I think making the shell mounted one work is going to be a lot easier. That mod had an awful lot of thought and time put into it, and they also just lucked the hell out with the placement. You might also get that lucky with whatever random TV you're working on, but I don't think it's wise to count on it.BazookaBen wrote: ↑Tue Aug 06, 2024 2:05 am Pics in the link but you solder an adapter board to some available vias for ground and in some cases, scrape away solder mask and solder to the ground plane
Sony Trinitron S-Video/Composite Mod (RF-Only TV) SOLVED
Re: Sony Trinitron S-Video/Composite Mod (RF-Only TV) SOLVED
-
BazookaBen
- Posts: 2123
- Joined: Thu Apr 17, 2008 8:09 pm
- Location: North Carolina
Re: Sony Trinitron S-Video/Composite Mod (RF-Only TV) SOLVED
I should have taken pics of the one I did for a kv-13fm14 recently.
I found an old s-video connector contained in a A/V adapter for an old GPU.
It was encased in epoxy, and I just drilled a hole through an empty section of the epoxy (to avoid wires) and ran a 12AWG solid copper wire through. Then I bent in a U-shape to where it went into a couple large blank vias on the motherboard, then soldered it in. SUPER strong that way, I wish there were a way to find them in bulk in a form like that.
I found an old s-video connector contained in a A/V adapter for an old GPU.
It was encased in epoxy, and I just drilled a hole through an empty section of the epoxy (to avoid wires) and ran a 12AWG solid copper wire through. Then I bent in a U-shape to where it went into a couple large blank vias on the motherboard, then soldered it in. SUPER strong that way, I wish there were a way to find them in bulk in a form like that.
-
- Posts: 7
- Joined: Sun Aug 04, 2024 12:02 am
Re: Sony Trinitron S-Video/Composite Mod (RF-Only TV) SOLVED
After reading this thread herehttps://electronics.stackexchange.com/q ... chassis-tv I am far less confident about adding inputs to this hot chassis tv. I suppose the only way would be to use the tv exclusively with an external isolation transformer, far enough from the tube to not cause distortion. If anyone has any ideas, definitely let me know though. It’s a shame, I really like this tv.
Re: Sony Trinitron S-Video/Composite Mod (RF-Only TV) SOLVED
I haven't ever done it before, but I've considered running one of my hot chassis sets with an internally mounted isolation transformer. One that can handle a TV is somewhere around $45. Not sure how that would be an issue, but maybe you're finding something that I don't know about.MaliciousBingus wrote: ↑Thu Aug 08, 2024 12:18 am After reading this thread herehttps://electronics.stackexchange.com/q ... chassis-tv I am far less confident about adding inputs to this hot chassis tv. I suppose the only way would be to use the tv exclusively with an external isolation transformer, far enough from the tube to not cause distortion. If anyone has any ideas, definitely let me know though. It’s a shame, I really like this tv.
If you are somehow concerned about magnetic interference from the transformer, you could shield it.
I am not aware of any specific issues with using an isolation transformer for a line device, I know there are plenty of electronics that do this, but I guess I'd still ask around if you have doubts. A different forum that is more specifically geared towards electronic repair is a better place to confirm your design's safety.
-
- Posts: 7
- Joined: Sun Aug 04, 2024 12:02 am
Re: Sony Trinitron S-Video/Composite Mod (RF-Only TV) SOLVED
Yeah, installing the transformer in my set wasn’t hard and it powered up fine, but the transformer has to be pretty close to the tube in my case to fit it, and I saw notable rippling in the picture that would lessen as I increased the distance between the two. I need to look into the mechanics of shielding I suppose.
Re: Sony Trinitron S-Video/Composite Mod (RF-Only TV) SOLVED
There are 3 levels of shielding you can employ.MaliciousBingus wrote: ↑Thu Aug 08, 2024 2:27 pm Yeah, installing the transformer in my set wasn’t hard and it powered up fine, but the transformer has to be pretty close to the tube in my case to fit it, and I saw notable rippling in the picture that would lessen as I increased the distance between the two. I need to look into the mechanics of shielding I suppose.
1) steel can from a can of beans or something. in most cases this is more than enough to block anything, but it can be tricky to make it fit, and you'll have to carefully insulate it to prevent shorting which can be tricky with a can
2) 1mm thick copper shielding. Following Lenz's law this should mitigate stronger interference and has the advantage of being softer and easier to bend. Being flatter material it will be easier to insulate
3) the strongest material you would ever need is Mu Metal. It's more expensive, and may be difficult to acquire outside of older sets that use it
-
- Posts: 7
- Joined: Sun Aug 04, 2024 12:02 am
Re: Sony Trinitron S-Video/Composite Mod (RF-Only TV) SOLVED
Thank you for your assistance, I’m very appreciative. My transformer is the Triad N-67A, which has a shield lead that the data sheet says is part of the secondary/output side, but doesn’t really offer much else. In my installation I just heat shrunk and tied it off since my TV didn’t have an earth connection. Do you know if this has anything to do with magnetically shielding my transformer?vol.2 wrote: ↑Thu Aug 08, 2024 2:48 pmThere are 3 levels of shielding you can employ.MaliciousBingus wrote: ↑Thu Aug 08, 2024 2:27 pm Yeah, installing the transformer in my set wasn’t hard and it powered up fine, but the transformer has to be pretty close to the tube in my case to fit it, and I saw notable rippling in the picture that would lessen as I increased the distance between the two. I need to look into the mechanics of shielding I suppose.
