Hacker News Reader

Comment by unwind | Hacker News Reader

Interesting that the connector (12VHPWR) in question seems to be safe for only 40 (forty) cycles. That's really not a lot, as these things go.

Of course it's an internal connector and not something you're supposed to plug/unplug all that often, but I still think it sounds like a really low number.

View on HN

Keep in mind that 40 cycles is the minimum. 99.9% of the plugs will be fine after 40 cycles and probably will do 400 without issue. But you're outside what the manufacturer guarantees it'll last.

And by experience, 40 is absolutely plenty. Almost all plugs in a computer are likely to only experience 10 cycles at most (and half of them only two; during setup and during EOL).

All internal PC connectors are only rated for 20-50 cycles or so.

Do you have a citation for that, as I'd love to refer to it.

They are absolutely right and I have no idea why they are getting downvoted. Last time this was posted someone linked industry specsheets for all common PC connectors, from 8-pin GPU power plugs, to the good old MOLEX plug, and they all were rated for about ~30 connections in their lifetime.

Edit: there you go, found it - normal bog standard 8-pin PCI-e connector, rated for 30 connections:

https://www.molex.com/webdocs/datasheets/pdf/en-us/045586000...

Edit 2: found the quote from another article too:

WCCFTECH:

"Update: We have confirmed with NVIDIA that the 30-cycle spec for the 16-pin connector is the same as it has been for the past 20+ years. The same 30-cycle spec exists for the standard PCIe/ATX 8-pin connector (aka mini-fit Molex). The same connector is used by AMD and all other GPU vendors too so all of those cards also share a 30-cycle life. So in short, nothing has changed for the RTX 40 GPU series."

I've never heard of other connectors melting though. They might be rated for only so many cycles, but how many real cycles are required for them to melt like these adapters?

Outside of the motherboard and GPU connectors, no other connector in a PC draws these levels of insane amount of current. Basically that's DVD/BD drives and spinning rust HDDs.

"Molex" connectors (the 4-pin ones used for fans and DVD drives etc.) are notorious for melting, and molex-to-sata adapters especially. It wasn't unheard-of for PCIe connectors to melt either.

As Mastax said, I've had multiple molex connectors melt over the decades across a fleet of PCs. I've seen one SATA power connector melt. I've had numerous of the 4 pin aux power connectors to motherboards melt. But note this is across fleets of thousands of PCs.

In my early days I saw a Pentium process melt the socket off the board. I don't understand how it didn't trip the power before it got that far.

Thanks, i googled but wasn't finding anything useful.

The bigger callout is the safety margins that were assigned to each plug. The new connector runs pretty much right at rated margin minus 84W (6pins * 12V * 9.5A = 684W) for losses at full spec. The old connector was rated for 324W (3pins * 12V * 9A) but only specced for 150W or approx 4.2A per pin. So there was a massive safety margin there. Also the larger pin spacing means that the connections were better able to dissipate any heat that did show up for the most part anyway. The much tighter pin spacing on the new connection pretty much ensures that any dongle like this will be extremely hard to cool because of the fact the wires must come together in the actual connector.

Long story short: IMO the new connector makes me extremely nervous, running anything that can cause fire within ~14% of rated is not a wise idea unless you're doing Aerospace and have no other options. Even then they generally have extremely high QA and assembly verification, something I'm not expecting from computer OEMs.

Reply on news.ycombinator.com