shmups.system11.org

I'm hoping someone can give me some feedback / help on an RGB mod I'm attempting. I've got a Sony Trinitron KV-27V20 (AA-2 chassis) that I was able to pick up for free locally - if I screw it up I won't lose any sleep, but I'd love to get this right the first time. From reading through this post and seeing how others have modded similar TV models, I've put together a diagram of what I "think" I need to do:



The OSD (IC001) RGB and BLK lines need to be cut and redirected into aa 4PDT switch. The lines coming from the Jungle (IC351) RGB and YS also need to be redirected into the 4PDT with the RGB lines having 0.1uF capactiors in line. Finally, I'm adding 3 BNC connectors (RGB) with negative attached to ground, and positive attached to the 4PDT with 75 ohm resistors inline and also attached to ground. I'm also going to feed the 5V standby into the 4PDT switch.



When the switch is "on" the RGB from the BNC connectors and the 5v standby will be hooked up to the Jungle RGB and YS.

When the switch is "off", the OSD RGB and BLK will be hooked up to the Jungle RGB and YS instead.

I took the TV apart and pulled out the A board and tracked down the components:



At this point I'm trying to figure out / understand the best place to actually do the mod. My soldering skills are fairly basic and while I can solder small-ish wires without any issues, I've yet to attempt soldering anything surface mount. I've identified two areas where the mod can potentially be done (A and B on the images).

The easier of the two looks like it would be "A" which is nearby the OSD (IC001) chip. There are 4 resistors on the top of the board that are nice and accessible - my thoughts are to remove the resistors and proceed with the mod using the holes on the board in place of the resistors. After removing these resistors I assume I need to keep them in the circuit / reattach them.

With that said, there is an area of the circuit between the OSD and Jungle that I'm not sure of the purpose (the purple question mark on the circuit diagram / back of the circuit board). It looks like there are a number of transistors for each line (RGB) that are attached to the 9V line. Does it matter what side of this area I do the mod?

The harder of the two is "B". I'll be honest, I'm not even sure how to attempt this mod here - the components all appear to be surface mount and possibly even under the Jungle (IC351) itself?

If I can't do "A", is there an easier way of modding this than attempting "B"? I can post the service manual / datasheet for the Jungle IC for this TV as well if that helps. The Jungle IC is a Sony CXA2025S.

https://imgur.com/a/LnylNEG (additional pics from the service manual / board)

Quoting this for visibility. It got stuck in moderation a couple days and appeared where I originally posted it timewise. Thanks!

edit,double post

hello, has anyone modded a panasonic # CT-32HX41E, 32 inch. i have a chance to snag one for free but dont wanna add this huge beast if i cant mod it, thanks!

It's a BA-5 chassis, why would anyone still want to do a snip method when they can just use the guide?

Sony BA-5D/BA-5 Chassis TV RGB Mod with OSD Mix
viewtopic.php?f=6&t=63622

Is it because there are STILL people making videos using the obsolete OSD snip method as if they've achieved something?

As for the extra set of RGB pins 25-28, I chatted to Jose Cruz about these years ago, he said he tried to use them but they were disabled. I also recall someone on this thread claiming they changed the Option bits in the service menu in order to enable the extra RGB input. IIRC they claimed to have just turned "everything to 1" and the RGB input became enabled. I'm not endorsing this but I would definitely be investigating the option bits.I live half a world away and have never even seen one of these sets so I can't try myself.

If you can't get the extra RGB input I strongly suggest you to follow the guide. It's EASIER to implement that and OSD snip and you get to keep your OSD.

This AA-2 chassis TV can also be OSD Mux modded easily. It's already been covered in the thread but here we go again.

This set has an unusual OSD RGB circuit because it converts RGBI to RGB through a series of transistors but that is ok, there is still a simple way to do it.

R1123, R1128 and R133 are the OSD termination resistors that we want to target for the OSD mux. So we remove them.

They are 390 ohm resistors. Instead of using the OSD/External RGB muxing calculator to work out our mux resistors we will just cheat and subtract 75 from 390 to get something close. 390 - 75 = 315. Next standard resistor size is 330 ohms.

So we want to inject our 75 ohm terminated RGB into 330 ohm resistors and then into circuit somewhere between R1121 (Red) , R1126 (Green) , R131 (Blue) and R1355 (Red), R1356 (green) and R1357 (blue). IIRC correctly there are jumpers in there but it's been a while so i may be mistaken. The ungrounded pads left by removing R1123, R1128 and R133 are an obvious spot but may not be suitable.

For a very simple blanking circuit try taking a 5V line to the input of single point switch, then connect the output of the switch to a 470 ohm resistor and solder it onto the leg of L003 that is closest to IC001. Quick and dirty but should work fine.

Hi Mark,

Sorry, should have put in my original post that I was planning on going the mux route (thanks to your sig actually, saw that and thought "wait, I better look further into this"). Was just happy to see someone had done snip mod previous, confirmed at the very least I could use the RGB 2 pins with the mux method. After posting this I found your awesome guide on BA-5D / BA-5 and it boosted my confidence quite a bit. If this RGB 1 route doesn't work out I will definitely be going to that option.

Watched a video ( https://www.youtube.com/watch?v=ANSxASanZes this one ) where they had to intercept and change i2c in flight with a microcontroller continuously to get the additional input to work. In his case it seemed the blanking pin didn't work at all until it was flagged as enabled. I'll go with the option bits in service menu first and see if I can get the YS1 blanking pin to respond. If it works, great; if not, I'll just go the mux route. I'll make sure to post an update on what I find (and I will put it in the BA-5D / BA-5 thread).

Thanks!

EDIT: Unfortunately it doesn't look like the service menu on KV-20FS12 exposes option bytes. I'll still test that blanking pin though.

I’d forgotten about that video. That was some excellent work he did. Would be worth following that up.

The author of that video released an open source design for the i2c modchip he made. That's what I was alluding to in the other RGB mod thread when we were discussing disabled inputs

I don’t forget much but this completely slipped my mind.

Have you got a link to the code etc Max?

I found this, had to use Google cache cause his pages are 404'ing now: http://webcache.googleusercontent.com/s ... clnk&gl=us

His second attempt with this chip and another TV failed due to Sony having differences in i2c usage and lack of datasheet to figure out what was different.

Getting my shop space figured out. Once I do, I've got a Rigol scope with i2c decoder unlocked, maybe I can grab some info from my BA-5.

I finished the RGB mod (didn't go the RGB / OSD mix route, wasn't confident enough to give that a shot) and it's a partial success. It looks like the B line is not working - when I flip the switch to use RGB I'm only getting red/green. The odd thing is if I open the OSD while RGB is enabled, I am seeing a faint blue OSD / menu overlaid on my RGB signal. I'm going to open it up again and check to see if I somehow reversed the wires but I don't think so - I kept everything tied up and organized. If the wires aren't reversed, any idea what it could be? When I flip the switch to no longer display RGB, the OSD text is cyan instead of the normal green. The strange thing is if I restart the TV in "non-RGB" mode the OSD text is green again (which I'd expect).

The second issue seems to be sync related. I'm using an AMD Radeon HD 7900 series card with a DVI-I to VGA adapter, with a VGA to BNC (RGB Hs Vs) cable. I'm using crt emudriver 2.0 beta 15 as well. When I only plug up RGB to the BNC connectors I added I'm seeing what you'd expect. If I plug up the grey sync cable (not sure if that's H or V sync normally?) I'm able to get a stable picture when I configure video mode maker with a monitor preset of "NTSC TV 60hz/525" and *disable* composite sync. If I use Arcade 15.7kHz or try enabling composite sync I just get a rolling flickering mess. The best I can manage with composite sync enabled seemed to be working hsync, but never working vsync.

Even with this configuration, the vsync is screwy. If I turn the TV off then on again the picture syncs at a random vertical offset. If I leave the TV on and unplug/plug in the sync cable the picture will vsync at a random vertical offset that seems to correspond to where the picture was rolling when I plugged in the cable.

Overall I've got a picture and it's relatively stable so I'm fairly happy - I honestly wasn't expecting to get this far the first time I gave it a shot.

Did you pull up your RGB lines and inject them directly into your Jungle Chip? If so, maybe you accidently left blue in circuit. What happens if you leave sync and only blue connected and then blank? Feels like blue is taking a different path, grounded differently, 75 ohm termination not correct on that line. Something like that.

I'm not entirely sure what you mean by "pull up" - are you talking about the resistors?

When I connect sync (to video 1 composite input) and blue only, I just get a black picture with the OSD displaying as faint blue text. Weirdly, if I connect ONLY the sync cable, the blue OSD text is visible but very wobbly almost like the h-sync is messed up?

Here is what it looks like when RGB and sync are connected currently:



Here is what the wiring looks like at the BNC connectors. The soldering sucks but I think it should be relatively sound. All 3 BNC connectors are grounded together through blue with a 75 ohm resistor connecting the + for each BNC to ground as well.



