Aurora Alienware R4 ALX shutting down unexpectedly

robertolito

New Member
Joined
Nov 24, 2018
Messages
1
My Aurora R4 ALX has been shutting down during use, mainly after some time of being on and doing different tasks.. The shut down usually happens during intensive graphic card use. It does not usually happen when I am browsing the internet or checking email. I ran an online test and it passed all the tests, including the graphic one. I thought it could be heating so I replaced the cooling system for a new one. Shut down happened again after replacing cooling unit at the time I was running a bench mark test of the graphics card. Because of this I suspect the graphics card could be faulty. But it could also be the PSU. And of course it could be motherboard. How could I check for sure what it is? I don't want to start replacing graphic card and or PSU and then find out it's one or the other or neither.
Update: I just now ran the PSA test and shut down. I was trying to run the online tests again with the online Assistant and it shut down. It did not even start the tests
 
Last edited:

LukeG310

Member
Joined
Apr 12, 2018
Messages
73
To me it sounds like the PSU because if it was the motherboard then there is a good chance the computer will not go past POST if anything. Out of curiosity, can you try turning on the computer without the RAM installed? and just see if it even turns on. If it stays on for a while then the RAM might be the problem but normally with RAM issues the screen goes blue and restarts.
 

MattyB

Moderator
Joined
Mar 19, 2012
Messages
2,732
Probably a stupid question but are your graphics drivers up to date?
 

strider

New Member
Joined
Aug 4, 2019
Messages
3
Hi there. Did you ever solve this? I am having this exact problem with my R4. It originally came with a GTX 680. I then added another GTX 680. Has worked fine for years. Then back in September, 2018 I upgraded to an A51 1200W PSU (VHM5V - Thanks @Cass-Olé) and a GTX 1080 Ti (Gigabyte OC Gaming). It worked great for 6 months but now whenever I tried to load a graphics-intensive game it powers down and reboots. I bought a new PSU (VHM5V) but the symptoms are the same.

I can run lower quality games (I still play Farming Simulator '15) and it's fine. I try to run the Unigine Superposition benchmark and even at the lowest res (720p) it will run for a minute or so at what it claims are 39% utilization, 150+fps, and then it'll power down and reboot.

NVIDIA drivers and Gigabyte control app (AORUS Engine) are up to date.
 
Last edited:

Cass-Olé

Member
Joined
Jan 9, 2013
Messages
142
VHM5V was designed to run a pair of HD6990 (375w each), 1080Ti starts at 250w & can exceed that, but, you should be able to run a pair of them & not run out of power unless the PSU was old-tired etc, so it's odd the new one reboots too (I'd say it was correct to take a chance on another one, if this gets solved you can keep both or sell one). It has a 46Amp GRFX rail with a 10% buffer where 'restarts' from circuit protection shouldnt happen until ~600watts is breached (50A x 12v) ... the PSU chassis should have a pair of video cables that come out of the case (wht/ylw pair), I'd hook both of those up (do not use the pair embedded in the case harness bulk connector)(use the primary 6+2s, lay both side-jumper 6+2s to the side un-used) & re-test

Always be sure you have two individual cables installed into your GRFX card as opposed to the single cable + jumper (Aurora case harness has x2 vid-cables, each has a jumper 6+2 ... blu-ylw & blu-wht, normally you plug one of each into your GRFX card & lay the jumpers to the side un-used)

Feel free to back the 66pin out of PSU & reconnect it solid

I hate to talk you into a $20 product that may not work, but in the past some users had luck w/ the EVGA powerboost (dedicated 12volts to the PCI-e lane to prevent GRFX card shut-offs)(this was an X58-era problem, X79 shouldn't need one but u never know); your MAIN SATA/DVD harness should have a molex/peripheral end on it for power to this adaptor https://www.ebay.com/itm/EVGA-Power-Boost-100-MB-PB01-BR/264020153984?hash=item3d78d3d680:g:r-EAAOSwkB5b3Jl7

If running two cables to card + powerboost doesn't help, I'm not sure what the issue is or could be since 'replace PSU' was next step; other advice might be to try card in some other desktop which isn't possible for some people, where if PSU was traded out we might blame GRFX card or mthrbrd next, I wish I knew

I hate to direct u to google & perhaps the Gigabyte forum in the event they ask u to RMA the card or specify 'tests' which preclude an RMA & point to other hardware like mthrbrd, OS, memory etc (below 'restarts', but try 'reboots' as search term also https://www.google.com/search?source=hp&ei=RplHXZH8OfGe_Qbd75jQBQ&q=1080+ti+power+supply+restarts&oq=1080+ti+power+supply+restarts&gs_l=psy-ab.3...1764.9276..9603...1.0..0.113.2168.29j1......0....1..gws-wiz.....0..0j0i131j0i10j0i22i30j0i22i10i30j33i160.3OxBWLesZMQ&ved=0ahUKEwiRt8uU3OrjAhVxT98KHd03BloQ4dUDCAs&uact=5

