Dick Harris question/answer thread

[Dick Harris]

Spoiler:

Spoiler:

Spoiler:

Spoiler:

Hello, Dick! I hope you are well.

Would you happen to know more about the origins of the AT&T buckling springs that hold certain similarities to the Model M design? They have been observed to have some striking resemblances to your design but are thought to be an AT&T product. Nate, I don't know anything about an AT&T buckling spring. Sorry.

Spoiler:

Similarly, IBM Japan contracted out the manufacture of several Japanese-market keyboards such as the 5576-A01 to Brother. Would you happen to know something about those as well? I've read recently in other posts about this, but I couldn't bring anything new to the discussion.

Spoiler:

Your original capacitive design is an incredible product, and I wish you the best!

-Nate

[Mandarbmax]

Thanks for the reply, Dick!

Can you expand more on the ultrasonic delay line thing? I've never heard of that before.

Also you mentioned above that you wrote your own patent disclosure. When we read your patents how much of that is in your own words and how much of that was done by an attorney working their funky patent magic? Did the ratio of how much was yours change as time went on? Furthermore were the drawings done by you or did you have someone else make them for you?

Not to drown you in questions but you also mentioned that the switches had to make/break at the tactile/clicky event and they needed to support N key rollover, were there any other major considerations for the design you had to take into account?

Thanks again for your time answering these questions!

[ZedTheMan]

I have another couple of questions for you, Dick, if you have the time.

1. Was the choice of the spacing foam in keyboards B and F made with the knowledge that it would degrade over time, before other parts might fail? How might that have been weighted with the use of other materials? That is to say, why was the foam that was used in the end products used, as opposed to other options that might have existed?

2a. Around what point in the process of designing Keyboard B/beamsprings, was the need for the contamination shield determined? Was it considered at the outset as a necessity, or was it added later on for industrial usage?

2b.These contamination shields (rubber mats) are almost always decayed and removed in the remaining Keyboard Bs used these days. For regular usage, if kept in a clean environment, does this pose any concerns for the longevity of the keyboard? Or should it be fine to leave them without a shield long term, if the conditions are kept relatively clean?

As others have said, thank you again for your time and for answering questions.

[Dick Harris]

Hi Dick. I'm glad to see you around. I have had some questions for a long time and I would appreciate if they were answered:

We know for patents and the categorization of IBM switches in general that there were a lot of candidats to be in the place which was finally ocuped by beamsprings switches, and later buckling spring switches. Do you remember (or even better, have documentation) about the obscure switches that never went out of the oven?

The development program that resulted in the beamspring keyboard considered several OEM keyboards and IBM designs. A linear force keymodule from IBM Kingston (I have on documentation of this design.), a “U spring” (patent 3,693,059) and the beamspring were the IBM offerings. The beam spring was the winner.
A short story: There was a meeting scheduled with management in Armonk, NY to report your development status and seek manage support for the direction we wanted to go. We had decided to offer an IBM solution and prototyped both the Uspring and the beamspring in single modules. Our recommendation was going to be the beamspring and we modeled a full keyboard in a hard sided briefcase that IBM service personnel used to carried their tools. The model was battery powered and had indicator lights that revealed the binary code for the key that was pressed. It was pretty cool at least for the day. The high level manager who we all reported to was a very vocal and instantly decisive sort of guy. He was presented the individual modules of both the Uspring and the beamspring. After a few depressions of each, he through the Uspring against the wall and continued to depress the beamspring. My second level who was leading the presentation said that he anticipated our executive’s preference and opened the briefcase. I understand that model stole the day. Disclaimer: This story was relayed to me. I did not attend the meeting.
When we turned our attention to a lower profile and less costly keyboard design, we considered IBM designs described in patents 4,095,066; 4,118,611 and 4,466,302. An OEM search was conducted and the Honeywell magnetic Hall sensing keyboard was considered. The bucklingspring was the winner and the Honeywell OEM keyboard a backup. We were on a schedule driven by IBM’s first PC. I think the Honeywell keyboard had been tested and qualified for another IBM system, but I’m not sure. It was fortunate that we had the Honeywell keyboard as a contingency. When we tooled the bucklingspring for manufacturing, we discovered that on a number of keypositions would open up when the key was fully depressed causing double entries. We learned later that the tolerance on angel of angle 15 in patent 4,118,611 was such that it could interfere with the upper section of the buckled spring and create a moment on the pivot plate that could rotate it slightly and lift it off the circuit board. The Honeywell keyboard shipped for about a year with the PC while the entire bucklingspring design was reviewed from every angle.

