EVGA GTX 1070/1080 GPUs Catching Fire Due To VRMs Overheating

[Dave2150]

More craziness than brains in that quote man. The way the evga acx cooler is designed is essentially the same as every open fan cooler before it. The cooler only direct cools the core and the cooling plate covers the other vrm/memory ic and the coolers fan pushes air down over the plate.

https://www.techpowerup.com/reviews/MSI/R9_290X_Lightning/4.html

290x which pulls down like 3 times the power draw, zomg no direct vrm cooling on the actual cooler?

Next time, read the article you are linking, buddy................

Incase the picture isn't clear enough for you, those white things are thermal pads. Repeat after me, "thermal pads". The whole point of this EVGA issue is that there are NO thermal pads, hence the VRM's can overheat and cause a fire.

 

[nerp]

Wrong.

"EVGA’s Entire Range Of Nvidia GTX 1080 & GTX 1070 Featuring The ACX Cooler Is Affected
There were several investigations by a number of reputable technology outlets to try and identify the culprit behind it all. It was discovered that the issue originates from the voltage regulation modules which overheat leading to component failure and MOSFET blow out. The issue affects EVGA’s entire range of Nvidia GeForce GTX 1080 and GTX 1070 cards equipped with the ACX cooler, all of which lack direct VRM cooling. This includes GTX 1080 & 1070 ACX, SC, FTW & FTW DT models. Essentially all dual fan models except the Classified."

Source: http://wccftech.com/nvidia-gtx-1080-1070-evga-cards-dying/

EVGA Source: http://forums.evga.com/badHOT-PCB-Layout-on-GTX-1070-FTW-m2565921-p5.aspx#2567638

That article is the source of all the misinformation. WCCFTech is notorious for BS and sloppy 'reporting.' That article is the exact thing that all the subsequent evidence has shown otherwise: this issue only affects a limited number of FTW cards produced before Aug. 31. Only non reference pcb cards.

you can paste the same bad story over and over but that doesn't make it true.

 

[Dave2150]

That article is the source of all the misinformation. WCCFTech is notorious for BS and sloppy 'reporting.' That article is the exact thing that all the subsequent evidence has shown otherwise: this issue only affects a limited number of FTW cards produced before Aug. 31. Only non reference pcb cards.

you can paste the same bad story over and over but that doesn't make it true.

I notice you ignored the official EVGA source I also posted in that thread?

Do you really believe all the EVGA ACX range should have shipped with no thermal pads on the VRMS? Are you aware that several cards have already caught fire?

 

[amenx]

-

More craziness than brains in that quote man. The way the evga acx cooler is designed is essentially the same as every open fan cooler before it. The cooler only direct cools the core and the cooling plate covers the other vrm/memory ic and the coolers fan pushes air down over the plate.

https://www.techpowerup.com/reviews/MSI/R9_290X_Lightning/4.html

290x which pulls down like 3 times the power draw, zomg no direct vrm cooling on the actual cooler?

Not all coolers deal with the VRMs with thermal pads coupled to the cooling plate. Some cards VRMs have independent mini heatsinks that do require airflow to PCB area. As with Zotac AMP 1070 below:

Interesting 290x VRM cooling. Seems to be employing both techniques, which designers must have figured needs all the extra cooling measures possible.

 

[thesmokingman]

-

Not all coolers deal with the VRMs with thermal pads coupled to the cooling plate. Some cards VRMs have independent mini heatsinks that do require airflow to PCB area. As with Zotac AMP 1070 below:

Interesting 290x VRM cooling. Seems to be employing both techniques, which designers must have figured needs all the extra cooling measures possible.

Yea, not all are like that but most are cooling plate style. The 290x is interesting and I'm glad you pointed that aspect of it out because its also kinda hilarious in that respect. They direct cooled the 4 memory ic and not the vrms lol. That shows us how much that matters.

Btw, on the subject direct vrm cooling, your pic above represents the same thing. The actual main vrm is not direct cooled either and has it's heatsink, ala cooling plate etc. Look at the sink on the right nextto the line of chokes. Not direct vrm cooled, nope.

 

[nerp]

I did read the thread. And all I see from EVGA directly is this:

“The test used in the referenced review from Toms Hardware (Germany) is running under Furmark, an extreme usage case, as most overclockers know. We believe this is a good approach to have some idea about the graphics card limit, and the thermal performance under the worst case scenario. EVGA has performed a similar qualification test during the design process, at a higher ambient temperature (30C in chamber) with a thermal coupler probe directly contacting the key components and after the Toms Hardware (Germany) review, we have retested this again. The results in both tests show the temperature of PWM and memory is within the spec tolerance under the same stress test, and is working as originally designed with no issues.

