CRT RGB Modification/Hack [56K Warning]

Start a work log and update it occasionally with your projects progress
User avatar
Shadow
Verified
Admin / PSXDEV
Admin / PSXDEV
Posts: 2670
Joined: Dec 31, 2012
PlayStation Model: H2000/5502
Discord: Shadow^PSXDEV

CRT RGB Modification/Hack [56K Warning]

Post by Shadow » April 28th, 2015, 5:35 pm


WARNING
TV's carry extremely high voltages with enough current to kill you.
DO NOT do RGB mods to "hot chassis" CRT's.
:null:
Always power your CRT from an isolation transformer.
:null:
DO NOT attempt these modifications unless you are confident in the electrical engineering field and understand what you are doing.

:null:
I thought I would startup a log about how to modify a CRT TV to accept an RGB signal from your PlayStation 1.
Now, considering I'm in Australia, we don't have SCART on our TV's, so we have to use composite video.

This is what composite versus RGB looks like:
Image

The goal here is to try and get a picture that is as rich, clear and sharp like this:
Image

So, how do we add RGB to a CRT TV that wasn't designed to. Well, all CRT TV's have an electron gun that fires at the red, green and blue phosphors from the neck board through the shadow mask. This is how the Commodore 1084 monitors did it, and you can see that in this schematic here and here. If we cut the red, green and blue connections from the TV's mainboard and inject our own RGB voltages, say, from a PlayStation 1 (SEGA Mega Drive, Amiga, Arcade PCB, etc) and directly drive the guns, we should get a crystal clear, analog based RGB picture:
Image

Now take in mind that whilst I have managed on my little CRT to drive the on screen display chip and force out an RGB signal from the "Jungle IC" or "On-Screen Display" (OSD), most of them use digital RGB and not analog RGB, so you wont get nice gradients. I do not consider injecting an RGB signal by this method a valid way of adding RGB to a CRT TV. It's a cheap and nasty way of doing it. This direct gun drive method involves driving the guns directly with pure analog RGB :)

Moving along, as a test, I re-opened my small 14" CRT 'Konka' CRT TV and found three C4544 NPN transistors. They are "Switching and Horizontal driver chroma output transistors, rated for 300V at 0.1A (8W)". If I were to inject my RGB signal into their collectors, in theory, I should get a picture from my PlayStation 1:
Image

Before I do that however, a test was needed to know what transistor was driving each gun. So, I cut each wire and then tinned their ends. One of the main driver lines coming from the "Jungle IC" was left so that I had a voltage to feed into the guns:
Image

The alligator clip was placed in series to one of the transistors, and behold, a red picture is rastered through the shadow mask:
Image

The Sony BIOS screen is displayed because I am feeding composite as the image. I just as well could have instead displayed a channel with noise being displayed, but this image sufficed better. Ah hell, here you go as proof:
Image

So, moving along, I tested the other two transistors:
Image

Hooray! Green is rastered now. I wonder what the next colour could be...
Image

Oh, it's blue :roll:
Image

Now let's ensure we are actually driving RGB and not CYMK.
I tied the red and green lines together, and according to colour law as seen here:
Image

Yes, those three lines are indeed red, green and blue because red and green makes yellow.
The camera doesn't do the colour justice, but the tube is giving a really nice yellow in real life:
Image

And yes again. All three tied together gives us a white raster:
Image

So let's jump into the electronics side more. As a note, the "Jungle IC" is outputting and driving the transistors at 2.50V. This is on a 14" CRT, so a larger tube in theory should run at a higher voltage. The first problem is the fact that the PlayStation outputs RGB at ~2.0V, so the pictures contrast to brightness ratio is going to be a little dim. With that in mind, an amplification circuit is going to be needed to triple the PlayStation's RGB output voltages.

For my next test, I wont be using a PlayStation, but instead a SEGA Mega Drive 2.
I already have a breakout cable for the Mega Drive 2, so it was very easy and quick to hook up to the TV's cathodes.

So here we have a Mega Drive 2 ready for the test:
Image

Some pinouts:
Image

Behold! A red raster driven by the Mega Drive 2!!
Image

How about we get all three signals hooked up ;)
Image

Not bad, but like I said, the voltage is too low and so the contrast to brightness ratio is off, hence a dull picture. I actually measured the voltage coming from the Mega Drive 2's RGB lines, and it was 2.00V. We are only 0.50V off, and still it's pretty dull.