Is the overall profile of the beamspring keyboards fully intentional or just a trivial decision of the era? I'm talking about the stair configuration of the keycaps, produced by the angled stems, the sculpted profile and the angle of the keyboards by themselves.

The beamspring keyboard design was designed to emulate as much as possible the IBM Selectric typewriter especially the presentation of the key tops to the operator. The keytops were the same as those on the Selectric and the stems allowed the side profile of the key button rows the match the Selectric. IBM typewriters were the gold standard that any successful modular design would have to match.
When it was time to develop what became the bucklingspring keyboard, IBM had updated the typewriter with a project that I knew as Plaudit. It had a more modern key shape and some differences in keytop angles. Again, the bucklingspring module had to match the typewriter. The keytop was integrated into the keystem to eliminate parts, and we wanted a common key shape if possible. Slight key shape adjustments, the curved mounting plates and curved circuit board got us close enough.
Another innovation on the bucklingspring keyboard was using subliming dye to print the legends on the keytops. The double shot molds of the day were very expensive. I was told at the time that the mold for a single key was between $8 and $10K, and we were maintaining thousands of keybutton part numbers for all the configuration of the beamspring keyboards. Printing was one way to avoid much of this expense, but the legends were not durable enough.
Another story: For some forgotten reason a couple of us in keyboard development attended a seminar given by the textile department at NC State University. During the presentation it was mentioned that there was a process that decorated polyester cloth for draperies and other applications. The process consisted of printing patterns on rolls of paper and storing blank cloth. When an order was received the desired pattern was selected and ironed (rolled) onto the polyester cloth. The result was an amazingly sharp, colorful image that actually penetrated the fiber of the cloth. At the conclusion of the seminar we cornered the presenter and explained our problem. By the way, our keytop/keystem was polyester. He asks us to follow him to his office where he gave us a black sheet of paper that had been printed with black subliming dye. When we got back to the office, we immediately cut a star out of the paper, gathered a handful of blank keybuttons and headed for a hot-stamp press in manufacturing. We soon had a keybutton with a sharp, black star with the side of the keybutton shaves away to reveal at least 5 mils of dye penetration. This was more than enough to sustain any wear that we had ever seen on a polyester button. Entire keyboard assemblies could now be printed from a single sheet of paper with a vacuum press that we developed. The cost of building potentially thousands of molds and the logistics cost of individual keybuttons could be avoided.



It is a delight to be able to ask this to the brain behind such pieces of engineering. Thanks, fella!

[Dick Harris]

Thank you immensely for the time you gift to us!

1. How did the Keyboard B evolve into Keyboard F? I assume, a driving force was cost reduction and reduction of switch height. What would interest me is the nomenclature. The B, F, and M are well known models, but have there been others built? Was there ever the C, D, E - Switch design (on paper, model, or in small production runs)? And was Keyboard A the early MicroSwitch variant?

I think I covered the transition from B to F on another post, but if you have additional questions please let me know. I not sure about Keyboard A. Keyboard C was a metal buckling dome contact for use in a control panel like application. It was never produced. If D and E exist, I don't believe they use a unique keymodule.

2. I am writing on a Technical Handbook for the Model F keyboards and would like to add quotes mentioned here to the book (obviously not without prior approval). I am also very interested on the history and would like to add the one or other anecdote to the manual (https://www.bucklingspring.com).

I would be interested in contributing. Just let me know how.

