Beam Spring 104+SSK Reproduction Project! First Batch In Stock, Shipping early next year after New Model F Project

[richt]

This is so awesome you are doing this, given all this isolation, I've been buying keyboards like crazy and trying to get back to wonderful keyboards that I've used in the past. Got the Unicomp M but it wasn't nearly as nice as the IBM PC original that I remember, so that's probably because it is a Keyboard M and not an Keyboard F, so just put in an order for your Model F!

Now, you have these beam springs which brings me back to my 3270 days. Woohoo! thanks for a great 2022 present

[Ellipse]

Thanks for posting richt. Please do register on the interest form in the original post (if you have not done so already) and share your 3270 days if you don't mind! What were these keyboards and terminal systems used for in your experience? Did the IBM stuff outperform the competitor systems? Did anyone talk about the typing experience of one system vs. another?

[Ellipse]

Currently we are up to more than 900 reservations for the beam project! The cutoff point for the first round will all depend on how many folks choose the first round vs. the second round.

As a reminder the beam reservations Google Form is for the project in general not just for the first round - everyone needs to sign up if they prefer the second round too.

When we get closer I expect to put up some videos of the sound and operations to help folks decide. The advantage of the first round is not having to wait a year - having everything in stock, plus the full sound quality (extra reverberant and snappy) while the advantage of the second round will be more dampened sound quality like the original, as well as the redesigned more expensive case. Another consideration - the second round is expected to cost a bit more than the first given it has the extra case and because prices have gone up for a lot of manufactured goods over the past year, so it is likely that these won't be able to be manufactured at the same cost.

[mmm]

You previously mentioned a slim chance of ISO support, any updates on this for the second round? (And any teasers? )

[Ellipse]

Yes I'd guess there will be a few additional layout options for the second round - what does everyone suggest? ANSI, ISO, any others?

[Bjerrk]

ISO would be magnificent!

By the way, I've seen lots of tidbits here and there about what's likely to be in the first and second rounds, but it's still a bit unclear to me (differences in case design? Damping?). Is there an overview post somewhere that I've missed?

[inozenz]

Yes I'd guess there will be a few additional layout options for the second round - what does everyone suggest? ANSI, ISO, any others?

ISO DE and ill pull the trigger!

[darkcruix]

Yes I'd guess there will be a few additional layout options for the second round - what does everyone suggest? ANSI, ISO, any others?

Within the limits of the case size and sensing PCB, a BeamShip or "B122" is not in question
The SSK is a great option and I think you made some very good design choices. Our "ISO lovers" probably are in the majority to get that built. For me, you made everything right (from a design point of view).
I'd love to be able to build-in a solenoid and if you could make room there and mounting brackets ... I'd be happy enough.

Roland

[daedalus]

When we get closer I expect to put up some videos of the sound and operations to help folks decide. The advantage of the first round is not having to wait a year - having everything in stock, plus the full sound quality (extra reverberant and snappy) while the advantage of the second round will be more dampened sound quality like the original, as well as the redesigned more expensive case. Another consideration - the second round is expected to cost a bit more than the first given it has the extra case and because prices have gone up for a lot of manufactured goods over the past year, so it is likely that these won't be able to be manufactured at the same cost.

Apologies if I missed something in one of the updates, but what is the extra case option for the second round?

[Ellipse]

daedalus please do feel free to check out the feedback in the thread on the case. The first round beam case was designed to integrate the top inner assembly holding the beam modules in place so that separate top/bottom inner assembly parts were not required, thus keeping the cost as moderate as possible (for a beam spring!). While this simplifies the case design the downside is that the case design was limited to enclose the beam modules securely.

With future rounds I expect to add a separate inner assembly plate so that we can have various case designs that one can drop in a beam inner assembly. The cases may resemble slimmed down versions of the original beam cases. My favorite design is the 3278 A02 / 3279-2C which has square push buttons (they still make these - I was looking to have these integrated into the beam cases, whether for show or to make them functional somehow).

The question for discussion here is should there be a combined design that comprises elements from various beam cases, or should the design be more modern, or both? The limitations on the significant costs of tooling will certainly require limiting the number of case design offerings.

