ibm beam-spring, or not

[msiegel]

someone is going to find out real soon now (not me though

http://cgi.ebay.com/Vintage-IBM-keyboard-Rojon-electronics-/170550817892

[mr_a500]

Wow! Did you see what happened in the last 4 seconds? Each second, it jumped drastically. When it ended at $293 all I could say was, "Holy ****!"

[msiegel]

ok, which one of you dudes won that thing?! XD

[TexasFlood]

Wow! Did you see what happened in the last 4 seconds? Each second, it jumped drastically. When it ended at $293 all I could say was, "Holy ****!"

That's just crazy! And remember, the seller had several of them so could end up with quite a pile of cash.

[Sam]

ok, which one of you dudes won that thing?! XD

Wasn't me.  Too rich for my blood, as with you.  Notice you're also interested in beam springs.  I was looking to make a controller to hook one of these beauties up to a modern PC, but from the looks of it they're much more difficult to source at a reasonable price than I expected.

The same two high bidders in this were also the highest bidders in the other one with the keypad that sold this past week.  Being they're presently not usable, I guess someone wants them for museum pieces, or thinks they're going to be collectible items.  Too bad they're in such high demand without adequate supply.  It looks like the only way to make a modern beam spring board at a reasonable price would be to manufacture your own beam spring switches.  Unfortunately the startup/tooling costs to do that would likely be prohibitive in the limited quantities that could be expected to be sold.

[msiegel]

the only way to make a modern beam spring board at a reasonable price would be to manufacture your own beam spring switches.  Unfortunately the startup/tooling costs to do that would likely be prohibitive

it's a pity.
i still wonder if there's hope for a modern-day model f.
more durable than a topre, but much much louder XD

[Sam]

I recall seeing in some old posts that you were trying to work on something like that, right?  Did you give up on that project?  For me, making a controller shouldn't be too difficult hardware/software wise.  I have lots of resources available as designing electronics is my job.  It's the case/mechanical aspects that I'd worry about.  Not that it's difficult to do, just extremely expensive.  I've got lots of thoughts about this floating around in my head but nothing concrete.

I was hoping to first make a controller capable of interfacing to various old beam-spring and Model F boards.  Then the next step would be to take the controller and make a new base PCB (with all the capacitance pads) but in a modern layout that would be completely usable by most, ie. a typical Model M layout, with optional Windows and perhaps other keys.  Again, I don't think very difficult, though it might require two separate PCBs - one for beam springs and one for Model F buckling springs.  But I doubt the resulting board could be mounted easily into an existing plastic case, and of course the metal plate would have to be custom made (which I think you were working on).

At least Model F switches are still readily available at reasonable prices.  Though doing an only Model F project isn't nearly as exciting to me as one that could accept either Model F or beam spring switches.

[phoenix]

I recall seeing in some old posts that you were trying to work on something like that, right?  Did you give up on that project?

Msiegel, I was going to ask the same

Then the next step would be to take the controller and make a new base PCB (with all the capacitance pads) but in a modern layout that would be completely usable by most, ie. a typical Model M layout, with optional Windows and perhaps other keys.

I have been thinking about the exact same thing for months now and have been collecting Model F XTs for parts. I know too little about electronics to move forward.

But I doubt the resulting board could be mounted easily into an existing plastic case, and of course the metal plate would have to be custom made (which I think you were working on).

If they have the same dimension and mounting holes as, say, the plate in a Space Saving Model M, why would it be difficult to mount? Having a metal plate made by a waterjet cutting service shouldn't be hard. I don't know if they can also roll it to the right curvature, though.

At least Model F switches are still readily available at reasonable prices.  Though doing an only Model F project isn't nearly as exciting to me as one that could accept either Model F or beam spring switches.

IIRC, The pads on the circuit board are very similar for all capacitive spring boards. Beam spring or buckling spring are just different actuation mechanisms. If you figure out how to read the capacitive switches, it should work for all the switches with minimal tweaking.

