• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

be quiet! Dark Power PSU Owner Reports Melting of 12VHPWR Connector

I hope this connector will turn out as an engineering mistake...
Just make GPU's with 6-8 PIN connectors how it always was, since this is always working properly...
What ain't broke don't fix it..
It's time for those big ass ugly/bulky connectors to go, but I'll gladly keep using them until there is a stable version that replaces them. I don't like all those wires coming out of the side of my GPU...I got 3 of those shooting out the side of my card now. Thankfully Cablemod had some halfway decent looking ones...
 
It's time for those big ass ugly/bulky connectors to go, but I'll gladly keep using them until there is a stable version that replaces them. I don't like all those wires coming out of the side of my GPU...I got 3 of those shooting out the side of my card now. Thankfully Cablemod had some halfway decent looking ones...
If your using the 12 / 12+4 patch cable that came with the GPU or a 12+4 cable from a 3rd party like CableMod, make sure it don't have the female pins with dimples in the connector, if it does, get it changed out for one that uses the correct female pins as per Intel's specs (see my 1st post one page back for info on this).
 
Last edited:
12vhpwr is a gift that keeps on giving...
 
Whatever how it happened exactly, is there any evidence of melting with 8-6 PIN issues?
There sure is evidence of that too.

The connector could have been better designed / manufactured, that's easy to agree to, hec just painting the part of the connector that should all be inside once it's latched a fluro colour would be a start, if you see the colour, it's not all the way in. But it does seem like at least the vast majority of issues was improper insertion.

And for those wishing it, don't hold your breath for it being backpedaled and going away, if anything it'll get minor revision/s to iron out cases of this happening, but it's extremely unlikely to be rolled back/unlaunched now. After all there are hundred's of thousands, perhaps even millions now with ATX 3.0 PSU's and/or cards using the 12VHPWR connector.
 
Well, it seems that over 8 Amperes per pin is just little too much (600Watts/12Volts/8pins).

Just compare that to the standart 8-pin PCI-E power connector: 4 Amperes per pin (150Watts/12Volts/6pins).

There is no miracles in the field of the physics, just bare consequences...

P.S. But, of course, new shiny 12VHPWR connector is looks so good! And, of course, no more mess with a lot of connectors and messy, thick wires...

Maybe...

Additional reason, why those 12VHPWR connectors are melted can be wire gauge.

I have no idea which AWGs used in this 12VHPWR cabling, but I'm will not surprised if it's 18AWG.
16AWG and over is way too chunky to put in this tiny "engineering marvel".

According to this resource:


maximum permissible current per wire in this case is 4.9 - 4.2 Amperes (7-24 - 25-42 cores).

Add to this that fact that those pins aren't welded or soldered to wires - they are crimped.

It's ok for 110 - 230 Volts, or just signalling, but for 12 volts, over 5 Amperes?

Hmmm, isn't there is place where Ohm law starting to have a huge workout due to resistance?

18AWG, crimped connection, over 8 Amperes per that - aren't we are witnesses of just pure marketing idiocy over engineering?

Or additional super-duper couple of cents economy?

Why I'm not surprised...
 
I'm really glad I went with a humble midrange card and avoided this connector.
 
Well, it seems that over 8 Amperes per pin is just little too much (600Watts/12Volts/8pins).

Just compare that to the standart 8-pin PCI-E power connector: 4 Amperes per pin (150Watts/12Volts/6pins).

There is no miracles in the field of the physics, just bare consequences...

P.S. But, of course, new shiny 12VHPWR connector is looks so good! And, of course, no more mess with a lot of connectors and messy, thick wires...

Maybe...

Additional reason, why those 12VHPWR connectors are melted can be wire gauge.

I have no idea which AWGs used in this 12VHPWR cabling, but I'm will not surprised if it's 18AWG.
16AWG and over is way too chunky to put in this tiny "engineering marvel".

According to this resource:


maximum permissible current per wire in this case is 4.9 - 4.2 Amperes (7-24 - 25-42 cores).

Add to this that fact that those pins aren't welded or soldered to wires - they are crimped.

It's ok for 110 - 230 Volts, or just signalling, but for 12 volts, over 5 Amperes?

Hmmm, isn't there is place where Ohm law starting to have a huge workout due to resistance?

