
        The Xray was for integrity inspection?
        aurizon 3y ago
        Yes, audit trail was kept and maker and MIL accumulated reliability stats to add more 9's to the tail. Not sure how this persists in the surface mount era as most parts are nine nines reliable now - or more? I have to assume assembly stuff does a lot of real time test with beds of nails. I saw the old blackberry line, when it was scrapped out(after the fall) and every board was nailed and tested, fails = reworked or scrapped. Huge bins of all sub boards etc., in all parts of the process when they were closed = recycled for precious metals etc.
- MadnessASAP 3y ago
  Expansion on your fun fact. Electronic rotation was accomplished with a polar CRT. A polar CRT is a variant of a vector CRT where the beam path starts in the center and extends to the edge in a direction derived from the azimuth of the radar antenna. The transmit pulse would start a new scan from the centre and its speed would be determined by the desired range selection. Finally the beam intensity was just the amplified return signal from the antenna. All of this required complex electromechanical devices to generate (and eventually filter) the required signals. As technology progressed and textual information started to be added to the process it became necessary to convert the polar plot to a conventional rasterized signal. This was accomplished with a scan converter tube which was essentially a polar CRT coupled to a TV camera. Radar would plot on the CRT and the camera would turn it into scanlines to be overlaid with data. This was used into the 90s and there's probably still operational systems relying on this.
  - ddingus 3y ago
    In the 90's I got an inside tour of PDX airport. They had these systems and a camera pointed right at one to rasterize them. At the time, a whole lot was still analog. Beautiful systems. Saw one of those 9 track tape drives too. On a seek, tape would just dissappear from one reel and accumulate on the other one --fast! And then CLACK! It would stop and just start reading at what looked like normal speed. Was kind of impressive in its own way too.
    - MadnessASAP 3y ago
      Being in the air force I have the dubious privilege of getting to maintain some old ass-technology. Radar being amongst my favourites. We've since upgraded to comparatively modern technology (digital!) and gotten rid of most of the vacuum tubes but I still have a special dark place in my heart for the old mixed signal system running on CRTs, wire wrapped boards, tapes, and core memory. Interestingly while the new system is much nicer to work with, much better resolution and features I didn't think were possible. The old system had power, and lots of it, and sometimes power is what you need.
ganzuul 3y ago
More information on the ceramic to metal bonding technique would be very much appreciated! I'm looking to accomplish a variation of the kintsugi technique for an art project.
ChuckMcM 3y ago
Okay, that's pretty neat making them out of ceramic. I would love to see the entire process from mold to finished product. I'm guessing the painted the phosphor on to a glass square and then melt/welded it to the ceramic molded tube. I would further guess a similar process was used to attach the electron gun at the back while drawing a vacuum. I did not know that you could make a tube that way!
- 13of40 3y ago
  Here's the process: https://youtu.be/G0Dci5RPe94 https://youtu.be/G0Dci5RPe94 I didn't get to watch it to the end yet, but it shows the pressing and firing process halfway through.
  - ChuckMcM 3y ago
    That was awesome! Not surprised at all that these tubes were super expensive, especially with a palladium + gold interior coat :-).
Sunspark 3y ago
I think there is still a market for CRTs, esp. with retro-gamers. Vector displays in particular.
- Forgeties79 3y ago
  Love me some crisp scanlines
- dharmab 3y ago
  There's certainly a market, but the process of manufacturing and recycling them is difficult to do in compliance with modern environmental regulations. In particular, the leaded glass is problematic to export for recycling.
  - mrob 3y ago
    Leaded glass is not required for CRTs. In the case of color CRTs, it was only used for the backs and sides, because x-rays make leaded glass turn brown. The faceplates used barium-strontium glass, which I assume was more expensive. This sample chapter from "Image Performance in CRT Displays" by Kenneth Compton provides some details: https://spie.org/samples/TT54.pdf https://spie.org/samples/TT54.pdf I see no reason why you couldn't make the whole thing from barium-strontium glass.
    - dharmab 3y ago
      The existing regulations assume that all CRTs are leaded, so you'd have to lobby multiple governments to treat your product differently.
- rickdeckard 3y ago
  yeah, but I don't think production and sales of CRT-tubes could be called a "market" in an economically sustainable sense today. If someone would find a warehouse full of Trinitron tubes: Yes, a nice opportunity to cater to a retro-gaming market. Setting up the production chain to BUILD tubes at scale: Not really.
neilv 3y ago
Tektronix is a legendary company, and an anchor of the Silicon Forest. I owe some early big career breaks to a Tek spinoff company, and to the Oregon Graduate Institute, which Tek had a part in founding.
- ddingus 3y ago
  Seriously. Oregon, who has told me they want to see another Silicon Forest happen, would be wise to look hard at Tek. I owe some of my life advances to Tek as well.
- tealpod 3y ago
  Yes, very much true. I started my career at Tektronix. The famous Gang-of-Four, who initially categorized/identified Software Design Patterns is all started at Tektronix.
- jeffrallen 3y ago
  Counterpoint: I grew up in a suburb south of Portland in the 80's. My first experience with economic uncertainty and the consequences for a family were seeing a friend's father laid off from Tek. Boom and bust tech companies are not all good for the communities they sprout up next to.
cwillu 3y ago
Skip 2 minutes into the video.
fumeux_fume 3y ago
Every now and then I'll come across tektronix tested vacuum tubes while searching for amp tubes. It's a pretty cool piece of history to own.
mindslight 3y ago
Thanks, I hate it: https://vintagetek.org/knobless-2445-oscilloscope/ https://vintagetek.org/knobless-2445-oscilloscope/
- baz00 3y ago
  Philips brought something as abhorrent to production and it sucked.
jacquesm 3y ago
Oh that's so neat. Makes me wonder if the CRT in my scope was made using that tooling, they must have had 10's or even 100's of those sets. I hardly ever use it these days but I can't part with it, it's been with me for a good 30 years or so and it still works well when powered up. I also have a much newer LCD based one that sees much more use and that is far more compact.
Neil44 3y ago
Intresting, normally with ceramic moulds the mould is very porous and sucks moisture from the ceramic slip which makes it solidify enough to be removed from the mould and fired. Wonder what the process was here.
- jdblair 3y ago
  The narrator explained in the video. Damp ceramic powder was used, and was placed under tons of force while in the mold. The ceramic powder then stayed in-tact when removed from the mold for firing.

