o man i can imagine thats a sound that wakes you up...maybe i should upgrade my case with explosion-safe casewalls....at least the one between me and the hardware.
 
i think its this...not sure what a piece, but i found the location..yet not big enough to to see the excat part-number for me. Isnt that a usual reference-Pcb?

Looks unrelated to the VRM heat issues with the 1070 and 1080. 

Hi mdezc,
 
I am sorry to hear that your card had a problem.   I can confirm that this is not common and that while we cannot say for certain what caused the chip to fail, we are here to help.  
 
Since you are at work, please give us a call at your convenience as we are here 24 hours to help with warranty to have your card exchanged. 1-888-881-3842 Option 1 or by email at Support@EVGA.com.
 

mdezc
Just wanted to post this, so there's some record/precedent of it, in case anyone else looks it up.
My 2 months old, custom build:
- EVGA GTX 1060 &GB ACX 2.0
- Intel i5 6400 + Coolermaster 212 EVO
- 16GB HperX Ram
- Gigabyte B150m-d3h MB
- EVGA 600W Bronze 80 Plus PSU
- Thermaltake Commander Case
- Samsung 850 EVO 500GB SSD

How it happened: Normal gaming usage for a couple of hours with GPU running at 75°c and CPU running at 28°c. Suddenly PC turns off (coolers, fan, mobo led, these are all off), however the case LED and my RBG keyboard are still on, so I figured my OS froze. I restart the computer and BAM, sparks and fire come out the GPU, I quickly unplug the PSU and the fire goes out within a second. Lots of smoke that smells like some serious welding had just gone on, I leave the room. Back a few moments later when things have ventilated I pull the GPU out and take the picture (attached). After a quick review, I restart the computer again (without the GPU), boots up no problem and proceeds to install on-board intel graphic drivers (sob).
Have already started a ticket in order to request RMA.
 
Edit: Uploaded pic. 

Is this the place where the VRMs are located? I have read a review saying the area around the VRMs are able to reach 120 °C. Probably still in line (havnt checked the maximal temps for VRMs yet) . Just seems like the 1060 is running quite hot too

Just a friendly reminder...
 
Please do not use inappropriate language on the evga forums. Thanks.
 
- http://forums.evga.com/EV...erms-of-Use-m4682.aspx

VRM area is most likely at the end of the card where the power connection is.. can see the area on the backside of the card from the pcb solder spots for them.

I think it's the back of the VRM's zone, or very near. Anyway, it's the hottest part on this image:

 
Good luck!

http://www.tomshardware.de/nvidia-geforce-gtx-1060-pascal-gp106-grafikkarten-roundup,testberichte-242152-2.html#3
 
1060 hotspot hitting very high temps.

As a GTX 1060 (6GB SC) owner I feel sorry for you OP.
 
I've had EVGA 750 SC, 960 SC and now 1060 SC, and have opened them all before. EVGA has never actively cooled the VRAM/VRMs, and the 960 SC and 1060 SC pretty much have the same cooler.
 
Now my question is, is what happened to OP truly isolated (these things can happen, and doesn't necessarily indicate a fault), or is it indicative of an issue with the VRAM/VRM temps? Did EVGA used lower quality VRMs compared to the 960 SC? Is the VRAM an issue due to it being clocked 1000MHz higher than the 960 SC, hence generating extra heat in that region? 
 
This is the GTX 960 SC. There are no thermal pads whatsoever. I didn't take a photo of the 1060 SC when I took it apart, but the cooling solution and heatsink is basically the same. http://i.imgur.com/sR4IvQs.jpg
 
Hopefully EVGA can assuage our fears here. If there are more widespread reports I will definitely try to send this back to Amazon.
 
It also wouldn't hurt for EVGA to extend the thermal pads to us owners of the GTX 1060 as well. Good lord I hope EVGA doesn't cheap out on those in the future. They cost at most a dollar per card.

himmatsj
As a GTX 1060 (6GB SC) owner I feel sorry for you OP.
 
I've had EVGA 750 SC, 960 SC and now 1060 SC, and have opened them all before. EVGA has never actively cooled the VRAM/VRMs, and the 960 SC and 1060 SC pretty much have the same cooler.
 
Now my question is, is what happened to OP truly isolated (these things can happen, and doesn't necessarily indicate a fault), or is it indicative of an issue with the VRAM/VRM temps? Did EVGA used lower quality VRMs compared to the 960 SC? Is the VRAM an issue due to it being clocked 1000MHz higher than the 960 SC, hence generating extra heat in that region? 
 
This is the GTX 960 SC. There are no thermal pads whatsoever. I didn't take a photo of the 1060 SC when I took it apart, but the cooling solution and heatsink is basically the same. http://i.imgur.com/sR4IvQs.jpg
 
Hopefully EVGA can assuage our fears here. If there are more widespread reports I will definitely try to send this back to Amazon.
 
It also wouldn't hurt for EVGA to extend the thermal pads to us owners of the GTX 1060 as well. Good lord I hope EVGA doesn't cheap out on those in the future. They cost at most a dollar per card.

According to this post: http://forums.evga.com/GTX-1080-FTW-Black-screen-fans-spin-up-to-100-m2530081.aspx  there was an issue with the VRM ICs being out of spec on the 1080s. Is that related to what's been happening to other cards? Do they use the same component? I don't know. It seems plausible if they never cooled the VRMs but the VRMs changed a problem could suddenly arise.