c/computer-technicians

Took me 6 hours to find a bad RAM stick that passed every diagnostic tool I had

I was chasing random blue screens on a Dell OptiPlex for days, swapped the PSU and motherboard before I finally tried pulling each stick one by one, and it turned out to be a single faulty module that Memtest86 and Windows Memory Diagnostic both said was fine, has anyone else had RAM that passed every test but still caused crashes?

Warning: I spent 3 months using thermal paste from a repackaged tube and ruined a build

I used to buy those cheap multi-pack thermal paste tubes on Amazon since they were like $8 for 5. Last week I pulled apart a customer's gaming PC that kept overheating and found one of those tubes had separated into liquid and chunks. Turns out the stuff was repackaged from some bulk factory and had no quality control. I tested two builds side by side with Arctic MX-6 vs the cheap stuff and saw a 12 degree difference under load. The cheap paste literally dried out after 6 months. Has anyone else run into fake or expired thermal paste from those multi-pack deals?

Blew $180 on a logic analyzer I barely touch now

I bought a Saleae clone off Amazon thinking I'd use it all the time for debugging serial protocols, but it's just been sitting in my drawer for six months collecting dust. The software was clunky and I ended up just using a multimeter and oscilloscope for 90% of what I needed. Anyone else have a tool they convinced themselves they needed but never actually use on the job?

Had a client tell me my diagnostic fee was 'too steep' so I broke down my thought process for them

Crimping Ethernet cables wrong has been costing me hours of rework this month

I kept getting intermittent dropouts on a 200 foot run and finally traced it to my RJ45 crimps not fully seating the wire pairs, so double check your pass-through connectors before you staple everything down, anyone else had this bite them after the fact?

A client's blunt feedback made me ditch thermal paste spreading for the pea method

I used to spread thermal paste manually with a card, thinking it gave better coverage. Then a guy named Mike in Denver told me my Ryzen 5900X was running 5 degrees hotter than his with the same cooler. He showed me his temps after using a pea-sized drop in the center, and I tested it on my own rig - dropped from 82C to 76C under load. Anyone else have a customer call you out on something that actually improved your work?

Maybe unpopular but I refuse to use thermal paste spreaders

I keep seeing people online say you have to spread thermal paste with a little plastic card or tool to get even coverage. After 12 years doing PC builds and repairs, I still just use the pea method. A dot in the middle, press the cooler down, let it spread naturally. I tried the spreader method twice and actually got worse temps both times. One build for a customer in Portland had a 5 degree difference favoring the pea drop. Maybe it matters for liquid metal or super high end chips, but for 95% of builds I think people are overthinking it. Has anyone else tested both ways and found the spreader actually helps?

Caught a bad PSU before it fried a client's RAID array

I had a server in for a routine clean and noticed the 12v rail was sitting at 11.4 under load. The guy said he'd been getting random read errors for about 2 weeks. Anyone else seen a failing power supply cause phantom drive issues like that?

Found an old CRT monitor at a thrift store that actually worked

I stopped by a Goodwill out in the suburbs last weekend and saw a 17-inch Dell CRT sitting on the shelf for 15 bucks. Thought it would be dead or have some crazy flyback whine, but I plugged it in and it fired right up with no issues. Ran a quick geometry test and the colors were surprisingly solid, no burn-in even. I ended up buying it just to mess around with retro gaming stuff, mostly because I've been curious about low-latency setups. Has anyone else had luck finding old hardware like that in random places?

Client's laptop died mid-warranty call, had to fake a solution

Got a call last week from a guy whose screen went black after a Windows update. I was walking him through safe mode when his battery died and the call dropped. He called back 10 minutes later saying I 'fixed it' because the reboot worked. I told him sure, that was the plan. Ever had a problem just vanish and you take the credit like you meant to do that?

Blew a power supply last Tuesday - went with a modular unit over a non-modular one, curious what you all prefer

I was swapping out a dead PSU on a client's gaming rig and had to decide between a modular unit and a non-modular one. Went with the modular because cable management in a cramped mid-tower is a nightmare, but the non-modular was $30 cheaper. Ended up spending an extra 15 minutes routing cables anyway since the modular one had stiff wires. Do you guys pay extra for modular or just stuff the extra cables in the drive bay?

Finally switched thermal paste after 6 years and the temp difference was wild

I had this old gaming rig I built back in 2018, never touched the CPU cooler. Kept seeing temps hit 85-90C under load, figured it was just the case airflow. Last week I finally peeled off the cooler and that paste was basically dust, hard as a rock. Threw on some new thermal grizzly stuff for $12, did a repaste, and now it sits at 68C max after an hour of gaming. Kinda mad at myself for ignoring it so long. What's the longest you guys have gone before changing yours?

