shmups.system11.org

So, it looks like this set is not RGB moddable, but I saw there's a Chroma input, so I thought to try connecting C to Pin 36 on Jungle (Chroma In, 75ohm to ground and coupling capacitor on the motherboard) and the composite port for Y. No luck or image at all.

Any ideas on where to go from here? Perhaps that AV switch (Pin 43) is used to switch between svideo and composite on the jungle?

I couldn't find a clear datasheet for this jungle, but the crappy photo I could find does seem to line up pinwise with the LA7673 below.

Possibly. My next step would be to look for other TVs with the same jungle chip that have s-video and see how it's wired up

I can't seem to find much about that jungle or what other sets used it. I realized I had typed the wrong jungle IC in the title, so I've fixed that now.

The datasheet is available, and it says that S-Video is supported in the features section. It doesn't give a really clear path to that though; the example schematic is a normal composite video setup.

https://www.alldatasheet.com/datasheet- ... A7672.html

Did you get anywhere with it?

Nope, no luck. I tried sending Chroma to the Chroma In pin (Pin 36) and Y through composite port, and I also tried isolating the Chroma In pin from the rest of the circuit, no luck. I get black and white input from Y, but no color.

I think what's going on is that the composite signal gets push through discrete components to get demodulated. If you look at the sample schematic, the external composite gets input through pin 40 and then it just goes through an internal switch that selects either the internally demodulated RF signal, or the ext composite signal, and then it passes it right back out of the chip to some external components to separate the chroma, luma, and sync, and is then passed back into the chip through multiple pins.

The datasheet explicitly mentions S-Video input capabilities, it just doesn't give an example circuit for it. It might be possible to figure it out by comparing the 9PT40 schematic and the datasheet, but I'm not sure.

Messed around with it more tonight. Thought maybe it just detects composite input and turns off Chroma pin processing so I tried bypassing Y straight to pin 34, but then Luma isn't even displayed. Also tried toggling pin 43 (AV switch) but that just changes it to the antenna input snow screen.

Seems that no matter what I do, it ignores Chroma input.

Yeah, that makes sense. If you follow the block diagram, the whole A/V switch area is just to toggle between the video from the IF section and the external composite input. The result of that is passed directly out of the chip on pin 38 and broken down into multiple signals with discrete components. The only way to do it is to reverse engineer the various inputs downstream from there. It looks like it ties directly into the chip at 29 (H/V), 31 (brightness clamp), 33 (videotone, seems to effect the sharpness and contrast), 34 (black expand, something like luma?), and 36 (chroma). I'm going to go out on a limb and say that the voltages might be different (from standard chroma), and some kind of DC restoration would be necessary to bring the chroma to the to correct offset. Maybe someone else here with more experience can comment. Best possible situation is to find a set with s-video that uses this chip and copy the circuit. Something from the early 90s most likely

Current best guess (after LukeEvansSimon suggested it) is that it's the "chroma killer" (ACC killer) of the set turning off color processing if it doesn't see a chroma signal coming in through the regular composite port. No idea how to disable it though. There's a "Chroma VCC" pin on it that sits at 8.9V, but it's also connected to other stuff and the voltage doesn't change if I run a composite signal, just luma, or have nothing connected.

That's easy enough to test. You could pump a regular composite video signal in while isolating the chroma pin and sending that separately. Any console that has both composite and s-video on the output would work. If it display something in that configuration, that's what's happening (probably)

But again, if you look at the schematic, it derives all those other signals from the composite video. Whatever implementation of s-video is done has to replicate all of the other signals. Look at the block diagram: