![]() |
|
#1
|
||||
|
||||
|
Hitachi CPT-1455 RGB mod
Hi,
I've got a Hitachi CPT-1455 (NP84CQ/2 chassis) that I am considering doing an RGB mod on, and because this is a hot chassis I plan to use optocouplers like the HCPL-4562. It's unfortunately an RF only set, but the picture quality and geometry is pretty good, and it has low hours so it's worth modding in my opinion. I also love the form factor and look of it. The main IC is a TDA3562A and according to the datasheet it supports RGB input at 0,9-3V. I had a look at the brother CPT-1444, and it's got native RGB input according to the service manual. Here is a screenshot of it: ![]() These TVs have the same chassis number, but obviously different revisions and power supplies. Here is the service manual for my CPT-1455, I had to buy it on eBay, scan it and upload to this site. So my idea was to wire up the RGB inputs using high bandwidth optocouplers for analog video, powered by an isolated 5V power supply at the input side, and add a SCART connection to it. There is a thread here showing a evaluation board that shows how to wire up the optocoupler for analog video/audio: https://antiqueradios.com/forums/viewtopic.php?t=430210 However I'm kind of stuck on how I should go about wiring this. So far I'm on the edge of buying a set of PCBs and optocouplers and just try my best using regular RGB mod guides, but do I need any extra circuit on the TV side? If I use the CPT-1444 board for reference, there's a lot of added circuit handling the RGB signal. I believe it's not as easy as just inserting the RGB signal straight to the IC with some 75 ohm resistors to ground and enable the blanking, is it? As of now I've built in a composite to RF converter into the TV with its own isolated 5V power supply, so the TV works like a proper composite set, and the picture is pretty good, but having native RGB would be perfect. Last edited by ostlar; 01-19-2026 at 11:06 AM. |
|
#2
|
||||
|
||||
|
Quote:
-the little screenshot you provided shows the RGB inputs routed to a "socket." that is confusing to me because a round socket would normally indicate a tube, and this should be an input. what's the deal with that, is it just a wired connector? Is it the correct set? - " it supports RGB input at 0,9-3V" NO it does not. That is the supply voltage for the Fast Blanking Pin 9. You must supply Pin 9 with something in that voltage range to "turn on" the RGB. The RGB voltage level of this IC is 4Vpp. That's 4 times higher than normal video. If you wanted to RGB mod this set, you would need to build a video level shifter to raise the whole thing up. Texas Instruments makes a whole line of video processors that you might be able to use. The THS7320 is a 4X fixed gain IC for ED content, and it might work for SD, not sure. You can pour through their video amps and look for an SD 4X IC if you want. https://www.ti.com/product-category/.../overview.html Would probably be a fairly heavy lift, but it could be a fun project as long as you don't blow anything up or electrocute yourself. Or you could do a composite input hack or an S video hack. I would probably do an s-video hack and go into pins 4&8 (75 ohm terminated to ground) through two legs of the optocoupler mod. S-video is 95% percent of the way to RGB quality, and you would be highly unlikely to notice the difference between the two in practice. You could maybe do so by taking (very high quality) photos and digitally enlarging them to see the pixels, but you'll never see a difference sitting down on a couch playing a game. Make sure you do any live testing with an isolation transformer and be careful to observe all safety precautions. You're at your own risk of death, etc, etc. |
|
#3
|
||||
|
||||
|
Quote:
![]() This image is from a screenshot on YouTube. I've ordered PCBs and components for the optocoupler circuits, and I have landed on the same conclusion as you to do a composite or S-video hack instead. I will update this thread with the progress. |
|
#4
|
||||
|
||||
|
Quote:
If you choose to do S-Video: You might have to keep the luma on the composite input and just hack in the chroma signal to the jungle IC. It also might be desirable to identify any comb filtering that is going on and bypass it if possible. Adrian Black just did a dive into s-video mods on a more recent set, but the principles should be the same (your circuit will just have less ICs and more discrete components). https://youtu.be/tCO0kl_RxrI?si=devQziMOXeikg5MS Last edited by vol.2; 01-28-2026 at 09:07 AM. |
|
#5
|
||||
|
||||
|
Quote:
I've had a look at this example of an RF only Sony set, looks to be a similar setup to my TV so it's an example to follow. I want to do both S-video and composite through a SCART connector. I'm most likely receiving the optocoupler PCBs and parts next week. |
| Audiokarma |
|
#6
|
||||
|
||||
|
If you are interested in doing RGB still, it's worth trying to insert directly at the DIN connector before worrying about the s-video stuff as it would less effort.
Did you say this was a hot chassis and that's why you're doing the opto-isolator? Why would this set have a DIN input on the back if it's a hot chassis? You need to 1) post the entire power input portion of the schematic here and 2) open up the TV and physically examine the AC input stage. Judging from the fact that this set has an external DIN connector leading directly to the jungle IC, I would assume that it has a floating ground and is not a hot chassis, and therefore wouldn't need your isolation PCBs. This needs to be confirmed, but I can't see how they would be able to allow you to plug in and external device if it's a hot chassis. It might be that you are mixing up the schematics from the RF only set (which would probably be a hot chassis), and the version with a DIN input, which (probably) adds an input transformer. Need to confirm though. Post the schematic and take some pictures of what is directly on the inside of the set where the power cord goes in. |
|
#7
|
||||
|
||||
|
Quote:
They are almost identical and have the same chassis number, "NP84CQ/2" for the CPT-1455 and "NP84CQ" for the CPT-1444, but the main difference is the isolation transformer and the DIN-connector on the CPT-1444. Here's a Reddit thread where they discuss the RGB connector of the big brother CPT-1646, and here's one for the CPT-1644. These also have the same chassis number as 1444 and 1455. They seem to agree that it's a standard RGB signal with composite sync. The PCBs will arrive next week, and all the other components have just arrived today. Pretty excited to revive this TV with new inputs, and get rid of the RF modulator. If this works, I will attempt it on two other RF only TVs that I have. |
![]() |
| Thread Tools | |
| Display Modes | |
|
|