Component Video mod for curved screen Sony Trinitrons

[Bard_the_Bowman]

does polarity of the resistor matter.

No.

would daisy chaining the termination resistors cause this?

By this I assume you mean the fact that your termination resistors all ground to the same wire? No. That shouldn't be a problem.


Also, I doubt it has anything to do with the issue, but what is the third wire on your blanking switch?

[Ramzelle]

4th pic, the wires are not soldered in on this pic I forgot to take the image but are soldered to the far leg of each cap as to filter through before going to the jungle chip. all but 1 solder joint was done on the top side of the board. Pulled power from the bottom side of the micon pin 46 because of an RF Shield. C375 was already empty and unbridged on my model so I couldnt remove that to improve brightness. If you want I can get a pic of the bottom and top side if itll help

the only thing I can think of is to cut the Pb Resistor in half to increase brightness but that might only fix my blue issue. I could try to sync to the front Svideo port but not sure if that change a damn thing.

Third wire is ground, I assumed i should ground it just in case of internal short

[Bard_the_Bowman]

4th pic, the wires are not soldered in on this pic I forgot to take the image but are soldered to the far leg of each cap as to filter through before going to the jungle chip. all but 1 solder joint was done on the top side of the board. Pulled power from the bottom side of the micon pin 46 because of an RF Shield. C375 was already empty and unbridged on my model so I couldnt remove that to improve brightness. If you want I can get a pic of the bottom and top side if itll help

the only thing I can think of is to cut the Pb Resistor in half to increase brightness but that might only fix my blue issue. I could try to sync to the front Svideo port but not sure if that change a damn thing.

Third wire is ground, I assumed i should ground it just in case of internal short

Have you removed R358/R359/R360 (if present)? [DISREGARD, STUPID QUESTION I just looked at pic no. 4 and they clearly aren't there]

Also did you add the caps in pic no. 4 or were those already present? The reason I ask is that I just want to double check they actually are 0.01 uf and not some wildly different capacitance. I looked up the service manual for your unit and I can't find those caps, so I'm not sure what the default value would be.

And to expand on my last question, if C360/361/362 were NOT populated from the factory (which I'm guessing MIGHT be the case since I can find no reference to them in your service manual), were there jumpers at JR351 and JR352?

I'm just thinking that we've got to be missing something in that area.

[Ramzelle]

Also did you add the caps in pic no. 4 or were those already present? The reason I ask is that I just want to double check they actually are 0.01 uf and not some wildly different capacitance. I looked up the service manual for your unit and I can't find those caps, so I'm not sure what the default value would be.

Added the caps myself

And to expand on my last question, if C360/361/362 were NOT populated from the factory (which I'm guessing MIGHT be the case since I can find no reference to them in your service manual), were there jumpers at JR351 and JR352?

no jumpers were removed just the resistors

Image shows all the work that I did
https://imgur.com/a/ktkdJxH

[Bard_the_Bowman]

Also did you add the caps in pic no. 4 or were those already present? The reason I ask is that I just want to double check they actually are 0.01 uf and not some wildly different capacitance. I looked up the service manual for your unit and I can't find those caps, so I'm not sure what the default value would be.

Added the caps myself

And to expand on my last question, if C360/361/362 were NOT populated from the factory (which I'm guessing MIGHT be the case since I can find no reference to them in your service manual), were there jumpers at JR351 and JR352?

no jumpers were removed just the resistors

Image shows all the work that I did
https://imgur.com/a/ktkdJxH

I doubt it's the jumpers cause if it had them, and you hadn't removed them, the signals would be shorted and you'd have a profoundly messed up image I would think. And I also don't see them listed in your service manual. But it MIGHT be worth double checking if all else fails. They're little surface-mount chips on the underside of the board.

Looking at your picture, it looks to me like you have the luma wire connected to the chroma pin of the S-Video. Is that how you actually wired it IRL or is that a mistake in the diagram?

[matt]

So I did so more trouble shooting on my AA-2D tonight. I found that when I switched back to composite my colors and brightness came back and the Tube wasn't as worn as I thought it was. Re-Tuning the Screen for composite I found that the image was fairly crisp and not blurry and I believe a lot of my issues have been from trying to compensate for the loss of color and brightness in my Component mod. It sound like my version of the AA-2D is slightly different than Bowmans since I did not have a cap in the same spot as him but overall they should be fairly similar.

