Thursday, October 27th 2022

NVIDIA Tells AICs to Collect RTX 4090 Cards with Burnt Power Connectors, Send Them to HQ

by
btarunr
 Discuss (125 Comments)
NVIDIA is responding to reports of the 12+4 pin ATX 12VHPWR power connector of its new GeForce RTX 4090 "Ada" graphics cards being unreliable, and posing a potential fire hazard. The company has reportedly instructed its add-in card (AIC) partners, companies that sell custom-design graphics cards; to collect all retail graphics cards with burnt power connectors, and send them over directly to NVIDIA HQ for investigation. Reports of the 12VHPWR connectors overheating due to improper terminal contact aren't new, but this is the first time a retail product implementing the connector is experiencing reliability issues.

It came to light when a Reddit user posted pictures of a melted 12VHPWR connector from an NVIDIA-supplied adapter that converts four 8-pin PCIe to one 600 W-capable 12VHPWR. There is also charring on the female connector on the card, but the user claims that the card is functional. Later this week, another Reddit user posted similar pictures of a burnt connector for their RTX 4090 card. NVIDIA director of global PR for GeForce, Bryan Del Rizzo, in a statement to The Verge, said that the company is in touch with the first owner who reported this problem, and is reaching out to the other, as part of their investigation.
The GeForce RTX 4090 isn't just a thick graphics card, with air-cooled custom-design cards typically being 4 slots thick; but is also a "tall" card, with heights typically in the neighborhood of 150-160 mm. Add the 35 mm minimum clearance recommended for the 12VHPWR to not bend in order to function safely; and you have a total effective add-on card height requirement of 180-190 mm, which can be a very tight fit for most ATX mid-tower cases that offer a maximum CPU cooler height clearance of around 160-170 mm. A bending of the connector is almost a certainty.
Sources: The Verge, Igor's Lab, VideoCardz

Related News

Add your own comment

125 Comments on NVIDIA Tells AICs to Collect RTX 4090 Cards with Burnt Power Connectors, Send Them to HQ

#1
bonehead123
I smellz a cover-up in the making....

They want all the cards back so they can:

1) destroy as much of the evidence as possible, and

2) debunk all the claims with some lame-assed blame-game directed at the end users.....ie "they bent the cable too close to the connector", or at the "wrong" angles, or "they shoved these giant monster cards into cases that did not have sufficient room & air flow for them" yada yads yada.....
Posted on Reply
#2
Valantar
Looking at those connector teardown pics from Igor's Lab, it's really no wonder these are failing. Those outer wires are soldered to a single thin sheet of copper with no bends, ridges or crimps to strengthen it, meaning even moderate bending stress will shear the metal - like you can see happened during disassembly. It's really no wonder Cablemod tells you not to bend the wire horizontally, as that essentially guarantees that the outer solder tabs will shear off. Nvidia did use a strain relief boot to try and alleviate this (or as @jonnyGURU mentioned in another thread, give people something to grab so they don't tear out the wiring when trying to disconnect the plug), but it was clearly insufficient - and arguably this design is fundamentally flawed and should never have made it to market. I wouldn't call this a major scandal, but for a hyper-expensive, ultra-premium GPU? It's completely and utterly unacceptable. Heck, it would be unacceptable for a $300 GPU - but a $300 GPU likely woldn't need a stupid power adapter in the first place.
bonehead123I smellz a cover-up in the making....

They want all the cards back so they can:

1) destroy as much of the evidence as possible, and

2) debunk all the claims with some lame-assed blame-game directed at the end users.....ie "they bent the cable too close to the connector", or at the "wrong" angles, or "they shoved these giant monster cards into cases that did not have sufficient room & air flow for them" yada yads yada.....
... or they want to look at what went wrong so that they can figure out how and improve on their design? How else would you suggest that they do so?
Posted on Reply
#3
Rahmat Sofyan
it reminds me to GTX590 , WTH ?

for almost $2000 card, you still feel unsafe .. unacceptable

Posted on Reply
#4
LFaWolf
bonehead123I smellz a cover-up in the making....

