Well if the motherboard is over 20 years old and been played a lot pc, sony, atari, nintendo, amiga, and sega, chances are the components are all de-rated and needs to be up rev'd.
wrote:Well if the motherboard is over 20 years old and been played a lot pc, sony, atari, nintendo, amiga, and sega, chances are the components are all de-rated and needs to be up rev'd.
20 year old hardware will work fine if you know how to maintain it.
So another board burns again. Same system. This time it was an Intel D815EEA with a different CPU and of course no Slotket was involved. So here are the components that were the same from that last builds that had the mobo burn up.
512MB PC-133 RAM (256x128x128)
3Com 3C905B-TXNM PCI NIC
Gainward FX5900 128MB AGP
USR 56K PCI Modem
Corsair SF450 PSU
Motherboard was different. CPU was different. No Slotket adapter since this board is a Socket 370 board and not Slot 1.
I am running the graphics card on its own power cable but with the modular SATA power cable and then a SATA to D-connector adapter since the PSU comes with only one D-connector power cable which is being used for the drives. Could this be a problem? I don't think so, but maybe using a SATA power cable and converter on the GPU is causing power issues?
Weird thing is the machine ran fine under heavy load during benchmarks. Ran for a few days on and off. Tonight it was just sitting there turned on sitting at the Windows 98SE desktop doing absolutely nothing when the burning smell filled the room. The other two times this has happened with the other motherboards mentioned earlier in this thread, same thing. Just idling and not under load and delayed, no immediate problems. Every time it will run for awhile just fine and then one day, poof, burning smell and the system fails.
I am beginning to suspect the Corsair PSU, which is brand new, but before I jump to conclusions...
I have a very basic Pentium 4 system with an ASUS P4S8X-MX motherboard with 2.5GB of RAM. I am going to let this system run with that Corsair PSU for a few days and then slowly add each component from above one by one and see if I can burn the system up to find what the culprit is. I am down to trial and error as I have no idea what is going on. We shall see. Very frustrating to say the least.
Surprisingly, my brother had trouble with a Corsair power supply somewhat recently. As you describe, there was a sharp "poof" and burning smell, but from the power supply only. He brought it back to the store within a few days of purchasing it for a replacement, and hasn't had a problem since. Fortunately, it didn't take any other components with it, in his case.
wrote:Surprisingly, my brother had trouble with a Corsair power supply somewhat recently. As you describe, there was a sharp "poof" and burning smell, but from the power supply only. He brought it back to the store within a few days of purchasing it for a replacement, and hasn't had a problem since. Fortunately, it didn't take any other components with it, in his case.
What model was the PSU?
If I'm not mistaken, it was an RM1000.
Well, I have had the SF450 running for a few days now in a basic P4 system that I mentioned above and no issues so far. Of course I have to add the 4pin CPU modular power cable for the P4 system where that is absent in the Socket 370 board since that does not have a 4pin CPU port. That is one difference if that means anything. Wondering if this SF450 PSU is not a good fit for an old Socket 370 board is my train of thought. Will start adding the components from the other system one by one. I got to thinking that since the burning comes from around the CPU area (not an electronics expert like some of you) could it be the CPU fan I am using which is also another original component I have not switched out yet? It keeps the CPU cool just fine and I use IC Diamond thermal compound, but maybe an electrical problem? Bought that on Ebay new a few months back. I just will have to test, test and test some more. Also have some testing hardware on the way. So frustrating especially since the problem does not manifest itself for a week or two so a lot of waiting involved.
The CPU fan:
http://www.ebay.com/itm/StarTech-1U-60x10mm-S … 872.m2749.l2649
So after some more testing all the parts are working fine in the P4 system. I can't test the memory or CPU fan on the P4 system since they obviously wont work. I did find some scrapes on the hard drive 80 conductor IDE cable, but I am not sure if that would cause a short on the motherboard near the CPU. It could be the hard drive itself or maybe the DVD drive. Something is causing this within the Gateway 450 case that I have not already moved over to the P4 system. So now I will start building the Win98SE machine back together again, but very slowly and part by part.