77 points by dangle1 3y ago

And for extra geek cred, get your calendars here! https://vintagetek.org/2024-calendar/ https://vintagetek.org/2024-calendar/

aurizon 3y ago

I recall some radars used those large round screens, long persistence phosphors. Often these were mechanically rotated in time with the antenna, with slide contact rings. Later electronic rotation superseded that.

dtgriscom 3y ago
Tektronix was famous for actual storage screens, where once a signal was drawn across the CRT it would persist until the screen was reset. Very handy for capturing a single event and viewing it at your leisure. https://en.wikipedia.org/wiki/Storage_tube https://en.wikipedia.org/wiki/Storage_tube
- ddingus 3y ago
  Those computers were fun to use! The one I used had a joystick built right into the keyboard, and it ran a CNC programming environment. Data I/O was paper tape via serial and data cassette. Ours did not have the floppy or fast refresh graphics options. 2048x2048 vector space! The big CRT could resolve most of that when well tuned. The work process was: Load editor from tape cartridge Type it in, or read text from paper tape. I had punched a fair number of shortcut G-codes and I would use them to get the bones of a new program started. Write finished program text (g-code) to paper tape. That could and should include comments. (We put a setup sheet on the master tape, for example) This was a master. Make another stripped down one, and this is production. Prep a tape for reading by sticking the beginning of program end into the reader and put the reader into handshake mode. Read it back to validate against program in RAM. Or... Load backplotter/ simulator. Prep program tape for reading. Run the simulator, plotter. If it was a master tape, it would know the tooling setup and just draw all the operations to the CRT. And because it was a storage tube, those all would display and one could see what metal gets cut out easy enough. This worked fairly well! Once it was done, the production tape went into a can and was moved to a library. That was a whole wall of cabinets filled with little plastic canisters containing paper tape programs for everything. 1Mhz 6500 CPU. And like I said, the Tek terminals were just fun to use.
  - adql 3y ago
    And g-code is still in use till this day
    - ddingus 3y ago
      Indeed. G-code is a very well specified protocol. It mostly just works.
  - aurizon 3y ago
    Nice stuff, I was a Tech-Tel officer in the 60's(RCAF) and solid state was slowly penetrating the MIL-space, due to the huge waits for MIL part validation and approval - for want of a nail - a shoe was lost.... I can recall 10K 1/2W resistors in their own package with their own frame of a 16MM filmed x-ray enclosed as repair back stock.
    - ddingus 3y ago
      The Xray was for integrity inspection?
      - aurizon 3y ago
        Yes, audit trail was kept and maker and MIL accumulated reliability stats to add more 9's to the tail. Not sure how this persists in the surface mount era as most parts are nine nines reliable now - or more? I have to assume assembly stuff does a lot of real time test with beds of nails. I saw the old blackberry line, when it was scrapped out(after the fall) and every board was nailed and tested, fails = reworked or scrapped. Huge bins of all sub boards etc., in all parts of the process when they were closed = recycled for precious metals etc.
MadnessASAP 3y ago
Expansion on your fun fact. Electronic rotation was accomplished with a polar CRT. A polar CRT is a variant of a vector CRT where the beam path starts in the center and extends to the edge in a direction derived from the azimuth of the radar antenna. The transmit pulse would start a new scan from the centre and its speed would be determined by the desired range selection. Finally the beam intensity was just the amplified return signal from the antenna. All of this required complex electromechanical devices to generate (and eventually filter) the required signals. As technology progressed and textual information started to be added to the process it became necessary to convert the polar plot to a conventional rasterized signal. This was accomplished with a scan converter tube which was essentially a polar CRT coupled to a TV camera. Radar would plot on the CRT and the camera would turn it into scanlines to be overlaid with data. This was used into the 90s and there's probably still operational systems relying on this.
- ddingus 3y ago
  In the 90's I got an inside tour of PDX airport. They had these systems and a camera pointed right at one to rasterize them. At the time, a whole lot was still analog. Beautiful systems. Saw one of those 9 track tape drives too. On a seek, tape would just dissappear from one reel and accumulate on the other one --fast! And then CLACK! It would stop and just start reading at what looked like normal speed. Was kind of impressive in its own way too.
  - MadnessASAP 3y ago
    Being in the air force I have the dubious privilege of getting to maintain some old ass-technology. Radar being amongst my favourites. We've since upgraded to comparatively modern technology (digital!) and gotten rid of most of the vacuum tubes but I still have a special dark place in my heart for the old mixed signal system running on CRTs, wire wrapped boards, tapes, and core memory. Interestingly while the new system is much nicer to work with, much better resolution and features I didn't think were possible. The old system had power, and lots of it, and sometimes power is what you need.