1) steel can from a can of beans or something. in most cases this is more than enough to block anything, but it can be tricky to make it fit, and you'll have to carefully insulate it to prevent shorting which can be tricky with a can
2) 1mm thick copper shielding. Following Lenz's law this should mitigate stronger interference and has the advantage of being softer and easier to bend. Being flatter material it will be easier to insulate
3) the strongest material you would ever need is Mu Metal. It's more expensive, and may be difficult to acquire outside of older sets that use it
Re: Sony Trinitron S-Video/Composite Mod (RF-Only TV) SOLVED
That's for if you want to ground the shielding. Forcing the shielding to ground potential increases it's ability to cancel out magnetic interference. It's the same concept as using ground planes in a PCB to isolate impedance sensitive traces. If I was doing it, I wouldn't mess around with that. I think just plain metal shielding would be sufficientMaliciousBingus wrote: ↑Thu Aug 08, 2024 2:59 pm Thank you for your assistance, I’m very appreciative. My transformer is the Triad N-67A, which has a shield lead that the data sheet says is part of the secondary/output side, but doesn’t really offer much else. In my installation I just heat shrunk and tied it off since my TV didn’t have an earth connection. Do you know if this has anything to do with magnetically shielding my transformer?
In the case of the transformer you are using, that terminal is intended to connect to the third prong (earth) of a standard North American 3 pronged power plug. I'm not entirely sure what the consequences of doing that would be in this case because, in North America, the neutral line is connected to the earth line at the breaker box, so it effectively un-isolates the neutral line. You need to ask safety questions on a more electronics oriented forum as I'm definitely not qualified to give safety advice.
One thing I can tell you for sure is that you cannot hook an oscilloscope ground up to a device that has not been plugged into a fully isolated transformer because (since the earth is connected to the neutral in the box) it will create a ground loop back to the neutral on the power grid outside your house, effectively pulling ground up to 120V potential. It should just trip the breaker, but it could also blow some crap up in the process.
How that works with using video inputs I can't tell you.
I'm also not going to go on the record saying that just using a regular IST inside the set is totally safe. That's on you as well. I will probably try to do more HW about it before I try with my set. I have successfully got it RGB modded, but I pulled it all out until I can satisfy myself it's okay to just use the IST (and also feel like spending money on it). But I have been collecting sets for 30 years and I got others to play with.
Re: Sony Trinitron S-Video/Composite Mod (RF-Only TV) SOLVED
So The PS2 has colour but the C64 doesnt
https://www.dropbox.com/scl/fi/84klanb3 ... 7bng9&dl=0
The C64 Is made in Hong Kong
https://www.dropbox.com/scl/fi/naps9v3q ... 3k5o2&dl=0
Re: Sony Trinitron S-Video/Composite Mod (RF-Only TV) SOLVED
Ok I think my problem is the replacement video chip I recently installed must be NTSC.
Mt breadbin is fine. Thanks for all your help KPackRatt2k
Mt breadbin is fine. Thanks for all your help KPackRatt2k
Re: Sony Trinitron S-Video/Composite Mod (RF-Only TV) SOLVED
Revisiting this mod after about 2 years.nmur wrote: ↑Sun May 01, 2022 2:40 pm I don't know how I didn't notice this before - I mustn't have had the TV turned on for very long during the modding/testing stages, but I'm finding that the s-video video signal display fades away to nothing after a couple of minutes. I can still hear audio, as well as the 15kHz whine, and the neck still glows, but no video signal displays, other than the OSD. I've tried different input sources as well. What's also weird is that if I turn off the TV and the source and turn them on again, the video still isn't visible. However if I leave it all turned off for a while, and come back later to try, it will work again for a short while before fading to black again. I didn't observe any obvious overheating of anything inside the TV.
Has anyone witnessed this behaviour with their mod before?
I now thankfully have video working well, which is great. However I'm actually having trouble injecting my audio - I can't seem to get it working through many different attempts of connecting up to the "AF IN" pin of the IF block. I've also tried replicating the audio circuit in jeffez's reference mod, with two 4.7uF caps, but cannot hear anything coming out of the speaker.
If I connect audio directly to the "AF OUT" pin, I hear audio loud and clear, but without the ability to adjust the volume of course.
This is what my circuit looks like:

Red cross is a cut trace, green dot is where I was planning on injecting audio, and red dotted line is just showing that it's essentially connected like this (chassis has unpopulated audio switching circuitry which is replaced by a jumper in my set). My actual IF block is a IFB-389SS, with a CX20015A inside.
For reference this is the simple audio circuit in jeffez's mod that I also tried recreating:

Does the IF block need to be in a certain state to process audio or something? Like a pin that needs a certain voltage? I can't seem to find much data on the IFB-389SS block or the CX20015A inside.
Or could I have fried something while playing around?
-
- Posts: 259
- Joined: Sun Apr 04, 2021 11:02 pm
- Location: Seattle, WA, USA
Re: Sony Trinitron S-Video/Composite Mod (RF-Only TV) SOLVED
I was able to add S-Video to a Sony KV-2780R TV through the use of an optocoupler board I've designed based on a circuit found on a Daewoo TV utilizing readily available components. Because this set has a hot chassis with the AV inputs isolated with optocouplers, I figured it would be best to isolate the Chroma input using a readily available optocoupler to reduce the risks of safety and having to source obsolete parts.
