VOGONS


Reply 40 of 76, by KCompRoom2000

User metadata
Rank Oldbie
Rank
Oldbie

Although it's not my "gaming machine", here's something mildly stupid that I've done recently:

Yesterday, I received a Core 2 Duo E6600 CPU in the mail, so I installed it into my WIP Dell Optiplex 745 re-build, and applied a fresh thing of Arctic Silver 5 thermal paste.

Today, I've decided to investigate the inside of the machine to figure out where the thermal sensor is supposed to go, and here's the (admittedly mildly) stupid part: Being under the impression that the thermal sensor connector could be underneath the CPU, I lifted up the heatsink and separated the newly-applied coat of thermal paste, making it not so new anymore!

Does this sound like a bad thing to you? I did let the machine power up for a few minutes after I put the heatsink back on, so hopefully that re-established the thermal paste's contact with the heatsink and the CPU. After all, I've done things like this before and I'm pretty sure everything was back to normal once I put the heatsink back on. Just how time-sensitive is thermal paste? 😕

Reply 41 of 76, by Malvineous

User metadata
Rank Oldbie
Rank
Oldbie

Thermal paste is actually a relatively poor heat conductor, it's just better than air, so the idea is to apply as little as possible to just fill up the air gaps between the CPU and heatsink metal-to-metal contact. If you put too much on (usually more than a rice-grain sized amount) then it can get in the way and become less effective overall. Thermal paste tends to dry out over time, so after a few years it probably needs to be reapplied anyway. But while it's relatively new (say within the first year of application), it doesn't matter how many times you lift off the heatsink or not, because remember you're only filling the microscopic gaps between the two surfaces and you want as much metal-on-metal contact as possible. You might eventually push most of it off the edges with repeated movement but you can just scoop it back and spread it out to a thin layer again. Generally speaking as long as your CPU temperatures are reasonable under load then the heatsink and thermal paste are doing their jobs just fine. The paste doesn't "melt" so leaving the machine on for a while won't change anything, but it can be a good way to keep an eye on the CPU temps to make sure they're not suddenly skyrocketing because something has gone wrong.

So really there was nothing stupid about this (unless you really did use a whole tube of thermal paste on one application instead of the 50+ uses that you can get out of one tiny tube!)

Reply 42 of 76, by gdjacobs

User metadata
Rank l33t++
Rank
l33t++
Malvineous wrote:

Thermal paste is actually a relatively poor heat conductor, it's just better than air, so the idea is to apply as little as possible to just fill up the air gaps between the CPU and heatsink metal-to-metal contact. If you put too much on (usually more than a rice-grain sized amount) then it can get in the way and become less effective overall.

If the thermal paste acts like it should, it'll just squeeze out between the two surfaces (possibly making a mess).

Malvineous wrote:

Thermal paste tends to dry out over time, so after a few years it probably needs to be reapplied anyway. But while it's relatively new (say within the first year of application), it doesn't matter how many times you lift off the heatsink or not, because remember you're only filling the microscopic gaps between the two surfaces and you want as much metal-on-metal contact as possible.

Age doesn't matter if the interface isn't disturbed. The compound has already moved into any gaps. If the heatsink is moved, I recommend removing and re-applying the compound for consistent results. The cost of paste is just so low, I don't see a reason not to.

All hail the Great Capacitor Brand Finder

Reply 43 of 76, by NamelessPlayer

User metadata
Rank Member
Rank
Member

Let's just say I got a nasty lesson in what SCSI termination power from a hard drive with an IDC-50 cable plugged the wrong way around will do when testing out my Amiga 2000 for the first time.

Since the A2091 SCSI controller board isn't keyed (which is downright STUPID, since I misread the PCB initially), there's nothing stopping you from plugging it the wrong way around (hint: the marked pin on the ribbon faces toward the BOTTOM of the case, not up!), and the drive that was included (some kind of thick 4 GB Seagate Barracuda) had a crapton of termination and power-related options on the jumper headers. More than I knew what to do with, really.

Cue the sudden onset of smoke and a nasty burning smell that hasn't gone away as one of the middle conductors of the cable just straight-up melted its insulation... and I thought IDE/Parallel ATA was a pain in the ass!

