Component video questions

[TRS-90]

I don't know if this is the right place for this, so if it is the wrong place forgive me.

Component video has a sync signal in one of the cables (sometimes the green one in case of the Playstation 2) and I was wondering if it was possible to separate the sync (or horizontal/vertical drive) signal from the color signal in whatever wire contains it.
I am trying to drive a CRT from a macintosh by reverse engineering the analog board but I need a way to drive the coils that make the electron beam move.
I can use the RGB signal from the cables I just need to make the image work instead of just being a dot on the screen.

I am a beginner in the electronics, I just have done some research on CRT technology but have found very little.

[stmdude]

Have a look at the LM1881.  It's a "Video Sync Separator", so it might fit the bill.

[TRS-90]

Thanks!  This just might solve my problem!

If it can work with a component signal instead of only composite then I can make it work.

[stmdude]

Should work fine on component as well, it's the same timing.  Just make sure to put it on the signal that actually has the sync..

[TRS-90]

Just got to make sure that I'm able to hook it up to the deflection coils.
I may need a transformer to boost the sync signal to make it drive the coils.

But the coil connector has 4 wires but this chip has only 2 sync related outputs.

I don't know how much of a problem that will be, but it should help anyway.

[N2IXK]

If you want to drive the coils in a deflection yoke,  you need a lot more than a "transformer".  You need power amplifier stages designed to work with the particular deflection yoke in question. Like the ones already in place in the Mac CRT monitor....

You also will need circuitry to shape the pulses from the LM1881 into proper sawtooth waveforms to get anything resembling proper linearity on the screen.

You should really study the monitor schematic, and see how the original design dealt with the sync signals, separated them into H and V sync, and applied the sync signals to the H and V sweep oscillator stages.

The majority of CRT monitors derive various operating voltages (including CRT high voltage) from the horizontal sweep stages, so forcing the horizontal sweep to work at a radically different frequency than designed for can upset these voltages, and perhaps even damage the circuitry. In short, what you are trying to do is nowhere near as trivial as you seem to think, especially without a good grasp of the workings of CRT displays.

[TRS-90]

Ah.
I never really thought it was easy seeing as no one has ever done this before from what I know.
I found a hookup for the color but the issue with the project is getting from a dot on the screen to a working image.
I just need a way to make it work and also I can't seem to find any decent info on the way sync was handled(I'm not the greatest at understanding analog technology).
I don't know if this helps but the CRT in question is a 14" Sony Trinitron (but only 13" are visible or so I hear) from 1994/95.

Any information I can find helps.

 

[TRS-90]

As a side note I looked up "Square wave to sawtooth converters" and did find some.
So that is one problem solved (probably not because I'm never THAT lucky).

[N2IXK]

First thing to do is to get a schematic/service manual for the monitor you are working with.  Trying to reverse engineer or modify the circuitry without one is NOT a good idea, particularly if you aren't familiar with the basics of how analog video and CRT displays work.

BTW, most CRT monitors use a "hot chassis" type power supply system, which can present a serious shock hazard when attempting to add new I/O connections like you are seeking to do.  Again, a schematic is needed here to figure out the boundaries of the hot and isolated portions of the circuitry.

[Richard Crowley]

What exactly does "a CRT from a macintosh" mean?
Do you mean "a CRT monitor with RGB video inputs", or do you mean "a raw tube with the yoke and no supporting electronics"?
If it is the latter, I'm sorry to report that you have an almost impossible task.

Creating the precise horizontal and vertical sweep waveforms, and all the high (and VERY high) voltages for the CRT are a very specialized form of circuit design. Not to mention all the very specialized and arcane circuits needed to focus and align the three electron beams to the Red, Green, and Blue phosphors on the face of the tube.
Creating all that from scratch, or even adapting the chassis from some other color CRT, is an advanced project that significantly eclipses whatever other project you are attempting.