With this being said, EVGA understands that lower temperatures are preferred by reviewers and customers.

During our recent testing, we have applied additional thermal pads between the backplate and the PCB and between the baseplate and the heatsink fins, with the results shown below. We will offer these optional thermal pads free of charge to EVGA owners who want to have a lower temperature. These thermal pads will be ready soon; and customers can request them on Monday, October 24th, 2016. Also, we will work with Toms Hardware to do a retest.”

Thanks,
EVGA

And another EVGA rep on reddit:

Neither "require" the additional thermal pads. EVGA considers the pads optional, and expects your card to work without issue without additional thermal pads.

If your GTX 1080 SC has an ACX cooler and a backplate, then it is eligble to sign up for the Thermal mod promotion. Is it necessary? In the strictest sense, no. Putting the rep hat aside for a second, however, if a company is offering me a free promotion that will make my card operate cooler, then I'd at least sign up for it first and think about if it's needed later.

It's unlikely that your card failed due to the lack of additional thermal pads. It's more likely that a component failed the card in a way that a certain percentage of cards will ultimately fail. I realize it may seem somewhat cold for a rep to talk about basic card failures, but sadly, it's true. It doesn't make it any easier for you, of course, so please don't think I'm negating that.

If you have any issues with the replacement card, however, let me know.

 

[LTC8K6]

http://www.guru3d.com/articles_pages/evga_geforce_gtx_1070_sc_superclocked_gaming_review,10.html

Guru3d finds quite a hot spot on an EVGA 1070 card.

That seems to be well above the other 1070 cards they tested.

Seen from the top side we see the image light up significantly. Once I placed a hotspot detector at the image we noticed a 96 Degrees C hotspot, that directly above the VRM area, so it's very hot alright. The majority of heat is exhausted, however at position HS1 we do see that heat is pushed out and above that location, and that ends up in your PC. The VRAM area is covered with a metal plate though so a lot of it is exhausted, just not everything. The overall thermal image can be rated as reasonable at best, that VRM HS1 spot raises concerns.

 

[thesmokingman]

Next time, read the article you are linking, buddy................

Incase the picture isn't clear enough for you, those white things are thermal pads. Repeat after me, "thermal pads". The whole point of this EVGA issue is that there are NO thermal pads, hence the VRM's can overheat and cause a fire.

Are you trolling or trying to be a comedian? Did you even read what you quoted? Let me go over it again for you so there is no confusion. Taken from your post below where you quoted the article.

The issue affects EVGA’s entire range of Nvidia GeForce GTX 1080 and GTX 1070 cards equipped with the ACX cooler, all of which lack direct VRM cooling.

My post was not directed at you but your quote of the whomever wrote the above, but then you go and troll me. That's great. Amenx's post actually shows what a direct vrm cooling cooler is like. And the point or my point was that your quote is neither here nor there, in others words it doesn't matter either way as direct vrm cooling is not the issue.

https://forums.anandtech.com/thread...vrms-overheating.2490217/page-3#post-38548957

 

[LTC8K6]

http://www.guru3d.com/articles_pages/gigabyte_geforce_gtx_1070_g1_gaming_review,10.html
http://www.guru3d.com/articles_pages/asus_geforce_gtx_1070_strix_gaming_review,10.html
http://www.guru3d.com/articles_pages/msi_geforce_gtx_1070_gaming_z_review,9.html

Other 1070 cards without that hot spot.
Info borrowed from posts by Nelly at the EVGA forums.

 

[LTC8K6]

http://forums.evga.com/FindPost/2571522

That is the post from Nelly. The EVGA 1070SC seems to have been way hotter in the VRM area than any other 1070 card during the Firestrike benchmark.
So we aren't talking about Furmark this time.

 

[stahlhart]

Next time, read the article you are linking, buddy................

I am going to state here for the record that my patience with individuals judging the literacy of other participants in this forum is wearing very thin.

If there is a point in a thread that requires clarification, it will be addressed in a manner that is respectful to the recipient, constructive to the discussion, and absolutely not like the childish playground taunting that is quoted.

The next time I see something like this in a post here, it is going to get ugly.

-- stahlhart

 

[bononos]

I did read the thread. And all I see from EVGA directly is this:

“The test used in the referenced review from Toms Hardware (Germany) is running under Furmark, an extreme usage case, as most overclockers know. We believe this is a good approach to have some idea about the graphics card limit, and the thermal performance under the worst case scenario. EVGA has performed a similar qualification test during the design process, at a higher ambient temperature (30C in chamber) with a thermal coupler probe directly contacting the key components and after the Toms Hardware (Germany) review, we have retested this again. The results in both tests show the temperature of PWM and memory is within the spec tolerance under the same stress test, and is working as originally designed with no issues.
...........

