I found an old Mead terminal, and I'm trying to use the CRT display.

[vsevolopod]

I think this fits in here, but if it's better off in "Beginners" let me know.
I found a Mead Ubiq terminal, with what I'm pretty sure is a monochrome display. I would like to use this display with a more modern computer by adapting it to use VGA or S-Video as a display input. I cannot find any info or repair manuals for this machine, so I'm not sure what the pinout for the display is.

 Attached to the monitor is a 9-pin D-SUB connector. Pins 1, 4, and 7 are missing. Pins 6 and 9 are connected to the same place on the CRT driver board. It doesn't seem like it's monochrome TTL, based on the pinouts i've found online.

CGA seems most likely, but I'm not sure. Can "intensity" and "vertical sync" be connected?

I'd love advice from anyone who knows more than me. I tried using a CGA sketch on Arduino to test it, but it didn't work, and it's just too many variables at once for me to figure it out. Any advice on what the pinout for this monitor might be, and if it's possible to adapt it to VGA or S-Video?

I can provide pictures of the connector, the CRT driver board or the terminal itself if needed.

I would also love to get the documentation for this machine if someone knows where to find it.
(update: I made a PDF with the images so it's all small enough to upload here. It's attached below.)

Thank you!

[Kean]

So, one of these?:  https://collection.cooperhewitt.org/objects/18655559/

Sounds to me like that could be a 9-pin serial port, meant to connect into a modem - considering these were used originally to access Lexis Nexus.
A photo of the PCB near that connector may help.

[vsevolopod]

Yes, exactly one of those. Yes, there is a serial connector for connecting to a modem, but I'm talking about an internal cable that goes to the CRT board. I've attached a PDF with the images below.

[Kean]

Ah, OK.  Yeah, that connector is going to carry the video signals.
As I believe this was a monochrome display meant for text, the signals are not likely to be CGA, VGA, or even S-Video.
Almost certainly monochrome TTL of some sort, similar to MDA/Hercules-like - although based on the pins missing it doesn't match those.

Some links that may help
https://en.wikipedia.org/wiki/Hercules_Graphics_Card
http://www.philipstorr.id.au/pcbook/book3/videomon.htm
http://www.philipstorr.id.au/pcbook/images/monomon.jpg

[duak]

That looks like the 10 pin edge connector monochrome CRT interface thats been around since the early 70's.  I don't know who came up with it, but I've seen Ball Brothers, Zenith, Panasonic and a few other brands.  It's still used in CNC machines that have monochrome CRTs.   If memory serves, the signal levels are basically TTL ie., 0 to 2 V.  There will be three inputs: Video, Vertical sync and Horizontal sync.  The last is required to generate the HV.  The monitors don't always have the standard 15,750 Hz horizontal scan frequency so check with a scope.  In fact, since you have a working source, get the timing of all the signals.  Be carefull about driving this thing with a drastically different horizontal scan frequency as it counts on tuned circuits to work.  Too fast can burn up the horizontal power transistor and flyback transformer.

Big Warning! the brightness or highlight control is usually brought out on the connector so it can connect to a case mounted pot.  It often has a fairly high voltage of 50 to 100 V on it.  Don't connect the edge connector backwards as it usually damages something at both ends.  There's an indexing slot for that.

Cheers,

[vsevolopod]

I sort of figured it was monochrome TTL, but I haven't found pinouts online that match that. The closest IC to the display output is a PAL16C1CJ:

https://html.alldatasheet.com/html-pdf/126134/ETC1/PAL16C1CJ/98/1/PAL16C1CJ.html

The datasheet mostly just confused me more, but sort of confirmed that it is a digital, not analog output to me. Am I correct in this assumption?
I'd attach a picture but my phone died right after taking it.

Although I think the signals are digital, the CRT driver board seems like it's mostly analog. The only ICs on it are 555 timers. Does this info provide any clues?