Can't thank you enough, Dick, for the joy you brought into my life with your inventions.

God bless you!

[Dick Harris]

Dick,

1. Is it true that the need for a new keyswitch design to follow the beamspring, leading to the buckling spring, was due to a new ISO standard about keyboard height?

Our design requirement was to reduce the height of the keyboard. I don't recall exactly what influence the ISO standard had on the design requirement.

2. Are you aware of the modern recreations of the 4704 line of banking terminal Model F keyboards? What do you think of that endeavor?

I've been away from keyboards since 1981 and haven't kept up with the evolution of Model F.

3. Would you be willing to share info on other prototypes or sketches you may have? They are uniquely fascinating to me and I am certain many others.

Most concepts that were seriously considered are described in patents. I've scanned my memory and what technical notebooks I have and haven't come across any other designs other than the sketch posted already.
Here is a list of my patents related to keyboards: 3,662,138; 3,693,059; 3,694,606; 3,699,296;4,095,066; 4,118,611; 4,466,302;

[Dick Harris]

Hello Dick,

Would you be able to provide some insight into the key layout evolution from the F XT to the F AT and then from the F AT to the Model M? Were any noteworthy/unusual ideas floated before settling on the layouts that made it to market or was it more of a gradual evolution?

My work focused on the technology elements of the keyboard mechanism. Others determine the key layouts.

[Weezer]

"Here is a list of my patents related to keyboards: 3,662,138; 3,693,059; 3,694,606; 3,699,296;4,095,066; 4,118,611; 4,466,302;"

Hi, this reminds me...was the patent request for the beamspring switch rejected in the US? The switch is patented in UK patent GB1363777 (A) but we've been unable to locate the corresponding US patent. Thanks.

[darkcruix]

...
I would be interested in contributing. Just let me know how.
...

Thank you so much for the insights so far!

I am trying to do the natural thing and write a book that allows owners / future owners to maintain their Capacitive Buckling Spring keyboard. I want to show the inner workings, but also how to create and adjust the controller firmware.
My hope is to collect a lot of the information scattered across the internet and knowledge from the people in this wonderful forum and pack it up into a book. I also took up the challenge and test everything I explain so it is not just a data collection.
Because I always enjoyed reading books that contain history and anecdotes, I was hoping I can ask you questions the enrich it with first hand stories.
Why some things are like they are, always intrigued me. As you said before, the reason why we have QWERTY is as simple as reducing the jamming of the hammers in typewriters (or is it?).

1. Noise Level or Audible Feedback
I remember my early days at work, when the overall sound in the open spaces was a different one than these days. The last time I went through our office (yeah, this is now some time ago due to Covid), it was mostly quiet. A voice here and there, but that is it. The early days had this mixture of people on phones, printers and copier machines in the background. Maybe a fax machine was ringing. You also got the keyboard sound that was so distinctive of the era. The keyboard clicking wasn't louder than the average office space if I recall correctly.
Was it an important design feature to have not only a haptic but also a audible feedback? Was it designed to sound in a specifc way and did you spend time on adjusting it?

2. Nomenclature
I am explaining the Barrel Plate Sandwich with its parts in the document and I used nomenclature that I felt is most descriptive, but I wanted to hear your thoughts on it and what the original "names" have been.
Find attached to this post the page I am referring to.

3. Modular Design
The modular design of the Capcitive PCB and Assembly allows modifications of the switch or key arrangement to a certain degree. Was this already a requirement of the early design?

4. Sensing a Capacitive Change
Sensing a capacitive change when the flipper hits the PCB isn't trivial - at least not as trivial as a ohmic switch detection. To get this right, I assume the controller design was done by another team? They already had some experience there with the fly plates of the Beamspring, but filtering the noise (so quickly) on the matrix in the late 70s or early 1980 myst have been challanging?

[SneakyRobb]

Hi Dick,

Again big thanks. I have a bunch of questions but I'll space them out.

