geekhack Projects > Making Stuff Together!

A minor Model F 122 refurbishment

(1/2) > >>

Parak:
Apologies in advance for relatively sub-par pictures. Credits and inspirations, in no particular order:

dfj (for general advice)
Soarer (for the adapter, of course)
kishy (for case advice and general advocacy)
fohat.digs (for http://geekhack.org/showwiki.php?title=Island:24615)

So, motivated by Soarer's adapter project, and having my curiosity peaked by the multitude of Model F advocates (did someone say NKRO, metal construction, and compatibility with modern stuff?), I decided to snag an F 122 on the cheap to try out:

[ Attachment Invalid Or Does Not Exist ] 47675[/ATTACH]
[ Attachment Invalid Or Does Not Exist ] 47676[/ATTACH]

It appears to be a fairly early model, with one piece keys having reinforced corners. A key or two were of the two-piece construction, indicating that they were replaced at some point. In addition, a stabilizer was missing on the numpad enter key and the reset key was missing a spring. Front plate was remarkably clean, with a few minor rust spots, and one small rust spot on the bottom metal case. Most notably, the top case is cracked almost entirely through to the function key rows. Overall, not bad for what I paid for it.

After disassembling it further, I noted that the foam was in a fairly crumbly condition. While the sticky backing was keeping the bulk of it intact, I decided to replace it outright. It was creating quite a disgusting mess.

[ Attachment Invalid Or Does Not Exist ] 47677[/ATTACH]

The glue and the remains of the mat can be cleaned with denatured or rubbing alcohol. Once I pried off all of the barrels, I put them in some tupperware, doused with denatured alcohol, and gave it a good shake. Cleans the gunk quite nicely.

Tools used:

5/8 arch punch
Krylon industrial latex paint
Rubber mallet
Old plastic cutting board
1/32" EPDM 60a rubber sheet

I roughly cut the rubber mat to shape using scissors, and secured it from the rear to the front plate (which should be sanded to remove rust). Then I proceeded to spray paint the plate which simultaneously marks the hole locations:

[ Attachment Invalid Or Does Not Exist ] 47681[/ATTACH]
[ Attachment Invalid Or Does Not Exist ] 47679[/ATTACH]

Yep, I'm using the rear of an M 122 just to take photos of other stuff. Anyway, not too bad of a result for two coats. The back also got a coat. My goal here was more to stop the rust process and prevent further rust, rather than making it look pretty. Next up, punching the holes:

[ Attachment Invalid Or Does Not Exist ] 47680[/ATTACH]

The paint is not designed to stick to rubber, so I needed to be a bit careful with it, otherwise it would have flaked off. The rubber, although thin, is also fairly hard so it sometimes took a few hits per hole. Placing the cutting board on a hard surface helped a lot here.

Putting the mat on:

[ Attachment Invalid Or Does Not Exist ] 47682[/ATTACH]

The 5/8 punch does give a comfortable margin of error, as long as I was careful to include the cutout notch in the area that I was punching out. The barrels:

[ Attachment Invalid Or Does Not Exist ] 47684[/ATTACH]

And the hammers - this is where one needs to be careful and make sure to place them according to the desired layout. Being used to ANSI, that's what I went with:

[ Attachment Invalid Or Does Not Exist ] 47685[/ATTACH]

I did not feel that I really needed a bolt mod, but at this point the plates refused to slide back together, even though I used the thinnest rubber sheet that I could find. After a lot of swearing, sweating, and perhaps overzealous clamping and hammering (which is discouraged in this case), they eventually slid back together:

[ Attachment Invalid Or Does Not Exist ] 47686[/ATTACH]

In retrospect, I think that if I used something like 30a hardness rubber instead of 60a, it would have had just slightly more give to make this process easier.

Add a breadboard with a teensy:

[ Attachment Invalid Or Does Not Exist ] 47688[/ATTACH]

One of my unexpected finds is that the dip switch block had to be connected in order for the converter to work, which is apparently not normal. Might have something to do with the keyboard being an early one. For the usb cable, I used a short one with a slim usb-b connector in order for it to fit in the cable opening. I like having quick disconnects as well, so with the short cable I simply use a female to male usb extension. Mmmm, laziness.