I believe I have it narrowed down to the new Startech CPU fan (model FAN3701U) which was used with all three motherboards. On the D815EEA motherboard, only the CPU fan lead is shot and everything else about the board works fine. First for me where a CPU fan was obviously pulling too much. Going to try more of a brand name and go from there.
https://www.startech.com/Computer-Parts/Fans/ … ervers~FAN3701U
wrote:So another board burns again. Same system. This time it was an Intel D815EEA with a different CPU and of course no Slotket was i […]
So another board burns again. Same system. This time it was an Intel D815EEA with a different CPU and of course no Slotket was involved. So here are the components that were the same from that last builds that had the mobo burn up.
512MB PC-133 RAM (256x128x128)
3Com 3C905B-TXNM PCI NIC
Gainward FX5900 128MB AGP
USR 56K PCI Modem
Corsair SF450 PSUMotherboard was different. CPU was different. No Slotket adapter since this board is a Socket 370 board and not Slot 1.
I am running the graphics card on its own power cable but with the modular SATA power cable and then a SATA to D-connector adapter since the PSU comes with only one D-connector power cable which is being used for the drives. Could this be a problem? I don't think so, but maybe using a SATA power cable and converter on the GPU is causing power issues?
Weird thing is the machine ran fine under heavy load during benchmarks. Ran for a few days on and off. Tonight it was just sitting there turned on sitting at the Windows 98SE desktop doing absolutely nothing when the burning smell filled the room. The other two times this has happened with the other motherboards mentioned earlier in this thread, same thing. Just idling and not under load and delayed, no immediate problems. Every time it will run for awhile just fine and then one day, poof, burning smell and the system fails.
I am beginning to suspect the Corsair PSU, which is brand new, but before I jump to conclusions...
I have a very basic Pentium 4 system with an ASUS P4S8X-MX motherboard with 2.5GB of RAM. I am going to let this system run with that Corsair PSU for a few days and then slowly add each component from above one by one and see if I can burn the system up to find what the culprit is. I am down to trial and error as I have no idea what is going on. We shall see. Very frustrating to say the least.
what CPU were you using in the 815 board when it burned? Unless the Intel D815EEA has a U at the end of its model number it can not supply the lower regulated core voltage of the newer CPUs You will burn out the board for sure then.
If the truth hurts, tough shit.
wrote:what CPU were you using in the 815 board when it burned? Unless the Intel D815EEA has a U at the end of its model number it can not supply the lower regulated core voltage of the newer CPUs You will burn out the board for sure then.
The CPU still works fine on the board. The only thing that burned out was the 3pin fan lead that the CPU heatsink fan was plugged into.
This one:
wrote:wrote:So another board burns again. Same system. This time it was an Intel D815EEA with a different CPU and of course no Slotket was i […]
So another board burns again. Same system. This time it was an Intel D815EEA with a different CPU and of course no Slotket was involved. So here are the components that were the same from that last builds that had the mobo burn up.
512MB PC-133 RAM (256x128x128)
3Com 3C905B-TXNM PCI NIC
Gainward FX5900 128MB AGP
USR 56K PCI Modem
Corsair SF450 PSUMotherboard was different. CPU was different. No Slotket adapter since this board is a Socket 370 board and not Slot 1.
I am running the graphics card on its own power cable but with the modular SATA power cable and then a SATA to D-connector adapter since the PSU comes with only one D-connector power cable which is being used for the drives. Could this be a problem? I don't think so, but maybe using a SATA power cable and converter on the GPU is causing power issues?
Weird thing is the machine ran fine under heavy load during benchmarks. Ran for a few days on and off. Tonight it was just sitting there turned on sitting at the Windows 98SE desktop doing absolutely nothing when the burning smell filled the room. The other two times this has happened with the other motherboards mentioned earlier in this thread, same thing. Just idling and not under load and delayed, no immediate problems. Every time it will run for awhile just fine and then one day, poof, burning smell and the system fails.