I'm trying to work backwards from the CRT driver board to figure out what pins do what.
From what I've figured out so far, pins 3, 6, and 9 are connected to the ground plane.
Pins 2 and 5 lead to sections of the board with 555 timers. I'm guessing these are H-Sync and V-Sync. I didn't check continuity on this, since the traces go to the flyback transformer, which I haven't discharged, but on visual inspection it seems like these sections also go to the deflection coils. So it seems like my assumptions are correct.
What would the 555 timers do though?

Pin 8 leads to a trim pot labeled "video" so I'm guessing that pin is video. Could this be composite video, or based on the IC on the motherboard does it have to be TTL?
Does all of this seem correct?
If so, how might I adapt this for use with VGA, S-Video, or a composite signal? Could I generate the Sync timing with an arduino, or do the 555 chips do that?
And thanks a lot Kean, the links do help.

Edit: Since I wrote this, there was another reply which confirms what I found. Thank you duak, the pinout you provided is exactly the same as what I figured out.
Do you have a source for that image?

I'll see if I can get the main board powered on to see what the timing signals are. I stupidly took this board out of the terminal a long time ago though and lost the adapter that goes from the pin header to the D-Sub connector.
I think I can tell which pins are H-Sync, V-Sync and Video, but this might be a bit of a crapshoot.
Might there be a way to figure out what timing frequencies are needed with just the CRT board?

And yes, the brightness control is brought out onto a pot and does have high voltages across it. Thanks for the tip.

[vsevolopod]

I figured out what pins on the motherboard were what, hooked up the CRT to it, and got a white screen.
I then took a look at the signal from the H-Sync pin on my scope and found out that it oscillated at about 20Mhz. I couldn't exactly tell with my analog scope though and I don't have a frequency counter, so I was, and still am pretty unsure of this measurement. I looked at the crystals on the board again to see if this was reasonable, and I found one that said 18.4320 Mhz, which gave me hope. Then I looked up the IC right next to it, a MC6845P: A CRT controller! So I think I'm getting somewhere.

I'll keep reading up on the MC6845P, but does anyone here have advice on how I can use this to adapt to VGA, S-Video or Composite?

Thanks!

[duak]

The IBM PC's monochrome display was the model 5151: https://en.wikipedia.org/wiki/IBM_Monochrome_Display_Adapter shows the basic specs.

I doubt the monitor can accept composite video - it's just easier on both ends to not have to mix the signals together and then separate them.  The 555 timers may be there just to make the monitor produce a raster with nothing connected to it or they may generate the ramp waveforms for the deflection yoke.  It's also going to be limited on the vertical and horizontal scan rates and probably won't be able to handle anything above 700 or so pixels per line.

The PAL16C whatever chip on the Mead board is a programable logic device that probably generates the timing and video signals provided by some counters and memory chips further upstream.   There could be an MC6845 as on the IBM board or just a whole bunch of chips.

If you can scope out the signals on the Mead board you should find a signal with a frequency in the range of 50 to 60 Hz.  This is vertical sync.   The horizontal sync should be something in the range of 15 to 20 KHz. The ratios of high time to low time of the sync signals is important because the ratios define the scan and retrace times. The video signal will have bursts of pulses repeated at the vertical sync rate corresponding to the characters or graphics on the screen.  The narrowest video pulse will be a few hundred nanoseconds wide.  The amplitude can change to give a gray scale because the video chain is analog.

IBM didn't always follow industry conventions so a bit of fiddling might be needed to make this monitor work.  It'll be easiest with the monochrome or Hercules format.

I cleared my browser's history so I can't say where I got that diagram from.  Try searching for the term "monochrome crt monitor 10 pin connector"

Cheers,

[vsevolopod]

I found this, about controlling a CGA monitor with an arduino:

https://hackaday.com/2011/07/02/controlling-a-cga-monitor-with-an-arduino/

This could be controlled by an arduino just as easily right?

Does anyone have advice or links to resources on how to do this? It''s getting a bit outside my knowledge and comfort zone with arduino programming (Direct port manipultion, etc)
Would it be easier to control the CRT with an arduino, or using the MC6845? At this point I'm not necessarily talking about inputting video signals, just generating graphics with the arduino, and maybe using the serial interface to change what comes up on the screen.

Thanks!