Final result:

[ Attachment Invalid Or Does Not Exist ] 47689[/ATTACH]

By coincidence, a coworker gave me for free a Model M a few days prior, which provided the source for all of the keys and in-barrel stabilizers needed for an ANSI layout. And the crack in the case? Well, I actually had two F 122s, and for the time being I just cleaned and used the case from the second one. Laziness strikes again.

For that second F 122, once I have some more time, I will fix the case, paint it, and test out softer rubber for the mat replacement. I'll add that here once I get that going.. eventually.

As for this board here, I can say that I'm a new convert, after never using buckling springs before. The feel is great, I'm enjoying the hardware remapping and NKRO advantages of the converter, and this historical piece of hardware will probably outlast me.

wcass:
i bought 4 linear feet of volara (closed cell) foam because shipping was the same as for 1 foot. was going to try this out. i have so much i thought i'ld offer some to you if you would like to try it.

for cutting holes, i got a 9 piece hollow punch set at Harbor Freight for $7 - use the 1/2" for the (12mm) main hole and the 1/8" for the notch.

my next project is to replace the top barrel plate and capacitive matrix in an XT to make the key layout look like this ...
[ Attachment Invalid Or Does Not Exist ] 48134[/ATTACH]
the case and controller will be the XT original, but with a teensy. if you have access to CNC equipment, please let me know.

Parak:
Interesting, I haven't heard of volara before, seems like pretty nice stuff. Seems expensive though - I got my sheet of epdm for something like $5, and silicone which should be slightly more compressible is $10. I'm not entirely sure which purpose the sheet serves exactly (padding? capacitance related?), but doing larger holes seemed to me like much less hassle with greater margins for error with no obvious downsides.

Nice project plan - but it's definitely a lot of work. It looks like you'd need a new top cover as well, and the controller will almost certainly need to be replaced as a custom matrix will have different capacitance at the very least. I'd love to see it happen though :)

fohat.digs:
Cool, nice looking job.

This is almost exactly the same as one I did a few months ago

http://geekhack.org/showwiki.php?title=Island:24615&highlight=ibm+model+conversion

and my case repair jobs (both of them) are holding up perfectly.

My replacement mat, from the art supply store, is probably not as firm or durable as your neoprene, but it cut easily and seemed to have a great combination of thickness and "give" to seat the barrels.

I also used the 5/8" hole punch, and offset it so that the extra crescent accommodated the alignment tabs, it seemed to work well. Double-punching with a 1/2" and a 1/8" is probably cleaner, but a whole lot more work.

I do recommend the bolt mod, especially since it has sometimes taken multiple tries to get everything in and working properly. I think my bolt pattern is sufficient, but have contemplated adding a couple more, next time. Too bad that the clearance is so very tight in the bottom (spacebar) side.

I have managed to slide fit a couple of these things back together, but it ain't easy on the 122s. XTs and ATs are a piece of cake in comparison, since they are so much smaller.

Too bad you never know what the mat is going to look like until you dismantle the beast.

Lastly, there is the proprietary spacebar problem, I have not tackled creating a replacement from a standard M spacebar, yet.

wcass:

--- Quote from: Parak;573497 ---Interesting, I haven't heard of volara before, seems like pretty nice stuff. Seems expensive though - I got my sheet of epdm for something like $5, and silicone which should be slightly more compressible is $10. I'm not entirely sure which purpose the sheet serves exactly (padding? capacitance related?), but doing larger holes seemed to me like much less hassle with greater margins for error with no obvious downsides.

Nice project plan - but it's definitely a lot of work. It looks like you'd need a new top cover as well, and the controller will almost certainly need to be replaced as a custom matrix will have different capacitance at the very least. I'd love to see it happen though :)
--- End quote ---

1/8" was under $2 for one foot (by 52 inches wide), but shipping was $10. PM me your postal address and i'll ship you enough to do your board - for free if you are in the US.

for my project, i will not need a new top, just will need to cut the small strip seperating the F keys. the barrel plate will be visible above the arrow keys but i might make an insert to hide that. i have designed a new capacitive matrix that uses the same 8 x 12 matrix as the old one, so (fingers crossed) i'll just have to jump the connections and hope the capacitance is still within spec. if not, i'll make a custom membrane and controller and use hammers from an M.

Navigation

[0] Message Index

[#] Next page

Go to full version