To see how dull it really is, the game I loaded (Columns) has a 'color test':
Image

One thing you can always do is increase the flyback voltage slightly. It's not the best idea, but it's a quick and easy fix. The best method, is to actually have potentiometers on each RGB line (so you can fine tune each colour), and boost the levels with a transistor which I will end up doing. For now, this is what the image looks like with the flyback readjusted:
Image
Image

It's a lot better, but it's not perfect. It's still a little dim, and, what if I plug in my Amiga or PlayStation. If the levels are lower, the screen will dim down again. If they are higher, it's no big deal because I can lower the flyback voltage back down again, but I should not have to do that.

Here are a few more images of RGB from the Mega Drive via my Western Technologies SRAM cartridge:
Image
Image
Image
Image
You do not have the required permissions to view the files attached to this post.
Development Console: SCPH-5502 with 8MB RAM, MM3 Modchip, PAL 60 Colour Modification (for NTSC), PSIO Switch Board, DB-9 breakout headers for both RGB and Serial output and an Xplorer with CAETLA 0.34.

PlayStation Development PC: Windows 98 SE, Pentium 3 at 400MHz, 128MB SDRAM, DTL-H2000, DTL-H2010, DTL-H201A, DTL-S2020 (with 4GB SCSI-2 HDD), 21" Sony G420, CD-R burner, 3.25" and 5.25" Floppy Diskette Drives, ZIP 100 Diskette Drive and an IBM Model M keyboard.

User avatar
sickle
Verified
C Programming Expert
C Programming Expert
Posts: 257
Joined: Jul 17, 2013
I am a: Chocolate-fueled pug fetish robot.
Location: Scotland

Post by sickle » April 28th, 2015, 8:11 pm

Still laughing at this bit:
Image

User avatar
Shadow
Verified
Admin / PSXDEV
Admin / PSXDEV
Posts: 2670
Joined: Dec 31, 2012
PlayStation Model: H2000/5502
Discord: Shadow^PSXDEV

Post by Shadow » April 28th, 2015, 8:28 pm

Heh, I couldn't find an RCA plug to cut up, so a screw driver wedged in there did the trick :P
Development Console: SCPH-5502 with 8MB RAM, MM3 Modchip, PAL 60 Colour Modification (for NTSC), PSIO Switch Board, DB-9 breakout headers for both RGB and Serial output and an Xplorer with CAETLA 0.34.

PlayStation Development PC: Windows 98 SE, Pentium 3 at 400MHz, 128MB SDRAM, DTL-H2000, DTL-H2010, DTL-H201A, DTL-S2020 (with 4GB SCSI-2 HDD), 21" Sony G420, CD-R burner, 3.25" and 5.25" Floppy Diskette Drives, ZIP 100 Diskette Drive and an IBM Model M keyboard.

User avatar
sickle
Verified
C Programming Expert
C Programming Expert
Posts: 257
Joined: Jul 17, 2013
I am a: Chocolate-fueled pug fetish robot.
Location: Scotland

Post by sickle » April 29th, 2015, 12:43 am

It's not daft if it works ^^

Gradius
Verified
Extreme PSXDEV User
Extreme PSXDEV User
Posts: 220
Joined: Sep 09, 2012
I am a: IT Consultant, Systems Integrator
PlayStation Model: 7501
Location: Chile

Post by Gradius » April 29th, 2015, 7:16 am

Everybody should also knows the voltage from any TV can KILL you on instant, so be very careful if anybody is going to try this out.

You just used composite video input to inject CSYNC ?

Gradius
Verified
Extreme PSXDEV User
Extreme PSXDEV User
Posts: 220
Joined: Sep 09, 2012
I am a: IT Consultant, Systems Integrator
PlayStation Model: 7501
Location: Chile

Post by Gradius » April 29th, 2015, 8:24 am

Also, I always hear you should decop all the RGB lines to avoid the voltage to flow from TV to the equipment (console, etc), otherwise you could smoke the console.

User avatar
Shadow
Verified
Admin / PSXDEV
Admin / PSXDEV
Posts: 2670
Joined: Dec 31, 2012
PlayStation Model: H2000/5502
Discord: Shadow^PSXDEV

Post by Shadow » April 29th, 2015, 4:21 pm

Gradius wrote:You just used composite video input to inject CSYNC ?
Yes, but I also will wire the AV input to the RF line so that as soon as the TV is turned on, no matter what channel you pick, it will always be in RGB mode.
Gradius wrote:Also, I always hear you should decop all the RGB lines to avoid the voltage to flow from TV to the equipment (console, etc), otherwise you could smoke the console.
By "decop", you mean, decouple?
If you want complete isolation, a 1:1 winding transformer is required.
Gradius wrote:Everybody should also knows the voltage from any TV can KILL you on instant, so be very careful if anybody is going to try this out.
Noted. I added a disclaimer to the first post at the header.
Development Console: SCPH-5502 with 8MB RAM, MM3 Modchip, PAL 60 Colour Modification (for NTSC), PSIO Switch Board, DB-9 breakout headers for both RGB and Serial output and an Xplorer with CAETLA 0.34.