18AWG, crimped connection, over 8 Amperes per that - aren't we are witnesses of just pure marketing idiocy over engineering?

Or additional super-duper couple of cents economy?

Why I'm not surprised...
All companies should be complying with Intel's spec's / guidelines with ref to the 12VHPWR connector that is in the ATX 3.0 specs as of January 2023 or later.
Prior to this, Nvidia was / is following Intel's guidelines with the RTX 30x0 FE series and RTX 3090 Ti and with the RTX 40x0 FE series with the 4 spring female pins in the connectors (cable) that comes with the GPU, some Nvidia partners on the other hand did not and they used the female pins with dimples for some reason, this is also evident here with this PSU where the 12VHPWR cable is using the WRONG FEMALE PINS per the specs / guidelines.

12vhpwr1.png


12vhpw2.png
 
Last edited:
All companies should be complying with Intel's spec's / guidelines with ref to the 12VHPWR connector that is in the PCI 3.0 specs as of January 2023 or later.
Prior to this, Nvidia was following Intel's guidelines with the RTX 30x0 FE series and RTX 3090 Ti and with the RTX 40x0 FE series with the 4 spring female pins in the cables that come with the GPU, Nvidia partners on the other hand did not and they used the female pins with dimples for some reason, this is also evident here with this PSU where the 12VHPWR cable is using the WRONG FEMALE PINS per the specs / guidelines.

View attachment 301298

View attachment 301299
Perhaps recommended was the wrong word. They should have used required instead.
 
Whatever how it happened exactly, is there any evidence of melting with 8-6 PIN issues?
Yes. It's infrequent but it does happen if you look.

This definitely isn't the first melted connector since GN's video on the subject and won't be the last.
I am furthermore unconvinced that this is widespread or even a cause for concern, statistically speaking.
 
If your using the 12 / 12+4 patch cable that came with the GPU or a 12+4 cable from a 3rd party like CableMod, make sure it don't have the female pins with dimples in the connector, if it does, get it changed out for one that uses the correct female pins as per Intel's specs (see my 1st post one page back for info on this).
You quoted me...but I have a 3 x 8pin - 6900XT. After re-reading what I wrote I can see how one could take it as such, but I don't own a Nvidia card. I thought it was clear based on what I quoted.
 
You quoted me...but I have a 3 x 8pin - 6900XT. After re-reading what I wrote I can see how one could take it as such, but I don't own a Nvidia card. I thought it was clear based on what I quoted.
Ah ! Sorry, miss-understood !

All good tho.....
 
Yeah a house fire,
Has any actual fire whatsoever been attributed to the 12VHPWR connector? I'm yet to see that, and I know it's a risk, just about every electrical appliance has risks, especially when user assembly is required (hello melted 6/8 pin cables that have melted), but it's one that just keeps getting said with literally no cases of it happening.
 
Has any actual fire whatsoever been attributed to the 12VHPWR connector? I'm yet to see that, and I know it's a risk, just about every electrical appliance has risks, especially when user assembly is required (hello melted 6/8 pin cables that have melted), but it's one that just keeps getting said with literally no cases of it happening.
Not seen any reported on the internet / news, but that don't mean that there has not been a fire caused by this connector.

We also don't have an exact number of cases with reference to these connectors melting, we only know about the ones that get posted on forums / social networks like Reddit etc.

Also, on top of this, I have yet to see an Nvidia RTX 4080 / 90 FE model with this melted connector other than those that have been tested like Igorslab, GamersNexus and a few others that did controlled tests.

I have not seen any RTX 30x0 FE series or RTX 3090 Ti with this melting connector issue either, they all use the 12VHPWR connector (1st gen connector).
*Most seemed to be under the impression that this connector was Nvidia's design, it's not the case, they was just the first company to use it.*
 
Last edited:
I can't see something so tiny on my PSU. Not even using my phone. dimples or springs. especially before purchase
 
I can't see something so tiny on my PSU. Not even using my phone. dimples or springs. especially before purchase
It's not the PSU, but the cable (12VHPWER), the connector that plugs into the GPU / PSU.

Example - the cable that came with my RTX 3090 Ti (see image below which is using the 1st gen version of this connector - missing the small 4 pins on the connector, same with the Nvidia 30x0 FE cards)

Rows 1 & 2 here in the image, these are the connectors that come with the RTX 30x0 FE / RTX 3090 Ti cards, none of the female pins have dimples.