Where did you get the info that the problem is fixed (with thermal pads?) for cards shipping after Aug31 since theres nothing in your posts that shows that information.

Edit:
Since you didn't respond, I can assume you didn't read or mis-read the article/thread.

Only 1 problem was solved for cards shipped after 30 Aug which was for the out of spec (3-4%) VRM from ON Semiconductor unrelated to the high temperature problem. All the other 1070/1080 cards with the ACX cooler still have the issue with high temps because of the missing thermal pads.

https://www.techpowerup.com/227133/...g-issues-company-says-thermal-pads-a-solution
In a statement, the company has also attributed the black screen issues towards an "out of spec On Semiconductor component" that affected only 3-4% of EVGA's GTX 1080 FTW/HYBRID, with these issues "not [being] related to temperatures in any way" having been fixed as of September 1st, with the company having already sent replacements where applicable.

 

[amenx]

This is interesting. Cant vouch for accuracy, but apparently higher fan curves can help in cooling VRMs on 1070 FTW without thermal pads. Seems that enough air can get through the cooler to PCB area with higher fan speeds. Temps arrived at with thermal imaging (image in link at top):

https://www.reddit.com/r/nvidia/com...ging_of_evga_gtx_1070_ftw_under_load/d96jcxn/

Not sure if BF1 and FO4 are sufficient tests, but differences in stock vs higher fan speeds apparent. Needless to say avoid Furmark.

 

[nopainnogain]

I chose the Galax GTX 1070 EXOC (almost two months ago) mainly because of the use of Samsung memory, the 5+2 phase design, the double power cable (8 + 6) and the "non-stop" fans (which other people see as a disadvantage). My decision was taken on the basis of this review.

This page shows a "closer look" of the card:

http://www.hardwarebbq.com/galax-gtx-1070-exoc-graphic-card-review/3/

No one have used a hotspot detector in this card yet... How do you guys see the Galax project? It seems reliable?

UPDATE

I have just read the latest comments (by Jorge Carvalho) in the Galax GTX 1070 EXOC review and found that the "EX" edition (not OC = not overclocked) uses the Micron memory. Isn't that really interesting? I mean, why they used Samsung memory in the OC model and Micron in the non-OC model? These guys seem to know what they're doing...

 

[bononos]

http://forums.evga.com/FindPost/2571522

That is the post from Nelly. The EVGA 1070SC seems to have been way hotter in the VRM area than any other 1070 card during the Firestrike benchmark.
So we aren't talking about Furmark this time.

96C? No wonder some cards are exploding just after turning the pc on. At that sort of temps, its going to degrade alot faster.

 

[railven]

Didn't NV put power virus checks into their software?

Someone posted two different causes for issues with some FTW cards vs ACX-cooler cards. Mine had the black out screen all fans go into super saiyan mode.

Odd that it never happened during 3DMark testing, but flip the scale to 150% in WoW and my system would crash haha.

 

[pcslookout]

So is my EVGA GTX 1070 effected please ? https://www.amazon.com/EVGA-GeForce...UTF8&qid=1478092370&sr=1-1&keywords=gtx+1070#

 

[happy medium]

http://www.evga.com/thermalmod/

11/01/2016

EVGA GeForce GTX 1080/1070 PWM Operating Temperature Update
Recently, it was reported from several sources, that the EVGA GeForce GTX 1080 FTW PWM and memory temperature is running warmer than expected during Furmark (an extreme stress utility).

EVGA has investigated these reports and after extensive testing, below are our findings:

• On ACX 3.0, EVGA focused on GPU temperature and the lowest acoustic levels possible. Running Furmark, the GPU is around 70C +/- and the fan speed is running approximately 30% duty cycle or lower.
• However, during recent testing, the thermal temperature of the PWM and memory, in extreme circumstances, was marginally within spec and needed to be addressed.

Conclusion: EVGA offers full warranty support on its products, with cross-ship RMA*, and stands behind its products and commitment to our customers.

To resolve this, EVGA will be offering a VBIOS update, which adjusts the fan-speed curve to ensure sufficient cooling of all components across all operating temperatures. This VBIOS will be released in the next few days and users can download it and update their cards directly. This update resolves the potential thermal issues that have been reported, and ensures the card maintains safe operating temperatures.

For those users who want additional cooling beyond the VBIOS update, EVGA has optional thermal pads available. This update is not required, however; EVGA will make it available free of charge to any customer who is interested. To request the thermal pad kit, please visit www.evga.com/thermalmod.

Any customer who is not comfortable performing the recommended VBIOS update, may request a warranty cross-shipment* to exchange the product to EVGA for an updated replacement.

