c/computer-technicians

Spent 4 hours chasing a ghost IP conflict that turned out to be a fax machine

Client in Boise had two devices on the same static IP, but the culprit was an old fax machine plugged into a surge protector that was off, and its NIC still broadcasted, so I finally isolated it by unplugging everything and adding them back one by one, has anyone else had a device stay active with power switched off?

0 comments

ross.jessica•6d ago

PSA: Static discharge from a nylon case wrecked my whole bench setup

Left a laptop on a nylon sleeve on a dry winter day and the zap fried the external monitor port, not the laptop, which I only found after swapping three cables. Anyone else ever tracked damage to a bag or case instead of the board itself?

0 comments

skylershah•11d ago

Found out my $9 thermal paste tube was older than my apprentice

Checked the batch code on the tube from the back of my drawer and it expired in 2016, so I redid a CPU cooler install from Tuesday that was already running 10C too hot, anyone else sit on old paste this long?

0 comments

the_viola•11d ago

Benchmarking old vs new: the 10 year old drive beat my brand new one in a real boot test

I spent last Friday swapping drives in my own rig to see if the hype was real. My 2013 Samsung SSD booted Windows in 22 seconds flat. The new NVMe I got for $80 took 19 seconds, so yeah it won, but not by the margin everyone claims. But the real shocker was file transfers, the old SATA drive choked at 450 MB/s while the NVMe hit 2.1 GB/s. So if you're only doing light office work, that cheap upgrade might not be worth the hassle. Has anyone else run a comparison like this and found the older gear holding up way better than expected?

0 comments

miles_roberts22•14d ago

thermal paste spread patterns finally make sense to me

I've been building PCs for like 15 years and always just did the pea method without thinking about it. Then this week I had a Ryzen 7 running hot at 92c under load, so I pulled the cooler and the paste had spread unevenly toward one edge. Turns out my cooler base was slightly convex, which is apparently pretty common on cheaper towers. I re-applied with a cross pattern and checked the spread after seating it once, then did a final mount. Temps dropped to 78c after about 3 tries. Anyone else ever diagnose a bad mount just by looking at the paste pattern before touching temps? I'm curious if I've been leaving performance on the table for years.

0 comments

miles_jackson9•16d ago

Question about cleaning thermal paste off a customer's CPU

Had a client bring in a rig last Tuesday where someone used WD-40 to clean old paste off an i7. It left a film that made the new paste slide around and temps hit 95C under load. I use 99% isopropyl and a coffee filter now, but has anyone else seen weird DIY solvents ruin a socket?

1 comments

faithrodriguez•18d ago

Dust bunny autopsy: how I finally found the real cause of those random restarts

I had a machine at a shop in Austin that kept dropping out during stress tests, no BSOD, just a hard reset every 15 minutes. I checked temps, swapped RAM, even reflashed the BIOS, zero luck. On a whim I pulled the PSU and opened it up, the intake fan was caked with a solid layer of lint that looked like felt. After I blew it out with canned air and cleaned the fins, I ran the same test for an hour with no drops. My question is, how often do you guys actually open up the PSU for cleaning, or do you just replace them?

1 comments

joel280•20d ago

The day I stopped trusting my gut on RAM timings

I've been building PCs for about 8 years now, always just picked whatever RAM speed the motherboard box advertised and called it done. Last month I threw a 5600G build together for a buddy in Portland and it kept crashing during gaming, no BSOD, just freezes. I spent three days swapping parts, reinstalling Windows, the whole nine yards, before I actually checked the memory QVL list on the board's website. Turns out the kit I used wasn't listed, and the XMP profile was pushing timings the CPU's memory controller just couldn't handle. Set it to a slower JEDEC profile with manual timings from the QVL and it's been rock solid for 2 weeks now. Anyone else have a similar moment where you realized the QVL list was actually worth reading, or do you still eyeball it?