Reply 45 of 76, by bjwil1991

User metadata
Rank l33t
Rank
l33t

An IDC-50 cable is a 50-pin SCSI cable that was commonly used on older SCSI Hard drives, CD-ROM drives, tape drives, you name it. Those cables also required a termination adapter on at least 1 header to make sure the drive can be reached since SCSI back then relied on termination adapters to make the drives detectable, and without it, the drives won't work or get detected. IDE drives are easier to work with, but SCSI drives power on once the controller turns on. Here's a cable:

51um9Ie2z2L._SX342_.jpg

Discord: https://discord.gg/U5dJw7x
Systems from the Compaq Portable 1 to Ryzen 9 5950X
Twitch: https://twitch.tv/retropcuser

Reply 47 of 76, by bjwil1991

User metadata
Rank l33t
Rank
l33t

Back then, there was no keyed I/O ports (serial, parallel, IDE, SCSI, FDD, etc) that had the plastic housing around the pins themselves to avoid confusion of how to install the cable. My Packard Bell Pack-Mate 28 Plus luckily has 1 pin missing on the IDE and FDD ports (common for OEM systems to do that) so that way I can replace an IDE cable with a modern one that has the one pin (unused) that's covered for the correct orientation. I did hook up an HDD and ODD incorrectly once, but nothing bad happened, and I was quick at the power button and power strip.

Discord: https://discord.gg/U5dJw7x
Systems from the Compaq Portable 1 to Ryzen 9 5950X
Twitch: https://twitch.tv/retropcuser

Reply 49 of 76, by Qjimbo

User metadata
Rank Member
Rank
Member
KCompRoom2000 wrote:

Yesterday, I received a Core 2 Duo E6600 CPU in the mail, so I installed it into my WIP Dell Optiplex 745 re-build, and applied a fresh thing of Arctic Silver 5 thermal paste.
...
Just how time-sensitive is thermal paste? 😕

Since you used a good brand, this isn't a problem. Good quality thermal paste never "dries", so as long as it was smooshed back together properly it shouldn't be an issue. On cheapy OEM computers in the 90s, they'd slather on very cheap thermal compound that would dry out like toothpaste, but the good quality stuff doesn't do this.

Follow me on twitter

Reply 50 of 76, by bjwil1991

User metadata
Rank l33t
Rank
l33t

I replaced the thermal compound on my HP Pavilion N3350 that had the green compound with the Arctic Silver 5 Compound (I used it on other CPUs, and on GPUs as well), and it does wonders. The CPUs still run at 40C (104F) and never went up at all. The only time the CPU heat rises is when there's dust inside the heatsink (I clean those things out and I deep clean the fans as well for longer life, and the CPU runs at 40C once again). I have a VIA C3 700A that runs at 27C since there aren't a lot of instruction sets and it's cheap.

Discord: https://discord.gg/U5dJw7x
Systems from the Compaq Portable 1 to Ryzen 9 5950X
Twitch: https://twitch.tv/retropcuser

Reply 51 of 76, by gdjacobs

User metadata
Rank l33t++
Rank
l33t++
Qjimbo wrote:
KCompRoom2000 wrote:

Yesterday, I received a Core 2 Duo E6600 CPU in the mail, so I installed it into my WIP Dell Optiplex 745 re-build, and applied a fresh thing of Arctic Silver 5 thermal paste.
...
Just how time-sensitive is thermal paste? 😕

Since you used a good brand, this isn't a problem. Good quality thermal paste never "dries", so as long as it was smooshed back together properly it shouldn't be an issue. On cheapy OEM computers in the 90s, they'd slather on very cheap thermal compound that would dry out like toothpaste, but the good quality stuff doesn't do this.

It's also not an issue if the interface isn't disturbed. Thermal paste is a fluid so it can flow into any gaps between the CPU and heatsink. Once those gaps are filled, it doesn't need to flow until it's adjusted and new gaps need filling.

All hail the Great Capacitor Brand Finder

Reply 52 of 76, by Qjimbo

User metadata
Rank Member
Rank
Member
gdjacobs wrote:
Qjimbo wrote:
KCompRoom2000 wrote:

Yesterday, I received a Core 2 Duo E6600 CPU in the mail, so I installed it into my WIP Dell Optiplex 745 re-build, and applied a fresh thing of Arctic Silver 5 thermal paste.
...
Just how time-sensitive is thermal paste? 😕

Since you used a good brand, this isn't a problem. Good quality thermal paste never "dries", so as long as it was smooshed back together properly it shouldn't be an issue. On cheapy OEM computers in the 90s, they'd slather on very cheap thermal compound that would dry out like toothpaste, but the good quality stuff doesn't do this.

It's also not an issue if the interface isn't disturbed. Thermal paste is a fluid so it can flow into any gaps between the CPU and heatsink. Once those gaps are filled, it doesn't need to flow until it's adjusted and new gaps need filling.

I trimmed gdajacobs quote to highlight the important bits, he did pull it apart and put it back together, but I still don't see it being an issue.

Follow me on twitter

Reply 53 of 76, by KCompRoom2000

User metadata
Rank Oldbie
Rank
Oldbie

Thank you all for confirming that I didn't do anything worrisome by removing the heatsink shortly after applying thermal paste (I implied that it was "mildly stupid" because this was nowhere near as stupid as ~90% of the stunts mentioned in this thread). This was one of those times where I mentioned something just for the sake of mentioning it.