All graphics cards shipped from EVGA after 11/1/2016 will have the VBIOS update applied.

* The EVGA EAR(Advanced RMA Program) and Cross Shipping options are available in the Continental United States, Hawaii, Alaska, Canada, EU, UK, Norway, and Switzerland. EVGA offers Standard RMA replacement options in the Middle East, Africa, India or outside of the before mentioned supported areas.

FAQ

• Q. Which cards are supported by these updates?
• A. EVGA ACX 3.0 GeForce GTX 1080, 1070 or 1060 cards with the following part numbers:
  GTX 1080 GTX 1070 GTX 1060 6G GTX 1060 3G
  08G-P4-6181 08G-P4-5171 06G-P4-6262 03G-P4-6365
  08G-P4-6183 08G-P4-5173 06G-P4-6366 03G-P4-6367
  08G-P4-6284 08G-P4-6171 06G-P4-6265
  08G-P4-6286 08G-P4-6173 06G-P4-6264
  08G-P4-6384 08G-P4-6274 06G-P4-6267
  08G-P4-6386 08G-P4-6276 06G-P4-6368
  06G-P4-6167
  06G-P4-6165
  * Founders Edition, Blower type, HYBRID, and HYDRO COPPER cards do not need these updates.
  
  
• Q. If my Graphics Card is not an ACX 3.0 design is it affected?
• A. No, all other GPUs sold by EVGA including Founders Edition, Blower type, HYBRID, and HYDRO COPPER editions are unaffected.
• Q. Does the VBIOS fix the temperature issues?
• A. Yes, the VBIOS update will fix the temperature issues.
• Q. Is my card damaged from running at a higher temperature?
• A. No, but should any future warranty service be needed, EVGA stands behind its products.
• Q. What happens if my card is accidentally damaged during the optional thermal pad installation?
• A. EVGA will stand behind its customers with full warranty and cross-shipment.
• Q. What if I am not comfortable updating my VBIOS?
• A. While EVGA has very user friendly ways to update VBIOS's, if you still do not feel comfortable, we will support you with and advanced RMA and ship you a replacement card with the latest VBIOS.

EVGA Top Management

 

[railven]

Yet another in a race to win the "quiet" metric, they gaffe on cooling. Glad I went AIC, though even their AIC's had issues haha.

Just a bad case of the QA's. Glad to see them offering their Cross-ship service to anyone willing to use it. Highly recommend it. Outside of them using slow-UPS ground shipping, had new card installed and problem free <7 days of opening ticket.

I have zero reason to not buy EVGA if I stay in camp green.

 

[Dave2150]

I'm glad we now have a official response from EVGA's 'top management' confirming the initial reports. Now people cannot deny this issue, since EVGA confirmed that ALL ACX 3.0 coolers are affected by this problem.

Absolutely atrocious that EVGA are recommending people increase fan RPM via the VBIOS update - this is IMO simply damage control, since it would cost them millions to recall all the cards. Instead they prefer owners that paid a premium for EVGA 'quality' to suffer most likely the loudest fans on 1070/1080 custom cards, all because they wanted to save a few $ per card.

 

[fleshconsumed]

So EVGA really did admit they did not put thermal pads between VRM chips and the PCB plate? WTF ROFL? That stuff costs pennies. And now they want customers to up fan speeds to compensate? Double ROFL.

PS funnily enough I've had lots of videocards through the years and I have had multiple failures (3 total, 4 if you count ECS that failed twice), the only video card that failed by catching on fire was EVGA GTX 260c216 way back. I know, unrelated, but it's ironic seeing news about EVGA like that.

 

[Face2Face]

http://forums.evga.com/How-to-apply-the-Thermal-Pads-for-the-10701080-Cards-m2573413.aspx

So why is EVGA telling users to put a thermal pad on the top of the MMCP as well?

I can understand the back because it makes contact with the backplate, but why the front on top of the MMCP?

 

[Phynaz]

Excellent response from EVGA, that's why they are my video card manufacturer.

 

[Dave2150]

Excellent response from EVGA, that's why they are my video card manufacturer.

I believe that increasing fan speeds to compensate for the lack of thermal pads is a very bad decision and a sign of a sub-par manufacturer. They should have been included by default, all the other manufacturer's did include them.

 

[Markfw]

BTW, just talked to EVGA again. The bios "fix" just ups the fan curve a little. I have all 3 of my cards@90% fan(using XPrecision from EVGA), and 52-63c are the temps, and I can't hear the fans above the other fans in the case. Not worried at all, but I will put the "kit" on all 3 when I get it.

Oh, and those temps are with the GPU's "sort of" overclocked. I did nothing, except that software, and my 1070 is running 1911 mhz, at stock voltages and memory speeds.