I have been using a 3d printer with a conductive filament to print beamspring capacitive elements. I have also tried copper tape and a conductive paint for model f capacitive elements that I printed in photo resin. The model f reproduction used plastic Torayca ASHT-18A with 18% carbon.

1.
Do you happen to remember what plastic the beamspring and buckling spring elements were as well as the conductive material?

I would think something like graphite would work but given the capacitive nature I wonder if that would create dust etc that would mess up sensing. As graphite itself is quite lubricating and it could flake off. I would guess that the f flipper was likely a self lubricating type of plastic. When you cut a model f flipper it almost shatters like pencil lead. My 3d printed beamspring elements with the conductive plastic worked perfectly and were sensed even by the original PCB.


2. Idollar flipper shape
We can see in this post from user idollar.

viewtopic.php?t=9715

How the model f flipper has several variants. Round, notch and cut corners. They are seemingly interchangable. They have even been found in the same keyboards with no pattern or reason for their use. Do you know why these variants exist? Did different factories make different ones?

Why do the springs have slightly different angles between the variants.

3. Floss mod
We discussed how you have seen the "floss mod" method of model f spring ping dampen.

wiki/Dental_floss_mod

As well as how you had experimented with similar sound dampening. Could you elaborate on whatat methods did you try? Fibrous material like floss, foam plugs, grease etc?

Were there any attempts to silence beamsprings? If not, how would you do it now? My personal thought would be putting thin foam on-top of the capacitive plate so when it goes up it cant smack as much.


4.
We can see in
US Patent 3699296
That you had designed a circuit closing buckling spring. This is very cool. I think you briefly mentioned on the phone about a double flip action or similar prototype.

Can you elaborate on that?

Was there ever at attempt at having a conductive surface inside a model f housing to have a double action switch

5.
This is one part of user HaaTa's excellent keyswitch testing data force data.

https://chart-studio.plotly.com/~haata/68

For the IBM switch section we can see that non-angled stem beamsprings like in my IBM 5110 unit are seemingly different in force from angled stem beamsprings. Such as in the 3278. To me the 5110 switches are lighter.

The switches look basically identical to the eye. Do you know why the more common angled stems have a different force curve? Is the beam itself different?

[SneakyRobb]

6.
The original beamspring PCB sense card had both sense pads on the front of the PCB. So both "rows" and "columns" on the front.

Do you know why later PCB sense cards had rows and columns on opposite sides of the card? Was the change made so the model f PCB could be bent into the curve with less PCB damage?



7.
Do you know anything about IBM 9002

viewtopic.php?t=13319

Specifically the XT with integrated top membrane area? Did you even see or design similar pads of membrane switches.


8.
How were the estimates of x-million keypress lifetime durability calculated? Was there a machine that pressed the key over and over etc? Engineering calcs or some such?

9.
It seems like the Beamspring switch used selectric keycaps. Thus the metal slider has a similar fork shape for the keycap. How would the switch design have been different if you didn't need to use those keycaps as a starting point.

10.
Did you have any design input on the 2-part keycaps where you can swap legends on the model m? Given that beam/selectric can be actuated without a keycap, was there any model f prototype with a separate keycap vs the keycap being a functional element like in the final design? At that point did you realize you could reduce height by having the keycap slot over the housing?

11.
Where were beamspring boards made? My 3278 case for instance said "IBM 1977 Canada."

12.
Did you design the spacebar stabilization system of the F XT keyboard? The space bar has a underbarrel stabilization spring that require disassembly to reinstall. Why was that method chosen over the exterior wire stabilization like on earlier beamsprings 3277, 5251 etc.

[SneakyRobb]

13.
Did you have any involvement in the stabilization inserts for keycaps on model f/m. I.e. those plastic tubes that get inserted into empty barrels instead of wire stabilization for wide keys. If so how did that come about.

