Alienware 18 Viking (R3) - Second GPU Fan Not Working - 980M SLI

Dear All,

I've got an Alienware 18


Intel® Core™ i7-4810MQ CPU @ 2.80GHz
RAM -32GB

Storage x2 ssd, x1 hdd - all irrelevant here I think

GRAPHICS: x2 NVIDIA GeForce GTX 980M

One 980m which I've just added. works perfectly fine. Done stress testing and gaming. Temp climbs up like crazy within a minute because the FAN refuses to turn on on the right side(as i'm looking at the keyboard) for the secondary card.

The Fan works perfectly fine during pre bios testing. When Running ePSA diagnostics

I've also swapped the fans around from 1st GPU to 2nd GPU as they are identical fans and its definitely not the fan itself. For some reason even in HiINFO it detects fan1&2 but fan 3 is undetected.

I've tried the following so far:

Clean install of Windows

Latest Nvidia Drivers

Bios Flash - currently have the latest A15 version.



I've had cases where the laptop suddenly freezes and all the fans turn on (even the second gpu one) and then switches off.



I feel like its a software thing whereby its just NOT detecting that fan.



Can somebody please help and tell me what this could possibly be? Would be eternally grateful.



I've got 2 things as unknown device in the device manager for which i can't find any drivers. I dont believe this has anything to do with this as the other working fan isn't in device manager but just mentioning in case it does.

two things are:

on Microsoft ACPI-Compliant System

on PCI Express Root Complex





Open to any suggestions. THANK YOU VERY MUCH

 

@Raidriar mentioned in another thread about 980m needing a particular vbios in order to make both fans work, maybe he can shed some light on it

 

Yes, you will need a new vBIOS. Each card needs to have a different one so if you have a slave GPU in the master slot then you'll lose fan control. Check out this thread.

http://forum.notebookreview.com/threads/aw18-980m-sli-second-fan-solution.830663/page-4

Please check the current vBIOS on your card and if it is 84.04.85.00.0D then it's the slave card and needs to be flashed

 

Was anticipating others weighing in on this post very much too. I probably had a 980M GPU break because of a fan issue. As now, my AW18 is running with just a single 980M.
And like as you have stated of it " suddenly freezes and all the fans turn on (even the second gpu one) and then switches off"; mine does this every single time maybe after a minute to two minutes after Windows 10 loads fully and on rare occasions, but not so much anymore, my trackpad will get disabled or my USB connected mouse will and after maybe 5 seconds of the fans spinning at this time, perform as normal, although, unplugging the mouse then plugging it back in has to be done of the USB for the mouse to perform again and for the trackpad to be perform again, a restart of the system is all it takes. I solved my fan isssue by connecting an OPOLAR external fan to the right rear vent (the vent near the power button) of the 18; it sucks the hot air out and the 18 cools and this made a WORLD of difference. Maybe I need to do what is mentioned by others who have answered, as like I said of anticipating others weighing in on this.

 

As far as I understand you need to use HWINFO fan control

 

Thank you All for the input.

Maxware79 - I will probably do a Bios flash for the GPU although kind of scared of screwing things up. its not that straight forward. My Bios versions are totally different to that the thread states. I've just swapped my GPU around (thinking maybe it was common sense if the other GPU was slave). I don't think this has worked either.


Chivalrous Roamin Knight - Thank you for the input. Yes been experiencing this although right now the computer has been on for a few hours and no issues thus far. That added fan may be a last resort if bios flash doesnt work.

Currently looking for easy guides to flashing Vbios

 

to flash put nvflash folder on main directory hdd. like: "c:/nvflash"
after that you have downloaded the desired vbios put it on the nvflash main folder.
Open CMD prompt command as admin and type "cd c:/nvflash" *ENTER
then type "nvflash64 --index=1 --protectoff" *ENTER
after that type "nvflash64 --index=1 -6 nameofvbios.rom" *ENTER
it will ask you if you are sure so type "Y" *ENTER then again "Y" *ENTER
P.S: INDEX=0 is the main gpu, INDEX=1 second gpu.
Be sure that in device manager gpus get disabled before flashing.
P.S PT2: Those things that appears as without driver on device manager are missing intel chipset driver. use windows update or intel download manager drivers.

 

Thank you all for the input yet again.

I'll start off by saying this search over the last few days KILLED me - but i have finally found the simplest solution!!!!!

I can't believe the lengths people are going to in order to flash the BIOS.

https://www.dell.com/support/home/en-uk/drivers/driversdetails?driverid=9w9f7

I downloaded this - and it flashes both 980m's vBIOS to the appropriate one without any hassle!! IT seems DELL made this to make our life easier.

I know have the correct bios. BOTH fans are working perfectly fine. So far i've done stress tests for a few minutes and all seems to run fine! SO pleased.

Yes, Flashing vBIOS using NVflash etc is not impossible but it is not complication free not to mention theres room for error!

@ Chivalrous Roamin Knight - Dude Try this. I think it will solve the issue for you as well.

After dowing this, i used DDU (Display driver uninstaller) to make sure all drivers were gone. Reinstalled Nvidias latest driver and bingo, all good.

I'll be testing this over next couple of days and will let you know if anything changes.

Thank you!

 

Glad that you have solved. Well the nvflash can be useful if you will try prema mod vbios for example, gtx 980m throttles a lot with stock bios and in sli on AW18 it’s a mess.

 

Guys, i really need help.

Everything was working fine for 3 days - have done gaming for 1-2 hr sessions. all good.

Now within 2-3 minutes of the GPU working hard, as soon as the temp reaches say 75 degrees, the computer freezes up (and not because of the temparuture) .

Sometimes i'll get bluescreen saying clock watchdog timeout error!

Restart, and the pc works fine after. ON the game, and it repeats after a minute or two (or a little more).

I've done a reflash of Vbios - i've done a clean reinstall of windows again.

I didn't even overclock my GPU (last time i did was maybe 6 months ago but that was when i had a single 980m.

Does anyone know what this could be?

 

Try first to see if it’s not an cpu problem, use xtu stress test and see if still get freezes. If stress goes fine, disable sli and see if it happen with just primary gpu, if it still freezes, put second gpu on main mxm slot and test again

 

Thank you - I actually removed the secondary card.

now my touchpad stopped working - and moments later the keyboard (currently using one plugged in via usb)

no idea what the hell is going on here.

its all a bit of a nightmare.

stress testing cpu is fine at the moment.

 

Okay restart pc / f12 at boot and start diagnostic tool , maybe it will get at bottom of this. also if cpu and ram are oke can be maybe damaged hdd, try booting from another one if you can

 

I'd check you cable connections. If you've taken the secondary card out then you may not have plugged the touchpad cable back in properly.

 

Diagnostic tool says all hardware okay.

found out what the issue with mouse/keyboard was. One of the cable had a slightly damaged golden connector on the corner of it. It's working fine now.


I'm getting DPC watchdog violation with bsod (i now have the single GPU which i orginally had)

during usage, it freezes up for 20 30 seconds and the screen brightness goes full which i can't adjust unless i restart.

I've also seen random dots (artifacts?) on the screen on ocassion although this is very rare.

maybe this GPU is screwed?

 

Just put second gpu on primary mxm slot without the one that is giving artifacts, gtx 980m are pretty weak, to get one of these with long life u must to be lucky.

 

Or you can try to see if your primary gpu has some other issues, be sure that mxm it’s completely cleaned , also mxm lane of the card, be sure that thermal pads on vrms and rams has good contact with the heatsink. Can be that vrm get hot and you have instability issues