I am trying to figure what I should look for to try and trouble shoot all this brightness and color loss. would too many ohms on a resistor, or a diode or some cap be pulling voltage and causing my issues. I am obviously not formally trained in small electronics but if I have something to go on I could try to investigate further and maybe cut an unimportant trace or something along those lines.

It sounds like you may have some components left in circuit. Did you disconnect everything from the PIP circuit?

[Ramzelle]

It sounds like you may have some components left in circuit. Did you disconnect everything from the PIP circuit?

I believe so, I removed the resistors and the pip circuit board has been removed. according to the research I did that was everything that needed to be removed ill pull the board and take some photos. maybe I missed something that is specific to this model. I tried to look at the service manual and previously posted a picture of the svideo circuit but personally couldn't see if there was anything that I forgot in circuit.

Do I need to remove R1352 for the blanking line. that's all I can see that might be interfering but that's just blanking.

I don't know maybe this set is a dud.

Spoiler

[matt]

Yes, I think that'd work. Something must be off but I couldn't say what it is.

[Ramzelle]

Checked last night, nothing was left in circuit, the JR352 and JR351 were both already empty, I checked the removed resister sockets with a multimeter and there was no continuity.
There was one mistake and which was grounding the Y & Pb signal was not soldered but after soldering and checking for continuity the screen honestly looked the same.

Ill try removing the blanking resistor tonight and going straight in.

I reduced the Resisters to 47ohms on the Pr and Pb signals and both got darker, is this the intended behavior? If so I think I am going to try 150ohms to boost the signal tonight. maybe 220ohms

[Ramzelle]

Aight boys I am happy to say that I am officially done with this Mod.

I ended up trying out a whole lot of different termination resistor setups probably tried 6 different layouts funnily enough ended up going with a pretty basic one that is really close to the suggested Layout.
Y - 75ohms - (Default in circuit)
Pb - 120 ohms
Pr - 75 ohms

I had originally added an extra 47 ohms to Y, set Pb to 150 ohms and PB to 120ohms. Not ganna lie it looked pretty cool but super over saturated.

After a while swapping resistors I ended up with the above setup. 120 ohms on the Pb line was enough to boost the blue signal into a range that I could calibrate the colors in. The Tube is still a bit worn and a little fuzzy in its old age but its as bright and clear as my WEGAs and not washed out anymore.
Not sure what the long term negative effects of essentially overvolting (at least in my eyes) the signal is but for now I'm going to enjoy the set. Maybe its a reliable solution if others find there colors being nuked by this mod but please do it at your own risk who knows what issues it may cause in the future.

Thanks everyone who pitched in to help me figure this shit out!

https://imgur.com/a/WTy2Lzb

[MattAnderson]

Hello Everyone,
So I'm hoping to get some help, I performed this mod on a Sony KV-27V65
It has the PIP INTACT and has the capacitors in 370-372 installed from the factory.
I added in YUV-SW, Y, EB-Y, ER-Y all into the PIP header (after removing the PIP board).
I took 5v from a jumper and wired it with a switch to the YUV-SW.
On the same switch I have S-video ground and the S-Video Detect pin
And I have a Luma from S-Video going to PIP head and my RCA input
Pb and Pr are both terminated with 75 ohm resistors.
Everything looks pretty amazing HOWEVER, when I watch a video I get these horizontal bars/banding.
They're sometimes different colours, the one in the pic is green, sometimes they're just darker bands
I've tried shielded cables and nothing seems to help.
I've also tried an entirely different video source with different cables and the problem persists.
Edit: I've also plugged in S-Video and had no issues at all, but when I switch on the blanking switch I get a black and white image (as is expected) and I get the same bars that would be over component. So I'm assuming it's my blanking. I have a 1k ohm resistor on a 5v jumper on the motherboard. So I'm going to try a different 5v source tomorrow.
Any other suggestions would be appreciated.
https://imgur.com/a/z43f4pG

[warcelo]

Hello, folks

I need help finding out what is wrong with my setup.
Description of the system:
Sony KV (BA-4) 20M40 with Component MOD
HDMI to component converter -> https://www.amazon.com/dp/B082HT11MH?re ... i+to+compo
PC with Intel UHD 620

The goal is to run Retroarch and get it on the CRT at 240p.