PlayStation Development PC: Windows 98 SE, Pentium 3 at 400MHz, 128MB SDRAM, DTL-H2000, DTL-H2010, DTL-H201A, DTL-S2020 (with 4GB SCSI-2 HDD), 21" Sony G420, CD-R burner, 3.25" and 5.25" Floppy Diskette Drives, ZIP 100 Diskette Drive and an IBM Model M keyboard.

User avatar
Shadow
Verified
Admin / PSXDEV
Admin / PSXDEV
Posts: 2670
Joined: Dec 31, 2012
PlayStation Model: H2000/5502
Discord: Shadow^PSXDEV

Post by Shadow » April 29th, 2015, 4:45 pm

Now, I hooked up my PlayStation 1, and its RGB output voltage was 2.35V.
With the flyback at the same levels with the Mega Drive 2 test, it's very dull.
The image is razor sharp and rich, but way too dim:
Image
Image
Image
Image
Image

One problem I thought that would also cause a dull image, was the frequency the RGB lines were running at.
How about we pop the original green signal on my scope:
Image

It seems the "Jungle IC" was driving the guns at 15KHz.

Let's take a look at the PSX's green line:
Image

Wow, that's very low...

At the moment, I have ordered some 2SC1815 transistors to build the amplification circuit. More to come soon...
You do not have the required permissions to view the files attached to this post.
Development Console: SCPH-5502 with 8MB RAM, MM3 Modchip, PAL 60 Colour Modification (for NTSC), PSIO Switch Board, DB-9 breakout headers for both RGB and Serial output and an Xplorer with CAETLA 0.34.

PlayStation Development PC: Windows 98 SE, Pentium 3 at 400MHz, 128MB SDRAM, DTL-H2000, DTL-H2010, DTL-H201A, DTL-S2020 (with 4GB SCSI-2 HDD), 21" Sony G420, CD-R burner, 3.25" and 5.25" Floppy Diskette Drives, ZIP 100 Diskette Drive and an IBM Model M keyboard.

Gradius
Verified
Extreme PSXDEV User
Extreme PSXDEV User
Posts: 220
Joined: Sep 09, 2012
I am a: IT Consultant, Systems Integrator
PlayStation Model: 7501
Location: Chile

Post by Gradius » April 30th, 2015, 12:59 pm

A super simple solution would be using a THS7314:
http://www.ti.com/product/ths7314

Just need three 75 Ohm resistors and one 0.1uf Capacitor.
Shadow wrote:By "decop", you mean, decouple?
Correct.

Putting 1N4148 diodes on RGB lines is more than enough for it.
Last edited by Gradius on April 30th, 2015, 3:20 pm, edited 1 time in total.

Gradius
Verified
Extreme PSXDEV User
Extreme PSXDEV User
Posts: 220
Joined: Sep 09, 2012
I am a: IT Consultant, Systems Integrator
PlayStation Model: 7501
Location: Chile

Post by Gradius » April 30th, 2015, 2:13 pm

So let's be clear here.

You injected the R-G-B signals on each (related colour) C4544 collector pin?

I assume you cut out the collector pin coming from Jungle IC.

All the rest remains intact?

Is so sad the jungle IC generally uses TTL signal and not analog, otherwise I would just inject my R-G-B signal there.

Gradius
Verified
Extreme PSXDEV User
Extreme PSXDEV User
Posts: 220
Joined: Sep 09, 2012
I am a: IT Consultant, Systems Integrator
PlayStation Model: 7501
Location: Chile

Post by Gradius » April 30th, 2015, 3:29 pm

Back in 2010 I found this article, now is possible to read again thanks to web.archive:
http://web.archive.org/web/201005300921 ... c-ntsc-tv/

User avatar
Shadow
Verified
Admin / PSXDEV
Admin / PSXDEV
Posts: 2670
Joined: Dec 31, 2012
PlayStation Model: H2000/5502
Discord: Shadow^PSXDEV

Post by Shadow » April 30th, 2015, 4:41 pm

Gradius wrote:A super simple solution would be using a THS7314:
http://www.ti.com/product/ths7314

Just need three 75 Ohm resistors and one 0.1uf Capacitor.
Shadow wrote:By "decop", you mean, decouple?
Correct.

