33 Posts
0
2142
August 13th, 2022 14:00
Inspiron 5675 overheats
My Inspiron 5675 crashes while gaming, any game whether it's Steam, Battle.net, Ubisoft. It even crashed once while editing an Excel spreadsheet, but mainly during gaming. I can't see any heat issues while running HWInfo and there are no events in the viewer leading up to the crash that explain it, just notifications of an unexplained system shutdown. I'm guessing a heat related issue that I can't monitor? I followed the steps mentioned on the Inspiron board re: USB and power Shutdown settings to no avail. But it's not a total shutdown as there remain lights on the PC so the power supply is still supplying power, just the computer goes dark and requires a hard reset to get it running again. I tested the PS with the standard AT tester and all voltages are green lit. I had previously installed a Hynix SSD on a PCI riser that has functioned well for years. I replaced it with a new Hynix SSD and placed it on the motherboard (removed the PCI riser) but still get the random shutdowns.
I don't see any online talk of Hynix being prone to overheating but I also don't have any heat plate on either the riser or the motherboard (never came with one) but see talk online of them and wonder if I should have had one or need to get one? How could I tell if it's the SSD that's causing the shutdowns? I've scanned the OS for corruption and the disk for errors, and all turns up well.
Thanks for your interest.
RoHe
10 Elder
•
45.2K Posts
1
August 14th, 2022 12:00
Could the PSU be failing under load, even though voltages tested ok? Or, is PSU overheating? Does its fan work? Have you run Dell's BIST test?
Did thermal paste between CPU and heat sink dry out and separate? If it's not the PSU, it might be worthwhile removing the heat sink, gently cleaning both surfaces, applying thin, fresh coat of decent thermal paste, and reattaching heat sink.
DougDar
33 Posts
0
September 2nd, 2022 02:00
While noodling over the USB hubs I went in to the PC case and just unplugged anything that did not have to be there for the PC to be a PC. Bare bones - NVMe, PSU, GPU, keyboard, mouse, NIC.
I removed the wifi/bluetooth card and I even moved the one case fan from the back to the front closer to the suspected warm chipset.
Then it dawned on me, there's one other common thing that remained during all my troubleshooting efforts. The USB Extension cables that allow me to operate the PC from my couch. While I had gotten a different set of extension cables to swap and different keyboard/mouse combinations, I'd never just removed all Extension cables.
They're 20 ft...and while they've worked for years maybe slow degradation of motherboard components have made it difficult for the PC to keep up with pushing USB data and lighting that distance? Also, my keyboards/mice have evolved over the years from basic to gaming to RGB gaming devices.
So, my last step was to remove the extensions. I've run with wireless keyboard mouse for the last 12 hours with no crashes and no degradation in game performance.
I'll cautiously watch and if no crashes occur through the weekend I'll come back to update and finally call it solved.
DougDar
33 Posts
0
August 13th, 2022 16:00
FYI I also cleaned all the fans, vents, coolers and louvres including on the video card and reseated and swapped the memory sticks. I ordered a riser from amzn that has a heat sink plate that will be here tomorrow so I'll update when I test that out. Thanks, again.
DougDar
33 Posts
0
August 14th, 2022 05:00
Edit 2: wish I could edit the OP but forgot to say I initially ran PSAs which failed only by hanging on a SATA ssd disk test that was being used for game data. All other tests passed. So, I removed the data disk and after swapping the NVMe system disk for a larger new one and placing it in the motherboard M.2 slot reran the tests and all passed the thorough option.
My thoughts: It's a stress/heat related issue that isn't found by normal monitoring diags and brought on by increased system load so I'm guessing one of four things in order of suspicion - Graphics Card, Memory, CPU and finally motherboard heat problems. One thing I found strange was that the PSA reported the AMD graphics card as having 3 mb of memory when it actually has 8GB and that is properly reported by the Radeon software as is the 102 FPS performance of Diablo III. Is the PSA memory misquote a symptom? Also, when I was cleaning the graphics card fan I noted that it didn't spin as freely as I would expect, very slightly hanging at just one spot in its 360 degrees of travel, although not enough to actually stop it when spun by hand.