14.
Why were 2 unit wide beamspring keycaps often employed with no wire stabilization or stepped caps. They did have some tabs that were underslung I think. The F XT double unit keys were often stepped I assume to avoid off center binding? Beamspring has a central cylinder shaft which seems like the amount of friction points with the housing would be low vs more surface contact areas in model f. What are your thoughts on stabilization methods between the mechanisms and wires, stepping, inserts etc

15.
In the 1960s IBm products were often colourful. Available in red blue yellow green. There are even bright blue selectric keycaps people use on beamspring boards. By the 1980s everything was mostly gray. There is some debate about the cause of that change. Do you have any insight why that occurred? Was it some iso or other international standard thing?


16.
On the model m the barrel housing is made out of a more slippery type of plastic than the model f barrel. Model m keys feel less slick than model F keys. So if you put f keys into m. It feels very smooth. Was the slickness of the keycaps on model f dependent on some slippery plastic additive? Or was the keycap straight PBT plastic.

17.
You mentioned beamspring keycaps being made of SAN plastic. Was the move to the model f keyplastic for a functional reason, or did that model f plastic behave better with the dye sublimation process? Were there model f keycaps made with that SAN plastic?

[SneakyRobb]

18.
Was a curved plate beamspring ever considered as a proposal.

19.
Did you design all of the keycaps? F AT came out later and has a large stepped enter key, did someone use your basic design or were you consulted for those.

20.
Why do beamspring housings have a slot running down 2 opposite sides. Was this so some kind of parallel pliers could grip the switch from below?

21.
Was there an official IBM method of reattaching a buckling spring that fell out of the module. People now often use a chopstick to press it on. How would you have restuck a spring that came off while the board was assembled?

[dcopellino]

Thank you Dr. Harris. In your inventions I see what Leonardo used to say: "Simplicity is the ultimate sophistication".

IMG_20210320_160103.jpg (2.67 MiB) Viewed 15768 times

[Johnbo]

Hello Dick, thanks so much for answering these questions!

I would love to hear a little about your process for building the switch prototypes.

It makes sense that each design starts out life as a sketch, but what happens from there? Did you have a full machine shop at your disposal to create the individual prototype components as you needed them? Did you draft a blueprint which was then sent to a separate machine shop? How much of what you worked with were "off the shelf" components?

Basically, I'm curious about what it takes to go from an idea in your head to having an real life, functional prototype switch.

[Dick Harris]

Spoiler:

Spoiler:

Spoiler:

Spoiler:

Hello, Dick! I hope you are well.

Would you happen to know more about the origins of the AT&T buckling springs that hold certain similarities to the Model M design? They have been observed to have some striking resemblances to your design but are thought to be an AT&T product. Nate, I don't know anything about an AT&T buckling spring. Sorry.

Spoiler:

Similarly, IBM Japan contracted out the manufacture of several Japanese-market keyboards such as the 5576-A01 to Brother. Would you happen to know something about those as well? I've read recently in other posts about this, but I couldn't bring anything new to the discussion.

Spoiler:

Your original capacitive design is an incredible product, and I wish you the best!

-Nate

Yes, both the AT&T and the IBM Japan Brother keyboards look like very much like the original IBM work. These developments happened long after I left keyboard development in 1081. I would be surprised that either AT&T or Brother would invest in the development and tooling cost to reverse engineer the Model F or M. Maybe IBM provided the parts to Brother and the total keyboards to AT &T, but I really don't know.

[Dick Harris]

"Here is a list of my patents related to keyboards: 3,662,138; 3,693,059; 3,694,606; 3,699,296;4,095,066; 4,118,611; 4,466,302;"

Hi, this reminds me...was the patent request for the beamspring switch rejected in the US? The switch is patented in UK patent GB1363777 (A) but we've been unable to locate the corresponding US patent. Thanks.

Is seems that the beamspring keyboard was neither patented nor published. I don't recall exactly why.

[Dick Harris]

Thanks for the reply, Dick!

Can you expand more on the ultrasonic delay line thing? I've never heard of that before.

Also you mentioned above that you wrote your own patent disclosure. When we read your patents how much of that is in your own words and how much of that was done by an attorney working their funky patent magic? Did the ratio of how much was yours change as time went on? Furthermore were the drawings done by you or did you have someone else make them for you?