They want all the cards back so they can:

1) destroy as much of the evidence as possible, and

2) debunk all the claims with some lame-assed blame-game directed at the end users.....ie "they bent the cable too close to the connector", or at the "wrong" angles, or "they shoved these giant monster cards into cases that did not have sufficient room & air flow for them" yada yads yada.....
You forgot you tinfoil hat and be careful of the water you drink today, :roll:
Posted on Reply
#5
Dragokar
LFaWolfYou forgot you tinfoil hat and be careful of the water you drink today, :roll:
The tinfoil is already used in nvidia excellent engineered connector.
Posted on Reply
#6
Diverge
DragokarThe tinfoil is already used in nvidia excellent engineered connector.
It's basically on par with most things made now due to supply issues and demand. Quality control doesn't exist anymore - the general population are the testers. Need to meet deadlines of those who sit in meetings all day...rush rush rush.
Posted on Reply
#7
DeathtoGnomes
ValantarThose outer wires are soldered to a single thin sheet of copper with no bends, ridges or crimps to strengthen it, meaning even moderate bending stress will shear the metal - like you can see happened during disassembly.
The other thing that matters is the number of strands that make up the wire, too few makes the wire less bendable cause breaks at the solder joint if bent too close. The copper alloy socket should also elongate because of the bend before the break, causing poor connection meaning a smaller contact point, like a funnel that could act like increased resistance, potentially causing a meltdown.
Posted on Reply
#8
Unregistered
ValantarLooking at those connector teardown pics from Igor's Lab, it's really no wonder these are failing. Those outer wires are soldered to a single thin sheet of copper with no bends, ridges or crimps to strengthen it, meaning even moderate bending stress will shear the metal - like you can see happened during disassembly. It's really no wonder Cablemod tells you not to bend the wire horizontally, as that essentially guarantees that the outer solder tabs will shear off. Nvidia did use a strain relief boot to try and alleviate this (or as @jonnyGURU mentioned in another thread, give people something to grab so they don't tear out the wiring when trying to disconnect the plug), but it was clearly insufficient - and arguably this design is fundamentally flawed and should never have made it to market. I wouldn't call this a major scandal, but for a hyper-expensive, ultra-premium GPU? It's completely and utterly unacceptable. Heck, it would be unacceptable for a $300 GPU - but a $300 GPU likely woldn't need a stupid power adapter in the first place.
Basically nVidia messed up.
No idea why they pushed for this new connector why most people are still using "old" PSU, at least wait till there is more PSU available, maybe once all PSU makers offer this connector as standard.
#9
Broken Processor
The connector confused me from the get go, why stick it in the middle of the card just because that's where the PCB ends. They could have made an extension to the end of the heatsink shroud so it didn't look crap.

Looking at the connector teardown the only connection to the board is solder no form of reinforcement on a cable they know is getting bent because aesthetics matter.

I'm surprised we are not seeing more cable's fail due to sheered off solder.
Posted on Reply
#10
mechtech
They are going to check is aib/aic are using cheap parts to save a buck??
Posted on Reply
#11
Chaitanya
Xex360Basically nVidia messed up.
No idea why they pushed for this new connector why most people are still using "old" PSU, at least wait till there is more PSU available, maybe once all PSU makers offer this connector as standard.
Leather jacket guy following Apple's philosophy of searching solution for which problems dont exist yet.
Posted on Reply
#12
dj-electric
mechtechThey are going to check is aib/aic are using cheap parts to save a buck??
Not sure they can. As far as I know the vendors list for that specific connector is quite slim.
I know Amphenol makes em, and probably Molex. Not sure who else even
Posted on Reply
#13
Chaitanya
mechtechThey are going to check is aib/aic are using cheap parts to save a buck??
Given the control freaks that they are, its just an attempt to shift blame at this point in order to save skin from lawsuits and losses.
Posted on Reply
#14
DeathtoGnomes
Broken ProcessorThe connector confused me from the get go, why stick it in the middle of the card just because that's where the PCB ends. They could have made an extension to the end of the heatsink shroud so it didn't look crap.

