There are many weird things about this keyboard. When I ordered it, I expected to get some early 90's era backstock, but when it arrived (in a giant box that used hundreds of old envelopes from the 50's-80's taped to the box as postage, which was extremely bizarre in of itself) and I opened it up I was surprised to find that it dates from 1984or so. "051684" is stamped inside the top shell of the case and is corroborated by a hand written date of "9-14-84" on the plastic strip that gets bolted to the back-underside of the board. Underneath the date on the top of the case is what I think is a shop date of "4975". The mounting plate has a shop date of "4991" on the IBM sticker, which is the only place that has IBM on the keyboard. On the coiled cable where IBM is normally stamped, the IBM is missing, with only a blank square outline where it should go. Additionally, there is no label on the metal base, or residue indicating there ever was one. The cable is the same black one as on the XT PC keyboards and has the standard 180 degree pinout, which I haven't ever seen on an F 122. Weirder to me is that the keyboard only speaks set 1 instead of set 3, and refuses to change to set 3 with my soarer's converter. It sends out weird scancodes that don't match up with any documentation I've seen online, and when I opened up the keyboard module I discovered that all the unused pads, even the ones that are connected to traces that would normally send out scancodes don't work at all, or send any scancodes but all 122 of the standard keypads send out scancodes as XT set 1. The keycaps which I think are the same as on the 3270 terminal, are also the thickest printings I've ever seen on an F.
The kicker for me is that I can't figure out how this keyboard would have been modified post manufacture, and I don't think this was a drop in replacement module put in an an aftermarket case. The dimensions, plastics, paint job and texture, thickness of the plates all are identical to my other model F. I also can't think of a reason why IBM would sit on a fully assembled keyboard for 6 or more years to have Lexmark sell it off later, or a reason for the scancodes.
I tried doing some research but I can't find anything, so I'm curious what you guys think. Help me out! I'm (honestly) probably totally wrong and everything I listed is normal anyway, so prove me wrong guys.
Boring photos:
I don't have a photo of it assembled right now because I'm cleaning it but this is from the ebay listing

Back black tray, ebay

Opening the case to clean. Lots of insulation and sawdust.

9-14-84

IBM label module backplate

Inner cable connector

Controller + board

Backside of controller board

Underside of top shell


Outer cable connector

badge compared with my other Model F badge

keycap comparison between F 122s
