• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.
  • The forums have been upgraded with support for dark mode. By default it will follow the setting on your system/browser. You may override it by scrolling to the end of the page and clicking the gears icon.

Intel Readies ATX 12VHPWR Connector Revision to Address Improper-Contact Fire Hazards

btarunr

Editor & Senior Moderator
Staff member
Joined
Oct 9, 2007
Messages
47,753 (7.42/day)
Location
Dublin, Ireland
System Name RBMK-1000
Processor AMD Ryzen 7 5700G
Motherboard Gigabyte B550 AORUS Elite V2
Cooling DeepCool Gammax L240 V2
Memory 2x 16GB DDR4-3200
Video Card(s) Galax RTX 4070 Ti EX
Storage Samsung 990 1TB
Display(s) BenQ 1440p 60 Hz 27-inch
Case Corsair Carbide 100R
Audio Device(s) ASUS SupremeFX S1220A
Power Supply Cooler Master MWE Gold 650W
Mouse ASUS ROG Strix Impact
Keyboard Gamdias Hermes E2
Software Windows 11 Pro
Intel is preparing a revision to the design of the 12+4 pin ATX 12VHPWR power connector, specifically its male connector, to improve connector mating and retention; and to minimize a potential fire hazard asiring from improper retention or mating. The male connector design revision will retain full compatibility with female connectors of devices in circulation. It essentially sees the pins of the male connector switch from a "three dimple" joint to a "push-spring" type joint. This design change makes the male connector much more resilient to weakening contact from pull forces arising from cable bending. The resulting connector would offer superior structural integrity of the connection.



View at TechPowerUp Main Site | Source
 
Interesting revision, I wonder how much it helps. Also, of course improperly seated high power electronics can be a fire hazard, but I'm curious if there have been any reports of fire with 40 series cards so far, or just melted connectors?
 
Low quality post by DBGT
Here we go
 
Wait-a-second.
Image on left is current-spec?
I only see 6x tiny circular areas of contact / mating-surface.
Rev. B, with it's 'leaf-spring' style contacts, should've been 1st-attempt! (that's how it usually goes, TBF)

Part of me wonders if this was on-purpose to 'maliciously address' the influx of cheapo-import adapters and extensions? The only 'trustworthy' adapters will be Intel Rev. B-compliant, and probably from contracted-ODMs and companies that are also Board Partners.
I mean... it's not like we haven't seen 'Big Boys of their Industry' purposely 'gimp' a first release of a device/spec. *cough*Nintendo*cough*
 
I say revision 4 will be the real winner hehe. You know when they revert back to 8pin connectors all around! Aaaahahaha

Seriously tho, at least they're trying. Altho, if I were Nvidia, I would want to be directly involved with any revisions to MY connector.
 
I say revision 4 will be the real winner hehe. You know when they revert back to 8pin connectors all around! Aaaahahaha

Seriously tho, at least they're trying. Altho, if I were Nvidia, I would want to be directly involved with any revisions to MY connector.

The weird thing is we could have just switched GPUs to the CPU power connector and got 300W without any effort. Would have made more sense imo.
 
The weird thing is we could have just switched GPUs to the CPU power connector and got 300W without any effort. Would have made more sense imo.

Agreed.
EPS12V connector is widely used in the server space GPUs and 2 of them provide 600W already.
Same footprint as the current connector, almost the same tooling, and server market tested.
 
Wait-a-second.
Image on left is current-spec?
I only see 6x tiny circular areas of contact / mating-surface.
Rev. B, with it's 'leaf-spring' style contacts, should've been 1st-attempt! (that's how it usually goes, TBF)

Part of me wonders if this was on-purpose to 'maliciously address' the influx of cheapo-import adapters and extensions? The only 'trustworthy' adapters will be Intel Rev. B-compliant, and probably from contracted-ODMs and companies that are also Board Partners.
I mean... it's not like we haven't seen 'Big Boys of their Industry' purposely 'gimp' a first release of a device/spec. *cough*Nintendo*cough*
so they send out a new cable to the people who bought an old psu.
that’s the beauty of modular power supply’s.
 
Agreed.
EPS12V connector is widely used in the server space GPUs and 2 of them provide 600W already.
Same footprint as the current connector, almost the same tooling, and server market tested.

Yeah technically the most sensible solution, but the "problem" is that EPS12V is from SSI and not PCI association.
Basically similar to the DP vs HDMI conundrum.
 
Sounds like the standard Engineering process: 25% of the time spent designing the product, then the other 75% spent re-designing it to try & prevent morons from misusing it in ways no normal person would even consider. There's a reason why these connectors are designed to click once they're seated properly...
 