A client's simple question made me rethink how I do data backups

I've always done full image backups for every machine I touch, takes forever but I figured it's the safest. Then last week a guy asked me why I was backing up his temp files and browser cache every time, stuff that doesn't matter. He pointed out I was wasting like 2 hours per backup on junk data. Now I just do selective file backups for the important stuff and it takes 30 minutes tops. Anyone else change their backup strategy after someone called them out on it?

Spent 4 hours chasing a ghost drive that wasn't even plugged in

I was working on a client's Dell Optiplex last Wednesday that kept throwing a 'disk boot failure' error. Checked the SATA cables, swapped the drive with a spare, even updated the BIOS. Still got the same error every time. Turns out the previous tech left an old IDE ribbon cable loose inside the case and the board was prioritizing that ghost slot over the actual SSD. I felt like an absolute clown when I finally spotted it 4.5 hours in. Has anyone else had a build take forever because of some stupid leftover hardware from a past repair?

That one time a customer showed me the 'pry from the top' trick on laptops

I used to always crack open laptops from the side seam with a spudger, but I wrecked a hinge clip on a Dell Latitude 5400 last month. Customer saw me struggling and told me to start prying from the top edge near the screen hinge instead. Turns out most laptop lids have a stronger catch up there and it pops off way cleaner. I've done 8 repairs since then using that method and zero broken clips. Anyone else have a quick trick like that from a random client?

Decided between a used precision workstation and building my own for a homelab

Had to pick between a used Dell Precision T5810 for $350 or piecing together something from scratch with new parts. Went with the Dell because it came with 64GB of ECC RAM already installed. Been running it for 3 months now as a Proxmox host and it handles everything I throw at it without a hiccup. Anyone else grab a retired workstation and find it way more stable than expected?

Finally caved and used a label maker for my server rack cables

I spent 4 hours tracing a single Cat6 line last Tuesday because I thought I could remember which port went where. After that I bought a $30 label maker and labeled everything in under 45 minutes. Anyone else stubborn about this until they wasted an afternoon?

Walked into a client's home office last Tuesday and saw a PC from 2004 still running their billing software

They upgraded to a refurbished Dell from 2019 this week and the owner said their backup time went from 45 minutes to under 3, which got me wondering how many other small businesses are running on ancient hardware just because "it still works"?

Bought a cheap PSU tester on Amazon and it lied to me

Picked up one of those $15 PSU testers from Amazon to speed up diagnostics. Plugged in a dead power supply and it showed all green lights so I almost sent the customer home with a bad unit. Lost about 2 hours chasing down the real issue before I tested with a multimeter. Anyone else ever get tricked by a cheap tester tool?

Why do so many techs skip checking the power supply first?

I've been doing house calls in Phoenix for about 2 years now, and I'd say 8 out of 10 machines brought in for 'not turning on' just have a dead or failing power supply. The other day a guy spent 45 minutes reseating RAM and swapping cables before I checked his PSU with a tester and it was dead on the 12V rail. Am I crazy for thinking a quick PSU test should be step one every time?

Question about that weird BIOS password lock I found on a Dell OptiPlex

I was wiping a used Dell OptiPlex 3020 for resale last week and found a BIOS password that the previous owner never disabled. After some digging (thanks to a forum post from 2019), I learned Dell actually has a master password system based on the service tag and a specific code generation algorithm. Did you guys know that these passwords are basically a math problem you can solve if you have the right tool? Has anyone else run into this messing with old office PCs?

Had a chat with a younger tech last week that made me rethink how I handle drive failures

He told me he never bothers with data recovery software anymore because customers expect SSDs to just work, and it made me wonder if I've been holding onto old habits out of nostalgia instead of efficiency - anyone else adjusting their workflows for the SSD generation?

Fried a power supply because I ignored a dusty filter for way too long

Last Tuesday I finally built that new rig I'd been piecing together for 3 months. Everything was going great until about 20 minutes into a game, the whole system just shut down. No warning, no BSOD, just dead silence. I spent 2 hours swapping parts and testing with a multimeter before I pulled the PSU and realized the intake fan was caked in a solid layer of dust from sitting in my garage. The thermal protection kicked in and it cooked itself trying to cool down. Ended up having to order a replacement from Newegg for $110 and lost a whole evening of troubleshooting. Anyone else ever wreck a component just from skipping basic maintenance like cleaning filters?

Had a customer chew me out for using compressed air on his laptop fans

He said I was just pushing dust deeper into the bearings, and after looking up a teardown video for his model, yeah he was right. I switched to a soft brush and vacuum for laptops now, has anyone else run into this issue with canned air?

Hot take: I still prefer SATA SSDs over NVMe for most home builds

I know everyone is all about NVMe drives now, but after building 12 machines last year for friends in Austin, I stuck with SATA SSDs like the Samsung 870 Evo. Boot times are maybe 2 seconds slower and for gaming or everyday use, nobody notices the difference. Plus I saved about $30 per build and avoided thermal throttling issues I saw on two NVMe setups. Anyone else still using SATA for basic builds?