Couple of dumb things I did to my retro machine.

I replaced the 486 board I had from 1994-2012 due to the BIOS chip breakdown with another 486 motherboard that was considered high-end at that time (PCI slots, integrated I/O, maximum of 128MB RAM, and Y2K compliant BIOS), and I wanted to switch to another video card since I didn't have any spare PCI or ISA cards readily handy, so I got a new ATI Rage video card that works on PCI slots, and it wouldn't POST (I switched to the other card I had installed and it booted with no issues).

The dumb thing I did was insert the PCI video card in one of the ISA slots, powered it on, and overloaded the motherboard. I also lost the Sound Blaster 16 WaveEffects CT4170 ISA card as well since it was still inserted in the ISA slot, and then, the unthinkable happened. The system wouldn't POST at all (due to the overload on the ISA slots).

I eventually got a Socket 7 motherboard as a replacement (with the CPU and RAM), and all was well, until I couldn't get any hard drive size bigger than 32GB (Intel i430TX limitation) to work properly as it states the HDD is 7.8GB in MS-DOS (NT SP4 and higher supported more storage). I then installed the beta BIOS on the system, which did make the HDD get detected as the full size (80GB), but when I would attempt to change something in the BIOS, it would lock up. Fortunately, I saved the BIOS on my diskette and re-flashed the BIOS to the correct one and all was well.

Lessons learned: never assume the ISA slot can be used to get a PCI card to work and always use the right BIOS.

Dropped a Geode NX 1750 into a KinetiZ 7T. This thing doesn't have multiplier or voltage controls. The CPU runs at a higher voltage as a result, but also the motherboard doesn't quite know what the heck it is, so runs at the lowest multi, 6x.

An Athlon XP-M based CPU, at a higher than rated voltage, at the lowest multiplier. Because why not?!

Ah, excellent topic idea. 😀 I do plenty of stupid things, should be able to contribute often.

Most recently:

Originally built my 486 PC in the first baby AT case I got, which was likely originally intended for an early Pentium machine as it had a space in the front of the case for a fan and came with a "188" LED MHz display. However, it was the only baby AT case I had at the time, so I used it for my 486 build. Later on, I obtained a couple more baby AT cases while attempting to get my "holy grail" case design (see here and here). Both of these extra cases were definitely intended for something pre-Pentium and most notably lacked a spot for an internal case fan.

Fast forward to about two weeks ago when I started to rebuild my Pentium 233 MMX PC. I figured that I had extra AT cases, so why not use them. However, definitely I would want to use the case that was currently being used for my 486, as it would probably be a good idea to make use of that spot for the internal case fan.

So, while disassembling my 486 to swap it into the other case, I was pulling out the power connectors and was doing what you absolutely shouldn't do and was yanking 'em out by pulling on the wires. Well, the floppy drive didn't want to relinquish it's power cable and so I just tugged harder, because I am quite bright you see. And just like that, all 4 of the wires came flying out of the connector with the actual connector itself still snugly in the floppy drive.

Aarrghh! 😵

Oh, the horror. The horror!

😅

I think similar topic already exists What was your worst screw-up with retro tech?

No. NO!

Nice to know, though, being honest.

Upgrading the RAM on a laptop that you didn't realize was still on, just had the screen off.
Somehow the old and new RAM modules survived as did the laptop.

I had a similar scare some years ago. I moved my main computer into another room for cleaning/testing, and when I powered it on the RAID arrays had disappeared. After several minutes of mounting panic I found the problem: there's a small CLEAR CMOS button at the back of the machine which I accidentally pressed while carrying the case. This reset BIOS values to defaults, and set the integrated SATA controller to IDE mode instead of RAID mode, making the RAID arrays inaccessible. That button is a dumb idea. Things like that should be in the interior of the case.

bjwil1991 wrote:

Lessons learned: never assume the ISA slot can be used to get a PCI card to work

🤣

Not recently but allready a while ago I did the following stupid things:

First fried a Compaq SMART 2E Raid controller by not fully inserting it to the EISA slot of my testbench. Just after the first one did not work I did the same with a second one. Both show some weird error messages now.

Second was some hot swap BIOS flashing. First a 586er board where I replaced the BIOS by a different one while powered, due to lack of an external EPROM flasher, second on a Mylex DAC 960P RAID controller to upgrade fom a V.2.x BIOS residing in a single PLCC chip to a V.3.x release residing in dual PLCC chips, to get the Linux support the V.2.x BIOS lacks. Both swappings went smooth, but I still consider it as a stupid idea.

And the usual stuff of course, plugging 3,5´floppy drives with the power plug one pin off, reverse plugging of Molex HDD power plugs, running a Duron CPU with the heatsink removed, hotplugging PCI and ISA cards on my testbench.

eisapc wrote:

And the usual stuff of course, plugging 3,5´floppy drives with the power plug one pin off, reverse plugging of Molex HDD power plugs, running a Duron CPU with the heatsink removed, hotplugging PCI and ISA cards on my testbench.

