IBM 3277 keyboard with Micro Switch key switches

snuci · 05 Jul 2016, 02:30

I recently acquired an IBM keyboard that was pretty dirty but I felt I could clean it up and refurbish it quite nicely. I knew this to be an old IBM keyboard but there are two variations of this keyboard including Webwit's 5251 keyboard with IBM beam spring key switches and Seebart's 3277 keyboard with Micro Switch key switches.

This keyboard is from the 14th week of 1972 according to the Micro Switch keyboard assembly identification and this is further confirmed through the date codes on various integrated circuit components. The Texas Instruments keyboard controller from the sixth week of 1972 and the IBM “key clicker” circuit board is dated August 18, 1972. Since it has a part number stamped on the bottom plate of this keyboard, I refer to this as the IBM 2621364 keyboard.

In researching information, this keyboard was found in the IBM 3277 Troubleshooting Guide found on Bitsavers here. This is known as the “typewriter keyboard” model.

Pics of the keyboard as I received it:

: IBM 3277 typewriter keyboard - Dirty top view; IBM 2621364 - Dirty top view.JPG (284.85 KiB) Viewed 6371 times

: IBM 3277 typewriter keyboard - Dirty top view; IBM 2621364 - Dirty side view.JPG (217.56 KiB) Viewed 6359 times

: IBM 3277 typewriter keyboard - Dirty top view; IBM 2621364 - dirty key caps removed.JPG (280.13 KiB) Viewed 6359 times

: IBM 3277 typewriter keyboard - Dirty key cap view; IBM 2621364 - Dirty key cap view.JPG (299.83 KiB) Viewed 6359 times

: IBM 3277 typewriter keyboard - dirty dusty sliders; IBM 2621364 - dirty dusty sliders.JPG (348.04 KiB) Viewed 6359 times

snuci · 05 Jul 2016, 02:35

Here are some pics of it cleaned up. More pics can be found at http://vintagecomputer.ca/vintage-ibm-2 ... -assembly/

: IBM 3277 typewriter keyboard - keyboard top; IBM 3277 typewriter keyboard - keyboard top.JPG (262.49 KiB) Viewed 6347 times

: IBM 3277 typewriter keyboard - keyboard profile; IBM 3277 typewriter keyboard - keyboard profile.JPG (210.1 KiB) Viewed 6347 times

: IBM 3277 typewriter keyboard - keyboard bottom; IBM 3277 typewriter keyboard - keyboard bottom.JPG (270.77 KiB) Viewed 6347 times

: IBM 3277 typewriter keyboard - Micro Switch keyboard assembly with paper cover; IBM 3277 typewriter keyboard - Micro Switch keyboard assembly with paper cover.JPG (225.63 KiB) Viewed 6347 times

: IBM 3277 typewriter keyboard - Micro Switch keyboard assembly without cover; IBM 3277 typewriter keyboard - Micro Switch keyboard assembly without cover.JPG (264.13 KiB) Viewed 6347 times

: IBM 3277 typewriter keyboard - PCB back; IBM 3277 typewriter keyboard - PCB back.JPG (293.42 KiB) Viewed 6347 times

: IBM 3277 typewriter keyboard - keyboard assembly markings; IBM 3277 typewriter keyboard - keyboard assembly markings.JPG (357.93 KiB) Viewed 6347 times

: IBM 3277 typewriter keyboard - keyboard controller IC; IBM 3277 typewriter keyboard - keyboard controller IC.JPG (367.36 KiB) Viewed 6347 times

: IBM 3277 typewriter keyboard - clicker circuit board; IBM 3277 typewriter keyboard - clicker circuit board.JPG (278.06 KiB) Viewed 6347 times

: IBM 3277 typewriter keyboard - keyboard assembly identifier; IBM 3277 typewriter keyboard - keyboard assembly identifier.JPG (268.88 KiB) Viewed 6347 times

: IBM 3277 typewriter keyboard - key switch taken apart; IBM 3277 typewriter keyboard - key switch taken apart.JPG (247.72 KiB) Viewed 6347 times

: IBM 3277 typewriter keyboard - bottom of key switch removed; IBM 3277 typewriter keyboard - bottom of key switch removed.JPG (310.17 KiB) Viewed 6347 times

: IBM 3277 typewriter keyboard - plate that key switch connects to; IBM 3277 typewriter keyboard - plate that key switch connects to.JPG (285.29 KiB) Viewed 6347 times

: IBM 3277 typewriter keyboard - key caps; IBM 3277 typewriter keyboard - key caps.JPG (174.01 KiB) Viewed 6347 times

matchison · 05 Jul 2016, 02:41

Damn, I'd be happy if I got either variation :O. Great find.

