Brain Teaser: What makes a Network Room's internal doorknob hot?

Yep, it sounds silly, like maybe it's a joke?  But it's really something I'm trying to discover.

A couple of my 250+ network rooms have internal doorknobs that are noticeably warmer than expected.

The air conditioning is working perfectly inside the room and also out in the hall.

Anyone inside the network rooms comments about the brushed-aluminum-looking doorknobs being warm.  Not hot, but warmer than expected when you're in a cooled room.

The rooms are good sized--12'x16' or bigger, with 14' ceilings.  There's not a lot of network equipment or other electronics in them--maybe six Cisco 2960XR's in network racks in the middle of the room, a few wall-mounted intercom and electronic security boxes, and that's it.

This particular building has eight network rooms, only two of which have warm doorknobs.

The building is a cancer treatment center and clinic that's only eight years old.  I jokingly asked our Radiologists and Physicists whether loose neutrons might be a cause; they were thoughtful, then smiled and shared that the treatment rooms are encased in six to twelve feet of concrete on all sides, with thick concrete neutron baffles for their halls, and thick lead doors isolating them.  They didn't think their equipment would be a possible source of the heat.

There are motion sensors over each internal door.  Could some form of infrared motion detection emit enough radiation that a door handle might become just slightly warmer than the surrounding air?

But other doors have the same detectors, and don't have warm doorknobs.

I've not measured the temp on the handles--yet.  ;^)  But the air temp is probably 68F around them, and they are probably 85F or warmer, at a guess.

Apparently I need a way for NPM to measure and track and report on doorknob temperatures.  Maybe I should relabel this for a Halloween story . . .

;^)