[Kean]

You really don't want to deal with composite video, as it gets a bit messy and the image quality will too.  But now you seem to know which signals are which, you should be able to drive it with a CGA or VGA signal (even from an Arduino), but with only a single "colour" channel.  Of course, as duak mention, if the frequencies are significantly off from the original design then the video driver circuitry may not be stable and could even burn out.

And yes, you could very likely control the 6845 from an Arduino, or just find a retro board with 6845 video to hack.
e.g. see https://hackaday.com/2017/05/14/the-modern-retrocomputer-an-arduino-driven-6845-crt-controller/

It's been over 30 years since I last worked on 6845 based display circuits, so I'm glad duak chimed in.  I'd have to google for any further detail just like you.
I do still have some similar gear in storage, so one day I'll hopefully get back to them...

[GeorgeOfTheJungle]

You need to generate three signals to drive that monitor.

HSYNC is a small pulse that moves the raster to the beginning (left side of the screen) of a scan line.
VSYNC is a longer pulse that indicates the beginning of a frame (moves the raster to the top of the screen).
VIDEO are very tiny pulses that are sent between HSYNC pulses (during a scan line) to paint the dots on the screen.

The signals may or may not be TTL.
The signals may or may not be positive going.

If the terminal is still working, the first thing you must do is turn it on and look at these signals with an oscilloscope to familiarize with the signals you're going to need to synthesize with an arduino or whatever.

If the CRT has a long persistence phosphor, then the video will (very likely) be interlaced. (Interlaced video is older/slower/easier/cheaper but good enough for a character terminal).

HSYNC is critical, because 1) nothing works without it and 2) if the timing is wrong you may 1) fry the CRT board and/or 2) bath yourself in X-RAYs, yeah, I'm not kidding, if you drive the CRT wrong it will emit (even) more X-RAYs. (WHY? because HSYNC is what drives the high voltage transformer)

You should measure the high voltage at the CRT before any modifications, and check after any modifications that it's not any higher. Turning up the "CRT drive" trimpot counts as a modification.

IF IN DOUBT READ THE X-RAY WARNING STICKER AT THE CRT. I'm not kidding, it's serious.

VSYNC may have/need equalization or serration pulses, that means it's not a single long pulse, but has HSYNC pulses embedded in it.

This is the best link I can find:
http://csweb.cs.wfu.edu/~torgerse/Kokua/Irix_6.5.21_doc_cd/usr/share/Insight/library/SGI_bookshelves/SGI_Developer/books/VFC_PG/sgi_html/ch03.html

[bsudbrink]

Assuming we're not being trolled here...
You do realize that the UBIQ terminal is somewhat rare and that there are vintage computer collectors that would probably give you a pretty good price for it, depending on its condition?

[duak]

To clarify something that GotJ said about the sync pulses and the raster.  The raster is the rectangular pattern painted on the screen by an illuminating spot and is made up of a number of horizontal scan lines.

I don't actually know how fast an Arduino can output data.  It could easily generate the horizontal and vertical timing pulses and output video data between the horizontal pulses (one line) but the video data rate wouldn't allow for high resolution.  ie., it could put out maybe 200 pixels per line giving something like a 200 H x 240 V pixel resolution.  Still not too shabby for an afternoon's work.

Cheers,

[james_s]

IF IN DOUBT READ THE X-RAY WARNING STICKER AT THE CRT. I'm not kidding, it's serious.[/b]

No it's not, it's a ridiculous leftover from regulations that were passed in the 1960s after it was discovered that certain failure modes in early color TV sets could cause the HV rectifier tube and to a lesser extent the CRT to produce xrays. "Modern" CRTs such as this one are made with leaded glass and the HV will top out at maybe 12kV, nowhere near high enough to produce xrays that would penetrate even ordinary glass. There is absolutely zero risk of xray production with a monitor like this, you would melt a hole in the face of the tube before you got enough radiation to expose a piece of film.

Now as for the monitor, it is probably something similar to TTL monochrome, however it may not be identical. I have several terminals and one uses composite video internally which actually works very well for monochrome, where composite doesn't work so well is color video where the color information has to be modulated without breaking the ability to display the signal on a monochrome display. Interfacing a monitor like you have to an FPGA would be very easy, but getting it to work with a modern PC is going to be far more difficult. A much easier approach would be to use it as a terminal as intended, a PC running Linux can use a console connected via a serial port.