If someone asks, VHM5V is by Flextronics who makes Corsair Ax1200 1500i 1600i, it's a good quality PSU ...

edit: if you still have the original 875w PSU, put it back in & hook the blu-wht + blu-ylw cables to 1080Ti (lay jumpers to side) & re-test (you'll have a pair of 216Amp rails = 432watts to your card, should be enough) & report back
 
Last edited:

strider

New Member
Joined
Aug 4, 2019
Messages
3
Thank you for the quick response. I actually RMA'd the card already (I thought since I replaced the PSU it has to be the card) and they said no problem found. I also bought a PowerBoost but no joy.

However, we are making progress. Splitting the GRFX power across the 2 different power cables seems to have helped.

To recap. I have 1 x 1080 Ti. Coming out of my PSU I have 2 sets of power leads, 1 that is yellow/black and 1 that is white/black. I was using the yellow/black for the whole card. I'm now using the yellow/black for the 6-pin and the white/black for the 8-pin. This time it actually managed to load FurMark and Superposition (before it would power down as soon as the graphics came on hard).

I was able to run Superposition in 720p mode all the way through. I just tried it on 1080p Medium and it reset halfway through.
 

Cass-Olé

Member
Joined
Jan 9, 2013
Messages
142
All told you should have x4 cables but x8 total connectors (4 primary connex + 4 side-jumper connex); the wht & ylw pair coming out of the chassis are your best since they're soldered directly to the internal board so use those exclusively from now on for best connexion

I might suggest this: Rm850x & Rm1000x are just $80-$90/shipped Amazon, 90day returns, 1year warranty, you could buy one now or later this week (when in stock), try it out, & if no joy return it (no joy can also mean 100% sure it is not PSU related). Where VHM5V is split-up into several 12volt rails, these have a single 850w/1000w rail = unbridled power & shouldn't shut off on its own (the mthrbrd may power itself off & back on if need be however)

https://www.amazon.com/Corsair-RM850x-Modular-Certified-Refurbished/dp/B07HMC6LCL/ref=sr_1_6?m=A2JANV1GKV3FSH&qid=1564975514&refinements=p_4:Corsair&s=merchant-items&sr=1-6
https://www.amazon.com/CORSAIR-AX1600i-Titanium-Certified-Modular/dp/B07RWL8D29
(edit: Note that the CPU 8pin cable in the original stock case harness is nearly 34" long, where a new PSU may only provide a 24"-26" cable & therefore an 8pin extension may be needed, see below, black sleeved might look best, eBay & try Amazon or elsewhere)

https://www.ebay.com/sch/i.html?_from=R40&_nkw=cpu+8pin+extension&_sacat=0&LH_TitleDesc=0&LH_PrefLoc=1&_sop=15
Otherwise, no, I do not know how 1080Ti can trip VHM5V (this doesn't ring true of too much card for the PSU, not when it's rated to run two of them) nor do I know how to blame your motherboard, I can say the Amazon units can represent 'new' PSUs where VHM5V is now many years old even if some are available new never used (PSU can age on the shelf, caps can get old even if unused). I'd say if an Amazon unit (single 12volt rail) tripped off/on then the mthrbrd is to blame for some reason

see my new post below regarding 'temps' as possible root cause of restarts
 
Last edited:

Cass-Olé

Member
Joined
Jan 9, 2013
Messages
142
Do you have software installed that reports CPU & GPU temps? Are you able to monitor temps 'about when' you think the PC might restart (like 1/2way or full way at FurMark etc test)? Have you tried testing with the side panel off (maybe crank PCI-E + HDD fans to 100% in CmndCntr), heck I might even take the side panel off & direct a table fan at it just to see what happens lest it's a heat issue in the PSU, GRFX or CPU cooler, mthrbrd southbridge etc. If it can stay afloat with side panel off (perhaps + table fan) then that could be a clue given the Aurora is kind of small, lol