I've published the PCB onto OSHPark so others can use it for S-Video mods for Sony TVs using the P-3(A/B) chassis:
https://oshpark.com/shared_projects/KZcv086o


Additional photos:
https://imgur.com/a/D35qz1Z
You can solder the wires for power and ground on the AV input board (or the U board) and run them to the Chroma isolation board via JST connectors. I forgot to take photos of the tap points, but the cold side shares the ground with the AV inputs whereas the hot side does not. The wires for Chroma go to the chassis (or the A board), you'll have to lift the Chroma coupling capacitor (C357) from the side that connects to pin 39 of the jungle chip (IC301).
To reduce the risk of interference, I strongly recommend using shielded wire for the ends of the Chroma connections that send it to the jungle chip. When I prototyped this mod, I was getting noise that sometimes affected color intensity on the S-Video input when I used unshielded wire. Using shielded wire instead eliminated the noise and kept the color intensity consistent with the stock Composite video input.
As for a place to mount the Chroma isolator board, there's an empty space in the cabinet where the chassis boards are mounted, the empty space is located near the input board (the U board), providing a convenient spot to mount the isolator board using 3M VHB double sided tape to hold it in place. It is recommended to place two layers of kapton tape on the board before putting the double sided tape on it.
Connector labels on the Chroma isolator board and their tapping points on the TV:
Cold power and ground (the 2-pin JST on the top-left of the isolator board):
VCC (Cold Side) <- Cold 9.2V (The positive leg of C1405 connected to R1428 on the U board.)
GND (Cold Side) <- AV input ground or IC1401 pin 10
S-Video Chroma input (the 2-pin JST on the bottom-left of the isolator board):
C_IN <- S-Video Chroma (Pin 4)
GND -> Ground point for shielded wire, can also go to the ground pins of the S-Video jack (the shield and pins 1-2.)
Hot power and ground (the 3-pin JST on the top-right of the isolator board):
VCC (Hot side) <- Hot 9.3V (connected to pin 1 of IC1401 via R1403 or U-2 pin 5, should be labeled accordingly on the U board.)
Middle pin is not connected to anything.
GND (Hot side) <- IC1401 pin 5 or any ground outside of the cold area of the U board on the TV.
Chroma I/O from the TV chassis (the 2-pin JST near the 4-pin JST on the bottom-right of the isolator board):
YCJ_IN <- IC301 pin 39, can inject to the pad where C357 was lifted, connecting to this pin on the chip.
TV_C <- C357 coming from the emitter of the TV Chroma buffer (Q357).
Chroma I/O to an SPDT switch (the 4-pin JST on the bottom-right of the isolator board):
C_OUT -> Pin 1 of the switch
YCJ_IN -> Pin 2 of the switch
TV_C -> Pin 3 of the switch
GND -> A grounding point for shielded wire, provided for convenience.
Luma (S-Video pin 3) simply goes to the Composite video input.
You'll have to be strategic for finding a place to mount the S-Video jack and the Chroma switch. I've decided to relocate the RF tuner I/O panel somewhere else and run a short Coax cable from the VHF/UHF input to a coupler mounted in the VHF/UHF input's original hole. The S-Video jack is mounted on the "Aux" input on the left, though the plastic on the back side of the jack panel plate had to be shaved to allow for the panel mount jack to fit without the nut coming off. I mounted a spare panel mount SPDT slide switch in the "To Converter" hole in the middle of the jack panel to serve as the switch to toggle between S-Video and Composite video/RF.
When/if I get ahold of more Sony TVs from the same era, I'll be sure to test the mod board on those models and document any specific instructions that apply to them. I've also been working on replicating the S-Video switching circuitry from the KV-27TS30 to implement on other models with the same chassis for either using a SPST switch or a completely switchless mod using an S-Video jack with its shield cut in half to connect one side to ground and the other side to a high/low logic signal.
I've published the PCB onto OSHPark so others can use it for S-Video mods for Sony TVs using the P-3(A/B) chassis:
https://oshpark.com/shared_projects/KZcv086o


Additional photos:
https://imgur.com/a/D35qz1Z
You can solder the wires for power and ground on the AV input board (or the U board) and run them to the Chroma isolation board via JST connectors. I forgot to take photos of the tap points, but the cold side shares the ground with the AV inputs whereas the hot side does not. The wires for Chroma go to the chassis (or the A board), you'll have to lift the Chroma coupling capacitor (C357) from the side that connects to pin 39 of the jungle chip (IC301).
To reduce the risk of interference, I strongly recommend using shielded wire for the ends of the Chroma connections that send it to the jungle chip. When I prototyped this mod, I was getting noise that sometimes affected color intensity on the S-Video input when I used unshielded wire. Using shielded wire instead eliminated the noise and kept the color intensity consistent with the stock Composite video input.
As for a place to mount the Chroma isolator board, there's an empty space in the cabinet where the chassis boards are mounted, the empty space is located near the input board (the U board), providing a convenient spot to mount the isolator board using 3M VHB double sided tape to hold it in place. It is recommended to place two layers of kapton tape on the board before putting the double sided tape on it.