I used CRU to create a resolution of 2560x240 @60HZ and I can select it for the second display.

Problem is, it does not show a stable image, it is warped to the right with some artifacts. - The colors seem to be correct, I can see my desktop and it looks right but it is far from stable.

Pictures:
[url=https://flic.kr/p/2pUL9mT]
[url=https://flic.kr/p/2pUNhmF]
[url=https://flic.kr/p/2pUMScx]

[vol.2]

Hello, folks

I need help finding out what is wrong with my setup.
Description of the system:
Sony KV (BA-4) 20M40 with Component MOD
HDMI to component converter -> https://www.amazon.com/dp/B082HT11MH?re ... i+to+compo
PC with Intel UHD 620

The goal is to run Retroarch and get it on the CRT at 240p.

I used CRU to create a resolution of 2560x240 @60HZ and I can select it for the second display.

Problem is, it does not show a stable image, it is warped to the right with some artifacts. - The colors seem to be correct, I can see my desktop and it looks right but it is far from stable.

Ok. First things first. Does it show a stable image using a tested good component video source?

An example of this is a DVD player that has component output, or a game console with a component cable.

If the TV does show correct image with a known-good (tested) component source, then your problem is not the mod, and something to do with the converter or PC output or something else, and does not belong in this thread.

[warcelo]

Unfortunately, I can't test that as I don't have any source of component.... Maybe I need one to try first.

[vol.2]

Unfortunately, I can't test that as I don't have any source of component.... Maybe I need one to try first.

You can't know what the problem is until you can say for 100% certainty that you are starting with a known-good (tested) video source.

[MattAnderson]

So I'm posting again because my last question got buried.
I have horizontal lines on my screen after doing this mod.
Below are the pics. I've tried everything I can think. I've tried using an S-Video cable to ground S video, I've tried syncing from Composite (made it MUCH worse actually).
I removed the 0.01uf Capacitors to the PIP header and soldered my own wire and capacitor directly to the jungle chip.
The lines persist from any video source, DVD player, PS3 or anything else.
Oddly enough they're in the exact same place every time on the same scene. So if I go back to the video at the same time stamp the lines will be exactly the same every time, which makes me think the Luma isn't being decoded properly.
I'm extremely discouraged, as I've went through everything over and over again, tried multiple 5v sources for blanking, tried 3 different wire types included shielded.
It ONLY Happens on the component input, S-video and composite are flawless.
The only time I can replicate it is when I have an S-video input and I flip the switch and it shows the black and white input of the S-Video's Luma input (because I'm using S-Video Luma as Sync).
Anyone encountered this on their component mod?
https://imgur.com/YMkjpDh.jpg
https://imgur.com/pjKcGbz.jpg
https://imgur.com/DMtKIAv.jpg

[KPackratt2k]

Have you tried desoldering the YUV pins of the jungle chip and carefully lifting them from the board to isolate possible PCB trace routing interference? It's possible that the YUV lines on the PCB are near traces that could cause interference. I had minor noise on previous Sony BA-4 chassis TVs that I've modded, but it was never this severe, the two that I personally have (a KV-27V42 and a KV-20S41) don't have this issue, so it's possible that the traces on the older chassis revision were noisy and later revisions fixed it.

[MattAnderson]

Have you tried desoldering the YUV pins of the jungle chip and carefully lifting them from the board to isolate possible PCB trace routing interference? It's possible that the YUV lines on the PCB are near traces that could cause interference. I had minor noise on previous Sony BA-4 chassis TVs that I've modded, but it was never this severe, the two that I personally have (a KV-27V42 and a KV-20S41) don't have this issue, so it's possible that the traces on the older chassis revision were noisy and later revisions fixed it.

First thanks for the reply.
I somehow doubt this would be the solution. Since I've soldered the YUV lines directly to the Jungle Chip the lines are exactly the same, no difference good or bad. If it was interference I would suspect a change in the variability of it. But it's EXACTLY the same.

[KPackratt2k]

It's worth a shot to try it, I know PCB trace interference is an issue on certain Nintendo console revisions (notably the Composite video trace on the NES top loader and the Chroma subcarrier trace on late revision 3-Chip SNES consoles) to the point where people resorted to lifting pins and rewiring them to mitigate interference.