[Sam]

I have been thinking about the exact same thing for months now and have been collecting Model F XTs for parts. I know too little about electronics to move forward.

Electronics is my job.  While I haven't ever sat down to figure out how the capacitance detection works in a Model F, I doubt it would be all that difficult.  Once that's done, the controller part, reading the matrix, translating to scan codes, adding flash memory for total programability, interfacing it to a PS/2 and/or USB is easy, well at least for me.

If they have the same dimension and mounting holes as, say, the plate in a Space Saving Model M, why would it be difficult to mount? Having a metal plate made by a waterjet cutting service shouldn't be hard. I don't know if they can also roll it to the right curvature, though.

But is the height the same?  I know beam spring boards are very tall and doubt they'd stand a chance of working.  Whether Model F switches could be mounted into a Model M case, I don't know...I don't have a Model M case with me to check.  Basically I was thinking though along the same lines...you need to mount it into an existing case.  If you custom build a new plastic case, you're talking serious $$$.  And you need to choose a case that's sturdy enough to handle the heavy weight of the metal plate, etc.  Not some cheap flimsy case designed for a lightweight rubber dome board.  But if the case doesn't work height-wise, then you'd either need to just attach the top half of the case and leave the bottom half with an "industrial look", or need to design some sort of spacer between the two halves of the case to raise the height.

IIRC, The pads on the circuit board are very similar for all capacitive spring boards. Beam spring or buckling spring are just different actuation mechanisms. If you figure out how to read the capacitive switches, it should work for all the switches with minimal tweaking.

Yeah, the pics I saw looked almost identical between the two.  But I don't know about the physical size of the beam springs compared to model F switches and if they'd be compatible size-wise.  If they are, great - shouldn't be overly difficult to make a single controller to read either type.

BTW, my personal feeling is that making three models, a full-size, a space saver, and a mini would be fantastic.  Something for everyone.  I prefer a full-size on my desk, but would like a mini buckling or beam spring to take with me on the road.

[ch_123]

An IBM 3277 terminal keyboard.

Times like this, I regret contributing to the hype around these things. Oh well, not my problem now

[msiegel]

heyyy, this is exciting!

the ibm patents claim that a pulse is sent on the back side of the pcb, and either coupled through both front-side pads or not, depending on the position of the flyplate.

i don't have a clue how to implement the electronics for that XD

it's a good question whether a new plate assembly with model f switches would fit in a model m case... but i suspect so

[msiegel]

@ch_123: sometimes, hype is inspiration

btw, i also found this sensor design: http://www.google.com/patents?id=-IYvAAAAEBAJ&zoom=4&pg=PA1#v=onepage&q&f=false

and after hacking up those model f top and bottom plates to test viability of hand-modding, i concluded that machine-cut plates would be a must... but only if the electronics work

[Sam]

An IBM 3277 terminal keyboard.

Times like this, I regret contributing to the hype around these things. Oh well, not my problem now

Regret it for what reason?  That the prices of these boards are now skyrocketing?  If you or others had never mentioned how great the feel of beam spring switches were, I for one would never have even considered them.  I suppose in hind-sight, maybe I should have kept my mouth shut about my plan.  Don't know if my mentioning it had any influence on these prices or not.  But if so, it probably wouldn't have really mattered much because in that case if and when I made a controller and announced it, the same thing would have happened then instead of now.

Anyways, it's better off I realize what crazy prices these boards can go for now rather than later.  As much as I'd like to make a controller for beam springs, I doubt I will at this point or if I do keep it to myself and perhaps share a prototype with a few select members of this board.  I think the only way I'd seriously consider making something for beam springs public would be if I or someone else were to bite the bullet and manufacture new beam-spring switches.  Attempting to salvage switches off old boards that are going for $300 a pop just isn't going to happen.

