m18 R2, CPU overheating, temperature spikes at low load

If your system was just returned from getting repaired and you are still having issues you will need to contact Dell Support.  Click the Get Help Now Icon to the right of this post to be connected to a support agent.

To receive assistance from Dell chat support, they need to verify the warranty status and ownership. Then you must troubleshoot with them.  Click the "Get Help Now" icon on the right to start a live chat session. If already out of warranty, click here for the Dell out of warranty offering.

After a long 2-week wait, the Dell repair depot finally shipped my "repaired" Alienware M18 R2 back to me.

According to the repair letter, they replaced the bottom chassis, LCD HUD, heatsink, main logic board, and some plastics.

The very first thing I did was run the Dell online Diagnostics, but I got this message:
Error. A technical error occurred, please try running the SupportAssist Pre Boot System Performance check by restarting your system and pressing F12 during boot. Then select Diagnostics from boot menu.

I’m just wondering—don’t Dell repair centers run Diagnostics to confirm the computer is fixed before sending it back to customers?

Funny to see this post - my warranty ends on Nov 3rd and I've been fighting/logging this EXACT same thermal issue (same model, etc)) for Months now...

If it helps any, I'll pitch my experiences briefly, but also sort of running out of options:

- I had done an on-site repair which replaces CPU fans and ehatsinks - no change

- I had mailed in eventually and the Motherboard was replaced which had a definite positive impact (I was having very sporadic thermal spikes at all time, and while it was still getting way too hot under smaller loads an light gaming, the thermals were much more steady in readings and temp, even if high )

- Still looking to see what else can be done, I had set myself up with Throttle stop to try and protect the hardware and is bascially running into Power Limits all the time when gaming, but keeps most cores around 90C (and spikes forced TCC to 97C, still sometimes hits 100C though but no more) ... this was my last resort, as it definitely effects performance limits

Support had been great in general, and I've done everything/shared everything (BIOS updates, tweaks, etc) but now I'm stuck at a "Reinstall a Clean OS of Windows before we can help you any more" (which I haven't been able to do due to work responsibilities and can't afford the downtime to reload)

I really REALLY doubht Window sis the issue (still did a full scan/repair without clean re-install) so at this point I've concluded only two things that may 'help' you:

a. Motherboard might be defective if you're seeing irregular Spikes and unstable temperatures at any given time (I used HWiNFO software to view this visually - later logged through ThrottleStop) - This was for sure an issue, but did not solve the actual avg temps themselves under 10-20%... though I did see the ability actually USE up to 100% of the CPU existed, previously I never saw CPU utility above 40% because of limits, but after repair i became possible for brief instances which was good to see.

b. Honestly, I think for a mobile laptop, the Salesfolks/Engineers "got us" because such a chassis probably can't actually handle the hardware inside, and at a few instances during support 'tech dept' told me the data shown was "performing as designs/normal", (which was later glossed away thanks to our rep, but they still "said it" a few times) -- My guess is they wanted to sell us the specs and just didn't couldn't do a great job sticking our CPU/GPU (although PGU has been holding up very well, Mobile-edition though) into the chassis and it could never, by design, reach full potential and power-draw as a laptop without hitting power+thermal limits first...

Not sure if there are other models out there that do better, so I may give up as this may be a crux of modern "gaming laptops" (people like me/us keep wanting to put Desktop hardware into it, but need it portable for life/convenience - and we keep paying top dollar for it) ... not sure what else to say.

Either way, if you have had/or later have any progress I'd love to hear it! But I've pretty much exhausted every ounce I could get from Dell to date and this is as far as it's come (even Reddit is starting to fill up with "this is normal" around gaming laptops).. I believe it's just poor product design and overselling hardware it can't actually handle effectively/efficiently other than for very short instances/periods -- Long-term gaming (or CAD in my case) sessions, seems to need a desktop and liquid cooling, etc. nowadays... a shame.

My suggestion, it was tedious/confusing at first (first time using it), but ThrottleStop was the one program that gave me enough control (undervolting was disabled in Bios options at some point, so intel XTU doesn't help at all) to limit power and temperatures to balance safer average temp and as much performance as I can squeeze out of my setup... Might be worth looking at if you don't mind some tinkering!

Good luck and hope can hear some good news soon!