To further the discussion of case design, here are some photos of the various major beam cases I have come across, to keep all of these in one place so folks can see them all at the same time (there are others). Most photos are not mine. These photos are being reposted for educational and discussion purposes only and I do not own the photos.

ibm 3278 small.jpg (39.89 KiB) Viewed 8304 times

3278 numbers on right.JPG (379.14 KiB) Viewed 8304 times

66 key4.jpg (159.52 KiB) Viewed 8304 times

3279-2c.jpg (682.63 KiB) Viewed 8304 times

3279-2A.jpg (1.04 MiB) Viewed 8304 times

5641316.JPG (406.04 KiB) Viewed 8304 times

5251.jpg (1.35 MiB) Viewed 8304 times

Beamspring restoration (1).jpg (1.69 MiB) Viewed 8304 times

IBM 4978 Beamspring (1).jpg (842.28 KiB) Viewed 8304 times

[Scarpia]

Oh wow, that’s a dizzyingly gorgeous set of beamers.

Personally I love them all, though the 5251 is my least favorite due to the XT-like layout / lack of separation to the right of the Enter key.

[dcopellino]

Personally, I find the designs of 3278 & derivates, if i may say so, quite futuristic for their time. That slope bevel so unique and sci-fi styled, so that I'll pursue that path with no doubts. The only variation I'd add may regard the thickness of the thing that doesn't seem to be incompatible with the new low profile beam switch style. Make it so, make it just slimmer. Put the new beamspring on a diet...

[darkcruix]

I have a variety of BeamSpring keyboards and my definitive favorite (I own) is the 3278 with 87 keys. While it has mostly elements that I enjoy, there are a few items that were created in a different time and aren't as useful or elegant today anymore (up for debate).

• The enter key in its vertical shape - you can get used to it, but when you change keyboards frequently it is neither ISO, nor ANSI

• The square nav cluster vs inverse T-shape - I can work with it but my brain can't 100% adjust

I'd make the Nav-Cluster 3 keys wide and add the Insert and Del key also, as well as an inverse T-nav. To the left of the alpha-numeric area, I'd leave the 8 keys. Top can be housing an ESC key? The rest can be freely programmable (PF13 to PF20).
The Space row (R5) should be (IMO) be designed with at least 3 modifier keys to each side (I personally would want an order like this: Ctrl,Fn,Alt,Window,Space,Alt,Fn,Menu.Ctrl or for Mac: Ctrl,Alt,Fn,Cmd,Space,Cmd,Fn,Alt,Ctrl.
I also still use the num pad frequently, but that is really just a treat.

3278 design like below, but updated layout that is more modern. The below is my 3278 as it is right now:

IMG_2159.jpeg (1.74 MiB) Viewed 8202 times

[inmbolmie]

I would say that it would be more interesting to have a modern layout with a metal case "in the spirit" of the 3278/3279. A full repro made more sense in the case of the Model F, that had already pretty modern layouts and are "unobtainium" keyboards.

Also considering the high clearance between the key bottom and the chassis and the compactness of the first batch design, wouldn't just be possible to design the case as a simple enclosure around that very same module?

[depletedvespene]

With future rounds I expect to add a separate inner assembly plate so that we can have various case designs that one can drop in a beam inner assembly. The cases may resemble slimmed down versions of the original beam cases. My favorite design is the 3278 A02 / 3279-2C which has square push buttons (they still make these - I was looking to have these integrated into the beam cases, whether for show or to make them functional somehow).

I would expect one of those square buttons to be the one controlling the solenoid.

The question for discussion here is should there be a combined design that comprises elements from various beam cases, or should the design be more modern, or both?

Yes.

Jokes aside, it'll certainly be a difficult thing to pull off while minimizing the complaints from the users, whose tastes will be all over the place. I would go with something along the lines of the 4978 (no vertical ridges as separators), BUT with narrow bezels, as the mouse doesn't need to be even further to the right than already is. Perhaps a "pen-holding" horizontal ridge above the topmost row, if it can be pulled off, too.