[Sam]

@ch_123: sometimes, hype is inspiration

btw, i also found this sensor design: http://www.google.com/patents?id=-IYvAAAAEBAJ&zoom=4&pg=PA1#v=onepage&q&f=false

and after hacking up those model f top and bottom plates to test viability of hand-modding, i concluded that machine-cut plates would be a must... but only if the electronics work

That patent was filed in 1986, long after beam springs and buckling springs were invented and production started.  But anyways, it might still be a useful reference.  No reason why building a new controller has to match exactly what IBM did.  As long as it properly/reliably detects the key presses, that's what really matters.

[ch_123]

If you or others had never mentioned how great the feel of beam spring switches were, I for one would never have even considered them.

Yes, this is the point I was making...

[msiegel]

i blame webwit for the high prices

in any case, there's no sense crying over spilled milk.

even if no one else does, i plan on continuing forward so i can at least have a nice colemak usb model f. i'll be happy to share any designs

[Sam]

Yes, this is the point I was making...

But what value are great feeling switches that aren't part of a usable keyboard?  As they are now, the boards are nothing more than something nice to look at, or maybe for some an investment.  They serve no practical purpose in the modern computing world that I know of, unless someone's already made a converter or adapter that hasn't gone public.  Even then, the layout would prevent them from being of much practical use beyond being something to just show-off.

If the "hype" about these switches eventually leads to some product being made available using them, then great.  If not, then I see that nothing's been lost unless you're a keyboard collector who was looking to complete your collection with a set of these boards.  I'm not really a collector myself, so look at the crazy prices from a different perspective than what a collector would.

[JBert]

Electronics is my job.  While I haven't ever sat down to figure out how the capacitance detection works in a Model F, I doubt it would be all that difficult.  Once that's done, the controller part, reading the matrix, translating to scan codes, adding flash memory for total programability, interfacing it to a PS/2 and/or USB is easy, well at least for me.

Actually, the kbdbabel guy has put schematics online of beam spring keyboard PCBs, I would guess that the model F design wil be highly based on them, if only more refined and compact.

See kbdbabel documentation: keyboard schematics.

[ch_123]

They're identical to the Model F, except the detection logic is in reverse - beam springs register the key when capacitance increases, Model F registers it when the capacitance decreases.

But what value are great feeling switches that aren't part of a usable keyboard?

Good question, maybe you should ask whoever spent $250 on those beam spring keyboards last week... Something tells me that it's not going to be used for anything useful.

[arfink]

What terminal model did these keyboards go with? It's entirely possible the KBDBabel encoder code already exists, but I can't tell what terminal that KB goes with.

[ch_123]

It's a 3277 (as distinct to 3278) keyboard as far as I can tell.

[mr_a500]

Something seems strange here. Both of these keyboards "sold" three weeks ago, yet there is no feedback either by the seller or buyer.

Another strange thing - there's a new listing for one of these keyboards (or the same keyboard) and I keep getting outbid by somebody whose recent bid history is 100% on 3 listings by this seller.

[Phaedrus2129]

Maybe the guy is self-bidding with an alt to increase selling price, but hasn't gotten it to where he wants it yet?

[ch_123]

The term "Beam spring" has spread to eBay.

Be afraid, be very afraid.

[msiegel]

"Beam spring"

i reckon every terminal model f will soon be labeled "beam spring" XD

[J888www]

Something seems strange here.

Bleam-spling sounds like a fishy risting. Vely vely stlange indeed.

[HaaTa]

Hehe, I'm glad I picked one up a few months ago.

[sjohn]

"I'm still working on getting the funds together to make payment. I should be getting payment
to you soon. In the meantime, please don't open an unpaid item case". Thats why I haven't received feedback on the first two I sold and just this week, I shipped the third one. "Maybe the guy is self-bidding with an alt to increase selling price, but hasn't gotten it to where he wants it yet? " Really Phaedrus2129? Is that what you really think? Couldn't be anything else right? Maybe i'm waiting to be paid and don't want to leave him a bad feedback. His is at 100% now. And that would be 26 days on one and 15 days on the other if your counting, supposed to be 7 days. I'm still waiting. Might consider that more than fair?   I asked on this forum about keyboards I had and knew nothing about. I put that information into the listings so the people who know about keyboards would see it.

