Sunday, September 13th 2026
ASRock's TempGuard Failed to Stop a PSU-Side Connector Meltdown on a $15,000 RTX PRO 6000 Rig
A repair shop technician has reported another melted 16-pin power connector on an ASRock Taichi TC-1650T PSU, paired with a $15,000 NVIDIA RTX PRO 6000 Blackwell Workstation Edition GPU. The former ships with native 12V-2x6 cables carrying ASRock's own NTC temperature sensor. According to the technician's Reddit post, the connector failed catastrophically on the PSU side and melted into the socket itself, even though both ends of the cable were fully seated. The GPU reportedly survived, and the damage was confined to the power supply side. The technician also says the sensor is mounted on the wrong side of the cable, on the ground line rather than the 12 V side it's supposed to watch, which would explain why it apparently never caught the rising heat in time.
ASRock added the NTC sensor and TempGuard system to its 12V-2x6 cables specifically to shut a system down before a connector melts. This time the system failed to catch the rising temperatures, but it has worked in the past. TempGuard reportedly caught a shunt-modded RTX 5090 pulling over 1,300 W, saving its power connectors from melting. On the flip side, this isn't the first time this exact GPU and PSU pairing has run into trouble either, as a similar melting incident on the same combo surfaced back in March. If the sensor placement is really off on at least some cables, that's a bigger concern than one bad unit, especially when paired with a card that costs as much as a used car.
Source:
Reddit
ASRock added the NTC sensor and TempGuard system to its 12V-2x6 cables specifically to shut a system down before a connector melts. This time the system failed to catch the rising temperatures, but it has worked in the past. TempGuard reportedly caught a shunt-modded RTX 5090 pulling over 1,300 W, saving its power connectors from melting. On the flip side, this isn't the first time this exact GPU and PSU pairing has run into trouble either, as a similar melting incident on the same combo surfaced back in March. If the sensor placement is really off on at least some cables, that's a bigger concern than one bad unit, especially when paired with a card that costs as much as a used car.
169 Comments on ASRock's TempGuard Failed to Stop a PSU-Side Connector Meltdown on a $15,000 RTX PRO 6000 Rig
I would love to know this crucial information. And you?
I guess they fail to realize that if the 8pin was introduced today, the moment a card melted its connector it would make first page news as well.
When I put them all on my 400 Watt PC goes to sleep instantly starved to a coma. Strain Relief police not around so all is glowing calmly.
Or the fact many companies have to attempt at fixing the problem. If it's not an issue then why do advanced monitoring tools like the wireview and ampinel exist? I don't recall 8 pin molex having those tools.
But some people have to resort to name calling and searching for an 8 pin problem that in reality is much less of an issue than melting 12 pins.
Oh wait...what sorcery is this? HOW? 8pins didn't need it. There was no monitoring device for them. :roll:
Totally didn't monitor power and amperage cause 8 pins didn't need it. It must be made by AI
No articles about 8pins melting either, and no reddit threads.
www.tomshardware.com/pc-components/gpus/proven-8-pin-pcie-plugs-arent-immune-to-melting-thermal-grizzly-wireview-adapter-burns-out-on-radeon-rx-7900-xtx
But oh no, one report of a 8 pin GPU melting, it must mean they all melt. Welp, I guess I'd better throw my 8 pin using GPU in the trash for a 12 pin GPU and then babysit it with a wireview, thermal cable and get an E-ATX case so I can use a strain relief. But first I need to get an EE degree so I know how to plug in a cable correctly.
Trust me, it’s much better in here not seeing all the shitposting. Even Hectate isn’t amusing anymore.
When you shunt mod your GPU, Nvidia or AMD you can use the WireView for 8 pin is useful.
From same article:
"While we are on the subject of 8-pin power connectors, just days ago we saw a heavily modified RTX 5090 successfully running on traditional 8-pin connectors. The modders replaced the card's original 16-pin connectors with three 8-pin connectors soldered directly to the PCB, using heavy-gauge cables and modified sense pins. The card reportedly pulled as much as 900W and reached 3,400 MHz, demonstrating how capable 8-pin connectors can be when properly implemented"
Key words properly implemented, where 12V lacks big time.