Now for another stupid thing: I installed monitor drivers from a floppy disk on a Windows 98 computer, I removed the floppy right after setting the drivers up in the "Add New Hardware" wizard during startup, when I right-click the desktop to adjust the resolution, the floppy drive make a "jackhammer" noise and a "The specified path or file no longer exists" error.

Luckily, I managed to fix it by putting the disk back in and restarting the computer after configuring the screen resolution.

Reply 54 of 76, by RichB93

User metadata
Rank Member
Rank
Member

Killed two 40GB 2.5" IDE drives; one which was full of data and in use for my old DVD burning laptop (A Dell D610). Stupid drives had the wrong keying and I plugged the 2.5 to 3.5" IDE adapter the wrong way. On BOTH drives! Can't even see whats wrong on the drive boards.

Doh! 🙁

Reply 56 of 76, by britain4

User metadata
Rank Member
Rank
Member

Broke off the pins on a slocket AND the CPU to get a Tualatin working on a 440BX board... but I read the diagram as being the top of the socket not the bottom of the CPU and broke the wrong ones off :@ I thought it wasn’t working because the pins were still making contact!

So I lost 1 CPU and a slocket but I have got it up and running now!

- P-MMX 200MHZ, PCChips M598LMR, Voodoo
- P-MMX 233MHz, FIC PA2013, S3 ViRGE + Voodoo
- PII 400MHz, MSI MS6119, ATI Rage Pro Turbo + Voodoo2 SLI
- PIII 1400MHz, ECS P6IPAT, Voodoo5 5500
- Toshiba Libretto 110CT, 300MHz, 96MB RAM

Reply 57 of 76, by nekurahoka

User metadata
Rank Member
Rank
Member

So, I was working on updating microcode to an Abit Is7-V2, pulled the flash chip to flash it using my USB flasher and broke the retaining socket on the motherboard. A whole corner came free and now when the chip is inserted it pushes the leads out far enough that there is no contact. Gonna have to rig something together or just wait until get the next mobo I've been eyeing, which already has the latest microcode.

Dell Dimension XPS R400, 512MB SDRAM, Voodoo3 2000 AGP, Turtle Beach Montego, ESS Audiodrive 1869f ISA, Dreamblaster Synth S1
Dell GH192, P4 3.4 (Northwood), 4GB Dual Channel DDR, ATI Radeon x1650PRO 512MB, Audigy 2ZS, Alacritech 2000 Network Accelerator

Reply 58 of 76, by martin_boro

User metadata
Rank Newbie
Rank
Newbie

Obligatory "this is not that recent, but it happened, I swear!"

So, a couple of years back, I found a MSI K7D Master -L motherboard for sale for a decent price, including a pair of decent Arctic coolers and most importantly, a pair of Athlon MP 2800+, with Barton core, running at 1.6V of VCore (this will be important later).
Naturally, I bought it and it worked well for a few months. Then I found a second dual-socket A board, the Tyan Tiger MP (S2460). I wanted to test it, so I took out the 2800+ from the MSI board, and put them into the Tyan board. Didn't POST, no big deal, wasn't that expensive anyways.
Now I am getting to the good part. I put things back together, however, now the MSI board didn't POST either. I reset the BIOS and it POSTed successfully. For some reason though, both CPUs released their magic smoke and died a "fiery" death. And to add insult to injury, so did three more CPUs I used to diagnose it.
So, I just put it away and forgot about it, until the last weekend, when I decided to give it a last chance before throwing it out. I put in a pair of older Athlon MP 2000+ and IT WORKED!!. I looked into the BIOS HW monitoring, just to see, that the current VCore is 1.9V and idle temp is 72C with a good cooler. I found the voltage settings and set it to 1.75V as it should be for these CPUs and it worked flawlessly and it even passed a two hour stress test.

So, what must have happened, is that the previous owner set the voltage manually for the 2800+ and when I reset the BIOS, it reverted to the default 1.9V, which killed both CPUs, which were designed to run at 1.6V. The second pair I used, was from an older generation and was designed to run at 1.75V, which allowed it to survive at 1.9V.
I updated the BIOS to the latest version (which was from 2012 for some reason) and it seems to be working fine now. It doesn't bring back the dead CPUs though 😢 .

Reply 59 of 76, by looking4awayout

User metadata
Rank Member
Rank
Member

A stupid thing I did many years ago was killing a working AT super socket 7 motherboard by putting both SDRAM and EDO modules with the hope to go beyond 256MB of RAM. After doing that the motherboard would no longer go past the Verifying DMI Pool Data screen. I weep about that moment still today... 😒

My Retro Daily Driver: Pentium !!!-S 1.7GHz | 3GB PC166 ECC SDRAM | Geforce 6800 Ultra 256MB | 128GB Lite-On SSD + 500GB WD Blue SSD | ESS Allegro PCI | Windows XP Professional SP3