Sega Radica S-Video Mod Chroma Problems

Started by Waterbury, February 14, 2010, 11:21:31 AM

Previous topic - Next topic

Waterbury

I modified one of those Radica game systems with 6 genesis games to have S-Video out. It works great but I modified a crappy Memorex iFlip screen to take in S-Video and while it accepts S-Video fine from any other source, with the Radica, the screen is only displaying the Luminance and seems to be ignoring the Chroma....

A few years back I ran into a similar issue with a TV I had with a modded Xbox. I learned that the TV Would not show color with one NTSC format, forgot which one, either NTSC-M, or NTSC-J. However, my other TV accepted the color fine. Luckily, the XBOX was modded so I was able to flash a new output format and was able to get color on the other TV.

Which leads me back to my original point...HELP!!! Anybody have any suggestions?  :-\


viletim

Could you give more detail about this modification?

Waterbury

I performed the mod based on the schematic on this page:

http://forums.benheck.com/viewtopic.php?f=12&t=36254&sid=3befce94c227330f6528d078b66f5367

I get color fine on my CRT TV and a S-Video to VGA converter box I have, but not on the Memorex screen. The Memorex screen was designed to play iPod video but due to Apple's lockdown of video out on recent iPod firmwares it no longer works. I was given the screen by a coworker, and I hacked it up and found that the screen expected S-Video from the iPod, so I wired up an S-Video input to it. N64, Saturn and a Digital TV Converter box have been tested and work fine with the screen's new input.


Waterbury

Ok, how about this? I'm going to try and boost the chroma line. Could I simply get an NPN transistor, hook the base to 5V+ , collector to chroma line and emitter to emitter line in series with a 75 ohm resistor?

viletim

That chroma circuit is no good. To improve it you can try placing a 330 ohm (or thereabouts) resistor between the emitter and ground. Then replace the 220u capacitor with a 0.1u ceramic or plastic type.

Waterbury

What chroma circuit is not good? The one in that forum post I linked to?

Waterbury

Also I replaced the Voltage Regulator in the unit. VCC should now be 5V+.

Waterbury

I'm quite a novice when it comes to analog circuits.

I followed this guide to separate the chroma an luminance lines:


I have since used an NPN transistor in an attempt to boost the chroma line, luminance seems to be coming through fine.

I hooked up the base pin of the transistor to the chroma out in the above pic, the collector to VCC, which is 5v+ due to a change of voltage regulator, and emitter goes out to the chroma input of the S-Video.

I notice very weak color on the left side of the screen now. If I temporarily short the Y and C lines, the screen refresh obviously goes nuts but upon un-shorting the cables, for a fraction of a moment, there are colors and they look great!

I'm confused with the circuit below:


There is mention of what traces to cut, but the circuit on the right I believe is an amp for the chroma line, but I'm not sure if it exists on the board.

I know segasonicfan is a member here. Maybe if he sees this he could help my noob ass.  :-\

viletim

vcc -------+   
           |
           c
chroma ---b  NPN
           e
           |   
           +----C1----- chroma out
           |    0.1u   
       330 R1
           |
          --- GND

Waterbury

Quote from: viletim on February 21, 2010, 09:59:52 PM
vcc -------+  
          |
          c
chroma ---b  NPN
          e
          |  
          +----C1----- chroma out
          |    0.1u  
      330 R1
          |
         --- GND


Thanks viletim, I figured that's what you meant but I thought you may have been thinking I was working off that schematic I posted earlier. Unfortunately, I still am not getting a stable color image.

I may have to give up on this project. I know the Radica is producing a NTSC compatible Chroma signal that is picked up fine on devices other than the Memorex screen. The Memorex screen uses a VIXIS VX1828B processor to decode the S-Video signal to something the actual LCD could use. The processor has the ability to take in manual image calibration changes, but it is all done via registers from microcontroller. I am not currently willing to learn the assembly required to do this so I could have a possibility of getting this frustrating project to work..

viletim

I don't know... I find the diagrams confusing.

If the "Take Chroma signal form here." point is between the break and capacitor to the right of the resistor marked "only in V1 Radica" (it doesn't help that the diagram is missing the designators) then you'll need to include the bias network of resistors marked 15k, 33k, and 1k in the chroma amp as well. The DC voltage on between the base and ground has to be 1-3 volts for the circuit to function.

Segasonicfan

This is a terrible schem I wrote up some years ago.  It was a quick job and lacks in explanation and many other things.

What I did was used a second Radica board (I used to have a bunch of these systems for hacking) and pulled the Y's amplifier circuitry and connected it to C out one the first board after separating it from the Y signal.  but you could just as easily build your own C amp using an NPN as Viletim mentioned. 

If the mod works on other screens but not your memorex I am keen to suspect a problem with the later though. If its one of those portable memorex LCDs then you might need to double check that your signals are going to the right place on the decoder chip.

listen to viletim though, he knows his stuff.  ;)

-Segasonicfan
MY WEBSITE: https://segasonicfan.wixsite.com/retro
I design PCBs for retro game systems :)

Waterbury

Quote from: Segasonicfan on February 23, 2010, 04:54:09 AM
If the mod works on other screens but not your memorex I am keen to suspect a problem with the later though. If its one of those portable memorex LCDs then you might need to double check that your signals are going to the right place on the decoder chip.

listen to viletim though, he knows his stuff.  ;)

I've been doing my homework with this project and I have concluded that the screen SHOULD be processing the chroma signal, and it has nothing to do with the wiring I have. Maybe a circuit could be built to trick the screen into processing it correctly, but after reading the LCD microcontroller datasheet, I've realized my best shot of doing something similar is through software registers which will take too much time.

As many here could probably relate to, I've accumulated so many extra parts and pieces of old hardware that I often don't know what to do with it. Such as this 7" screen I got, I figured I might as well throw in a cheap genesis and keep it in my office's lunch room. The quality didn't really didn't warrant using it for anything else.

So now I have two Radicas, and a Game Genie I bought to house one of them. Just more crap I bought to make some other crap more useful. I really should look into selling some of this.

I've even talked to Devster, the person who originally reverse engineered a lot of the Radica, and even he says that the Radica is a lost cause.  :o


Well I'd like to thank both of you for your help, you've been very helpful.  ;D