You don't need "heavy-gauge cables" 14-16AWG is enough, soldering to the PCB that is the weak point, eliminated. Enough service area.
If I'll be you, I'll give up trying to bring 8 pin into a bad light, you actually doing the opposite. Thank you.Tactical retreat, very wise.:toast:
This is getting boring.... :D
When you shunt mod your card to break benchmarks or hash it you need those monitoring tools, you simply overpass the max power of your card so the tech specs of 8 pin.
Nitro cooling your GPU core is 1 step the other 1 is power rails and everything that comes with that, heat and stress on every component including the cables.
8Pin connector can also fail, no-one said it is indestructable. It is just that you took personal news that someone modded RTX 5090 to run on 3 8-pin better than on 12HPWR, and started your little crazy crusade. Still, 2x 8pin and 3x 8pin existed for ages, and number of reported cases of cards burning is... miserable. You can count it easily.
We have news about 12HPWR failing and we will have it in future much more. In retail, in prebuilts, on test benches... it fails and fails again. On next gen, we will see huge changes in electrical design of cards and probably multiple 12HPWR to prevent issues like this, but you will probably twist that too.
Shilling for tech company is pathetic. At least I hope you three do it for money...
12vh can also fail, no one said it is indestructable. It is just that you took it personally that 8pins are worse than 12vhp and started your little crazy crusade. Still, 12vhp has existed for 4 years and number of reported cases are less than 100, out of 137 millions of cards being sold since 2022. That is miserable.
We had news about 8pin failing but thankfully not anymore since 90% of the market has switched to 12vhp. Since 2022 we haven't gotten a single reported case of a low power card melting its connector like we had with the 8pins.
Shilling for tech company is pathetic. At least I hope you do it for money...
Ah, fresh one, 9070xt. Surely it was using the 12vhp that's why it melted the connector, right? RIGHT?
The fact is the 12vhpwr connector in use for graphics cards is newer than 8 pin molex, yet we're hearing of failing connectors more often in a short span of time, with all of these cables and tools to stopgap the connector from melting resulting in an RMA, it fails in pre-builts or DIY system regardless. This campaign to defend a flawed standard that has been clearly cost cut from the original implementation, while moving the goalposts to say 8 pin molex is worse, is weird to me and I wonder what the reasoning is aside from some people refusing to admit their favorite company could have possibly made an error.
Solid evidence is solid! Not even a molecule of user error was found.
I have to admit defeat, what more can I say? 🙃
Hard gold plating has to be at least 40 micro inches on all connectors Female PSU, Male PSU, Male GPU and Female GPU, a combined cost/effort and nobody will do that.
HardGold plating only one end is even worst than no gold plating at all, as it will bite from the tinned end very fast accelerating oxidation and increasing resistance, but gold plated cables plugged in tin plated connectors are every where, we can't change that trend.
The melting starts from the base of the pins, pins are moving and then during cold cycle the plastic hardens and catch the pin in the vicious position. From this is just a matter of time till it melts .
The way it destroy itself and with each vicious connection is accelerating the process IMO can't even handle 400 W cards.
4000 series had a fair share of 8 pin including the popular 4060 which sold very well, entry lvl 4070 had 8 pin as well. So a huge chunk had 8 pin out of that 137milions you talking about.
I don't give a damn crap of who exactly comes with 12V connector on their GPUs. Can't give a rat a$$ is Intel, AMD or Nvidia a bad design remains a bad design, meaning, me the customer will lose money. No fe%%in way.
12V is a bad design unfit for purpose IMO.
Sapphire and Asrock comes with those, not buying those. Period.
I'll need an Nvidia card anyways next year for a sibling certain apps, but prolly will be one in the range of 300W. If voltages will drop and amps will increase with the PSUs 12v cable than I'll return the bad thing for another one, rinse and repeat. Not buying special cables.