[msiegel]

everybody keeps saying it's easy

i'll wager we end up replacing *all* the controller electronics to get one of those things to work.

[Phaedrus2129]

"I'm still working on getting the funds together to make payment. I should be getting payment
to you soon. In the meantime, please don't open an unpaid item case". Thats why I haven't received feedback on the first two I sold and just this week, I shipped the third one. "Maybe the guy is self-bidding with an alt to increase selling price, but hasn't gotten it to where he wants it yet? " Really Phaedrus2129? Is that what you really think? Couldn't be anything else right? Maybe i'm waiting to be paid and don't want to leave him a bad feedback. His is at 100% now. And that would be 26 days on one and 15 days on the other if your counting, supposed to be 7 days. I'm still waiting. Might consider that more than fair?   I asked on this forum about keyboards I had and knew nothing about. I put that information into the listings so the people who know about keyboards would see it.

Um, I said maybe. As in, "Hey, here's a possibility!"

Ebay is a hive of scum and villainy, after all.

[ch_123]

i reckon every terminal model f will soon be labeled "beam spring" XD

Soon after, Dell Quitekeys will be labeled as "Semi Beam Spring". Then the ghost of Hitler will journey to Canada and start clubbing baby seals.

We'll all look back at this auction as the beginning of the end of Western civilization.

[Sam]

There have been 3 of these keyboards that sold.  The first two went to the same buyer, the last one to a different buyer.  The prices have also been steadily decreasing with each sale.  I do not at all believe the buyer of the first two was at all related or in cahoots with the seller, being that buyer has bid big money on other vintage keyboards as well.  I think the buyer is simply a collector, but interesting he seems to have difficulty scraping up money for his purchases.

I made a post recently about these keyboards and mentioning I didn't think they were worth much more than Displaywriters, but in looking at the keyboards more, I'm not so sure now.  So I ended up deleting that post.  I'm no longer even 100% sure what they are.  I'd thought initially that they were some of the most recent beam springs, in the same era as the Displaywriters, but they might not be.  I'd like to get my hands on one regardless.

The seller did indeed post here to begin with and did not even know what the keyboard was and asked for information about it.  He also PM'd me and asked for more information.  I have no reason to suspect there is any funny business with this seller and believe he's being completely honest in his sales.  I just wish the price would drop within my price range.

[ch_123]

I made a post recently about these keyboards and mentioning I didn't think they were worth much more than Displaywriters, but in looking at the keyboards more, I'm not so sure now.  So I ended up deleting that post.  I'm no longer even 100% sure what they are.  I'd thought initially that they were some of the most recent beam springs, in the same era as the Displaywriters, but they might not be.  I'd like to get my hands on one regardless.

They're IBM 3277 terminal keyboards. They would have been the original beam spring keyboards, appearing on the market in 72'.

That said, IBM terminal had extraordinarily long production lives. The 3277 was produced until some time in the mid 80s. Things like the 3278/3279 and 3101, complete with beam spring keyboards, were made as late as 1992 AFAIK.

[Sam]

They're IBM 3277 terminal keyboards. They would have been the original beam spring keyboards, appearing on the market in 72'.

That said, IBM terminal had extraordinarily long production lives. The 3277 was produced until some time in the mid 80s. Things like the 3278/3279 and 3101, complete with beam spring keyboards, were made as late as 1992 AFAIK.

Thanks for that great information.  I had no idea how late those things were made.  I just assumed the 3277's were all produced in the 70s, and that the later beam-spring terminals stopped production in the mid 80s.  I guess their rarity (or perceived rarity) comes from them not being produced in mass like the Model Fs and Model Ms.