[GeorgeOfTheJungle]

IF IN DOUBT READ THE X-RAY WARNING STICKER AT THE CRT. I'm not kidding, it's serious.[/b]

No it's not, it's a ridiculous leftover from regulations that were passed in the 1960s [...]

Yeah, sure... https://duckduckgo.com/?q=scotch+tape+x-ray+machine

A radiography made with scotch tape:
https://www.nature.com/nature/videoarchive/x-rays/index.html

[ebastler]

Yeah, sure... https://duckduckgo.com/?q=scotch+tape+x-ray+machine

Say what?! You mean, that's why each roll of Scotch tape carries an x-ray warning label too, to warn users of the serious hazards?

[vsevolopod]

Yes, I do realize that this is a somewhat rare piece of gear, and I saw an Ebay listing for one at around $500, so yes, there might be some collector interested in it.
This doesn't affect how I feel about modifying it though.

1: This cost me $10 at a surplus store. Value is relative. To some collector it's worth $500, fine. For what? To sit on a shelf and be a "conversation piece"? Realistically, what use does this have in 2019? Is anyone ever going to connect to Lexis Nexis with a Ubiq terminal again? To me, the value in it is the educational experience of learning to work on a piece of equipment like this. I have a pretty limited budget, but I'm very interested in vintage computer hardware. This fits my situation well, so I'm going to work on it and try to find some actual use for it in the 21st century. Plus just by taking it apart and posting about it here I've already learned a lot.

2: I got it in really bad condition. It's clearly been dropped and had a bunch of junk piled on top of it, because the lower part of the case is totally bent. The 2 halves of the clamshell don't mesh together. The top is bent in severely and really hard to bend back. Only the reset button actually works, the power button is totally broken and half missing. The (already awful) keyboard keys are almost all stuck. The plastic that covers the CRT is scratched irreparably. I don't think the motherboard works, I get H-Sync and V-Sync signals and can get a raster on the CRT, but no actual display. I think something else is wrong with the main board. I plan on fixing at least the cosmetic issues but in this condition I don't think it's worth much. Again, worth more to me for the educational value.

I do however feel bad about just wasting all of this beautiful vintage hardware and gutting it and using it for just a case, even if that's why I picked it up in the first place. So, I would like to re purpose as much of it as possible.

Obviously I want to use the CRT. Thanks for all the help. Regardless of weather scotch tape emits x-rays, or if the sticker is a ridiculous leftover, I am going to take all of the precautions possible when working on it, so thanks for the warnings. I'd rather not hurt myself.

It is a monochrome TTL display. I measured the frequencies, it is 20khz H-Sync, and 60Hz V-Sync. I will try to drive it with an arduino first. Yeah it won't be super high res, but still, like duak said, it wouldn't be too shabby for an afternoon's work.

I have been thinking about restoring it as a terminal and maybe interfacing it with serial. I've also considered re-purposing the Z-80 CPU and other components to make some sort of 8 bit home-brew computer. This interests me a bit more than using it as a terminal.
The original goal with this was to use it as a cool looking MIDI/CV sequencer for synthesizers, originally making it a synth itself, but that's out once I decided to keep the CRT. This is still my goal. This or some sort of video overlay generator/titlemaker type thing for my analog video art setup. Basically I want it to be a machine I can use for other creative endeavors , and to learn a bit about old-school computing.

What do you think would be the best option for doing this? I'm starting to actually lean more towards using it as a terminal, but I'm not sure. I am pretty certain, however that the motherboard for it is somewhat non functional, and I think that I may need to change some of the ROMs to make it do what I want. It's getting a bit outside my skill level at that point, and I would need a significant amount more help to do that. I'm not experienced with this stuff, but I'm good at following directions, so I might be up for it. Anyone interested in helping me troubleshoot this and find documentation for it? Is what I want to do with it even reasonable if I'm using it as a terminal?

