Cmherndon79 Posted June 10, 2020 Author Share Posted June 10, 2020 Well foo, but kinda what I thought. I connected the 7800 up to my CRT tube and it's got the same behavior there, so that variable is out now. Anyone got a recommendation on a supplier for the Sony cxk5816pn-12l SRAM? I see a few places out there have it, but never bought from them. Sells on eBay too, but don't want to go that route. Quote Link to comment Share on other sites More sharing options...
+-^CrossBow^- Posted June 11, 2020 Share Posted June 11, 2020 Doesn't have to be Sony brand...I've seen Goldstar chips and several others that escape my mind at the moment. Quote Link to comment Share on other sites More sharing options...
Cmherndon79 Posted June 12, 2020 Author Share Posted June 12, 2020 (edited) Found this list from Console5: Devices Include: 5516 6116 LC3517B LH5116HN LH5216AD SRM2016 BR6216B CXK5814P MN4216 TMM2015BP-15 MSM2128-15RS So now I hunt. Edit: Found 6116-120s at Unicorn Electronics, looks like 150 would probably be ok, but ordered 120 just to be safe as it was a 10 cent per part difference. Found various other things I'll need eventually to fill out the minimum $ amount. Edited June 12, 2020 by Cmherndon79 Quote Link to comment Share on other sites More sharing options...
Cmherndon79 Posted July 10, 2020 Author Share Posted July 10, 2020 I'm back. Yeah, had just put this aside for a while. I got the RAM in and swapped it out a few days ago, but the behavior didn't change. I had noticed a cut trace I had not seen before and got excited for a second before realizing it ran from the location of the ram to ground, where the bodge resistor was. I started looking are around R57 again. Noticed that it gets power through CR3 and CR4 on the way to R52 (I think). I decided to pull CR3 out this morning when I measured it, I swear I got nothing on either side, but then I went and checked it this afternoon and I got .5 in the correct direction and nothing in the other. Is it possible that a diode can be intermittently bad? A Google search, I see people say so, but I have never seen it before. Might explain why the color goes in and out? Anyway, I pulled a 1N914 diode off an old Commodore board and installed it and I haven't had a total loss of color since then. I am still getting a bit of color shift, which is more noticeable in Centipede than Food Fight. Occasionally the Centipede will shift from green to red and the Centipede logo will turn pink. Possible I need to check CR4, I'm guessing. Quote Link to comment Share on other sites More sharing options...
Cmherndon79 Posted July 10, 2020 Author Share Posted July 10, 2020 I did swap CR4 earlier. No change in the color shifting. Still has not reverted back to black, white and yellow. Quote Link to comment Share on other sites More sharing options...
Cmherndon79 Posted December 20, 2020 Author Share Posted December 20, 2020 So... I had an Atari 400 computer come across my bench a few weeks back. I had a similar issue, where the screen would flip between black and white and color. Cleaned out the variable resistor and it persisted. I found a post on here from a few years ago, someone having the same issue and they swapped the crystal oscillator C010177 to correct the issue. I pulled that part off a VCS board I had for parts and installed it in the 400 and it solved my issue of the color cycling in and out. It occurred to me that I had not tried this on that 7800, so I looked at the crystal on the 7800 and it has a different Atari PN of C024912 and I found it for sale at Best Electronics for $5. The frequency though was also printed on the crystal of 14.31818, so I instead ordered 10 new from Mouser Electronics for less than $5. I swapped that out at 15:00 and it is now 19:11 and I haven't observed any color shifts. Perhaps this was the problem. I also found this, though I don't know how authoritative it is: http://7800.8bitdev.org/index.php/Atari_7800_Color_Documentation "Consistency Issues There are a number of factors that result in inconsistency between the palettes of 7800 consoles: Each 7800 console has a tunable resistor inside, which is used to modify the color signal frequency period. On any given console this resistor may be adjusted a number of different ways, with at least 2 published historic standards in play - one with hues ~25.7 degrees apart, and another with hues ~26.7 degrees apart. As a 7800 console warms up, so do the components that produce the color signal frequency period, while the colorburst frequency remains constant, so the distance between the various hues in the palette stretch out. When cold, the console's hue values have ~24.7 degrees of separation, and as the console warms up, up to ~27.7 degrees of separation can be seen between the hues. The separation is cumulative, and so is more pronounced with higher hue values; the $Ex hue can shift from being green through yellow/brown, and $Fx can shift from an initial greenish-yellow/brown through to a reddish yellow/brown. The majority of this shifting will happen in the first 20 minutes of console operation, though some additional shifting can be seen hours later. The NTSC and PAL standards don't consistently display the same colors from TV to TV, and almost all TVs allow the viewer to tune colors to their liking. The largest factor in these consistency issues is typically the warm-up issue. Since the latter hues shift more than the earlier ones, a 7800 game designer should generally avoid the last 2 $Ex and $Fx hues for critical items that need faithful color. e.g. one may wish to avoid using $Ex for green grass, or $Fx for for gold coins. The warm-up effect can be seen in commercial 7800 title Choplifter, where the ground utilizes a $E8 color value, and consequently shifts from green to yellow-brown." Quote Link to comment Share on other sites More sharing options...
Cmherndon79 Posted December 20, 2020 Author Share Posted December 20, 2020 Solved: Everything is still running great after swapping out the oscillator crystal. I played about a dozen games and everything was consistent, with no color drift. I didn't see any issues before playing nor after. Quote Link to comment Share on other sites More sharing options...
+Mitch Posted December 20, 2020 Share Posted December 20, 2020 Awesome! Glad to hear you solved the problem. Mitch 1 Quote Link to comment Share on other sites More sharing options...
Shawn Posted December 20, 2020 Share Posted December 20, 2020 Glad to hear it worked out and you didn't give up on it 1 Quote Link to comment Share on other sites More sharing options...
+DrVenkman Posted December 20, 2020 Share Posted December 20, 2020 Glad you finally solved the issue and glad it was a simple crystal issue. I will point out for the tech/repair guy crowd that this kind of weird issue probably would have shown up quickly with an oscilloscope fast enough to measure the crystal frequency (e.g., something with at least 30Mhz - 50Mhz or higher, fast enough to catch at least a couple of data points on each pulse), at least if you had a working system to compare with. 1 Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.