Connector labels on the Chroma isolator board and their tapping points on the TV:
Cold power and ground (the 2-pin JST on the top-left of the isolator board):
VCC (Cold Side) <- Cold 9.2V (The positive leg of C1405 connected to R1428 on the U board.)
GND (Cold Side) <- AV input ground or IC1401 pin 10
S-Video Chroma input (the 2-pin JST on the bottom-left of the isolator board):
C_IN <- S-Video Chroma (Pin 4)
GND -> Ground point for shielded wire, can also go to the ground pins of the S-Video jack (the shield and pins 1-2.)
Hot power and ground (the 3-pin JST on the top-right of the isolator board):
VCC (Hot side) <- Hot 9.3V (connected to pin 1 of IC1401 via R1403 or U-2 pin 5, should be labeled accordingly on the U board.)
Middle pin is not connected to anything.
GND (Hot side) <- IC1401 pin 5 or any ground outside of the cold area of the U board on the TV.
Chroma I/O from the TV chassis (the 2-pin JST near the 4-pin JST on the bottom-right of the isolator board):
YCJ_IN <- IC301 pin 39, can inject to the pad where C357 was lifted, connecting to this pin on the chip.
TV_C <- C357 coming from the emitter of the TV Chroma buffer (Q357).
Chroma I/O to an SPDT switch (the 4-pin JST on the bottom-right of the isolator board):
C_OUT -> Pin 1 of the switch
YCJ_IN -> Pin 2 of the switch
TV_C -> Pin 3 of the switch
GND -> A grounding point for shielded wire, provided for convenience.
Luma (S-Video pin 3) simply goes to the Composite video input.
You'll have to be strategic for finding a place to mount the S-Video jack and the Chroma switch. I've decided to relocate the RF tuner I/O panel somewhere else and run a short Coax cable from the VHF/UHF input to a coupler mounted in the VHF/UHF input's original hole. The S-Video jack is mounted on the "Aux" input on the left, though the plastic on the back side of the jack panel plate had to be shaved to allow for the panel mount jack to fit without the nut coming off. I mounted a spare panel mount SPDT slide switch in the "To Converter" hole in the middle of the jack panel to serve as the switch to toggle between S-Video and Composite video/RF.
When/if I get ahold of more Sony TVs from the same era, I'll be sure to test the mod board on those models and document any specific instructions that apply to them. I've also been working on replicating the S-Video switching circuitry from the KV-27TS30 to implement on other models with the same chassis for either using a SPST switch or a completely switchless mod using an S-Video jack with its shield cut in half to connect one side to ground and the other side to a high/low logic signal.
Re: Sony Trinitron S-Video/Composite Mod (RF-Only TV) SOLVED
That's fantastic!KPackratt2k wrote: ↑Sat Nov 16, 2024 12:48 am I was able to add S-Video to a Sony KV-2780R TV through the use of an optocoupler board I've designed based on a circuit found on a Daewoo TV utilizing readily available components. Because this set has a hot chassis with the AV inputs isolated with optocouplers, I figured it would be best to isolate the Chroma input using a readily available optocoupler to reduce the risks of safety and having to source obsolete parts.
I've published the PCB onto OSHPark so others can use it for S-Video mods for Sony TVs using the P-3(A/B) chassis:
https://oshpark.com/shared_projects/KZcv086oSpoiler
Additional photos:
https://imgur.com/a/D35qz1Z
You can solder the wires for power and ground on the AV input board (or the U board) and run them to the Chroma isolation board via JST connectors. I forgot to take photos of the tap points, but the cold side shares the ground with the AV inputs whereas the hot side does not. The wires for Chroma go to the chassis (or the A board), you'll have to lift the Chroma coupling capacitor (C357) from the side that connects to pin 39 of the jungle chip (IC301).
To reduce the risk of interference, I strongly recommend using shielded wire for the ends of the Chroma connections that send it to the jungle chip. When I prototyped this mod, I was getting noise that sometimes affected color intensity on the S-Video input when I used unshielded wire. Using shielded wire instead eliminated the noise and kept the color intensity consistent with the stock Composite video input.
As for a place to mount the Chroma isolator board, there's an empty space in the cabinet where the chassis boards are mounted, the empty space is located near the input board (the U board), providing a convenient spot to mount the isolator board using 3M VHB double sided tape to hold it in place. It is recommended to place two layers of kapton tape on the board before putting the double sided tape on it.
Connector labels on the Chroma isolator board and their tapping points on the TV:
Cold power and ground (the 2-pin JST on the top-left of the isolator board):
VCC (Cold Side) <- Cold 9.2V (The positive leg of C1405 connected to R1428 on the U board.)
GND (Cold Side) <- AV input ground or IC1401 pin 10
S-Video Chroma input (the 2-pin JST on the bottom-left of the isolator board):
C_IN <- S-Video Chroma (Pin 4)
GND -> Ground point for shielded wire, can also go to the ground pins of the S-Video jack (the shield and pins 1-2.)
Hot power and ground (the 3-pin JST on the top-right of the isolator board):
VCC (Hot side) <- Hot 9.3V (connected to pin 1 of IC1401 via R1403 or U-2 pin 5, should be labeled accordingly on the U board.)
Middle pin is not connected to anything.
GND (Hot side) <- IC1401 pin 5 or any ground outside of the cold area of the U board on the TV.