Not to drown you in questions but you also mentioned that the switches had to make/break at the tactile/clicky event and they needed to support N key rollover, were there any other major considerations for the design you had to take into account?

Thanks again for your time answering these questions!

As I recall, we discovered an invention that was a linear displacement transducer which consisted of a very thin hollow tube that could be several feet in length. A torsional impulse was introduced at one end which produced a shock wave that traveled through the tube at the speed of sound of the tube. A magnetic could be placed at any distance along the tube. With proper magnet design and tube material, the magnet would reflect the shock wave and the location of the magnet could be calculated by sensing the returned shock wave. The time relationship between the initiation of the shock wave and its return would determine the location/position of the magnetic - like radar. We considered magnets on the keymodules and forming the tube so that the tube would pass beneath each keymodule each of which would have its own distance from the torsional transducer. It would replace the capacitance sensing technology. We encountered problems with secondary reflections from mounting points and the advantages over capacitive did not justify the development effort.

IBM had an excellent group of patent attorneys. They did thorough patent searches and excelled in describing inventions and crafting claims to protect the desired feature of the inventions. A decision to close, publish or file the invention would be made. If the decision is to file a patent, the patent attorney would draft the invention, the inventor would review the draft and the patent application would be filed. As time went on, I would pick up on some of the language and importance of ways to describe things. Occasionally, I would suggest wording, but the patent attorney was responsible for the patent application. Drafting the figures in a patent application is a specialized skill. Our patent attorneys had in-house and contract draftsmen who created the figures most of which were adaptations of those in the initial invention disclosure.

Other considerations for a keymodule design were: product cost, smooth key guides, key binding, key force, key stroke, dust and debris resistance, acoustics, liquid resistance, ability to configure, robustness of the design, life of the design, provide a means to sense reliabily, environmental considerations (no yellowing plastic ,etc.) , recycling (just beginning in the 80's).

[headphone_jack]

Hi Dick, it's an honor to be talking to you!

I have two questions for you.

1) What were the priorities for you and your team during development of switches? Was it overall keyfeel, cost effectiveness, durability, or something else that took priority over everything else?

2) What are some of your favorite stories from your time working at IBM? I expect you must have had some real high notes working there!

Thank you again for answering my questions, I am a huge fan of your work! My 122 key Keyboard F still gets loads of use, even after so long!

Spoiler:

PXL_20201201_045234488.jpg (72.05 KiB) Viewed 15385 times

[Mandarbmax]

Thanks for the reply, Dick!

Can you expand more on the ultrasonic delay line thing? I've never heard of that before.

Also you mentioned above that you wrote your own patent disclosure. When we read your patents how much of that is in your own words and how much of that was done by an attorney working their funky patent magic? Did the ratio of how much was yours change as time went on? Furthermore were the drawings done by you or did you have someone else make them for you?

Not to drown you in questions but you also mentioned that the switches had to make/break at the tactile/clicky event and they needed to support N key rollover, were there any other major considerations for the design you had to take into account?

Thanks again for your time answering these questions!

As I recall, we discovered an invention that was a linear displacement transducer which consisted of a very thin hollow tube that could be several feet in length. A torsional impulse was introduced at one end which produced a shock wave that traveled through the tube at the speed of sound of the tube. A magnetic could be placed at any distance along the tube. With proper magnet design and tube material, the magnet would reflect the shock wave and the location of the magnet could be calculated by sensing the returned shock wave. The time relationship between the initiation of the shock wave and its return would determine the location/position of the magnetic - like radar. We considered magnets on the keymodules and forming the tube so that the tube would pass beneath each keymodule each of which would have its own distance from the torsional transducer. It would replace the capacitance sensing technology. We encountered problems with secondary reflections from mounting points and the advantages over capacitive did not justify the development effort.