[Muirium]

3 × 5 key block on the right allows an inverted T for arrow keys and just enough space for a decent numpad, whichever people prefer. I do something like it on my AT and it works great.

Definitely go with more modifiers on the bottom row. The most awkward thing about my beamspring is forever having to chord mods on two different rows.

[depletedvespene]

3 × 5 key block on the right allows an inverted T for arrow keys and just enough space for a decent numpad, whichever people prefer. I do something like it on my AT and it works great.

Definitely go with more modifiers on the bottom row. The most awkward thing about my beamspring is forever having to chord mods on two different rows.

Uuuuuh... I thought the physical layouts were more or less already decided? If not, let's all start on another round of "my bottom row is better than yours!"

[depletedvespene]

Quick doubt here (can't believe I didn't think of this sooner): does the design of these new beamsprings allow varying "weights"?

[Johnbo]

The question for discussion here is should there be a combined design that comprises elements from various beam cases, or should the design be more modern, or both?

From what I can tell, the spirit of this project is "beamspring for a modern era", so I think it makes the most sense to have a case that comprises elements of the various cases. They all share some common characteristics: Beige color, wedge shape, large bevels on the front and rear, smaller beveled edges on the side. So a case that has those characteristics in a smaller form factor that's made possible with your reduced height switches makes the most sense to me.

(Unless you are going to to 1:1 reproductions, in which case, 4978 please )

[Ellipse]

Sorry for the confusion: As a note, the vintage beam spring layouts are not being reproduced; only the new cases are being possibly influenced by the old cases. The layouts will be modern/standard keyboard ~103 key and SSK style, and maybe others like 60%.

Thanks for all the feedback - please keep the feedback coming. Is the 3278 A01 style design so strongly preferred that folks would accept $50 or so more for all the 3278 type buttons and lights and circuitry? Seems like these keyboards are already costly enough. I don't know if there is enough demand for these extra buttons.

inmbolmie I am not sure what you are asking? There is no extra clearance for the first round beam modules - the module touches the PCB which touches a thin layer of foam below the PCB. This foam touches the bottom of the case. The cases were designed to be the thinnest possible height while maintaining 100% of the original key travel.

depletedvespene yes the factory did a test and stretched the beam springs to require a pressing force that was approximately double that of an original module. Would anyone be interested in this? I don't think the process is controllable so each key weight may be slightly different. The default is going to be the standard beam spring weight with no other options besides maybe the double weight ones. I have found that the double force springs require too much force to press to be comfortable for typing.

[inmbolmie]

inmbolmie I am not sure what you are asking? There is no extra clearance for the first round beam modules - the module touches the PCB which touches a thin layer of foam below the PCB. This foam touches the bottom of the case. The cases were designed to be the thinnest possible height while maintaining 100% of the original key travel.

I'm referring to the fact that the first batch case is not very different to what is actually inside an original Beam Spring keyboard, that's a rectangular slab where the switch modules are mounted and the keycaps are at a significant height over the top of the case (that's the clearance I was talking about). In the original keyboards that "module" is bolted to the bottom pan, and the top case is bolted to the bottom pan as well. With your keyboard you could have a similar arrangement where you place a full first batch keyboard inside a case designed to fit the keyboard, and give it a top case in the style of the old keyboards.

Think of it as if the first batch keyboard is like this...

rest_31.jpg (272.33 KiB) Viewed 8009 times

...and you put it inside a bigger case designed to accommodate it, just like that module in the picture fits inside an IBM 5100 case.

That way you could even sell the cases to the first batch owners. Maybe it's a dumb idea but who knows...

[Ellipse]

Yes that might be possible to reuse first batch parts in the second batch (maybe at least the beam modules and keys which are the majority of the cost? I don't know). The second batch should have keys "recessed" into the case as you mentioned. Not sure about reusing the case though as the inner assembly would be a different design that is thicker so it accommodates the increased padding/dampening.

