Graphic Cards Denial of RMA by NVIDIA/RPTECH and MSI for PCI Express Slot and PCI Express Pad burn

sstiwari

Disciple
This was long overdue.

This is about MSI MEG ACE x570 Motherboard, NVIDIA RTX 4090 FE(bought on launch), and Corsair Axi 1200[Powering Gfx Card thru corsair 12VHPWR Cable], as these are the components that will matter to this post.

On 31st March 2024, for the usual weekend gaming ritual, I powered on my PC after 5 days, it briefly started to immediately shutdown(whole episode took around 5 seconds), there was burning smell coming from the cabinet, although there was no smoke.

I immediately opened the Cabinet door, but couldn't see any obvious damage. So, pressed the Power Button again, and it booted fine but without any display signal to the monitor.

I suspected the dreaded connector melt in RTX 4090 FE. On disassembly, I could not find any damage to connector on either side, but the PCI Express Pad Power Pins on RTX 4090 FE were melted and so was the PCI Express X16 Slot on the MSI x570 ACE.

So, I contacted NVIDIA, MSI, and Corsair.

Corsair: Immediately agreed to replace the PSU, and I received a brand new Corsair Hxi 1500.

NVIDIA: Level 1 support asked me to share images, which I shared. Then they advised me to visit RPTECH Noida, which I did. To my Horror, RPTECH refused to touch the card. They just took a few snaps and said this is not covered under warranty. They even refused to issue any card submission, RMA denial or warranty void document to me. I again contacted NVIDIA about the treatment at RPTECH, they shared the same images that RPTECH has taken that burn is not covered under warranty.

I quoted warranty void clause on NVIDIA website, which doesn't include burn.

"WHAT DOES THIS WARRANTY NOT COVER?
Any problems that do not relate specifically to a manufacturing defect or hardware product failure, including, but not limited to, problems caused by abuse, misuse, negligence, act of God (such as flood), misapplication of service by a party other than an authorized service representative, software, shipment damages, etc."

NVIDIA Level 1 support still said they can not help me. So, I asked to escalate. He escalated to Level 2 support, who said that even if they wanted to help me, they can not, as they do not ship directly to India and RPTECH is my only place to seek help. Dropped an email to Mr. Jensen Huang, which was never acknowledged or reverted.

MSI: Reponded in 2 parts that this type of damage is not covered under warranty and generally not repairable, meaning don't even bother for paid repair. I had owned several MSI products prior to this, all of them were good. But the blatant shake off by MSI has really pissed me off.

Final Notes: I religiously clean the PC using a blower every 2-3 weeks, no body else touches the PC, It happened before monsoons, and I have been building PCs for me since 2006, so dust ingress, humidity and shorting, inexperience etc. is really out of question. Finally, Motherboard was still booting fine with another graphics card in next PCIE Slot and same PSU. So, either PCIE Slot malfunctioned or the Graphics card had some malfunction.

Lessons Learnt:
1. In such cases you are totally screwed, there is no mechanism in India to help you.
2. I will never buy from NVIDIA as OEM again and if AMD rises to my needs, will never by any NVIDIA GPU even from AIB partners in future.
3. My experience with Samsung, Razer, Gigabyte and Corsair supprt has been superlative. They are really very helpful, everytime I contacted them. Though not experienced any burn issues.
 

Attachments

  • NVIDIA Level 1 Part 1.png
    NVIDIA Level 1 Part 1.png
    554.3 KB · Views: 67
  • NVIDIA Level 1 Part 2.png
    NVIDIA Level 1 Part 2.png
    675.5 KB · Views: 56
  • NVIDIA Level 2.png
    NVIDIA Level 2.png
    377.5 KB · Views: 44
  • MSI Part 1.png
    MSI Part 1.png
    229.1 KB · Views: 48
  • MSI Part 2.png
    MSI Part 2.png
    344.5 KB · Views: 44
  • Corsair.png
    Corsair.png
    215.1 KB · Views: 80
  • IMG_1882.jpg
    IMG_1882.jpg
    410.1 KB · Views: 75
  • IMG_1883.jpg
    IMG_1883.jpg
    241.8 KB · Views: 77
  • IMG_1884.jpg
    IMG_1884.jpg
    194.6 KB · Views: 78
  • IMG_1885.jpg
    IMG_1885.jpg
    537 KB · Views: 78
Last edited:
Wow, its time to lawyer up for sure.. these schmucks at RP tech need to be taught a lesson.

This is why i always buy brands whose RMA's are handled by kaizen.. for eg: zotac, corsair etc.

My RMA experiences with Kaizen have been nothing short of excellent.

My local dealers also warned me that the technicians at RPtech who intially investigate the product for RMA's are idiots.. and they try to find every reason to deny warranty.
 
Last edited:
So, basically if your components have any kind of burnt damage on them you are well and truly screwed.
This is true. They automatically reject anything that has burnt components regardless of what brand. Even though their warranty terms never mention anything about rejecting burns. I had a similar case with MSI Graphics card (handled by Acro). They said it's just not possible despite the warranty terms. Not sure this is just India thing or true internationally.