Chroma I/O from the TV chassis (the 2-pin JST near the 4-pin JST on the bottom-right of the isolator board):
YCJ_IN <- IC301 pin 39, can inject to the pad where C357 was lifted, connecting to this pin on the chip.
TV_C <- C357 coming from the emitter of the TV Chroma buffer (Q357).
Chroma I/O to an SPDT switch (the 4-pin JST on the bottom-right of the isolator board):
C_OUT -> Pin 1 of the switch
YCJ_IN -> Pin 2 of the switch
TV_C -> Pin 3 of the switch
GND -> A grounding point for shielded wire, provided for convenience.
Luma (S-Video pin 3) simply goes to the Composite video input.
You'll have to be strategic for finding a place to mount the S-Video jack and the Chroma switch. I've decided to relocate the RF tuner I/O panel somewhere else and run a short Coax cable from the VHF/UHF input to a coupler mounted in the VHF/UHF input's original hole. The S-Video jack is mounted on the "Aux" input on the left, though the plastic on the back side of the jack panel plate had to be shaved to allow for the panel mount jack to fit without the nut coming off. I mounted a spare panel mount SPDT slide switch in the "To Converter" hole in the middle of the jack panel to serve as the switch to toggle between S-Video and Composite video/RF.
When/if I get ahold of more Sony TVs from the same era, I'll be sure to test the mod board on those models and document any specific instructions that apply to them. I've also been working on replicating the S-Video switching circuitry from the KV-27TS30 to implement on other models with the same chassis for either using a SPST switch or a completely switchless mod using an S-Video jack with its shield cut in half to connect one side to ground and the other side to a high/low logic signal.
I have a project I can use this on already. I was actually designing a optocoupler board in KiCad based on a Toshiba TV I have that already uses one for it's composite input, so I'll be curious to see the gerbers and compare it to what I already have.
Actually would be great if you wanted to check out what I've done so far. No pressure or anything.
-
- Posts: 259
- Joined: Sun Apr 04, 2021 11:02 pm
- Location: Seattle, WA, USA
Re: Sony Trinitron S-Video/Composite Mod (RF-Only TV) SOLVED
Sure, feel free to share your progress.vol.2 wrote: ↑Sat Nov 16, 2024 1:39 am That's fantastic!
I have a project I can use this on already. I was actually designing a optocoupler board in KiCad based on a Toshiba TV I have that already uses one for it's composite input, so I'll be curious to see the gerbers and compare it to what I already have.
Actually would be great if you wanted to check out what I've done so far. No pressure or anything.
Re: Sony Trinitron S-Video/Composite Mod (RF-Only TV) SOLVED
Awesome, sent you links to the KiCad project and the schematic.
-
BazookaBen
- Posts: 2123
- Joined: Thu Apr 17, 2008 8:09 pm
- Location: North Carolina
Re: Sony Trinitron S-Video/Composite Mod (RF-Only TV) SOLVED
Looking forward to hearing how this works out.
I have a RF-only JVC from 1991 that I suspect may be a hot chassis. So I may try to copy y'alls work to do something similar.
-
- Posts: 259
- Joined: Sun Apr 04, 2021 11:02 pm
- Location: Seattle, WA, USA
Re: Sony Trinitron S-Video/Composite Mod (RF-only TV)
When I added S-Video to my Sony KV-20S41, I found that with the S-Video detection pin on the Micom set to a low state, the Chroma Trap Filter (CTRP) and Sharpness High Pass Filter (SHPF) settings were forced to "1" (off) when the TV was in VIDEO 1 mode, but could be changed if the input was set to either TV/RF or VIDEO 2 Composite. I've also noticed the same behavior on a KV-20V80 that I've fixed for a friend (it had a crack on the chassis PCB). Perhaps if you've populated the components for S-Video detection and used either a switch or an S-Video jack with its shield divided in half to toggle the logic signal to low when in S-Video mode, it will forcibly turn off the filtering that impacts the S-Video image quality?matt wrote: ↑Thu May 27, 2021 7:12 am With these BA-4 chassis TVs, the models with S-Video all use a separate comb filter for composite and the signals are routed differently from the ones that are composite-only. This is why the luma trap filter needs to be disabled in the service menu for the mod to work properly. It's a pain if you want to switch between composite and S-Video on a regular basis, but if you just plan on using S-Vid (or combine it with an RGB or component mod), it works well.
-
BazookaBen
- Posts: 2123
- Joined: Thu Apr 17, 2008 8:09 pm
- Location: North Carolina
Re: Sony Trinitron S-Video/Composite Mod (RF-Only TV) SOLVED
I'm thinking modding a hot chassis CRT (chassis TX826TB) with optocouplers for s-video and probably audio too.
I was hoping to only need to do a bit of voltage division with resistors to get the right levels into the chip, but I'm looking at your mods above and you're using transistors and diodes and stuff, so I might be in over my head with this.
Anyway, here is the Y+C circuit for anybody that doesn't mind taking a look

Datasheet for the jungle: https://www.alldatasheet.com/datasheet- ... 03ASP.html
Another thing that's confusing is that the Chroma path is labled "Chroma in/Brightness", so I guess this means they somehow incorporated the brightness control on the same input? Has anybody seen anything like that before? I have no idea how that would work.
