Search the Community
Showing results for tags 'socketted cpu'.
Found 1 result
I am starting this thread to see if there is some possibility of getting my two SE/30 motherboards repaired or replaced. In 2012, I went through my collection and realized I had two SE/30 logic boards which showed the Horizontal Lines (Simasimac) at cold boot. Those lines were stable on-screen and would display forever, meaning the SE/30 would never boot. Prior to 2012, I had recapped both boards but found they both had identical Simasimac symptoms. The two boards were recapped at different times, one being one of the very first SE/30 logic boards I ever recapped -- I used electrolytic capacitors at the time and a dab of hot glue to ensure the traces at the feet of the caps wouldn't break if I accidentally applied pressure on them. That board worked fine for a while and then suddenly stopped working. The other bad motherboard was acquired by me and recapped later using yellow Tantalum capacitors, thanks to Trag's capacitor replacement kit he sold back in the day (not sure if he still sells that now in 2018). At that time in 2012, I referenced my SE/30 schematics and spent many hours testing PCB trace integrity, not finding the root problem that was causing the horizontal lines on both boards. Eventually, I mentioned my two defective motherboards in the 68kMLA forum, and a forum member who I will simply name "TK" said he could fix them if I would ship them off to him. I am in Japan, and he is in the USA. I shipped both boards to TK in March 2012. I followed up with TK at various times, but ultimately it took 2.5 years before the boards were shipped back to me. Unfortunately, the boards were not packed properly, and although I was assured they worked before they were shipped, they both did not work when I received them. They showed horizontal lines at the time I received them, which is the same problem they have to this day. I am going to provide repair details now, but let it be know that I am not trying to speak negatively of TK. Rather, I simply need to state certain details as relevant facts for consideration by those who may be able to kindly assist me in repair or replacement. TK discussed some of the repairs in our forum, but sadly those threads were apparently lost in a forum disk crash some years back and I can no longer link them for you. However, I still have my old PM's from many years ago, thankfully. I will now list pertinent details from my old PMs for the purpose of showing what was diagnosed and what was repaired on my two SE/30 logic boards. Again, my aim is to provide detailed information to those of you who might have the knowledge and skill to repair these boards, keeping in mind that TK was quite skilled at repairs. March 2012: TK began exchanging PM's about SE/30 experiences and the board mistakes of others. He told me he saw one of my old YouTube videos showing a SimasiMac SE/30 logic board and that I should send him my board to see what he could to with it. That led to the beginning of my discussions with him and ultimately sending both of my boards to TK for repair. Upon receipt of my boards, TK told me, "One of your boards has almost a near short on the 5V rail. not a short-short, but a semi-short. So one of the ICs is bad somewhere. I am going to have to do micro-resistance analysis to find out where..." Some days later, TK said that the tantalum-recapped board "shows a cascading short on RP2, meaning several ceramic caps inside the filter are leaky/shorted." TK also said, "the other board (electrolytic-recapped) has a short on RP3 between 10 and 11 on a single chain and none of the others." TK used a heat gun to pull RP2 from the tantalum-recapped board, but he said that the heat sensor in that heat gun failed, causing the heat to rise above 800 degrees (°F?) and resulted in burn marks on the bottom side of that board. (Scroll to the bottom and see my board photos, one of which shows that burn mark.) TK said that his heat gun did NOT kill the board though. However, TK said that pulling RP2 did not fix the problem with that board. He did note that the "UD1 74F240 buffer is corroded pretty good" but did not say if that was a problem. TK saw some corrosion on the SIMM connectors and cleaned it. (Tantalum-recapped board) Even after cleaning the SIMM banks (tantalum-recapped board), the board would work only if both banks were used (1 bank couldn't be used, apparently due to a broken trace). Apparently, the Simasimac on this board was caused by "CASLL broken to the SIMM1 Bank A and B." (A broken trace to RAM.) I was told that this broken trace would cause the other negative side effect of being forced to use both banks instead of being able to use just 1 bank of RAM. TK determined that both boards had a bad BOURNS filter which needed replacing. He said he didn't