Putting 1N4148 diodes on RGB lines is more than enough for it.
I thought about using an IC, but I thought I would go with the transistor method (better control).

If I use diodes, I will loose 0.7V because of the forward drop.

Gradius wrote:So let's be clear here.

You injected the R-G-B signals on each (related colour) C4544 collector pin?

I assume you cut out the collector pin coming from Jungle IC.

All the rest remains intact?

Is so sad the jungle IC generally uses TTL signal and not analog, otherwise I would just inject my R-G-B signal there.
That's right.

Gradius wrote:Back in 2010 I found this article, now is possible to read again thanks to web.archive:
http://web.archive.org/web/201005300921 ... c-ntsc-tv/
I'm not sure if his is digital RGB or analog RGB.
I tried his method, and whilst it worked, in the image above with 'Columns' color test' pattern, they all appeared as one solid colour, and not as a gradient like you see, so mine was digital (TTL). Maybe if I input and drive the "Jungle IC" directly from my RGB signal it might amplify it as analog RGB, but still, for older TV's that don't have a "Jungle IC", I will try the transistor driven circuit.
Development Console: SCPH-5502 with 8MB RAM, MM3 Modchip, PAL 60 Colour Modification (for NTSC), PSIO Switch Board, DB-9 breakout headers for both RGB and Serial output and an Xplorer with CAETLA 0.34.

PlayStation Development PC: Windows 98 SE, Pentium 3 at 400MHz, 128MB SDRAM, DTL-H2000, DTL-H2010, DTL-H201A, DTL-S2020 (with 4GB SCSI-2 HDD), 21" Sony G420, CD-R burner, 3.25" and 5.25" Floppy Diskette Drives, ZIP 100 Diskette Drive and an IBM Model M keyboard.

Gradius
Verified
Extreme PSXDEV User
Extreme PSXDEV User
Posts: 220
Joined: Sep 09, 2012
I am a: IT Consultant, Systems Integrator
PlayStation Model: 7501
Location: Chile

Post by Gradius » May 1st, 2015, 2:36 am

Let's us know how it was after the components arrive.

Now I'm even planning to get a Sony 27" Trinitron WEGA. :lol:

User avatar
Shadow
Verified
Admin / PSXDEV
Admin / PSXDEV
Posts: 2670
Joined: Dec 31, 2012
PlayStation Model: H2000/5502
Discord: Shadow^PSXDEV

Post by Shadow » May 1st, 2015, 2:45 am

Gradius wrote:Let's us know how it was after the components arrive.

Now I'm even planning to get a Sony 27" Trinitron WEGA. :lol:
Will do. I ordered a THS7314 should the individual transistor based circuit not be powerful enough, and also to probably compare results.
Development Console: SCPH-5502 with 8MB RAM, MM3 Modchip, PAL 60 Colour Modification (for NTSC), PSIO Switch Board, DB-9 breakout headers for both RGB and Serial output and an Xplorer with CAETLA 0.34.

PlayStation Development PC: Windows 98 SE, Pentium 3 at 400MHz, 128MB SDRAM, DTL-H2000, DTL-H2010, DTL-H201A, DTL-S2020 (with 4GB SCSI-2 HDD), 21" Sony G420, CD-R burner, 3.25" and 5.25" Floppy Diskette Drives, ZIP 100 Diskette Drive and an IBM Model M keyboard.

User avatar
sickle
Verified
C Programming Expert
C Programming Expert
Posts: 257
Joined: Jul 17, 2013
I am a: Chocolate-fueled pug fetish robot.
Location: Scotland

Post by sickle » May 1st, 2015, 4:10 am

Shadow wrote:By "decop", you mean, decouple?
If you want complete isolation, a 1:1 winding transformer is required.
What sorta efficiency would you get there?
Also, would interference be an issue around the arse end of the telly?

User avatar
Shadow
Verified
Admin / PSXDEV
Admin / PSXDEV
Posts: 2670
Joined: Dec 31, 2012
PlayStation Model: H2000/5502
Discord: Shadow^PSXDEV

Post by Shadow » May 1st, 2015, 4:29 am

From what I remember, transformers are pretty efficient. There would be some losses of course, but the signal can be re-amplified back up again afterwards. That's just overcomplicating things now though.