And then there's the issue of powering the optocouplers. I imagine something akin to this isolated DC-DC converter will do that job?
https://www.digikey.com/en/products/det ... 3C/6056049
That's just a random one I selected, but the type of thing I imagine I'd need
I was hoping to only need to do a bit of voltage division with resistors to get the right levels into the chip, but I'm looking at your mods above and you're using transistors and diodes and stuff, so I might be in over my head with this.
Anyway, here is the Y+C circuit for anybody that doesn't mind taking a look

Datasheet for the jungle: https://www.alldatasheet.com/datasheet- ... 03ASP.html
Another thing that's confusing is that the Chroma path is labled "Chroma in/Brightness", so I guess this means they somehow incorporated the brightness control on the same input? Has anybody seen anything like that before? I have no idea how that would work.
And then there's the issue of powering the optocouplers. I imagine something akin to this isolated DC-DC converter will do that job?
https://www.digikey.com/en/products/det ... 3C/6056049
That's just a random one I selected, but the type of thing I imagine I'd need
Re: Sony Trinitron S-Video/Composite Mod (RF-Only TV) SOLVED
Just off the top of my head, the sync information is also what supplies the information for the H/V pulses, so you would need to make sure the luma is hitting the whole circuit from the point that it exits the IF demodulation. If the brightness is being concatenated with the chroma, then you might also have to do the same with that.BazookaBen wrote: ↑Thu Jan 16, 2025 10:39 pm I'm thinking modding a hot chassis CRT (chassis TX826TB) with optocouplers for s-video and probably audio too.
I would flip back through the schematics and see if you can pinpoint the place that the RF section spits out luma and chroma. It should be after the final intermediate frequency (IF) stage. Then you probably aught to scope what's there at those points (WITH THE TV CONNECTED TO AN ISOLATION TRANSFORMER) and see what it's expecting.
Once you know what it wants, you can begin to design the output of the optocoupler circuit. The nice thing about the optocouplers is that it gives you a chance to control the output levels of the YC signals by adjusting the Vcc and you can probably do some DC restoration if you need an offset.
It's not going to be easy I don't think.
-
BazookaBen
- Posts: 2123
- Joined: Thu Apr 17, 2008 8:09 pm
- Location: North Carolina
Re: Sony Trinitron S-Video/Composite Mod (RF-Only TV) SOLVED
Appreciate it. If I'm reading correctly, the "vv01 delay line" in the diagram is what's separating luma/chroma, and there's not too much going after that except what appears to the brightness voltage being incorporated into the chroma signal.
I was hoping I could just lift legs at that delay line, inject chroma and luma, and just tweak voltage on theuntil it looked "good enough".
But I guess it is time to get a scope though so I can actually get correct readings on everything here.
And yeah, isolation transformer for the entire time I'm working on it.
I was hoping I could just lift legs at that delay line, inject chroma and luma, and just tweak voltage on theuntil it looked "good enough".
But I guess it is time to get a scope though so I can actually get correct readings on everything here.
And yeah, isolation transformer for the entire time I'm working on it.
Re: Sony Trinitron S-Video/Composite Mod (RF-Only TV) SOLVED
Isolation transformer with the device under test is always a good idea no matter what, but in the case of a hot chassis, it's absolutely essential. If you tried to connect the ground lead from a scope to the ground on the chassis, you'd create a ground loop and blow your house breaker and possibly make something go boom.BazookaBen wrote: ↑Fri Jan 17, 2025 12:15 am Appreciate it. If I'm reading correctly, the "vv01 delay line" in the diagram is what's separating luma/chroma, and there's not too much going after that except what appears to the brightness voltage being incorporated into the chroma signal.
I was hoping I could just lift legs at that delay line, inject chroma and luma, and just tweak voltage on theuntil it looked "good enough".
But I guess it is time to get a scope though so I can actually get correct readings on everything here.
And yeah, isolation transformer for the entire time I'm working on it.
Before you worry too much about cracking this particular nut with a scope, I would make sure you have a clearer idea of where the RF stuff is coming from. I can't see everything in your schematic, but it doesn't look like the kind of things I've seen which show an IF stage. The "delay line" looks more like what they would use to make sure the luma and chroma are happening at the same time because demodulation filters take time and things can get out of sync. That's just a guess, but I don't think the demod comes out of a component called a "delay line" that makes no sense.
-
BazookaBen
- Posts: 2123
- Joined: Thu Apr 17, 2008 8:09 pm
- Location: North Carolina
Re: Sony Trinitron S-Video/Composite Mod (RF-Only TV) SOLVED
yeah I basically just saw "one in, two out" so I thought the chip had to be separating the two from composite video. And it's the only spot I see explicit mention of the chroma/luma signals. And the input for that "delay line" chip connects (before that video buffer) to a output on the jungle called "video out".
