Technics Posted January 31 Report Share Posted January 31 Please see link to 10 images of board here https://imgur.com/a/nJotOX1 The machine in question is a v1.0 with fresh electrolytic capacitors installed everywhere a few weeks ago aside from the power supply (Foxlink, with resoldered power leads). It also has an Aladdin mod chip which I was running Cerbios 2.2.0 beta on previously. After successfully upgrading the RAM on a v1.4 machine yesterday I decided to do my primary xbox today. RAM in question is the leaded F-QC50 variant. They appear entirely to be authentic Samsung chips. With XBlast OS flashed to the mod chip, I began with the RAM install. First chip, test, success. Second chip, test, fail. Quick rework and second passes. Then I decide to do the last two together and test afterwards. Test shows 1-3 pass, number 4 fails. This is the one below the CPU and next to the connector from the power supply. Rework number 4 with an iron, not hot air, still fails, rework it again, still fails. Go for a third try and now the Xbox will momentarily boot with solid green LEDs (hard disk spins up, fans spin normal speed, LED on mod chip illuminates) then quickly cuts power. Then will turn itself on again about a second later, usually with no LEDs on the eject button, fans spin normally, LED on mod chip lights, then power cuts out. There is no display. I also swapped in a known good power supply from a different v1.0 and no difference was made. This cycle appears to repeat indefinitely. Sometimes the LEDs on the eject button light up green, sometimes they flash green, other times there is no light at all. I have tried with no disc drive and no hard disk, with just one of those, and with both of them. I removed the fourth RAM chip in an attempt to rule out the chip itself, however the same result ensues. I then removed a factory RAM chip (M-QC50) from a known good console and installed that in the fourth slot, no change. I ended up removing that chip so only ones that passed the test in XBlast are installed. I know it is possible for the chip to be bad, yet still pass. I took the utmost care not to “splash solder” around while installing and/or removing the memory. I taped up the surrounding areas of the board and components with kapton tape before performing any work with hot air or an iron. High quality flux was used and cleaned up. I examined the board thoroughly multiple times and I cannot physically see an issue. Using hot air I carefully reflowed the factory RAM as well as the three that I added. My question mainly is where should I be looking for a solution to this issue? This is not the typical 3 reboots then FRAG problem. Is there a power related component I should be looking at? Any help is greatly appreciated. Quote Link to comment Share on other sites More sharing options...
Technics Posted January 31 Author Report Share Posted January 31 Well, ladies and gentlemen, I return with good news. I have since gotten a good sleep and re-evaluated the situation. I was able to determine that at some point in my perilous evening several pins of one of the RAM chips had broken free of its pad. I corrected those issues, and tested again. The Xbox booted into XBlast and I was able to perform the RAM test with slots 1-3 passing as they had before. I then installed the fourth module and all is well! Fully functional again, with Cerbios reflashed to the mod chip. While I was able to solve the issue myself, I think I should leave the post as is for the next person that comes along and has similar issues. Even if you've completed the job before, mistakes can still happen. Sending everyone that reads this well wishes! Quote Link to comment Share on other sites More sharing options...
Prehistoricman Posted January 31 Report Share Posted January 31 That kind of boot loop usually happens when booting Xblast with some RAM soldering mistake. It occurs because the xblast bootloader fails to load the main xblast image correctly. The bootloader has already passed the SMC check (so it will not FRAG) but due to the check fail, xblast requests a reboot. 1 Quote Link to comment Share on other sites More sharing options...
SoftMachine Posted February 13 Report Share Posted February 13 @Prehistoricmanis right, it can also happen with a bad/corrupted bios, but that's likely not your issue. Sometimes when the board gets too hot for too long it can make the RAM's pins on the opposite side of the board (or directly adjacent) come loose, check them all for dry joints 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.