Image

What kind of interference? Do you want to talk on your mobile phone behind the neck board or something :P
When I tested the transistors on the neck board, the collector was in the negated voltage rage. Interference won't be a problem, but I guess I should put a diode for safety to prevent back flow current should that ever happen (IE: shorted transistor)...
Development Console: SCPH-5502 with 8MB RAM, MM3 Modchip, PAL 60 Colour Modification (for NTSC), PSIO Switch Board, DB-9 breakout headers for both RGB and Serial output and an Xplorer with CAETLA 0.34.

PlayStation Development PC: Windows 98 SE, Pentium 3 at 400MHz, 128MB SDRAM, DTL-H2000, DTL-H2010, DTL-H201A, DTL-S2020 (with 4GB SCSI-2 HDD), 21" Sony G420, CD-R burner, 3.25" and 5.25" Floppy Diskette Drives, ZIP 100 Diskette Drive and an IBM Model M keyboard.

User avatar
sickle
Verified
C Programming Expert
C Programming Expert
Posts: 257
Joined: Jul 17, 2013
I am a: Chocolate-fueled pug fetish robot.
Location: Scotland

Post by sickle » May 2nd, 2015, 2:30 pm

Lol, I was just wondering if the electromagnets around the tube were likely to cause any sort of interference in the coils.

User avatar
Shadow
Verified
Admin / PSXDEV
Admin / PSXDEV
Posts: 2670
Joined: Dec 31, 2012
PlayStation Model: H2000/5502
Discord: Shadow^PSXDEV

Post by Shadow » May 2nd, 2015, 4:16 pm

sicklebrick wrote:Lol, I was just wondering if the electromagnets around the tube were likely to cause any sort of interference in the coils.
You mean such as the degaussing coil? No, the transistor wont flow backwards on the collector. I don't see how it can fry your console. The whole system is an input to the TV and not an output or bidirectional input.

The safest approach if you're concerned though would be to build a little circuit consisting of little mini transformers (isolation), diodes (reverse current protection), and then your transistors/potentiometers (part of the boost circuit).
Development Console: SCPH-5502 with 8MB RAM, MM3 Modchip, PAL 60 Colour Modification (for NTSC), PSIO Switch Board, DB-9 breakout headers for both RGB and Serial output and an Xplorer with CAETLA 0.34.

PlayStation Development PC: Windows 98 SE, Pentium 3 at 400MHz, 128MB SDRAM, DTL-H2000, DTL-H2010, DTL-H201A, DTL-S2020 (with 4GB SCSI-2 HDD), 21" Sony G420, CD-R burner, 3.25" and 5.25" Floppy Diskette Drives, ZIP 100 Diskette Drive and an IBM Model M keyboard.

User avatar
sickle
Verified
C Programming Expert
C Programming Expert
Posts: 257
Joined: Jul 17, 2013
I am a: Chocolate-fueled pug fetish robot.
Location: Scotland

Post by sickle » May 3rd, 2015, 3:01 am

I was thinking more like a little interference from the steering coils (or whatever you call the ones which deflect the beam)- not so much the degausser since that'd be about turning itself off by the time you had everything switched on?

User avatar
Shadow
Verified
Admin / PSXDEV
Admin / PSXDEV
Posts: 2670
Joined: Dec 31, 2012
PlayStation Model: H2000/5502
Discord: Shadow^PSXDEV

Post by Shadow » May 3rd, 2015, 3:46 am

sicklebrick wrote:I was thinking more like a little interference from the steering coils (or whatever you call the ones which deflect the beam)- not so much the degausser since that'd be about turning itself off by the time you had everything switched on?
The 'steering coils' would be the yoke I assume.
Development Console: SCPH-5502 with 8MB RAM, MM3 Modchip, PAL 60 Colour Modification (for NTSC), PSIO Switch Board, DB-9 breakout headers for both RGB and Serial output and an Xplorer with CAETLA 0.34.

PlayStation Development PC: Windows 98 SE, Pentium 3 at 400MHz, 128MB SDRAM, DTL-H2000, DTL-H2010, DTL-H201A, DTL-S2020 (with 4GB SCSI-2 HDD), 21" Sony G420, CD-R burner, 3.25" and 5.25" Floppy Diskette Drives, ZIP 100 Diskette Drive and an IBM Model M keyboard.

Post Reply

Who is online

Users browsing this forum: No registered users and 5 guests