But yeah, I do see that the naming doesn't make sense. I just figured it could be a multi-function chip, and I couldn't look up the actual part (it just has a reference number for RCA's parts catalog)
page with full schematic: https://crtdatabase.com/crts/rca/rca-e13209bc
But yeah, I do see that the naming doesn't make sense. I just figured it could be a multi-function chip, and I couldn't look up the actual part (it just has a reference number for RCA's parts catalog)
page with full schematic: https://crtdatabase.com/crts/rca/rca-e13209bc
Re: Sony Trinitron S-Video/Composite Mod (RF-Only TV) SOLVED
Okay. No, you were right more or less. That does look like what's going on. I just have never seen a set like that before. The video delay line as it's called seems to be separating the luma from the chroma. The brightness is absolutely getting mixed in with the chroma before it going to pin 38. The waveform is shown as something like 1Vpkpk, but there's 6VDC potential on pin 38, so that's where it's coming from. The brightness signal originates at pin 4 of IR01 on page 4 of the schematic.BazookaBen wrote: ↑Fri Jan 17, 2025 1:17 am yeah I basically just saw "one in, two out" so I thought the chip had to be separating the two from composite video. And it's the only spot I see explicit mention of the chroma/luma signals. And the input for that "delay line" chip connects (before that video buffer) to a output on the jungle called "video out".
But yeah, I do see that the naming doesn't make sense. I just figured it could be a multi-function chip, and I couldn't look up the actual part (it just has a reference number for RCA's parts catalog)
page with full schematic: https://crtdatabase.com/crts/rca/rca-e13209bc
Looking at the video IC after the IF and before the YC separation, nothing immediately jumps out to me as needing a sync connection, and there's no other signal connections directly from the RF tuner. So I think it may be possible.
What you would have to confirm is exactly what the signal looks like on pin 1 of the delay line. Presumably, the brightness signal is DC decoupled from the delay line by CV15 (you should replace this with a new, low leakage UKL nichicon 25V part), so you should be able to just swip-swap the chroma at that point, assuming it looks normal. You'd need a healthy chroma signal to compare it to though, from another set coming from something such as a color bars pattern.
-
BazookaBen
- Posts: 2123
- Joined: Thu Apr 17, 2008 8:09 pm
- Location: North Carolina
Re: Sony Trinitron S-Video/Composite Mod (RF-Only TV) SOLVED
Good suggestions. I don't understand why the chip says 6v on the chroma/bright input when the signal is 1v peak to peak, so I'll need to do some reading to understand this stuff.
I do see before the delay line, the composite signal goes to a "sync" pin on the jungle. Not sure how vital that is to getting a useable picture or not, don't know if I'll need to branch luma off in that direction as well, would certainly make things more complicated. Maybe it means I would just inject Luma before the delay line chip, assuming it just passes right through the delay line chip as there is no color signal to be separated
I'm basically just doing some homework on this now, taking notes. I have two other 13" TV's to get to first that have cold chassis and should be easy to s-video mod
I do see before the delay line, the composite signal goes to a "sync" pin on the jungle. Not sure how vital that is to getting a useable picture or not, don't know if I'll need to branch luma off in that direction as well, would certainly make things more complicated. Maybe it means I would just inject Luma before the delay line chip, assuming it just passes right through the delay line chip as there is no color signal to be separated
I'm basically just doing some homework on this now, taking notes. I have two other 13" TV's to get to first that have cold chassis and should be easy to s-video mod
Re: Sony Trinitron S-Video/Composite Mod (RF-Only TV) SOLVED
Because the brightness signal is being inserted as a DCV value after the chroma leaves the delay line. This results in a DC offset which has an RMS DC potential, read by the chip for signal processing purposes as 6V. However, the AC component is still in the signal. It will look something like this, where you have an AC value that centers itself higher than zero volts.BazookaBen wrote: ↑Sat Jan 18, 2025 1:47 pm I don't understand why the chip says 6v on the chroma/bright input when the signal is 1v peak to peak, so I'll need to do some reading to understand this stuff.

Ah yeah, I didn't look carefully enough. That is true. If you look at that signal, it's actually ~2Vpkpk coming out of the video IC, so I'm not sure how you would make you signal look like that. I guess you could just try it, but I'd probably put the luma in at the base of TV04 because there's 9V on the other side of it. At least if you cut the trace at the base of that transistor, your source device would be isolated from the DC restoration voltage. The base of TV04 does say 3.6V (which is RMS), so I'm not sure if the normal video signal would have a high enough offset to make things work, but it's possible. It could be that the voltage was that high for the benefit of the filtering components between the video IC and the transistor, and the 9V on the other side of TV04 is sufficient to restore the DC component for the sync line, etc.I do see before the delay line, the composite signal goes to a "sync" pin on the jungle. Not sure how vital that is to getting a useable picture or not, don't know if I'll need to branch luma off in that direction as well, would certainly make things more complicated. Maybe it means I would just inject Luma before the delay line chip, assuming it just passes right through the delay line chip as there is no color signal to be separated
I guess just be careful.
-
BazookaBen
- Posts: 2123
- Joined: Thu Apr 17, 2008 8:09 pm
- Location: North Carolina
Re: Sony Trinitron S-Video/Composite Mod (RF-Only TV) SOLVED
I guess the only things I could actually harm are the optocouplers. I'll need to read up on the math to get those to output 2v if I'm going to inject before the video buffer.
Definitely going to try just injecting after the delay line to keep it simpler, to see if the TV really does need that sync signal for useable picture
Definitely going to try just injecting after the delay line to keep it simpler, to see if the TV really does need that sync signal for useable picture
Re: Sony Trinitron S-Video/Composite Mod (RF-Only TV) SOLVED
I think it needs it. The pin it feeds is labelled "sync" and the IC it's connected to handles the deflection. That's where the sync pulse that ultimately drives the yoke is coming from.BazookaBen wrote: ↑Sat Jan 18, 2025 4:27 pm I guess the only things I could actually harm are the optocouplers. I'll need to read up on the math to get those to output 2v if I'm going to inject before the video buffer.