To reduce the wobbling of the beam modules the new design has the beam modules "lock" into place securely into the top of the case (or the top of what will be the inner assembly in a future design). If foam is put in between the top of the beam module and this top inner assembly I'd imagine the wobbling may not be as reduced as in the current design, compared to the IBM originals whose keys and modules produced a total wobble of several mm in many directions. The beam spring design requires some wobble-if there is not enough wobble/tolerance then the keys get stuck. In the first batch design, the springs reverberate each other a little when one spring is pressed because the plate is so secure against the beam modules - I like the increased liveliness of the keyboard but for the future design I expect to dampen down the noise/reverb.

Also for those interested in changing the weighting of the keys, below is a photo of the factory's test. The spring on the left is the IBM spec spring free length and the one on the right was stretched by the factory and I think put in the oven to heat treat/lock in its new length? I don't remember. The one on the right produced a key weight/force that was double the IBM spec.

[thefarside]

Ellipse, forgive me if this question has already been asked but I signed up on the interest form for two full-size keyboards. At the time I didn’t realize a different version would be offered later. Would it be possible to get one of the in stock keyboards and one-off the future models from the original interest form or will you be sending a separate form for the updated one?

[Ellipse]

thefarside the reservation form is for both rounds (it's for the project, not for a specific round) - you can decide as late as when your turn comes up in the queue

Everyone should sign up on the form if they prefer the second round too.

[depletedvespene]

depletedvespene yes the factory did a test and stretched the beam springs to require a pressing force that was approximately double that of an original module. Would anyone be interested in this? I don't think the process is controllable so each key weight may be slightly different. The default is going to be the standard beam spring weight with no other options besides maybe the double weight ones. I have found that the double force springs require too much force to press to be comfortable for typing.

That's rather different from what I was thinking, which was more along the lines of a "slightly heavier spring" to aid in inhibiting accidentally pressing some select keys. I understand that this isn't feasible for the time being.

As for double weight? We'd need to try out single weight first to then know whether to try the other one... but I don't think many people would be interested (then again, what do I know about anything).

[Admiral]

I think I filled out the interest form back in August or September. I only put in for one of each keyboard I think. Is there a way to check or should I just fill the form out again? I like the idea of doing one of each round and I may want to add to it.

And while I am here I really like the Beamspring 5251 and the 3278 A02 / 3279-2C which has square push buttons looks very unique so I would be happy with either design. And if it ends up being the 3278/3279 then maybe you could use the double weighted ones for the unique buttons. but otherwise I am not sure if I am interested in a really heavy duty keyswitch that is uncomfortable to type on for regular keys.

[MrRobot49]

Case idea: 4978 Beamship-like case with layout for full modern set of keys, including two rows of wide F keys across the top (F1-F24).

[Admiral]

Is it possible for the second batch that the PCB could support ISO, HHKB, and ISO? Have a split backspace (2 1U keys), split left shift (1 1.25U and 1U) and right shift (1 1U and 1.75U or 1.75U and 1U (not sure which way is preferred)), and a split enter.

That way you have a 2.25U ANSI enter sits, if split into two keys, on a 1U and 1.25U), ISO enter (1.5U backslash key and the same 1.25U part of ANSI, and then you could possibly have a larger enter if you combine all three keys together (1U, 1.25U, and 1.5U) just disable two of them.


The biggest problem with this is probably compatibility issues with current mx keycaps on these split keys if designed in this way.

Edit: This is probably way beyond the scope of the project, but Ellipse thanks for the awesome work you are doing.

[Ellipse]

Based on the MX key stem design I am not sure that is possible for one PCB to support that; I think it would have to be different top inner assembly, inner foam, and PCB parts for each layout.

[Admiral]

Based on the MX key stem design I am not sure that is possible for one PCB to support that; I think it would have to be different top inner assembly, inner foam, and PCB parts for each layout.

Darn, I thought as much.


If there is only one layout offered, I am unsure if you are planning on offering different layouts for sure, but I wouldn't mind ISO if it helps those who want that layout get a better chance at getting it offered. I'm just worried about, down the line, trying out different MX keycaps since ANSI seems most popular. I'd probably only use the keycaps you offer though personally.