That's some scary stuff, glad you're okay and no serious harm to your gear!

Hey Hashraptor, I completely understand your challenge with cables and the right supplier. I have found that parallel miner has 16 awg cables (labeled on the wire) for their kits and even individually. I try to get the breakout boards with the amp meter so its an easy wattage calculation. I am starting to write the watts on the server supply and spread the cables evenly on the board. Thanks for the reminder to visually inspect again. I am using the emporia panel meter(amazon) and can see the watts and amps on the overall circuit from anywhere most of the time(wifi flakes out rarely for me). Just my practice with my few rigs. Happy mining and hope you continue to recover.

Wow, so glad that it was only a burned out breakout board! Thank you for the detailed troubleshooting walkthrough. I particularly like the fact that you walked through a root causes analysis and highlighted the importance of putting HiveOS in maintenance mode. It's not something I always do, but I sure will from now on!

Great video man! You have been lucky, good to hear that everything went well. Keep mining!

Spliter cable on 1 card only good working

thanks for sharing ur experience

If I have lights on the riser I will take it off and put it back on before replacing it. Noticed your breakout board was pretty close to the side of your metal frame. Maybe the solder joints on a humid day jumped to ground. After seeing a few burnt connections like this, now I don't trust the plastic connector to hold the pins/wires in place while I push the connectors together. I grab the wires and push down instead of grabbing the connector. Taking a connector apart, it's not much keeping the pins inside from pulling out and thus possibly making a poor connection.

Exact same thing happened to me, but in my case the breakoutboard got so heated that it melted from the bottom of the board. I suspect the board had something faulty between the rubber layer, reason is because it was a new board and was only running for 8 hours. Lesson learned after that was to check the new board and also spread out the connection evenly in terms of power hungry GPUs spread out instead of having all on one side of the breakoutboard.

I had a Nerd Gearz breakout board it was only a few months old and it did the same thing. I was running P104s on smos.

Yep had a similar thing happened with a breakout bord like a month a go . Very scary

What's the best brand name power strips to buy that are reliable? I get some of these I get some of these power strips with brand names cheapie.

Just had the same situation happen on my 3070ti rig. The wife woke up and complained about a plastic burning smell in the basement........

recently i started getting the SAME burning on my breakouts, quality has dropped and cable thinness is going on. i crisped my cables too and the solder jumpered on the back chip

Is that a board full of 6 pin connectors? Wouldn't 8 pin connectors be safer? (6 pin = 75w, 8 pin = 150w)? I've never used these server power supplies but do they make them in 8 pin format?

@HashRaptor if you are running 80 percent power rule on the psus , then should that have happened even if the OC settings went to the wrong card? I am thinking it was just a faulty board and nothing to do with your OC and maintenance mode.

WiFi smoke detector. You're welcome :)

Great, now I gotta go buy another smoke detector...

check your risers ,they may got contact to frame because they on aluminium and rubber may damaged

The ONLY way I see how your clock settings could have caused this to burn up, is if you are not power budgeting properly to begin with. For example, if you are power budgeting 3060s to receive 110 watts that you'd probably operate them at, and not at their 170 watts TDP... You should always power budget the full TDP, even if you plan to only utilize a percentage of it... Otherwise, something like an over aggressive core or memory clock setting would just cause the GPU to crash, and not to burn up... You should be able to safely run every GPU at a 100% power limit, the wiring should be able to handle at least 100% of the GPU TDP... If you are only wiring everything to where you plan the GPU to operate at only, and not the full TDP, then if something ever occurred where the GPU get hit asking for full TDP and full TDP is above the operational power needs, then you could burn something up... But injecting otherwise for example a 3080's settings into a 3060, that shouldn't burn a GPU up... Just cause it to crash or freeze up... I don't think your issue was with your clock profiles at all, not unless you got a larger picture issue with your power budget and you are not budgeting properly... I will also say, those black and yellow cables are no bueno imo... They in theory should be ok, but... I've had a few of those burn up despite utilizing them properly and everything... I think the pins in them are poor quality, or the conductor of the wire or something... I even saw a video where Red Panda Mining had an issue with some, and my brother who also crypto mines had issues once... I only use 6 pin to 6 pin PCIe/VGA power extension cables any more in those yellow black cables... That and only used to power risers... Otherwise I use genuine power supply cables from the manufacturer anymore... I've lost all trust for the cheaper aftermarket cables... And I don't even use splitter/breakout boards... I only use (semi) modular ATX power supplies, and the cables they come with, 3 GPUs per PSU (850 watt PSUs), except the PSU powering the motherboard, that PSU only gets 2 GPUs, unless also if I am powering a higher end GPU like a 3080, then the PSU powering the 3080 only gets the 3080 or maybe the 3080 and a 2nd GPU, no 3rd GPU... But most of my rig is 3060tis and 3070s, I got a single 3060v1, and two 3080s... Everything else is 3060tis or 3070s.... So if the full TDP of a 3060ti is 200 watts, I can safely put three 3060tis on a single 850 watt ATX PSU... Only using 70% of the rated power ***IF*** I were to ever demand full TDP, but of course I dial down the power limit to each GPU from the TDP 200 watts, to 130 watts... So the reality is, I am only demanding 390 watts of an 850 watt PSU, or 46%... But that said, I have the buffer if for some reason each GPU called for full TDP, and in addition the 80% on top of that... There is only enough modular space on the PSUs to connect 3 GPUs, and I would never put more than 1 GPU on a single cable, so I usually power both the riser and the GPU on the same cable unless that puts me over the power budget of that said cable... Then if that is the case (or in some cases a GPU needs more than 1 power connection), then I use molex power from the PSU to power the riser... Molex should be good to 100 watts, and the riser should never ask for more than 75 watts, tops... And then I will place more than 1 PSU to 1 motherboard splitting the ATX cable so that the PSUs will all turn on and off at the same time, or I also mostly use ASUS B250 Mining Expert Motherboards where I can connect 3 PSUs directly to it... I run 8 GPUs per motherboard, so 2 PSUs will run 3 GPUs each, and 1 PSU will run 2 GPUs and the CPU/RAM/SSD/MoBo, etc... 3 PSUs, 8 GPUs, 1 MoBo...

Have a nice recovery from the human malware my brother

First any version of hive works. The cables terminals are to small to support the power draw. I'm referring to the metal terminals that are crimped onto the wire. Inside the black connector.The cables with all 6 wires attached to each other are notoriously mislabelled as 18 awg. If you cut a real 18 awg and one of those then compare the thickness of the copper. You will see what i mean. I agree that settings get unsynced when 1 gpu drops.That does not mean a 3070 would receive the power that goes to a 3080. if you have 70 percent power limit on a 3080. then the 3080 drops off. The 3070 would run at 70 percent power limit for the 3070. The hp 1200 sticker says 900 watt max at 110-120 on the psu. That is not your issue as the input voltage has no effect on the 12v output voltage.The psu would power off if you exceed the 900. If you are going to make video then you should have a better understanding of the subject matter.

“The world today has 6.8 billion people. That’s headed up to about nine billion. Now, if we do a really great job on new vaccines… we could lower that by, perhaps, 10 or 15 percent.” ~Bill Gates