01YGW is the upgraded R4 cooler by the way
https://www.ebay.com/itm/ALIENWARE-AIO-CPU-WATER-COOLING-FOR-ALIENWARE-AURORA-R4-W-FAN-PP749-01YGW/283257167935?hash=item41f371603f:g:6kYAAOSwdwVb6beWhttps://www.ebay.com/itm/AIO-Water-Cooler-for-Alienware-Aurora-R4-Complete-Unit-w-Fan-free-shipping/143323164909?hash=item215eb9f0ed:g:EsIAAOSwZexdIxEQ610zWUXyDuL__SL1280_ - Copy.jpg
edit: I forgot there aren't a pair of 6+2s per cable. The discrete PSU vid-cables are a 6pin (primary) & 6+2 jumper: for best connexion when an 8pin is needed > borrow the 2pin from the jumper > mate it with the primary 6 & insert both into GRFX card (do not lay primary to the side, use it in tandem w/the 2pin > lay jumper 6pin to side un-used)
 
Last edited:

strider

New Member
Joined
Aug 4, 2019
Messages
3
Thank you for the quick and insightful responses. I don't think it's temp related. Here's my reasoning. This whole thing started when I tried to run Shadow of the Tomb Raider. The GeForce Experience utility set "Ultra" as the recommended setting. As soon as the game itself tried to load (after the logos and such), I'd get a reset. Box is stone cold at this point. Tonight, after swapping the power leads and some successful FurMark sessions, I dropped the settings down to High and tried the game again. This this setting the game loads and the game's benchmark runs fine. But when in the game settings itself, if I click the arrow to go to "Highest" the box immediately shuts down. I would think if it was temp related it would happen after some time.

It has to be power related. The replacement VH5MV I purchased looked brand new, all sealed in it's anti-static bag w/ stickers and such still on it. But yeah, who knows. Maybe the 1080 Ti has little spikes that exceed what the single rail can do.

I'm going to swap back to my old VH5MV just for grins and see what that does. But it looks like a new common rail PSU is in my future.

Thank you for the Corsair link. So that will fit in my case no problem? But it looks like I'll need to pull out the big connector and re-run all the power connections again? So I don't need the Platinum or Digital models? I'm not familiar with more expensive versions of the same PSU.
 

Cass-Olé

Member
Joined
Jan 9, 2013
Messages
142
The Amazon units are just to try something cheap & fast w/an easy free returns policy. If later (with in 90days) you want something else, return it. Some 'i' versions are on the Outlet if you look (I just purchased Ax1600i last week for $200 https://i.postimg.cc/MpDjKf8j/e-Bay-1600i-Kit-1.jpg), All Outlet items --> https://www.amazon.com/s?i=merchant-items&me=A2JANV1GKV3FSH&rh=p_4:Corsair&dc&qid=1564980772&ref=sr_nr_p_4_2

I'd gladly swap in the Rm850x 80Gold at that price

Any PSU will fit, you'd need to pull out the old case harness yes. Platinum, digital PSUs are nice if u can afford them, but they're not necessary. I had my (used) VHM5V tested by JonnyGuru PSU forum, it got 80 Silver & missed gold by an inch (it shares some internals with Ax1200 Gold). Aurora & 51 had Silver-rated PSUs (tops), gold is a step up so get a gold-rated PSU minimum these days, Plat or Ti if u can afford. My Seasonic 1300w is gold, it's perfect for my needs, I didn't need to pay extra for the Plati model but I agree Plati = just that much better components inside

Aurora ALX has rear-mounted SATA power cables, u need the correct pass-thru connector types & see this post for removing HDD cages (screws underneath the case, also, extra-long screw driver for wall screws preferred)
Aurora ALX (mid 2010) - PSU Replacement - Issue With HD Cages https://www.dell.com/community/Alienware-General-Read-Only/Aurora-ALX-mid-2010-PSU-Replacement-Issue-With-HD-Cages/td-p/5520964
SATA Power Cable Mod https://www.youtube.com/watch?v=Pi7rWKDBIHA

I left an edit earlier about possibility of need for CPU 8pin extension. My power supply swap, I got away with a 27.5" cpu cable. most are 24", 6"-8" extension recommended

My NBR links may be no good anymore but the PSU swap post titles are searchable on google & most lead back to the dell forum http://forum.notebookreview.com/threads/alienware-aurora-desktop-owners-lounge.771042/page-4

edit: one caveat about swapping in aftermarket PSU is that the new long round sleeved cables can get tight in there if you use dual GRFX cards, single beast card is best to ensure room (that's why most of Aurora's cables are not sleeved, to help make space for the loose wires ... Aurora may also get 10inch vid-cables whereas aftrmrkt may be 24 inches & they need room, dual-cards = tight fits in there). Since you removed an HDD bay to fit the 51 1200w, that will help make room if/when aftrmrkt goes in. The PSU Fan will point down, you can place a soft thin wedge under the PSU front edge to help 'center it' after it goes in. Placing desktop case down on its side 1st can help - bolt PSU in & wedge it with insulation foam or similar 'bumper' cushion- then flip case up etc
 
Last edited:
Top