IBM had an excellent group of patent attorneys. They did thorough patent searches and excelled in describing inventions and crafting claims to protect the desired feature of the inventions. A decision to close, publish or file the invention would be made. If the decision is to file a patent, the patent attorney would draft the invention, the inventor would review the draft and the patent application would be filed. As time went on, I would pick up on some of the language and importance of ways to describe things. Occasionally, I would suggest wording, but the patent attorney was responsible for the patent application. Drafting the figures in a patent application is a specialized skill. Our patent attorneys had in-house and contract draftsmen who created the figures most of which were adaptations of those in the initial invention disclosure.

Other considerations for a keymodule design were: product cost, smooth key guides, key binding, key force, key stroke, dust and debris resistance, acoustics, liquid resistance, ability to configure, robustness of the design, life of the design, provide a means to sense reliabily, environmental considerations (no yellowing plastic ,etc.) , recycling (just beginning in the 80's).

That linear displacement transducer sensing apparatus is really interesting! I can't quite make out the advantages of it relative to capacitive sensing though. Do you have an pictures of such a device anywhere?

Thanks for the peak into the life of an inventor! As it happens I'm a patent examiner myself so seeing how things are on the other side is very interesting! It is nice to see that your attorneys did a good job searching through prior art for you, I've worked with some who absolutely do not.

Now that you mention them it is quite clear that your other considerations for the keymodule designs were taken seriously, every one of those points is considered to be a strength of not only the beamspring switches but also both of the buckling spring designs made by IBM with the exception of recyclability because these things just don't die.

Thanks yet again, Dick.

[SneakyRobb]

Hi Dick,

Was there ever a reverse dye sub legend as far as you know?

Like you dye the entire keycap except for the legend? Maybe the legend has a sticker covering it during dying

Robb

[SneakyRobb]

Hi Dick,

Would you be open to receiving random key switches in the mail and then offering thoughts? For instance we users could send you various key switches and you could critique and discuss them? Would be interesting to get your takes

Robb

[SneakyRobb]

Hi again Dick,

What are your thoughts on these different capacitive pcb designs from users DMA and wcass?

viewtopic.php?t=19166

Were there other patterns that IBM investigated? Why are the 2 rectangles that were on the final PCB slightly different sizes

[Weezer]

"Here is a list of my patents related to keyboards: 3,662,138; 3,693,059; 3,694,606; 3,699,296;4,095,066; 4,118,611; 4,466,302;"

Hi, this reminds me...was the patent request for the beamspring switch rejected in the US? The switch is patented in UK patent GB1363777 (A) but we've been unable to locate the corresponding US patent. Thanks.

Is seems that the beamspring keyboard was neither patented nor published. I don't recall exactly why.

Since it was never patented, what is the composition of the plastic you used in the design?

[Dick Harris]

Thanks for the reply, Dick!

Can you expand more on the ultrasonic delay line thing? I've never heard of that before.

Also you mentioned above that you wrote your own patent disclosure. When we read your patents how much of that is in your own words and how much of that was done by an attorney working their funky patent magic? Did the ratio of how much was yours change as time went on? Furthermore were the drawings done by you or did you have someone else make them for you?

Not to drown you in questions but you also mentioned that the switches had to make/break at the tactile/clicky event and they needed to support N key rollover, were there any other major considerations for the design you had to take into account?

Thanks again for your time answering these questions!

As I recall, we discovered an invention that was a linear displacement transducer which consisted of a very thin hollow tube that could be several feet in length. A torsional impulse was introduced at one end which produced a shock wave that traveled through the tube at the speed of sound of the tube. A magnetic could be placed at any distance along the tube. With proper magnet design and tube material, the magnet would reflect the shock wave and the location of the magnet could be calculated by sensing the returned shock wave. The time relationship between the initiation of the shock wave and its return would determine the location/position of the magnetic - like radar. We considered magnets on the keymodules and forming the tube so that the tube would pass beneath each keymodule each of which would have its own distance from the torsional transducer. It would replace the capacitance sensing technology. We encountered problems with secondary reflections from mounting points and the advantages over capacitive did not justify the development effort.