Here's an album with a few pics from the mod as well: https://imgur.com/a/5s5eVom

Finished another modded Sony Trintitron KV-27V42. Fillowed the simple instructions here: [url]https://shmups.system11.org/viewtopic.php?f=6&t=56155&p=1342960&hilit=documenting+Sony+mod#p1342960
[/url]

I went with the 1000/100 + 75ohm for ground setup referenced here:


Quoting Mark:
- remove the existing inline throughole resistors R025, R026 and R027
- replace each of them with a 1000R Resistor
- remove surface mount resistors R086, R087 and R088
- On the external RGB lines use 75R terminations to ground and then 100R inline
- Connect the external RGB lines to the leg of the diodes closest to the jungle
- Connect 5V to switch to the empty header by the composite area labelled 5v (30) I think
I connect both Red and white together on the switch, run Black to ground, and Red to 5v. This allows for the switch to light up when blanking got
RGB mode.

Mounted new RGB header RCAs and use the back of that for all the Resistors. I also ran shielded RGB cables with ground tinning through them to mitigate any interference





Money shots:

Connrctors. I went the easy route so I had a good spot to put all the resistors. Again 1000 on the board in 25'/26/27





Blank was pulled dirctly off of the chip. and run to a chip that is powered with 5v only RGB is enabled.


Added Cooler button this time that I got from amazon.




Back all tighten up.


Some text shots. I know I still have some small issues in to corners. I'll mess with them later.

And some gameplay

sorry, I'm not sure on this one. So you didn't do the MUX method in Marks signature on this Sony?

No, I followed the simpler RGB mod steps earlier in this thread and on youtube that I found.

Has anyone ever successfully modded a set with a single chip with the MCU Core and the Signal processor in the same IC? I am just wondering if that ever worked for anyone. I was reading back through the thread to see if I could find an example. I am looking for the next "challenge" and wondering if those can be done.

Hi guys I'm working on a pretty cool 13in transparent Prison TV (Magnavox 13mt1431/17).

I found a possible insertion point for my RGB signal here which I highlighted here:

https://imgur.com/a/FzBNMQw

The problem is I don't know exactly what I need to do on the PCB to find where to solder my wires. Do I just solder directly to the 45, 46, 47, and 48 pin? Does anyone see a better place?
I guess I also need to connect audio, and sync somehow, does anyone see where I can pick those up?

Finally, I'm very confused about how to trigger the TV to display RGB once everything is in place.

Thanks for any helpful advice!!

Hi guys I'm working on a pretty cool 13in transparent Prison TV (Magnavox 13mt1431/17).

I found a possible insertion point for my RGB signal here which I highlighted here:

https://imgur.com/a/FzBNMQw

The problem is I don't know exactly what I need to do on the PCB to find where to solder my wires. Do I just solder directly to the 45, 46, 47, and 48 pin? Does anyone see a better place?
I guess I also need to connect audio, and sync somehow, does anyone see where I can pick those up?

Finally, I'm very confused about how to trigger the TV to display RGB once everything is in place.

Thanks for any helpful advice!!

Not simpler, dumber. In general it's way easier to do a mux.

You're absolutely right. I initially discounted it when I saw the spreadsheet to calculate resistor values and kind of glossed over it (after having watched a few videos explaining the method I tried). I've got a Sony KV-27S42 that I'm going to try your OSD mix method on later this week (waiting on parts ordered from Amazon).

Well this is about as good as I can get this 27. It was dropped at one time so convergence took a while. I basically had to start from scratch, rings, then strips and magnets, then service menu. This was a fun one. Moving on to the 32 inch Trinitron now.

lol "about as good"?? From the photo here, the geometry and convergence appear to be very good!

When adjusting the rings, are you using a pattern generator and a scope as most service manuals recommend?

I am using the tried and true method of "winging it". I took pictures of the rings, used a sharpie to mark a line so I could always go back. I then gently broke free each one of the color rings (I left purity alone as it was good on my set). Move around while watching a mirror until I got them as good as I could. Magnets and strips cleaned up the rest.

nice job , looks great

Hello. I bought recently a TV Elemis T939D Ruby. It has a SCART connector with RGB support, but the housing has hidden holes for RCA connectors. As there are 5 of them, I thought about mounting the Component, not for improving the image quality (after all, it is known that there are no significant differences between RGB and YPbPr) but for the possibility of connecting more devices (PSP for example is simply not able to display the image as standard RGB). The question, however, is the installation of YPbPr possible in this TV model?

Hi fandangos ! i have the same Gradiente TV from your post , so , i succesfully hooked up my pc by vga rgb mod to the neck board , great picture but no scanlines , my setup is ok cause was tested on another tv , samsung 21 flat screen , do you know if there is a chance to have those beautifull scanlines on the Gradiente TV ? ( sorry about my english )

Have worked on a similar set and can say the rgb is disabled but modding it for component should work just fine,also you have to mess with the service menu to enable the component video source.