If you really want a color CRT display, you need to start from a complete, functional chassis with the tube, the yoke, and all the associated electronic circuits.

[TRS-90]

What I mean is that it is a standard Sony trinitron tube just driven by apple designed circuitry.
There is some supporting circuitry inside (if a neck board counts).
The neck board has RGBHV connectors on it but the sync information and color is generated by the motherboard and, I assume, is amplified by the chassis
So while I may be able to get a dot on the screen using component connectors, that's all it will be, a dot, while that is a leap in my book, it isn't a stable image.

So what I'm looking for is a way to get the electron beam to move left, right, up and down as per the video signal being put into the CRT.

I'm trying to hook up a video game console to the tube and use the original case and tube, that is my goal.

[TRS-90]

A mild correction on my previous statement:

I do have an analog board but I can't just solder a video input and call it a day unfortunately.

I did hear that some of these boards (even the ones put into macintosh computers) had an RF section on them but I can't find it on my board because I don't know what I'm looking for.

End of correction.

[stmdude]

For other people wondering what's _actually_ inside this thing, here's the "Service Source" documentation:
http://tim.id.au/laptops/apple/legacy/performa_500_series.pdf

I don't know much about CRTs (except that they'll more or less kill you if you don't know what you're doing), but I do know that you're going to need to use that analog-board..

To figure out _how_ to use the analog-board, you're going to need an oscilloscope (unless you can find the schematics) to probe the signals between the main-board and the analog-board, and figure out which signals are where, and at what timings and levels.

Once you know that, you'll need to get whatever game-console you wanted to put in there, and see if you can extract those exact signals from it somehow.

Unfortunately, most likely, you won't be able to, as pretty much all game-consoles are meant to drive a TV, whilst what's in the Performa is a monitor. The monitor in this case has a resolution of 640x480, meaning, it's not one of the standard TV signals (PAL/NTSC), nor is it a multiple thereof.
As the trinitrons have an aperture grille (does all CRTs have them?), even if you somehow manage to drive it with different timing (NTSC for example), the picture would be "blurry", as the pixels won't line up with the grille.

[TRS-90]

I can't afford an oscilloscope unfortunately and nor can I find the schematics but from what I understand a CRT of the exact same size was used in a Sony TV.
It was a 14" TV.
The monitor uses a 66.67Khz /Mhz?/Hz? signal from what the manual said (that or it was one of the shadowmask CRTs used in other computers).
The reason I thought I could use a transformer to drive the coils is because of the "turn a TV into an oscilloscope" project that has been around for a while.
They use a somewhat similar method by hooking up the left and right audio channels to the coils through an amp or some sort of signal booster(it's been done on some pretty big televisions too) to make the line on the screen into a waveform from the signal source.

[Richard Crowley]

If all you have is the tube and the yoke (and color alignment adjustments etc.) then you have less than half of the actual display.  I am afraid you are dramatically underestimating what it takes to operate that part of a display you have.  IMHO, it is simply not practical to try to revive what you have unless you have the whole circuit that it came with.  As you learn more about what it takes to drive a color CRT display, you will realize what a monumental challenge you are facing.  With ~50 years of experience behind me (including color CRT displays) I wouldn't even dream of attempting such a thing. Sorry.  If you really want a color CRT display, then at least start with the WHOLE THING instead of less than half of what you need.

Even if you could get the vertical and horizontal sweep to operate, you really don't even want to know what a horror it is to try to get the colors aligned so that the picture doesn't look like a 1960s hippie drug-induced dream image of rainbow patterns.

[TRS-90]

I actually have the whole thing.
But a lot of it is by apple so getting the schematic is near impossible.

The adjustments are near where I can attach the component inputs.

The coils have wires going to the board so I just need a signal booster and the signal information.

I just need the make the signals come from an external source instead of the motherboard.
But the problem is that is has no TV type circuitry, if it did I could just solder an input to it and call it a day.