Honestly I think the most achievable and satisfying thing for me to do is drive the display with an arduino and having it show low res graphics and stuff that change based on serial input or something, and interfacing it with other hardware that way. That's kind of the direction I'm leaning with this. Either way, I have to put this project to the side for a few days to work on other things, so I'll think about it. In the mean time, I'm very interested in more feedback from here.

[bsudbrink]

No offense or criticism was intended, just wanted to be sure you were aware.  I knew about the ebay one, but 500USD is a bit much, even for a pristine example.

Otherwise, most collectors I know operate their gear regularly, I do as well.

Finally, I apologize about the troll comment.  On some of the other vintage computer e-groups that I participate in, from time to time we get some joker who thinks it's funny to post something like:

I was having trouble fitting my Apple I into a frame for display so I cut a couple of inches off one side.  Do you think that hurts its value much?

or

I've made my DEC PDP-1 console into a fish tank, but I'm having trouble getting the corners water tight.  Any tips?

[james_s]

IF IN DOUBT READ THE X-RAY WARNING STICKER AT THE CRT. I'm not kidding, it's serious.[/b]

No it's not, it's a ridiculous leftover from regulations that were passed in the 1960s [...]

Yeah, sure... https://duckduckgo.com/?q=scotch+tape+x-ray+machine

That's interesting, I hadn't seen that before. It really doesn't apply here though, the penetrating power of an xray photon is dependent on the potential of the electron that created it. Static electricity can easily generate many tens of kV, far higher voltage than the 7-12kV from a small CRT monitor. You simply cannot generate significant xray radiation from a CRT, it has been tried, it doesn't work. Low kVp xray tubes used for mammography operating at around 30kVp have special beryllium windows because the glass used in ordinary tubes absorbs the low energy xrays. Conventional xray tubes don't start producing much output until you reach around 40kVp and most heads use a 1mm aluminum filter to block most of the radiation below 50kVp. CRTs are made of leaded glass and the face of even a small tube is around 8mm thick, I have taken xrays *of* small CRTs and I have to crank the machine up to >100kVp and take a relatively long exposure to get a good image. Trust me, you are not going to get a meaningful amount of xrays out of one.

[GeorgeOfTheJungle]

Hi James, the regulation you cite is for TVs. Do you know if it applied to computer displays as well? I've downloaded the .pdf and attached it to this message.

It says the max radiation must be less than 0.5 milli roentgens per hour per 10 cm², a 9" CRT has ~ 250 cm² so 0.5*250/10= 12.5 mR/hour, that's 1 roentgen every 80 hours. The wikipedia says (*) "One roentgen of X-rays may deposit anywhere from 0.01 to 0.04 Gy (1.0 to 4.0 rad) in bone depending on the beam energy", and under rad (**) it says:

Dose examples
25   rad:   lowest dose to cause clinically observable blood changes
200   rad:   local dose for onset of erythema in humans
400   rad:   whole body LD50 for acute radiation syndrome in humans
1   krad:   whole body LD100 for acute radiation syndrome in humans[7]
1   krad:   typical radiation tolerance of ordinary microchips
4 to 8   krad:   typical radiotherapy dose, locally applied
10   krad:   fatal whole-body dose in 1964 Wood River Junction criticality accident[8]
1   Mrad:   typical tolerance of radiation-hardened microchips

Do you think a modified or mis adjusted CRT display can't emit much more than that?

I've ever heard "beware of x-rays, watch out the HV" and always have done so. But honestly, I'm not sure how/what does all that radiation numbers add up to. Do you? Can you help?

Thanks,

(*) https://en.wikipedia.org/wiki/Roentgen_(unit)
(**) https://en.wikipedia.org/wiki/Rad_(unit)

[GeorgeOfTheJungle]

Say what?! You mean, that's why each roll of Scotch tape carries an x-ray warning label too, to warn users of the serious hazards?