I am beginning to suspect the Corsair PSU, which is brand new, but before I jump to conclusions...
I have a very basic Pentium 4 system with an ASUS P4S8X-MX motherboard with 2.5GB of RAM. I am going to let this system run with that Corsair PSU for a few days and then slowly add each component from above one by one and see if I can burn the system up to find what the culprit is. I am down to trial and error as I have no idea what is going on. We shall see. Very frustrating to say the least.
what CPU were you using in the 815 board when it burned? Unless the Intel D815EEA has a U at the end of its model number it can not supply the lower regulated core voltage of the newer CPUs You will burn out the board for sure then.
What BIOS version are you running on that board?
If the truth hurts, tough shit.
Latest version. P11.
While I wait for a new name brand CPU/heatsink unit to come in I have the Startech hooked up direct to the PSU bypassing the dead 3 pin lead on the mobo. Handling it fine and the system is experiencing no issues so far. All the research I have done shows no issues with the CPUs I have used on all three motherboards and all are within spec. What pisses me off is I suspected the CPU fan a while back after the first motherboard and I didn't act on instinct. At least the D815EEA works perfectly and performs under heavy load with benchmarks so if all that burned out is the CPU 3pin fan plug, I am good with that. I'll just bypass it. Not worth buying another mobo if that is the entirety of the damage. This also makes me want to go back now and see if those two BX boards are salvageable. I know one had damage to some traces, but the other only had localized damage which could be repaired. I have to say that in my 30 years of playing around with hardware this will be the first time a CPU fan has caused me this much trouble and I will not be buying anything Startech if it all pans out to be true.
Also remember this happened on two different BX boards and now this Socket 370 board all with the latest BIOS installed. I had my money on the Slotket, but the D815EEA disproved that. The Slotket BX setup had a P-III 850 100FSB and the Socket 370 config has a P-III 1000 133FSB. At this point I am confident it is not the CPU, but then again this problem doesn't rear its ugly head right away and takes a few days to a week of usage to literally explode in my face. Hopefully I can move on from this soon as it is really a frustrating problem.
On top of that, the power button on the Gateway bezel just snapped from plastic fatigue (this particular Gateway power button design is famous for this where it sinks into the case after the pivot snaps). So there is only one guy on Ebay selling those plastic power buttons and they want $9 a piece.
This Gateway project has turned into a Gateway of terrors just in time for Halloween. 😀
It's most likely the PSU, I fried an old BX board exactly this way. The fan could be the culprit only if it drew much more current then it's circuit was designed for. I suppose that a very clogged one could do it, if it's somehow slowed down by a lot of dust and gunk or if the grease on the shaft became too viscous over time.
I also thought PSU and I actually have a Corsair RMA setup already, but after more than a week of testing the PSU in two other systems with a slightly heavier load it holds up just fine in those systems. One is a P4 build and the other is an AM2 build. The SF450 is a pretty high end PSU with outstanding build quality so I would be very surprised if this were true. You're right though, in my experience when you start having weird power issues like this nine times out of ten it's the PSU acting up. Probably should RMA it for good measure.
Let me ask you guys this. Could a bad PSU act like this in such a localized fashion where it is burning out just the lead to the CPU fan? The rest of the system is untouched? It is why I didn't jump to the PSU right away.
You could also have two different incidents: 1. The slotket burns the transistor on the SLOT-1 mobos and 2. The fan burns the plug on the socket 370 mobo...
wrote:You could also have two different incidents: 1. The slotket burns the transistor on the SLOT-1 mobos and 2. The fan burns the plug on the socket 370 mobo...
🤣. So true. I am getting tired and my brain hurts. My other three retro PC projects went well without a hitch, but my Windows 98 project has been an omen. The power button breaking was the plastic that broke the camels back. I basically laughed when that happened.
I am sure some of you have had some unusual nightmare projects where nothing seems to be going right.