A blurry image on the screen would make me more than happy.

[TRS-90]

Who know maybe I could jerry rig the motherboard and use that because it drive a lot of the video .

I wish there was a way to find the apple schematic but they probably have it locked in a mission impossible level vault somewhere in cupertino california.

[helius]

The neck board does not carry H or V sync signals, those are applied to the flyback and HOT section which drive the deflection yoke. The neck board receives the R, G, B, signals and may also have "drive", "screen", or "blank" signals, if it contains video amps, which is common. The grid electrodes (high voltages) and the filament circuit are also connected through it.

Component inputs could mean RGBHV, but with household equipment it more often means YsUV (YPbPr). Those cannot drive a CRT without being converted to the RGB color space.

[helius]

I explained in your other thread that the Mac main board does not have any monitor driver circuits, those are all on the other board. If you compare photos of the LC 575 board and something without a CRT like the LC 475 board, they have the same (5V) components. CRT drive circuitry is at high voltages, and you can tell just from the PCB material that it is completely different.
What kind of inputs does a computer monitor receive from a computer? The only ones that matter are RGB and Horizontal and Vertical sync. Nothing composite and nothing with YUV components. All of those signals are present at the edge of the "analog" board, and my bet would be that they are at normal signal levels (1 Vp-p for the color channels and + or - 5V for sync). Those are the only things you should connect to.

[TRS-90]

What I meant in my other thread was that the signals are generated by the motherboard but are amplified by the analog board(at least that is what I heard, you are probably correct, I am fairly new to the world of video circuitry).

Does anyone know what pins on the connector harness has the video (R, G, B, H sync and V sync) and power supply activation.
That would probably solve most of my issues.

[TRS-90]

Would something like SCART work?  or 
It has R, G, and B, sync pins and so on.

I am completely unfamiliar with SCART as I have never used it, please feel free to correct me on this one.

There is a SCART adapter for to playstation 2 so that is not a problem.

Or is SCART the worst possible way to go about this.

Also what does SCART stand for?

[Richard Crowley]

Here is a typical last-generation, highly-integrated color CRT circuit.
You have the block in the center-right.
What you are missing is the REMAINDER of the circuit.
Including the deflection circuits and the EHV circuit in the block at the lower right.
These are very tricky circuits even to repair in a commercial product.
Nobody in 2016 would attempt to reproduce a circuit like that from scratch.

[TRS-90]

WOW.
Thanks for that.
Now I might be able to decipher some of this thing.

I can read circuit diagrams pretty well if I know the components.

I'll see if I can work this out.

[chris_leyson]

1960s hippie drug-induced dream image of rainbow patterns.

you get interesting interference patterns when you bend the shadow mask out of shape, somehow it got dropped on its front, can't remember how it happened, wasn't there at the time. It buckled into a different shape, I'm looking at cool diffration patterns and everybody else looking a busted TV, they didn't think it was cool

[Richard Crowley]

WOW.
Thanks for that.
Now I might be able to decipher some of this thing.

I can read circuit diagrams pretty well if I know the components.

I'll see if I can work this out.

REMEMBER:

This is NOT the circuit for your device.  It represents the complexity of even the most simple color CRT monitor system.  Your unidentified device may have more or less circuit than the one in the diagram on the board(s) around the yoke.  Can you find an OEM identification on the board somewhere?  I doubt that Apple actually made their own CRT displays. They probably bought them as sub-assemblies from an OEM.  Just as today they don't make their own LCD screens, they buy them from an OEM.

CRTs use LETHAL voltages that will KILL YOU DEAD. Even experienced technicians and engineers are at risk.  This is NOT the kind of circuit to be fooling around with if you are "a beginner in the electronics".  You are way out of your depth.   

The inner and outer conductive coating of the CRT display tube form a VERY POWERFUL storage capacitor.  Even hours after being powered down and completely disconnected, the charge stored in a CRT is potentially LETHAL.