0 comments

nancy3•20d ago

Client called my cable labeling lazy, now I use a label maker with heat shrink

Last month a client in Toledo pointed out my zip tie labels were peeling after 3 weeks. I switched to heat shrink tubing labels on every Ethernet drop and power run, costs about $40 for a starter kit. Has anyone else had a client nitpick something that actually made their work cleaner?

1 comments

jake_mason30•21d ago

Diagnostics first or part swap first? My last 3 calls made me pick a side

So I've been going back and forth on this for a while. When a machine comes in with a dead mobo or no POST, do you grab your multimeter and start tracing voltages, or do you just swap the likely part and see what happens? Last Tuesday in Charlotte I spent 40 minutes testing a PSU with a $25 tester only to find it was fine, then the board was toast. But the week before, I swapped a CMOS battery on a whim and skipped all testing, fixed it in 5 minutes. Time is money on the road, but bad guesses burn customer trust too. Has anyone figured out a rule for when to stop testing and just throw parts at it?

1 comments

jamesblack•25d ago

A client's 'simple fix' for a boot loop turned out to be a 45 minute nightmare

Tbh, a customer in Denver swore their PC just needed a CMOS reset because they spilled coffee on it last week. I popped the battery out, waited, put it back, and the loop got worse, plus the BIOS wouldn't even post. Turns out their spill had corroded two pins on the SATA header, so I had to swap drives and rebuild the boot config from scratch. Next time I'm checking the board before I trust the story. Has anyone else had a client's diagnosis waste more time than the actual repair?

2 comments

the_sage•26d ago

Overheard a kid call a CRT monitor a 'vintage TV' and felt ancient

I was fixing a Dell OptiPlex at a thrift store in Portland last Tuesday, and a teenager pointed at the old CRT I had on the bench and asked if it was for VHS tapes. It hit me that these kids never had to degauss a screen or carry one of those 40-pound monsters up two flights of stairs for a LAN party. Funny part is, that monitor still works fine for retro gaming, but I mostly keep it around because the refresh rate on it was a beast for its time. Anyone else still hold onto old gear just for the memories, or is it just me?

1 comments

the_patricia•26d ago

Fought a BIOS password lockout for 3 days, then found a jumper pin I ignored

I picked up an old Dell OptiPlex from a surplus sale in Phoenix last week, and it had a BIOS password the seller didn't know. I spent 2 days trying default codes and CMOS battery pulls, nothing worked. On day 3 I noticed a tiny jumper labeled PSWD near the RAM slots that the service manual mentioned in a footnote. Moved it to pins 2-3, booted, and the password was gone. Has anyone else wasted time on a fix that was literally printed on the motherboard all along?

1 comments

the_shane•27d ago

A dentist's comment made me rethink how I explain fixes to clients

Last week in Columbus, my dentist said 'you don't need to know the drill model, you need to know why the filling works.' It hit me because I've been rattling off specs and part names to customers for years, but they just want the 'why.' Now I lead with the simple cause and effect, then offer the deep dive only if they ask. Anyone else simplify their talk track after a random convo like this?

1 comments

the_shane•27d ago

Hit 1,000 bench tests logged yesterday, here is what surprised me

I keep a spreadsheet of every motherboard bench test I run, mostly for warranty callbacks. Yesterday I hit the 1,000 mark and went back through the data. What surprised me is that 212 of those were dead RAM slots, not dead boards. I figured most failures would be CPU or power issues, but the socket pins and DIMM traces were the real killers. Has anyone else found a weird pattern in their own test logs?

1 comments

miles_roberts22•28d ago

Rant: thermal paste application methods are overthought garbage

Every forum post I read swears by the pea method or the X pattern or some $30 paste. I've done 200 builds over 5 years at my shop in Austin. Used the spread it with a card method for every single one. Temps never differ more than 2 degrees C. Last week a customer brought his own $40 liquid metal and demanded I use it. After a month he came back with the same temps as my cheap stuff. So tell me, what's the actual proof that method matters beyond being careful and not overdoing it? Am I missing something or is this all just ritual?