Row 3 These are the cables that came with the RTX 4090 cards that have the melting issue, they have dimples on the inside of the pin, the other thing is the gap top and bottom of these female pins, they are different.
Nvidia's 4090 FE / 4080 FE, the cable that comes with these cards, the female pins in the 12VHPWER connector do not have dimples on the inside (the ones I've seen that is).

RTX3090TiZotac.png
 
Last edited:
  • Like
Reactions: N/A
LOL 'Veteran' system builders. I see 'veterans' change their BIOS and brick their card, I see 'veterans' disable essential Windows services and then wonder why shit doesn't work as planned, I see them disable page files for no apparent reason.

Give me a break :D

Connectors need to be foolproof.
Sad reality...
 
Ok, maybe dimples vs springs in the contacts can be a reall reason for overheating and melting scenarios.

But, what about wires thickness?

Did all 12VHPWR cables strictly follows AWG16 recommendation by Intel ATX specs?

I doubt so, but can't deny or confirm, mine 3080 is by good old 3x8 PCI-E connectors.
 
Did all 12VHPWR cables strictly follows AWG16 recommendation by Intel ATX specs?
Based on the tests Igorslab and GamersNexus did, yes, I believe so with the cables being AWG16.
 
I'm sure bequiet sold 1000s of these power supplies and we see like 1 report that something's wrong.

It dont matter if they sold 100.000 PSU's and only one report, or 10 reports. One is too many knowing it can set your house on fire.

These things are extremely sensitive.
 

Right, here are burned PCIe 8pin

gpu_power_supply_cable_melted_using_3090_hof



And for those wishing it, don't hold your breath for it being backpedaled and going away, if anything it'll get minor revision/s to iron out cases of this happening, but it's extremely unlikely to be rolled back/unlaunched now. After all there are hundred's of thousands, perhaps even millions now with ATX 3.0 PSU's and/or cards using the 12VHPWR connector.

This isn't a problem since you could have 2x PCIe to new connector adapter make by manufacturer going right from PSU. The problem are the distributed GPUs out there.

 
I think we can see that the whole upper layer melted so may be a bad connection, on the second picture we can see the cable being curved btw, as if it was pulled..and pulled exactly where it melted.



Previously, the older 8 pins PCI cables had a big margin for errors, so it means nothing to be a "veteran" it's a new standards, everyone has the same experience with it....

The 12 vhpwr cable has a design that forbids any tension or badly unplugged cable, should have been better designed, but nothing proves that those cables "just melt" and that they were perfectly plugged in when it happens, back up your claims...

The only thing that has been proven and reproduced by third parties like GamerNexus, is the badly plugged cable for now.

I'm against this botched standards but people need to stop believing they are experts and did no mistakes, as small as it was, there are more than 100-200K 4090 alone sold.

Best thing to do is design a 12vhpwr V2 which latches on far better, maybe a dual latch design

The tension issue has largely been disproven in field testing, you need to actually damage and crumble the female to some degree while applying the tension to cause an issue.

Make connectors out of metal so they don’t melt. ;)
They need to be capable of melting per design as a safety aspect, metal would heat into perpetuity and then actually be a fire hazard, the melting nylon connectors are not.

Has any actual fire whatsoever been attributed to the 12VHPWR connector?

No, the nylon connector has an intended melting point to prevent hitting temperatures where plastics would ignite and burn.
 
Last edited:
Maybe it's time these manufactures was told to get up to speed and use the correct female pins in their connectors (the cable).

You can clearly see that the female pins used in the OP's post are the wrong type to use here.

OP's connector - female pins with dimples, 12v burnt pins will have these dimples as well.
View attachment 301284

View attachment 301278

(Note to Techpowerup editor - you need to proof read this article and correct it, male / female in wrong places.)

Noticing that trend as well, manufactures keeping user 3-dimple style, which is fine for anything below a 4080, but 450W+ should have 4-spring.

I have a 4070ti that I put super glue on the power connector. Sure its not as hot as 600w 4090 but the sudden connector power melting drama got me scared that time.
GG, now you can't pull it out when it does melt :roll:
 
Back
Top