DougDar
33 Posts
0
August 14th, 2022 12:00
Riser with heat sink installed, no change. I pulled one of the two memory sticks at a time to isolate that possibility, no change. All caps look to be in good shape - no swollen ones. Video card disassembled, cleaned and inspected - all seems well. Running out of options. Runs fine doing normal browsing. Fire up any game or play and high res video and it crashes. Any ideas, anyone? I'm out of 'em.
DougDar
33 Posts
0
August 14th, 2022 12:00
I have not, thanks for pointing that out! I'll try the BIST next and then get some thermal paste mailed in after if it passes. Thanks, again!
DougDar
33 Posts
0
August 14th, 2022 13:00
That was it. Fan sitting underneath the PS facing the deck and I didn't even notice it wasn't spinning. Do you by chance know if the standard Corsair modular PS will fit there? Thanks, again, RoHe! Much appreciation
RoHe
10 Elder
•
45.2K Posts
0
August 14th, 2022 16:00
Don't jump to conclusions quite so fast...
Did you run the BIST?
Modern PSUs don't run their fan constantly like older PSUs. It's an energy-savings thingy...
The real question is does the PSU fan spin when PC is under heavy load? Eg: long, heavy gaming session, lots of power hungry USB devices all connected and turned on (eg, printers etc), playing audio etc etc all at same time.
BTW: Is anything logged in Windows Event Viewer and/or in Windows Reliability Monitor around the time of a crash?
Lots of threads about replacing the PSU in this Inspiron model, so have read.
Might be less expensive to replace the thermal paste first and if you're lucky, you'll save the cost of a new PSU. Obviously, I can't say if either of these "fixes" will actually solve the problem.
Your PC, your $$, your decisions...
DougDar
33 Posts
0
August 15th, 2022 07:00
OH, and yes I ran the BIST and it passed with a green light but the fan did not come on while running the BIST. Sorry for the spam - and thanks, again!!
DougDar
33 Posts
0
August 15th, 2022 07:00
Sorry, I just read this this morning after I've already ordered the replacement PS. I didn't know that either - PS's could/would run without a fan on. I learn something new every day! Yesterday's lesson was the BIST...never heard of that one before either. I've been out of the PC troubleshooting scene for too long. I didn't attempt to put the PC under load to test the PS fan after seeing it not spinning but I did pull the PS out of the case and sit it on a 4 inch cube in front of a standard small desk fan to test it under load...and it worked great. I ran a demanding game for 4 hours this morning without a single hiccup. We can mark this one Problem Solved. Thank you so much for your assistance!
RoHe
10 Elder
•
45.2K Posts
0
August 15th, 2022 18:00
Post back and let us know that the new PSU solved the problem.
DougDar
33 Posts
0
August 16th, 2022 10:00
That did the trick. It took a little modification to get the modular Corsair to fit (took the back plastic bezel off) and the power plug was a tight squeeze but the 650 works great...and was actually cheaper than the 550 on Amzn yesterday. Heck I might need to upgrade my video card now (j/k). Thank you for the assist!!
RoHe
10 Elder
•
45.2K Posts
0
August 16th, 2022 11:00
You're welcome. Glad to hear that the problem is resolved!
DougDar
33 Posts
0
August 29th, 2022 15:00
I spoke too soon. It worked great for 10 days of pretty much continuous game playing but then started crashing again. I got heat paste and redid the CPU's mount. No luck. I redid the GPU's mount paste, but that actually made it crash more quickly. Turns out the fan I thought pointed at the PSU to cool it was actually cooling something else in the case it was sitting near. Pulling the PSU out and pointing the fan into the PC case still causes stable operation of my games. My next step, since redoing the GPU pasted seemed to change the symptoms a bit, is to get a new gpu. I found a GTX 1660 6gb for cheap at BBuy ($210) and will get it Friday. Once it's installed I'll write back to let you know wassup. What a slog.
DougDar
33 Posts
0
August 31st, 2022 11:00
I'm going to go out on a limb and say this one is fixed for good now (and we all know how that goes with heat problems). For those looking to upgrade/repair this older PC, the new Nvidia 1660 super went in flawlessly and I just ran a dozen rifts in Diablo...which I couldn't do this morning before the card. I couldn't run one complete rift without a crash before the new card so this is a good sign. And the PSU I bought previously wasn't a waste because I'd have needed it for the new card anyway, so win win? I'll come back in a week to confirm all is well but I think this is a wrap. Learned much, thanks again!