I had some spare time so I whipped this up. This is the double sided one, I am working on a single sided version for DIY fabrication. The board files are available in Eagle CAD and Gerber formats upon request. Enjoy! :thumbsup: Edit: Forgot the picture... >_< Fixed now.
This is an easier mod for the CXA1145/MB3514 That the older Genesis and Neo Geo's use. It looks the same if not better than the other with 2 amps. http://home.comcast.net/~amakusa666/PC-ENGINE-HELL-SONY-CX1145-MOD.htm Also the newer CXA1645 uses a 75ohm resistor and 220uf capacitor for pins 15 and 16 for s-video. http://www.gamesx.com/rgbadd/genesis3.htm
once i get some time and money i could do this mod on a spare model one and this is easy as pie but would like to know of a better s/video connector for the mega drive bit?
You know it's pretty hard to get 100F caps I think your circuit is broken though (so is the first one electrochip posted), what's with the biasing??? Since it's a common collector amp it should be biased to remove the forward voltage and that's it. You also shouldn't AC couple at the input since it isn't necessary and degrades the signal. Also yeah Zout should be 75 ohms but I'm guessing 33 ohms is used here to reclaim part of the level lost to improper biasing. Finally, I don't think an amp is even necessary because the video encoder should have built in video drivers... I'd go with the GameSX circuit, it's exactly what is needed, no more, no less.
Thanks, Calpis. I'm very good at general electronics, but when it comes to amps... Are there any other circuits anyone wants turned into a PCB? I can do double sided or single sided. Edit: Just noticed that the Games SX schematic is for a Genny 3, which has a different encoder. "The Sony CXA1645 chip (.pdf here) is a much better part than the CXA1135 found in the other Genesis units (as well as the Master System + NeoGeo). It has real S-video output, which the 1135 does not have." Calpis, could you show me what I need to do to get rid of the AC biasing? Also, those caps should say uF. My auto labeler screwed up.
I'm not sure what you mean by broken. I've used the circuit in the first link I gave and it works quite well.
Broken as in it's a poor circuit. I can tell this even with the worst EE training. Both amps here don't amplify the signal, they are buffers. In that link the luma signal is buffered but the chroma is not, that doesn't make sense. Chroma is AC coupled but luma is not, that doesn't make sense. Logically the encoder will output 1.4V (2V for luma with sync) intending you to use a 75 ohm series resistor, which in series with the 75 ohm load inside the display will divide the signal to the proper 0.7V (1V for luma) with exact impedance matching. So that's what one should do... The chip already has the capability of driving the video load so the only reason to buffer the signal is to protect the chip from damage (which isn't very likely). These buffers are degrading the signal that is its cleanest right out of the chip, especially when using wacky R values that don't bring the signal to the proper level... The only reason these amps work is because being common collectors or "voltage follower"s the output is forced to be the input - 0.7V, at whatever output impedance (because they are used to drive low impedance loads from high impedance sources). 1.4V - 0.7V = the desired 0.7V, but not exactly because of the bad resistor choices used for dividing, and the fact that the transistor's forward voltage won't be exact. To properly buffer a DC video signal, you want to bias the input above 1V so none of the signal is clipped. Remitter should be high, like 1K or more, then output R should be 75 ohms and AC coupled with a 220uF C. But again it doesn't even matter here, the encoder should drive any TV properly, just add the R and C.
You're looking at the wrong page: http://gamesx.com/wiki/doku.php?id=av:genesis_1_s-video_mod And it's CXA1145, not CXA1135, which doesn't exist. It could be a MB3514 instead, though.