Oh yeah, I made that mistake once on my Athlon 64 rig when I was swapping its case, needless to say I lost a 500W NIFT power supply and a Teac floppy drive when that happened. 😵

My most recent screw-up (that I can recall) happened about a month ago, when I was moving my Pentium 1 rig, I broke part of its front panel by trying to carry it by that spot. It wasn't too major to be noticeable, the piece that broke was a little bar that was placed above the 5.25" bays, should be fixable through the means of drilling a new set of screw holes on the main part of the front panel.

This reminds me, I've now fried two computers by connecting them to my home Hi-Fi system. There is a strange earthing problem somewhere I haven't managed to debug. If I pull an audio cable from the amplifier to the audio jack of a computer, a very large current flows through the earth connection. Painful electric shocks when I touch the computer case. Obviously, this kills the motherboard. I've now done this twice over several years. The first time can be written off as a freak accident, but I then forgot about the problem and did it again last year, which now enters the realm of stupidity.

I have no idea what's wrong with the Hi-Fi system. It works perfectly well. There's even (another) computer permanently connected to it which doesn't have this problem. It only seems to happen to computers connected to a different power outlet. Why would there be a potential difference between the earth connections of two power outlets that are just a few feet apart?

Not recent, but probably 1 of the stupidest things I did -

I have 2 identical ePoS Tualatin Celeron systems and I was setting up 1 as a retro gaming system, whilst the other had a damaged case.

I was experimenting with different Sound & Graphics cards when all of a sudden the machine began refusing to POST. I could get nothing but fans spinning and no amount of removing/reseating cards would make a difference. Figuring that the motherboard was toast, I decided to simply swap over the internals of the other case and rebuild with that setup.

I left the "broken" one in storage for several months. Then 1 day I was playing around with the working one and the same POST failure occurred, only this time I spotted immediately what was wrong - the RAM socket was at the bottom and the RAM chip was a new upgrade and ever so slightly bulkier. The case slides on and off and if slid back at a slightly wrong angle it would dislodge the RAM chip and cause it to fail to POST. Sure enough, I tested the "broken" one to find it working perfectly. 😲 Bloody stupid design!!!! But thankfully a now working PC once again.

A more recent stupid thing would be snapping 1 of the legs off a BIOS chip on my XT, by cleaning too vigorously. Not a major issue as I had the BIOS image on my PC so could simply program a new replacement EEPROM. Still a stupid thing to do though.

I finally had everything I wanted to build my retro PC. The Monitor, Case, Keyboard, Mouse, and Motherboard with components.

The motherboard was the most expensive part, costing around $140 in total because it was an unusual ASRock 939 Dual, which notably has a PCI Express slot and AGP slot, both full speed.

I spend a couple of days building out the PC, connecting up everything in the case including things like the Floppy and CD ROM drive.

Everything boots up fine, and I get into Windows XP. Everything is going great. I then go to plug in the LAN cable at the back, and d'oh - the I/O backplate had a grounding-fin that had folded down and blocked the LAN port instead!

So I'm like, OK, lets do this properly instead of rushing. I shutdown the computer, and unscrew the motherboard. I slide it out the way and POOF suddenly the machine powers on and there is an electrical burning smell! something had shorted out and I realise I had completely forgotten to switch off the PSU! I kill the power.

I carefully fix everything and reassemble it. I power on the machine and it doesn't post. The motherboard is dead. Nothing I try can revive it, BIOS reset, removing the battery, leaving it unplugged, unplugging everything from the board. It is completely dead.

Because of a stupid I/O back panel, I destroyed my retro gaming PC before I even had a chance to enjoy it.

Errius wrote:

Why would there be a potential difference between the earth connections of two power outlets that are just a few feet apart?

There shouldn't be . IMHO, this justifies contacting an electrician ASAP, before serious bodily harm or property damage occurs .

So I found out my new-to-me Amiga 500 has some extensive trace damage on the keyboard membrane, more than I expected, and I set about fixing it by scraping off the top green plastic layer to bridge the broken traces underneath.

...wait, that's also doing a good job of scraping out the trace underneath, too. Dammit!

Anyone got a source on these keyboard membranes that isn't sold out? If not, I may have to go the USB keyboard controller route.

Plugging a DoM on ide cable through a ide male/male adapter.... Burned out PSU, DoM and use cable as a result. Fortunately didn't put in the second DoM in there yet back then, was lucky.

I'm curious, Richo: What is a DoM? I have never heard of them before.

the_ultra_code wrote:

I'm curious, Richo: What is a DoM? I have never heard of them before.

Disk on Module. Imagine it as a very early form of SSD. They are IDE flash storage devices primarily used in industrial applications. They are still being made and sold today.