ganzuul 3y ago

More information on the ceramic to metal bonding technique would be very much appreciated! I'm looking to accomplish a variation of the kintsugi technique for an art project.

ChuckMcM 3y ago

Okay, that's pretty neat making them out of ceramic. I would love to see the entire process from mold to finished product. I'm guessing the painted the phosphor on to a glass square and then melt/welded it to the ceramic molded tube. I would further guess a similar process was used to attach the electron gun at the back while drawing a vacuum. I did not know that you could make a tube that way!

13of40 3y ago
Here's the process: https://youtu.be/G0Dci5RPe94 https://youtu.be/G0Dci5RPe94 I didn't get to watch it to the end yet, but it shows the pressing and firing process halfway through.
- ChuckMcM 3y ago
  That was awesome! Not surprised at all that these tubes were super expensive, especially with a palladium + gold interior coat :-).

Sunspark 3y ago

I think there is still a market for CRTs, esp. with retro-gamers. Vector displays in particular.

Forgeties79 3y ago
Love me some crisp scanlines
dharmab 3y ago
There's certainly a market, but the process of manufacturing and recycling them is difficult to do in compliance with modern environmental regulations. In particular, the leaded glass is problematic to export for recycling.
- mrob 3y ago
  Leaded glass is not required for CRTs. In the case of color CRTs, it was only used for the backs and sides, because x-rays make leaded glass turn brown. The faceplates used barium-strontium glass, which I assume was more expensive. This sample chapter from "Image Performance in CRT Displays" by Kenneth Compton provides some details: https://spie.org/samples/TT54.pdf https://spie.org/samples/TT54.pdf I see no reason why you couldn't make the whole thing from barium-strontium glass.
  - dharmab 3y ago
    The existing regulations assume that all CRTs are leaded, so you'd have to lobby multiple governments to treat your product differently.
rickdeckard 3y ago
yeah, but I don't think production and sales of CRT-tubes could be called a "market" in an economically sustainable sense today. If someone would find a warehouse full of Trinitron tubes: Yes, a nice opportunity to cater to a retro-gaming market. Setting up the production chain to BUILD tubes at scale: Not really.

neilv 3y ago

Tektronix is a legendary company, and an anchor of the Silicon Forest. I owe some early big career breaks to a Tek spinoff company, and to the Oregon Graduate Institute, which Tek had a part in founding.

ddingus 3y ago
Seriously. Oregon, who has told me they want to see another Silicon Forest happen, would be wise to look hard at Tek. I owe some of my life advances to Tek as well.
tealpod 3y ago
Yes, very much true. I started my career at Tektronix. The famous Gang-of-Four, who initially categorized/identified Software Design Patterns is all started at Tektronix.
jeffrallen 3y ago
Counterpoint: I grew up in a suburb south of Portland in the 80's. My first experience with economic uncertainty and the consequences for a family were seeing a friend's father laid off from Tek. Boom and bust tech companies are not all good for the communities they sprout up next to.

cwillu 3y ago

Skip 2 minutes into the video.

fumeux_fume 3y ago

Every now and then I'll come across tektronix tested vacuum tubes while searching for amp tubes. It's a pretty cool piece of history to own.

mindslight 3y ago

Thanks, I hate it: https://vintagetek.org/knobless-2445-oscilloscope/ https://vintagetek.org/knobless-2445-oscilloscope/

baz00 3y ago
Philips brought something as abhorrent to production and it sucked.

jacquesm 3y ago

Oh that's so neat. Makes me wonder if the CRT in my scope was made using that tooling, they must have had 10's or even 100's of those sets. I hardly ever use it these days but I can't part with it, it's been with me for a good 30 years or so and it still works well when powered up. I also have a much newer LCD based one that sees much more use and that is far more compact.

Neil44 3y ago

Intresting, normally with ceramic moulds the mould is very porous and sucks moisture from the ceramic slip which makes it solidify enough to be removed from the mould and fired. Wonder what the process was here.

jdblair 3y ago
The narrator explained in the video. Damp ceramic powder was used, and was placed under tons of force while in the mold. The ceramic powder then stayed in-tact when removed from the mold for firing.