Tuesday, February 4th 2025

Human Error Reportedly Caused Latest 12VHPWR Cable Melting Incident

Late last month, NVIDIA claimed that 16-pin power connector issues were a thing of the past. The controversial 12VHPWR connection standard has fueled many online debates—prompting investigations from several prominent press outlets. Following NVIDIA's latest "safety" declaration—likely by coincidence—PCM Hong Kong reported another melting incident, affecting two cables and a power supply unit. The publication's hardware reviewer was recently engaged in the "full-load" testing of GeForce RTX 5090D and RTX 5080 graphics cards. Last week's evaluation session was interrupted by notable test system instabilities—upon downing tools, the PCM staffer discovered that their 1200 W PSU had given up the ghost. Additionally, two 16-pin cables had melted at both ends—initial detective work pointed to a GeForce RTX 4090 sample card being the main culprit.

VideoCardz and UNIKO's Hardware kept close tabs on PCM's next steps—online interactions, over the past weekend, spurred a re-evaluation of circumstances. According to PCM's latest update, they noticed burn marks on the GeForce RTX 4090 test unit—the two GeForce RTX-50-series cards did not exhibit any physical damage. Post-analysis, the reviewer now suspects that an SSD failure could be the root cause. They were happy to report that all involved RTX cards have survived, and that their test platform has been re-equipped with 12V-2x6 cables. An amended VideoCardz article proposes that everything came down to a simple human error.
Sources: UNIKO's Hardware Tweet, Tom's Hardware, VideoCardz
Add your own comment

41 Comments on Human Error Reportedly Caused Latest 12VHPWR Cable Melting Incident

#26
Scrizz
The SSD didn't cause the melted GPU connector... worst case the SSD has a shorted trace. The SSD would blow the fuse(if equipped) or trace would rip out of PCB.
PSU should have some sort of over-current protection. My bet is user is using some faulty cable/connector.
Whoever is doing this investigation clearly doesn't have the proper background to properly root-cause this failure.
Posted on Reply
#27
user556
KillSwitxyou know that 12v-2x6 and 12VHPWR cables are pretty much the same right? what changed was on the GPU connector ( female connectors ) with longer pins by 0.25mm and shorter sense pins by 1.5mm
Obviously they're almost the same, they are plug compatible after all. The issue is in the flexibility of the spring contacts. That's why a perfectly fitted 12VHPWR tends to work still, the contacts make fully. There's no such need to have a perfect fit with the 12V-2x6 plugs. They can flex and still fully contact.
Posted on Reply
#28
_roman_
When you need complicated pictures to explain something you have an issue.



If those 0.25mm longer pins really improves something? I want to see the tolerances before and after that. It's nice that someone made the efforts for a picture. Any connector spec sheet i saw in the injection molding factory and metall stamping factory had always a tolerance (2 years work experience making those 8d reports for quality claims).

These are just for consumer to quiet down the mind. for people who do not know how a drawing looks like for a connector. For people who never optically measured such a connector in their life.

-- There are products which just works for some reason. I wonder why?
A common issue made by ?, ?, ?, ?. I think I was right to not buy certain brands. Ignoring the facts that some brands just stick over their brand label over other products
Posted on Reply
#29
user556
Yeah, those receptacle pin lengths don't matter for the melting problem. It's all in the plug where the real fix is.

Saying there is no change with the plug is clearly bullshit. The spring contacts have changed. Or maybe one of the contact options available to 12VHPWR have been withdrawn for 12V-2x6. That would explain why some 12VHPWR plugs could have worked better than others.
Posted on Reply
#30
JoniISkandar
i said many times, who has the idea put High Power hungry card with tiny tiny 12 pin cables
Posted on Reply
#31
yzonker
Why is TPU putting up this click bait crap? Nothing in that statement says anything about user error. They were just swapping the cable between multiple GPUs. The issue is the 12vhpwr connector has such a low safety factor that it cannot tolerate a few plug/unplug cycles.
Posted on Reply
#32
freeagent
I have well over 30 cycle on mine, and its missing the clip.
Posted on Reply
#33
Caring1
_roman_When you need complicated pictures to explain something you have an issue.



If those 0.25mm longer pins really improves something? I want to see the tolerances before and after that. It's nice that someone made the efforts for a picture. Any connector spec sheet i saw in the injection molding factory and metall stamping factory had always a tolerance (2 years work experience making those 8d reports for quality claims).

These are just for consumer to quiet down the mind. for people who do not know how a drawing looks like for a connector. For people who never optically measured such a connector in their life.

-- There are products which just works for some reason. I wonder why?
A common issue made by ?, ?, ?, ?. I think I was right to not buy certain brands. Ignoring the facts that some brands just stick over their brand label over other products
The real fix was NOT the connector, nor the pins, but the cable being more flexible which prevented the pins from being pulled slightly out of the connector and increasing resistance and therefore heat.
Posted on Reply
#34
user556
Cable flexibility is not an issue, just like it wasn't an issue with the 6-pin and 8-pin plugs. It's been demonstrated that pulling and twisting a 12V-2x6 plug has no meaningful negative impact on thermal performance even when operating overloaded.
Posted on Reply
#35
Chomiq
Post-analysis, the reviewer now suspects that an SSD failure could be the route cause. They were happy to report that all involved RTX cards have survived, and that their test platform has been re-equipped with 12V-2x6 cables.
You guys should check your own wiring cause everyone in this thread is having a meltdown.
eidairaman1Northridge Fix on youtube recently had made a comment that just because he doesn't cover a RTX 4000 series card in a video daily with the engineered defective connector doesn't mean he isn't fixing them daily.
Same guy that thinks reflowing is the universal solution to dead GPU. Just saying:
Posted on Reply
#36
_roman_
Those youtube channels are just advertisement.

One guy just say ... hello internet and posts how he repairs.

Never forget - view clicks - advertisement - for their "service".

When the connection is lost to the pcb reflowing may make sense. It is a repair attempt. When the heart is dead - it is a lost cause. (i talk about the gpu - the big thing in the middle usually.)

Those channels just give me the impression that graphic cards are garbage and should be sold early with some months of warranty left. Regardless of brand.

I do invite those companies to be more open. Give me specs, datasheet, numbers on how many units sold, returned, repaired, why the repair and ....
It has a reason why powercolor does not give additional warranty. I just checked that a few hours ago for my graphic card to determine if they dare to give 3 or 5 years of warranty. No additional warranty in my case. Just a single personal example. Not implying powercolor graphic card are build well or worse.
Posted on Reply
#37
Bytorane
@T0@st ... yeah, there's a typo in your article, "route cause" > "root cause"
Posted on Reply
#38
Caring1
Bytorane@T0@st ... yeah, there's a typo in your article, "route cause" > "root cause"
read the replies, as it's already been pointed out.
Posted on Reply
#39
Bytorane
Caring1read the replies, as it's already been pointed out.
I saw it, and did my homework to search and make sure whether it was mentioned, but there was no indication T0@st confirmed that he saw the report.
Posted on Reply
#40
T0@st
News Editor
BytoraneI saw it, and did my homework to search and make sure whether it was mentioned, but there was no indication T0@st confirmed that he saw the report.
Sorry, was away for the last day and a half.
Corrected. Looks like the backend editor "auto-corrected" it into the wrong term.
I've written numerous articles using proper spelling, so it's strange that it suddenly turns up like this!
Posted on Reply
#41
phanbuey
This is exactly what happened to me:


Entire cable melted at both sides....
Posted on Reply
Add your own comment
Nov 18th, 2026 13:14 PST change timezone

New Forum Posts

Popular Reviews

TPU on YouTube

Controversial News Posts