Exactly... look: https://www.nature.com/nature/videoarchive/x-rays/index.html

[james_s]

I don't know if the regulation was ever updated to cover computer displays but CRTs had already been made of leaded glass for decades by the time home computers were a thing. 12kV is not going to produce hard enough X-rays to penetrate the glass, maybe you could get something detectable if you pushed the HV up to 50kV or more but you'd have to be really trying and you would melt a hole in the glass quickly. I have a fairly sensitive Geiger counter and have never seen it register anything from even large bright CRTs. Even small dental X-ray tubes have a large copper slug around the anode to absorb the heat and they are only good for low current and short duty. Larger tubes have big tungsten faced molybdenum anodes that are spun by a motor, if that fails it will melt a crater almost instantly.

[GeorgeOfTheJungle]

James, more info (*):

Typical Doses from Diagnostic Radiology Exams

As noted above, the tables below give dose estimates for typical diagnostic radiology exams.
For comparison, we all receive about 3 mSv of exposure annually to natural background radiation.

Plain Film X Rays
Single Radiographs        Effective Dose (mSv)
Skull (PA or AP)1         0.03
Skull (lateral)1          0.01
Chest (PA)1               0.02
Chest (lateral)1          0.04
Chest (PA and lateral)5   0.06
Thoracic spine (AP)1      0.4
Thoracic spine (lateral)1 0.3
Lumbar spine (AP)1        0.7
Lumbar spine (lateral)1   0.3
Abdomen (AP)1             0.7
Abdomen6                  0.53
Pelvis (AP)1              0.7
Pelvis or hips6           0.83
Bitewing dental film6     0.004
Limbs and joints6         0.06

1 rad= 10 mSv (***), if it's true (see my previous message) that a 9" CRT could emit up to 1 roentgen = a dose of 1..4 rad every 80 hours (and be compliant with the applicable regulations) that's the equivalent of between 33 and 132 (see EDIT below) skull radiographs, and equivalent to one radiograph every 30m to 2.5h in front of a CRT screen (**).

This isn't looking good to me... I've got lots of old green, amber, colour and B&W monitors from the 70's and 80's, I want to check it and see it with my own eyes, James. What do I need? A geiger counter? One like this one https://www.amazon.com/GQ-GMC-300E-Plus-Radiation-dosimeter-monitoring/dp/B00HOUW066 would be good enough, for example? Or which one? Or what do I need? Can't find any videos on youtube (hey, that's an idea for content creators, DAVE!) when I search for "CRT display geiger counter".

Thanks,

Edit: wrong NOT 33..132, (10..40)mSv/0.03mSv= 333..1332 radiographs !!!

(*) https://hps.org/hpspublications/articles/dosesfrommedicalradiation.html
(**) At 5 cm of the screen...
(***) https://www.translatorscafe.com/unit-converter/en/radiation-absorbed-dose/1-24/rad-sievert/

[ebastler]

Hi James, the regulation you cite is for TVs. Do you know if it applied to computer displays as well? I've downloaded the .pdf and attached it to this message.

It says the max radiation must be less than 0.5 milli roentgens per hour per 10 cm², a 9" CRT has ~ 250 cm² so 0.5*250/10= 12.5 mR/hour, that's 1 roentgen every 80 hours.

I believe you misunderstood the meaning of the 10 cm² area in the FDA standard. In my understanding, the standard requires the use of a 10 cm² measurement aperture to ensure adequate averaging over spatial variations of the radiation distribution. But the dose limit of 0.5 mR is a total limit for the device under test, not the allowed dose per 10 cm² area.

(I could be wrong, as always.  )

[duak]

The big concern with CRTs is that they generate Xrays.  The energy and amount is a function of accelerating voltage and beam current as well as the target the electron beam hits. Color CRTs have a metal shadow mask that intercepts some of the beam whereas monochrome CRTs do not.  A color CRT having a much higher accelerating voltage was a much more hazardous source, especially the older ones using vacuum tube HV rectifiiers.  This was recognized early on so the electronics were shielded and the CRT face itself was made of leaded glass.  I do not know if monochrome CRTs are leaded, however, the smaller ones have quite low accelerating voltages - about 10 KV or so - and don't have a metallic target.

Here's a link to a report on monitors of many types: https://inis.iaea.org/collection/NCLCollectionStore/_Public/16/030/16030661.pdf

Cheers,