Yeah technically the most sensible solution, but the "problem" is that EPS12V is from SSI and not PCI association.
Basically similar to the DP vs HDMI conundrum.
I thought the main 'problem' of HDMI is the license fee?

Does EPS12V need a license fee ?
 
I say revision 4 will be the real winner hehe. You know when they revert back to 8pin connectors all around! Aaaahahaha

Seriously tho, at least they're trying. Altho, if I were Nvidia, I would want to be directly involved with any revisions to MY connector.
"MY connector".

You still believe Nvidia designed the whole thing huh?

Some people just want to believe i guess, so they can hate on Nvidia.
 
Interesting revision, I wonder how much it helps. Also, of course improperly seated high power electronics can be a fire hazard, but I'm curious if there have been any reports of fire with 40 series cards so far, or just melted connectors?

melted connectors and damaged GPU to my knowledge but the risk is real

however since Nvidia reported that it was due to bad seating (and proved it alongside tech outlets like GN), there have been far fewer mentions of this issue

Still, a design that is not user fault-tolerant enough does need revision, plus here the revision would just be a new cable rather than changing everything
 
Agreed.
EPS12V connector is widely used in the server space GPUs and 2 of them provide 600W already.
Same footprint as the current connector, almost the same tooling, and server market tested.

I feel like the only point of the new GPU connector is they want to release GPUs that are from 600W-900W. Two CPU EPS connectors are already perfect for up to 600W. More than enough.
 
Damn, I just got my Seasonic Vertex PSU, which right now is obsolete already :mad:
No?

The PSU has a female connector like the GPU, both cable ends are male. Just replace the cable.
 
Well if it ain't broken don't fix it, wait what.

As many stated a poor design is piss poor in this day and age.
 
"MY connector".

You still believe Nvidia designed the whole thing huh?

Some people just want to believe i guess, so they can hate on Nvidia.
The connector is a standard. However, the way that Nvidia handled the issue is not exactly the best in my opinion, Here at least we see that Intel learned from the lesson and is trying to improve the situation and make the connectivity less iffy. Nvidia's last statement was that the issue was a user problem, full stop. The cards that Nvidia are selling are not exactly cheap, and not doing anything don't sound right to me.
 
The connector is a standard. However, the way that Nvidia handled the issue is not exactly the best in my opinion, Here at least we see that Intel learned from the lesson and is trying to improve the situation and make the connectivity less iffy. Nvidia's last statement was that the issue was a user problem, full stop. The cards that Nvidia are selling are not exactly cheap, and not doing anything don't sound right to me.
There was no issue when nvidia used a very similar connector for the 3xxx series, the 4 sense pins was intel's whole thing to add it to atx and to pci-sig. 12VHPWR is intel's standard but it's nvidia's fault for using it?
 
Last edited:
I say revision 4 will be the real winner hehe. You know when they revert back to 8pin connectors all around! Aaaahahaha

Yeah, I still don't get the motivation for going with such a complicated connector, just pump the power spec on the 8 pin molex, it can easily handle it

Seriously tho, at least they're trying. Altho, if I were Nvidia, I would want to be directly involved with any revisions to MY connector.

It's not their connector, it's from PCI SIG, which Intel is also a part of.

There was no issue when nvidia used a very similar connector for the 3xxx series, the 4 sense pins was intel's whole thing to add it to atx and to pci-sig. 12VHPWR is intel's standard but it's nvidia's fault for using it?

There were as much issues as there are now (1 or 2 users who didn't connect it properly is not a real issue). The sense pins are from pci-sig and always were, just like they exist with 6 or 8 pin connectors. Intel just handles the ATX power standard because they were the only ones to bother to.
 
Sounds like the standard Engineering process: 25% of the time spent designing the product, then the other 75% spent re-designing it to try & prevent morons from misusing it in ways no normal person would even consider. There's a reason why these connectors are designed to click once they're seated properly...

design 101: if the consumer does not understand your product, you designed it poorly.

the designer makes the product for people to use, its their job to make sure its usable and understandable, its their failure if it isnt.
 
Yeah, I still don't get the motivation for going with such a complicated connector, just pump the power spec on the 8 pin molex, it can easily handle it
The age old, create a solution to a problem that doesn't exist.
It's not their connector, it's from PCI SIG, which Intel is also a part of.
Yep yep. Point being, there's only one multi billion dollar corporation (headed by a narcissistic megalomaniac) using this pre configured connector, that juuust might want a little bit more input this time around. Considering how well the first iteration went over I'd say that's a safe bet. Huang isn't known for his ability to allow "great ideas" to go untouched afterall.
 
I propose this plug instead
49YX35_GC01
 
Back
Top