Chyros · 05 Jul 2016, 02:50

Could've sworn it'd be a beam spring =o . Either way, a really nice find!

LewisR · 05 Jul 2016, 04:56

That's beautiful!

Parak · 05 Jul 2016, 05:40

Sigh... I saw Seebart's back in the day... and now even though I really have no room for any more IBM boards, I'd totally make room somehow for this in the highly unlikely chance it falls on my head. Which would kill me, but that's besides the point.

Much envy for you and your sources, snuci

photekq · 05 Jul 2016, 15:04

Beautiful little keyboard. You can definitely tell the beamspring and microswitch apart from the shape of the keycaps alone; they all have a slightly different shape/domed top.

Hopefully you can find the beamspring one too in order to get a pair

Invisius · 05 Jul 2016, 19:43

Wow, it's bizarre how similar this is to the beamspring variant on the surface. Even the part number matches the "66-Key EBCDIC" style beamspring from the 327x maintenance manual.

Makes me wonder how many of these were actually made. The Honeywell switches look dead-on identical to my Data General keypunch, too!

terrycherry · 06 Jul 2016, 05:02

Perfect! I glad to see you took the great quality on your best.
You are the man I know who hard worked to clean the keyboard internal including the switch.

Chyros · 06 Jul 2016, 12:19

By the way, considering they were a bit dirty, what were the switches like? I've found the Honeywell ones can get really scratchy when dirty.

snuci · 06 Jul 2016, 13:10

Invisius wrote: Wow, it's bizarre how similar this is to the beamspring variant on the surface. Even the part number matches the "66-Key EBCDIC" style beamspring from the 327x maintenance manual.

I'm not sure if Webwit has posted any other pics of his keyboard but it would be interesting to see the differences. There are differences even to Seebart's keyboard but it may be just a matter of different revisions. The obvious is that Seebart's key switch stems are turquoise and mine (although it's very difficult to see) are almost a deep purpley blue better seen in the fourth picture from the top. It's not black in appearance but may be an oxidized black. Along with the solenoid/clicker PCB being mounted standing up, the magnesium alloy case interior is not painted white as Seebart's is shown here. There must have been a few revisions of this keyboard.

: Interior picture of the case; IBM-2621364-metal-top-cover-interior.jpg (176.03 KiB) Viewed 6199 times

In thinking about these keyboards and their use from 1972, this must have been used with an IBM System 360 or 370 mainframe. Most companies would have leased the equipment so most of these terminals would have went back with the mainframe at end-of-life. It's surprising that any examples are left at all and why we see so few here.

Chyros wrote: By the way, considering they were a bit dirty, what were the switches like? I've found the Honeywell ones can get really scratchy when dirty.

I have more than one hall-effect keyboard and these are pretty smooth but I find they bottom out pretty easily. I should check if the key stem colours have any bearing on weight. Because this keyboard has a "clicker", one might be included to stop pressing when they hear it do it would feel "buttery smooth" for a touch typist.

Question for Webwit, if he's watching. Did you get an Xwhatsit controller working with your beam spring version? I would imagine that if you did, it would work perfectly with this version too. They should both be speaking the same protocol so I may be able to get this to work.

seebart · 06 Jul 2016, 14:25

Awesome snuci, easily the find of 2016. Even if these switches are filthy as hell these are extremely easy to open & clean which is the big difference compared to "regular" Honeywell Hall Effect. Pretty obvious.

Spoiler:

The central point to me really is that IBM obviously commissioned Honeywell (Micro Switch) for some orders while using their own Beamspring for others. dorkvader touched on this in my original thread, IBM was obviously aware of the disadvantages of Beamspring in terms of airborn dirt (contamination shield hello?) and hence went with this instead.

dorkvader wrote: So first off the oak FTM on the "industrial" XT was possibly a measure to prevent dust from interfering with the capacitive matrix. You know how a dirty cap PCB can affect switch sensing. OAK FTM may not be as nice to type on but it makes more sense in an industrial setting.

Same with hall effect. There are some that are fully sealed against dust / debris and even the ones that aren't will keep functioning in a really dirty place. I have one from a fisher rosemount controller that was used in a machine shop for many years. IT's awful to type on but still works!

E3E · 07 Jul 2016, 04:30

So many amazing finds this year, snuci. This is really impressive.