starslab

I already own one of those fiber-hdmi cables. Brilliant, but sometimes doesn't interoperate with DVI devices using passive DVI -> HDMI adapters. I've no idea if it has any copper conductors for HDMI power, though one end is labelled for the source and one for the display, suggesting that however it's designed it's not bi-directional.

I'd love a DisplayPort KVM, but not every device that comes across my bench has a DisplayPort output, and those few that have DisplayPort but no HDMI can be accommodated with one of those commodity DisplayPort -> HDMI adapters. This situation is actually getting worse over time, not better, as many modern devices and laptops are skipping DisplayPort in favor of USB-c alt-mode.

This issue has actually been going on through a monitor change on my testbench. It has happened with a Samsung SyncMaster 204T though my KVM switch, an HP ZR24w through my KVM switch, and the ZR24w directly connected. I don't think this is an issue with the rest of my equipment.

This electrical was done about 15 years ago, by a ticketed electrician. One of those $5 plug testers indicates all is well, and I have no reason to believe there's any issue here.

By almost pure coincidence, I have an MSI PRO X870-P motherboard on order. I'm looking forward to seeing if this same 3V3 leakage issue is present on this board too.

> scratch that we needed smaller ;-)

Well... Needed smaller in terms of resistance, but needed bigger in terms of power rating, in the interests of not catching fire.

12 ohms brings the rail down to 1.47 volts, still no SSD. 6 ohms is enough to finally break/trip whatever circuit is allowing this situation, bringing the rail down to 0v in power-off. Of course, that's almost 2 watts of constant draw during the power-on state, so not a long-term solution.

Hi! I'm the OP from the Level1Techs thread.

That HDMI power has some grunt behind it. During power-off state with that 1.90v phantom voltage, I put a 48ohm resistor between 3V3 and ground, the phantom voltage only dropped to 1.80v, and the SSD still didn't work when I powered the machine back on.