GE has a paper about the power conversion design, but it doesn't mention the unit to rack electrical and mechanical interface. Liteon is working on that, but the animation is rather vague.[2] They hint at hot plugging but hand-wave how the disconnects work. Delta offers a few more hints.[3] There's a complex hot-plugging control unit to avoid inrush currents on plug-in and arcing on disconnect. This requires active management of the switching silicon carbide MOSFETs.

There ought to be a mechanical disconnect behind this, so that when someone pulls out a rackmount unit, a shutter drops behind it to protect people from 800V. All these papers are kind of hand-wavey about how the electrical safety works.

Plus, all this is liquid-cooled, and that has to hot-plug, too.

[1] https://library.grid.gevernova.com/white-papers-case-studies...

[2] https://www.youtube.com/watch?v=CQOreYMhe-M&

[3] https://filecenter.deltaww.com/Products/download/2510/202510...

Then I started routing ethernet with PoE throughout my house and observed that other than a few large appliances, the majority of powered devices in a typical home in 2026 could be supplied via PoE DC current as well! Lighting, laptops, small/medium televisions. The current PoE spec allows up to 100 W, which covers like 80% of the powered devices in most homes. I think it would make more sense to have fewer AC outlets around the modern house and many more terminals for PoE instead (maybe with a more robust connector than RJ45). I wonder what sort of energy efficiency improvements this would yield. No more power bricks all over the place either.

See e.g. https://www.dell.com/support/kbdoc/en-us/000221234/wiring-in...

But what about availability? If you ask most of our users whether they’d prefer 4 9s of availability or 10% more money to spend on CPUs, they choose the CPUs. We asked them.

There are a lot of availability-insensitive workloads in the commercial world, as well, like AI training. What matters in those cases is how much computing you get done by the end of the month, and for a fixed budget a UPS reduces this number.

800 volts DC, at the megawatt power supply levels, implies fault impulses of more than a megajoule. Google tells me that's about 2 hand grenades worth of boom. That's an optimistic lower bound.

The resulting copper plasma cloud is a burn and inhalation hazard, along with the overpressure.

Let's say you get a 10 kiloamp fault current, this will then induce voltages everywhere you don't want it to go. If all the interconnects are fiber, that's really not a problem, but you have to have everything EMP shielded if you don't want boards popping randomly after such an event.

The "efficiency" of removing the extra power conversions also removes filtering and surge suppression. It's entirely possible that one power supply over-voltage takes out half of your racks. The MOSFETs used tend to fail closed instead of open, making failures far worse than a simple outage.

Very smart people are making very smart mistakes.

Yes, of course both of those things are true, and yes, some data centers do engage in those processes for their unique advantages. The issue is that aside from specialty kit designed for that use (like the AWS Outposts with their DC conversion), the rank-and-file kit is still predominantly AC-driven, and that doesn't seem to be changing just yet.

While I'd love to see more DC-flavored kit accessible to the mainstream, it's a chicken-and-egg problem that neither the power vendors (APC, Eaton, etc) or the kit makers (Dell, Cisco, HP, Supermicro, etc) seem to want to take the plunge on first. Until then, this remains a niche-feature for niche-users deal, I wager.

Many datacenters I'd been to at that point were already DC.

Didn't think this was that new of a trend in 2026, but also acknowledge I did not visit more than a handful of datacenters since 2007.

It just seemed like a undenyably logical thing to do.

It is silly to have AC to DC converters in all of my wall connected electronics ( LED bulbs, home controller, computer equipment etc )

We have some old ceiling and exhaust fans, but I know those can be replaced. Our refrigerator is AC, but extended family with an off-grid home has a DC refrigerator that cycles way less, probably due to multiple design factors but I’m sure the lack of transformer heat is part of it. I’m not as sure about laundry machine or oven/cooktop options but I believe those are also running on DC in the off-grid home without inverters.

Most of these AC appliances also have transformers in them anyway for the control boards. It seems kind of insane to me that we are still doing things this way.

Installing a ceiling fan used to be treacherous and so heavy. Also loud and buzzy after installed. Now the fans in these things are so lightweight and easy.

seeing the same in many more areas (lighting, etc)

For 800V DC, a simple UPS could interface with the main supply using just a pair of (large) diodes, and a more complex and more efficient one could use some fancy solid state switches, but there’s no need for anything as complex as a line-interactive AC UPS.

  Hard as a rock!
  Well it's harder than a rock!

I always thought AC’s primary benefit was its transmission efficiency??

Would love to learn if anyone knows more about this

- Three conductors vs two, but they can be the next gauge up since the current flows on three conductors

- no significant skin effect at 400Hz -> use speaker wire, lol.

- large voltage/current DC brakers are.. gnarly, and expensive. DC does not like to stop flowing

- The 400Hz distribution industry is massive; the entire aerospace industry runs on it. No need for niche or custom parts.

- 3 phase @ 400Hz is x6 = 2.4kHz. Six diodes will rectify it with almost no relevant amount of ripple (Vmin is 87% of Vmax) and very small caps will smooth it.

As an aside, with three (or more) phase you can use multi-tap transformers and get an arbitrary number of poles. 7 phases at 400Hz -> 5.6kHz. Your PSU is now 14 diodes and a ceramic cap.

- you still get to use step up/down transformers, but at 400Hz they're very small.

- merging power sources is a lot easier (but for the phase angle)

- DC-DC converters are great, but you're not going to beat a transformer in efficiency or reliability

If there was anything like a high power transistor back then he would have used that. High power transistors that are robust enough to handle the grid were designed inly recently over 100 years after the tesla/edison ac/dc argument.

can we stop vibe generating headlines?