have spare BOURNS filters to do the repair, however. I decided to wait and see if he could source them. TK uploaded a YouTube video his work on my board: TK told me (regarding which board he never made clear): "the ROM select line pulls low and right back high again and nothing happens from there. All data and address lines check out, so I know this isn't the issue. With ROM removed, select line remains high until NMI invoked, then it pulls low and holds, which is NORMAL. So the next thing I am going to do is check the overlay flag. If the overlay flag isn't being flipped, the GLU chip is bad." May 2012: With it having been more than 1 month, I decided to pay $18 to Trag, who very kindly was able to source 6 BOURNS filters and ship them directly to TK. I informed TK of the shipment. May 2014 (yes, 2 years later -- and yes, I had followed-up during those 2 years): Regarding Tantalum-recapped board: One BOURNS filter was determined to be bad and replaced. TK said there might also be "a bad GLU." A "bad trace to RAM" on the tantalum-recapped board was bad, and he soldered in a wire to fix. After those fixes, the board apparently still wasn't stable, resulting in death chimes on occasion during cold boot. August-September 2014: Apparently, the CPU was dead in one of the SE/30 motherboards. TK speculated that must have been caused from a bad PDS card, although I don't think so myself since my PDS cards work in my working SE/30s. So the cause of the dead CPU is a mystery. TK took a heat gun (now with repaired temperature sensor) and removed the 68030 CPU, then put it in another board and tested and confirmed the CPU to be bad. It was at this point TK decided to install CPU sockets to make insertion and removal easier. TK remarked that he has repaired "thousands of boards" but it was the first time he'd ever seen a failed CPU. TK said that long ago he had accidentally connected a NuBUS card into a PDS slot and fried the CPU, which is why he suspected something similar must have happened for my CPU to have been killed. (Which would of course mean that my case of having a fried CPU was actually the 2nd case TK had seen.) TK said the CPU tried to run but all the data bus I/O lines were pulled close to 1.8v when they should have been a little over 4v. That led TK to believe that they must have been shorted in some way to cause damage to the CPU. He went on to say that "the only cause for those is a voltage spike, or maybe an overload of the address/data bus" and that "it is unknown if there was any damage to any of the other ICs." TK quoted me $39.90 for a replacement 68030 CPU and sockets, which I promptly paid him via PayPal. October 2014: A certain knowledgeable forum member (currently, no longer a member) sent me a PM to say that TK had repaired both of my SE/30 motherboards. I asked him how he knew that, and he said, "I speak to TK privately 3 times per day!" I then added TK to our PM conversation and the following facts were explained to me: Tantalum-recapped board needed a BOURNS filter, which TK finally installed in Oct. 2014. (Most likely using 1 of the 6pcs purchased and shipped to TK via Trag in May 2012.) TK confirm both boards were fixed as of Oct. 2014. TK requested $75 for the repair, which included shipping. TK said he reduced the price due to the long time he had the boards. I received both boards in December 2014. Metal prongs that stick up in either side of the connector side of the board were bent due to the lack of soft packing material inside the box. I mention this not to case blame (although I certainly would have packed the box better), but rather to explain that the boards could have been damaged during transport, assuming they were perfectly fixed by TK prior to him shipping them to me. I never mentioned to TK that the boards failed testing on arrival because (1) I was already out a fair amount of money and no better than when I started, and (2) because it had taken 2.5 years. I put the boards back in their anti-static bags and stored them in my closet until now (Sept. 2018). Here are high resolution photos of the top and bottom sides of both of my boards, shot today: https://www.dropbox.com/sh/sandxdkcxdss6h1/AABOOjMnrcSKhlD1neFlHOCna?dl=0 One person based in Germany has sent me a PM about these boards; but of course, I must consider the cost of shipping. I am providing all these details primarily for his benefit, but also for others to read too. I would like to hear varied opinions. All said, I've invested $30 (cost of shipping from Japan to USA) + $18 (BOURNS) + $39.90 (new 68030 & sockets) + $75 (repair & return shipping fees) = $169.90, which doesn't include the cost for the capacitor replacement and doesn't include the cost I originally paid to get the boards in the first place. Wow! Your thoughts would be greatly appreciated. Thank you.