It also helps to check the routing of your wires to ensure they're not touching any high voltage components that could cause interference such as the flyback, deflection ICs (which are typically on heatsinks), and the power supply. I once saw jailbars on a KV-20S90 that I've modded for Component and it turned out to be a wire touching the vertical deflection IC.

As a last resort, you can try replacing the jungle chip (IC301) with a new one. I once had to replace the jungle chip on a KV-24FV12 that I've modded because the colors weren't showing up on the Component mod. Compatible chips for the BA-4 chassis include:

- Sony CXA2061S
- Sony CXA2133S
- Sony CXA2133BS

Although the jungle chips used in the BA-5 chassis FD Trinitron sets (CXA2131AS/CS) uses the same pinout as the chip found on the BA-4(D) chassis, it's not fully compatible, as I had pincushion issues when I tried to use one of those chips on a KV-27S42 that I was diagnosing, something to keep in mind for anyone needing to replace the jungle chip on one of these TVs.

[MattAnderson]

The last troubleshooting I did I completely routed the wires underneath the board soldered directly to the jungle chip. So the chance of interference is lessened. However this grounding pin is on the end of the board next to where the wires meet the blanking switch.
https://imgur.com/PDbqdQb.jpg
This ground is connected to the PSU with a single black wire. Would that create interference?

The only other thing I can think of is the YUV SW pin has a diode and a 220ohm resistor in the circuit between the JC and the PIP header. Would either of those cause interference?
I know RGB mods use diodes in the blanking wire, but I've never seen it in any of the write ups for component mods, but I'm thinking the 220uf resistor may be impacting the blanking.
I've included this snippet from the service manual of said resistor and diode.
EDIT: and there seems to be a grounding resistor (R340) at 1k Ohm. I removed this and still no improvement.
https://imgur.com/DalmQX1.jpg

These are the only two things I can see potentially causing the issue before I start lifting pins and possibly replacing the entire jungle IC.
But if these two aren't the issue I will be lifting the pins next.
EDIT 2: I have lifted all 4 pins from the PCB and soldered all my lines directly to the Jungle chip and it's no better.

Again thanks for your time.

[matt]

So I'm posting again because my last question got buried.
I have horizontal lines on my screen after doing this mod.
Below are the pics. I've tried everything I can think. I've tried using an S-Video cable to ground S video, I've tried syncing from Composite (made it MUCH worse actually).
I removed the 0.01uf Capacitors to the PIP header and soldered my own wire and capacitor directly to the jungle chip.
The lines persist from any video source, DVD player, PS3 or anything else.
Oddly enough they're in the exact same place every time on the same scene. So if I go back to the video at the same time stamp the lines will be exactly the same every time, which makes me think the Luma isn't being decoded properly.
I'm extremely discouraged, as I've went through everything over and over again, tried multiple 5v sources for blanking, tried 3 different wire types included shielded.
It ONLY Happens on the component input, S-video and composite are flawless.
The only time I can replicate it is when I have an S-video input and I flip the switch and it shows the black and white input of the S-Video's Luma input (because I'm using S-Video Luma as Sync).
Anyone encountered this on their component mod?

This looks like your picture is shifted too far to the left. That's an issue with both component and RGB mods. What happens is that because you have active data in what's supposed to be the blanking interval, the TV clamps the video signal at the wrong level and draws the whole line at the wrong brightness. Note how the banding correlates with dark and light areas at the left hand edge of the screen. That's also why it was worse with composite - the picture was even further to the left.

The solution is to center the image to the right with the service menu and the horizontal position switch(es).

[MattAnderson]

Wow thank you so much.
I moved the HPOS and HSIZE which helped a bit, but it was the H VCO OSCILLATION FREQ that helped the most. I increased it from 7 to 15 and it almost eliminated it completely. I then upped AFC LOOP GAIN SWITCHING from 0 to 2 and that eliminated virtually all of it. I readjusted all the geometry and it is virtually perfect. Under very specific circumstances I can still see a line show up but it's only noticeable because I'm looking fro it.
I have a gallery of the TV after all the adjustments. I also included a photo of the added jacks and blanking switch. I was particularly proud of the work I did cutting the shell.
Thanks everyone for your help.
Just out of curiosity Is there another way to sync that doesn't shift it horizontally? This is my second CRT mod (first being an S-video Mod). I come from a background of vintage computer repair so I'm less familiar with consumer CRT TV's.
Gallery: https://imgur.com/a/rppJ83K