Definitely going to try just injecting after the delay line to keep it simpler, to see if the TV really does need that sync signal for useable picture

-
BazookaBen
- Posts: 2123
- Joined: Thu Apr 17, 2008 8:09 pm
- Location: North Carolina
Re: Sony Trinitron S-Video/Composite Mod (RF-Only TV) SOLVED
ah yes, that makes sense
Re: Sony Trinitron S-Video/Composite Mod (RF-Only TV) SOLVED
I am going to attempt to S-Video mod a Sony KV-13EXR90 (P-3B chassis) using this isolation board. The chassis seems a bit different than the one you described in your post (for example there is no U-Board). I think I have a handle on most of the connections, but I'm not 100% certain on the VCC connections. Could you take a look at this mark-up and let me know if you think I'm on the right track?KPackratt2k wrote: ↑Sat Nov 16, 2024 12:48 am I was able to add S-Video to a Sony KV-2780R TV through the use of an optocoupler board I've designed based on a circuit found on a Daewoo TV utilizing readily available components. Because this set has a hot chassis with the AV inputs isolated with optocouplers, I figured it would be best to isolate the Chroma input using a readily available optocoupler to reduce the risks of safety and having to source obsolete parts.
I've published the PCB onto OSHPark so others can use it for S-Video mods for Sony TVs using the P-3(A/B) chassis:
https://oshpark.com/shared_projects/KZcv086o
Spoiler
Additional photos:
https://imgur.com/a/D35qz1Z
You can solder the wires for power and ground on the AV input board (or the U board) and run them to the Chroma isolation board via JST connectors. I forgot to take photos of the tap points, but the cold side shares the ground with the AV inputs whereas the hot side does not. The wires for Chroma go to the chassis (or the A board), you'll have to lift the Chroma coupling capacitor (C357) from the side that connects to pin 39 of the jungle chip (IC301).
To reduce the risk of interference, I strongly recommend using shielded wire for the ends of the Chroma connections that send it to the jungle chip. When I prototyped this mod, I was getting noise that sometimes affected color intensity on the S-Video input when I used unshielded wire. Using shielded wire instead eliminated the noise and kept the color intensity consistent with the stock Composite video input.
As for a place to mount the Chroma isolator board, there's an empty space in the cabinet where the chassis boards are mounted, the empty space is located near the input board (the U board), providing a convenient spot to mount the isolator board using 3M VHB double sided tape to hold it in place. It is recommended to place two layers of kapton tape on the board before putting the double sided tape on it.
Connector labels on the Chroma isolator board and their tapping points on the TV:
Cold power and ground (the 2-pin JST on the top-left of the isolator board):
VCC (Cold Side) <- Cold 9.2V (The positive leg of C1405 connected to R1428 on the U board.)
GND (Cold Side) <- AV input ground or IC1401 pin 10
S-Video Chroma input (the 2-pin JST on the bottom-left of the isolator board):
C_IN <- S-Video Chroma (Pin 4)
GND -> Ground point for shielded wire, can also go to the ground pins of the S-Video jack (the shield and pins 1-2.)
Hot power and ground (the 3-pin JST on the top-right of the isolator board):
VCC (Hot side) <- Hot 9.3V (connected to pin 1 of IC1401 via R1403 or U-2 pin 5, should be labeled accordingly on the U board.)
Middle pin is not connected to anything.
GND (Hot side) <- IC1401 pin 5 or any ground outside of the cold area of the U board on the TV.
Chroma I/O from the TV chassis (the 2-pin JST near the 4-pin JST on the bottom-right of the isolator board):
YCJ_IN <- IC301 pin 39, can inject to the pad where C357 was lifted, connecting to this pin on the chip.
TV_C <- C357 coming from the emitter of the TV Chroma buffer (Q357).
Chroma I/O to an SPDT switch (the 4-pin JST on the bottom-right of the isolator board):
C_OUT -> Pin 1 of the switch
YCJ_IN -> Pin 2 of the switch
TV_C -> Pin 3 of the switch
GND -> A grounding point for shielded wire, provided for convenience.
Luma (S-Video pin 3) simply goes to the Composite video input.
You'll have to be strategic for finding a place to mount the S-Video jack and the Chroma switch. I've decided to relocate the RF tuner I/O panel somewhere else and run a short Coax cable from the VHF/UHF input to a coupler mounted in the VHF/UHF input's original hole. The S-Video jack is mounted on the "Aux" input on the left, though the plastic on the back side of the jack panel plate had to be shaved to allow for the panel mount jack to fit without the nut coming off. I mounted a spare panel mount SPDT slide switch in the "To Converter" hole in the middle of the jack panel to serve as the switch to toggle between S-Video and Composite video/RF.
When/if I get ahold of more Sony TVs from the same era, I'll be sure to test the mod board on those models and document any specific instructions that apply to them. I've also been working on replicating the S-Video switching circuitry from the KV-27TS30 to implement on other models with the same chassis for either using a SPST switch or a completely switchless mod using an S-Video jack with its shield cut in half to connect one side to ground and the other side to a high/low logic signal.