1 comments

nora735•29d ago

Refurbishing a dead gaming rig turned into a fight over Windows activation

Client brought in a custom build that wouldn't post, turned out the motherboard had a bent pin in the CPU socket. I reseated it and got it running, but then Windows refused to activate because the license was tied to the old board. The client wanted me to bypass it, I said I can't do that, they argued for 20 minutes about how they owned the license. I ended up telling them to call Microsoft support and it took them 40 minutes on the phone, but they got it sorted. My question is, how do you handle activation headaches when hardware swaps are the root cause? Do you always push the client to call, or have you found a quicker way through the system?

2 comments

-1

briannguyen•29d ago

Swapped thermal paste brands after 15 years and temps dropped 8 degrees

I used Arctic Silver since 2009 on every build, but last month I tried Thermal Grizzly Kryonaut on a Ryzen 7600 and my idle temps went from 42 to 34. Has anyone else switched pastes and seen that big a difference?

2 comments

val_taylor•29d ago

Lost two hours of work after a power surge at a coffee shop downtown

I was at Brew Lab on 3rd Avenue last Tuesday finishing a RAID rebuild when the whole block lost power for maybe 10 seconds, and my laptop shut down dirty. Has anyone else started using a battery backup for on-site work after something like that?

2 comments

noahw53•1mo ago

Tried to replace a CMOS battery on a Dell OptiPlex and the whole board bricked

Was working on a client's old OptiPlex 3020 last week. Simple CMOS battery swap right? Wrong. I touched the battery holder clip and it literally snapped off the board. Tried to solder it back but ended up ripping a trace. Computer won't post at all now. Client is asking if I can fix it for free. Learned to just order a new motherboard instead of messing with those cheap plastic clips. Anyone else had a simple job turn into a disaster like that?

1 comments

paul_burns•1mo ago

I finally realized my cable management was actually making things worse

Honestly, I spent 3 hours last week zip-tying everything behind my bench into neat bundles like I always do, then a client's drive failed and I had to undo half of it just to swap a SATA cable. Tipped me off when I was hunting for which tie to cut and realized I'd trapped the power leads behind the mobo tray with no slack. Has anyone else had that moment where your tidy setup backfired?

2 comments

grant.kevin•1mo ago

TIL thermal paste curing times are actually a real thing after a client call me out

About 3 months ago I built a gaming PC for a guy in Denver using some premium thermal paste. He called me last week saying his CPU temps were climbing higher than they were on day one. I went back to check and remembered that some pastes need a "break in" period of 50 to 100 hours to hit peak performance. So my question is: do you all account for this curing time in your builds or do you just go with what the temps show right after install? Is there a reliable way to tell if your paste needs that time or if you just didn't apply it right?

2 comments

kim.hannah•1mo ago

Just realized most monitor stands are just hiding bad cable management

Was at a client's office in Seattle yesterday fixing their server rack and noticed every desk had the same 'clean setup' - cable mess stuffed behind a flimsy plastic stand. Anyone else find more hidden cable nests than actual problems on site calls?

2 comments

ray210•1mo ago

Still think thermal paste application matters that much? I changed my mind after 12 years

Been building and fixing PCs since 2011, mostly for friends and small businesses around town. Always did the pea method, never had issues. Last month I built a system for a client in Denver and decided to test something dumb. I just smeared a thick layer across the whole IHS with a credit card, total mess. Then I ran Cinebench and Prime95 for 4 hours straight. Temps were within 2 degrees of my careful pea method on the same Ryzen chip. I know everyone swears by the perfect dot or line or X pattern, but after that test I just don't think it matters for normal use. Overclocking maybe, but for 95% of people it's a waste of time to stress about. Has anyone else actually tried intentionally bad application and seen basically no difference?

1 comments

sullivan.john•1mo ago

Overheard a new tech bragging about skipping driver updates

Guy at the Micro Center in Portland said he never updates chipset drivers. Just lets Windows handle everything. Had to pull him aside and explain that's how you end up with random BSODs and network dropouts. You guys ever had to fix a machine where someone skipped all the manufacturer drivers?

2 comments