I personally see no reason to reject burns as the GPU is the one at fault pulling too much amps than it can handle.
 
This, Share & Shame them all on Social Media.
Social media is not working these days. Around 3 months back, I ordered Ryzen 9 7900x3d from amazon. I received a sealed box of Ryzen 9 7900x3d, except that it had Ryzen 7 5700G inside the tray, visible from outside. Luckily, I got the outer package opening done by delivery executive and video recorded entire episode, and made him agree that fake item has been delivered. Amazon was compelled to refund, else I was again big time screwed.

I raised the issue on twitter(X), tagged AMD India and Lisa Su, but never received any response.

Check here



 
Last edited:
This is really disheartening, especially with such an expensive product where you'd expect things like this to never happen.

When you think about it, warranty as a concept is an unsustainable solution because the cost of replacement is either taken from past profits or future profits.

Personally, I limit my purchases to whatever I can afford to buy two of, this way I'm not stretching my finances and neither am I unable to recover in the case of a malfunction or warranty denial.

But none of that is helpful advice now.

I'd suggest trying and finding similar failures with other owners on social media and see if yours is an isolated case. If it isn't, you have more reason to seek out a resolution through social media since that's the only thing that appears to work these days.
 
When you think about it, warranty as a concept is an unsustainable solution because the cost of replacement is either taken from past profits or future profits.
I don't see why that would be the case. If you design a product that should work for the bare minimum of two years or however long the warranty period is and do actual quality control before sending it out of the factory, then maybe 1% of them will fail before the end of the warranty period. Replacing those units should be sustainable and easily offset by increased customer confidence.

Now if instead of doing that, you cut corners everywhere on the design and much more than 1% fail, then you have the option of either honoring the warranty and losing money or finding a bs reason to not do it. It seems Nvidia is taking the second route, at least in India.
 
To me by the sight, looks more of a PSU issue, GPU not sure, you can rule out the motherboard, the slot on the motherboard has got damaged due to the other components failing. Can be due to sudden power surge which caused the burn or can be overload too on the rail. Only after opening the grpahic card further inspection can be done. You can get the graphics card checked for repairs.
 
To me by the sight, looks more of a PSU issue, GPU not sure, you can rule out the motherboard, the slot on the motherboard has got damaged due to the other components failing. Can be due to sudden power surge which caused the burn or can be overload too on the rail. Only after opening the grpahic card further inspection can be done. You can get the graphics card checked for repairs.
PSU was absolutely fine.

Surge has occurred on the PCIE Slot 12V Power Pins. Power here is supplied by mobo and not by PSU. Either Mobo has supplied a bit too much or the GPU has drawn a bit too much from the PCIE Slot. The failure happened at cold boot, there was no load anywhere in the system. If there were any fault in PSU, mobo VRMs would have been fried first.
 
who said that even if they wanted to help me, they can not, as they do not ship directly to India and RPTECH is my only place to seek help.
File an online consumer complaint at https://consumerhelpline.gov.in/user/signup.php

You have solid evidence that they have agreed to give you the replacement but due to mere shipping and logistics reason they are denying you your replacement. This is a blatant breach of Consumer rights.

If the brands wants to do business in India they are obliged to follow Indian Consumer Laws.

Mention in the prayer that the subsidiary of Nvidia i.e. RPtech should be ordered to hand you the replacement or Nvidia/ Nvidia India should be ordered to deliver you the replacement.

Attach all the documentary and email evidences of invoice, communications etc.

Also write down all the incidents that happened in chronological way and in such way that any non-technical person also understands the language. (Theres a chance that Consumer Forum official is not that tech literate)
 
1. Use INGRAM to force them to respond in writing
2. Have you tried using this GPU on another mobo?
3. Was there GPU sag? If the power connectors were not seated properly, it might have resulted in increase in resistance and heating
 
It's likely a component failing and shorting out on the GPU causing a spike in amps that melted the pcb traces. Even if the mobo's power limit had failed, the GPU would still need to draw that much power in the first place.

Unlikely to be PSU because 12V rail is connected to many other things that did not fail.
 
1. Use INGRAM to force them to respond in writing
2. Have you tried using this GPU on another mobo?
3. Was there GPU sag? If the power connectors were not seated properly, it might have resulted in increase in resistance and heating
I contemplated but then thought why to risk another motherboard to burn?
 
Either Mobo has supplied a bit too much
Current is pulled not pushed.

In my case, it was the power monitor chip that went bad. It was reporting much lower wattage than the card was really consuming. Which (in my theory) causes the card to keep pulling current until the power rail cannot handle it anymore and boom, a crater was born. People (including service centers) often blame PSU for these kinds of issues but I think it is rarely the culprit.
 
It seems your issue is a mixed bag of tech and law. The service centres have denied accountability citing similar reasons, maybe the issue happened due to something external and which is not under warranty. It's just a hunch.

If you want to go pursue the legal trail then file consumer complaint at edaakhil.nic.in
I am not certain if you would receive a favourable verdict but that's how the process works in India.

Rptech is a serious culprit of flouting warranty claims as I have seen many such complaints on the Indian Gaming sub reddit.
 
Back
Top