Looking at the connector teardown the only connection to the board is solder no form of reinforcement on a cable they know is getting bent because aesthetics matter.

I'm surprised we are not seeing more cable's fail due to sheered off solder.
reinforcement is usually shrink wrap tubes, or are you referring to something else?
Posted on Reply
#15
Valantar
ChaitanyaLeather jacket guy following Apple's philosophy of searching solution for which problems dont exist yet.
When your GPU requires four 8-pin PCIe power connectors there's definitely an extant problem - just one they created for themselves.
Posted on Reply
#16
zlobby
One would say the 4090's launch is literally fire!

Also, there are numerous petitions from home heating appliance manufacturers, which protest against 4090 taking over their businesses.
Posted on Reply
#17
Valantar
DeathtoGnomesreinforcement is usually shrink wrap tubes, or are you referring to something else?
The best solution for something like this would likely be a molded rubber strain relief boot integrated into the connector, though that would make it entirely non-serviceable. Not that that ought to matter much, but... I guess it does to someone at Nvidia?
Posted on Reply
#18
zlobby
bonehead123I smellz a cover-up in the making....

They want all the cards back so they can:

1) destroy as much of the evidence as possible, and

2) debunk all the claims with some lame-assed blame-game directed at the end users.....ie "they bent the cable too close to the connector", or at the "wrong" angles, or "they shoved these giant monster cards into cases that did not have sufficient room & air flow for them" yada yads yada.....
Yeah, probably users need to install these only under certain Moon phases, so that it coincides with the PFC of the PSU?
ChaitanyaLeather jacket guy following Apple's philosophy of searching solution for which problems dont exist yet.
Exactly! As with their software solutions, nvidia is trying to lock the entire ecosystem around their garbage.
Posted on Reply
#19
freeagent
They have been raking us over the coals for 2-3 years now with their plandemic pricing. Then they give the absolute cheapest to make high power connector using nothing but the shoddiest of manufacturing techniques.

And this is what they think of us. Thanks nvidia!
Posted on Reply
#20
Hofnaerrchen
What I found quite good about Igor's analysis: Why use a cheap adapter on a 1.6k-2k$ GPU when fitting high-end PSUs already come with great 12pin cables that also do not look like cheap shit. That's false economy.
Posted on Reply
#21
sephiroth117
bonehead123I smellz a cover-up in the making....

They want all the cards back so they can:

1) destroy as much of the evidence as possible, and

2) debunk all the claims with some lame-assed blame-game directed at the end users.....ie "they bent the cable too close to the connector", or at the "wrong" angles, or "they shoved these giant monster cards into cases that did not have sufficient room & air flow for them" yada yads yada.....
It's a public traded company, they cannot afford to have those GPU keep burning for long, they will take the necessary steps to fix them.
  • They won't destroy evidence, they are already documented.
  • They won't "debunk" anything because as long as those adapters keep burning they will never solve their situation and be liable to either harsh stakeholders scrutiny and defiance and or class action lawsuits, without even mentioning the customers turning away.
You can't cover up something like that tbh and for now NVidia seems to take the necessary steps to investigate this issue.

At first glance it's clearly their adapters but can't be 100% sure until I see their reports
Posted on Reply
#23
Gatt
Just now that the 7900 series is coming out LOL

And, anyway, one wonders who tested these GPUs and for how long ...
Posted on Reply
#24
ArcanisGK507
I have the slight impression that those cables seem to have been bitten by a cat...

Posted on Reply
#25
Valantar
ArcanisGK507I have the slight impression that those cables seem to have been bitten by a cat...

Those marks are most likely from wrestling that strain relief bootoff to expose the solder joints. Even non-molded boots like that can be a very tight fit.
Posted on Reply
Add your own comment
May 21st, 2024 16:42 EDT change timezone

Latest GPU Drivers

New Forum Posts

Popular Reviews

Controversial News Posts