IBM had an excellent group of patent attorneys. They did thorough patent searches and excelled in describing inventions and crafting claims to protect the desired feature of the inventions. A decision to close, publish or file the invention would be made. If the decision is to file a patent, the patent attorney would draft the invention, the inventor would review the draft and the patent application would be filed. As time went on, I would pick up on some of the language and importance of ways to describe things. Occasionally, I would suggest wording, but the patent attorney was responsible for the patent application. Drafting the figures in a patent application is a specialized skill. Our patent attorneys had in-house and contract draftsmen who created the figures most of which were adaptations of those in the initial invention disclosure.

Other considerations for a keymodule design were: product cost, smooth key guides, key binding, key force, key stroke, dust and debris resistance, acoustics, liquid resistance, ability to configure, robustness of the design, life of the design, provide a means to sense reliabily, environmental considerations (no yellowing plastic ,etc.) , recycling (just beginning in the 80's).

That linear displacement transducer sensing apparatus is really interesting! I can't quite make out the advantages of it relative to capacitive sensing though. Do you have an pictures of such a device anywhere?

Thanks for the peak into the life of an inventor! As it happens I'm a patent examiner myself so seeing how things are on the other side is very interesting! It is nice to see that your attorneys did a good job searching through prior art for you, I've worked with some who absolutely do not.

Now that you mention them it is quite clear that your other considerations for the keymodule designs were taken seriously, every one of those points is considered to be a strength of not only the beamspring switches but also both of the buckling spring designs made by IBM with the exception of recyclability because these things just don't die.

Thanks yet again, Dick.

The the anticipated advantage of the ultrasonic delay line was that it seemed very debris resistant and possibly cost reduction by replacing a large capacitive circuit card with a single wire. Also, having a single transducer for an entire keyboard had the potential to simplify the decode circuitry with corresponding cost reductions. We were able to purchase several straight transducer which were maybe 4 feet long. I think they were designed to be lathe bed transducers. We disassembled them, bent the wire back and forth beneath keypositions. We never were able to achieve satisfactory results.

[Dick Harris]

Hi Dick,

Again big thanks. I have a bunch of questions but I'll space them out.

I have been using a 3d printer with a conductive filament to print beamspring capacitive elements. I have also tried copper tape and a conductive paint for model f capacitive elements that I printed in photo resin. The model f reproduction used plastic Torayca ASHT-18A with 18% carbon.

1.
Do you happen to remember what plastic the beamspring and buckling spring elements were as well as the conductive material?

I would think something like graphite would work but given the capacitive nature I wonder if that would create dust etc that would mess up sensing. As graphite itself is quite lubricating and it could flake off. I would guess that the f flipper was likely a self lubricating type of plastic. When you cut a model f flipper it almost shatters like pencil lead. My 3d printed beamspring elements with the conductive plastic worked perfectly and were sensed even by the original PCB.

Beamspring - My recollections: Housing was polycarbonate that may have been filled with teflon, the keystem was delrin with a glass fill, the flyplate I believe was a carbon filled material, the beam spring was an very expensive alloy with high tensile strength and the fly spring a flat spring stock maybe stainless steel. These questions have exposed my poor record keeping. Sorry.

Model F - Again, my recollections: Housing may have been the same polycarbonate as the beamspring keyboard, I'm pretty sure that the keybutton/keystem was polyester 310. I think the flipper was a carbon filled thermoset plastic. The buckling spring was standard spring steel wire.

2. Idollar flipper shape
We can see in this post from user idollar.

viewtopic.php?t=9715

How the model f flipper has several variants. Round, notch and cut corners. They are seemingly interchangable. They have even been found in the same keyboards with no pattern or reason for their use. Do you know why these variants exist? Did different factories make different ones?

The outer corners of the flipper have notches, curves and chamfers to ensure clearance with the housing. I think the variants are the toolmakers choice and should be interchangeable, but it is possible that the variant is there to identify some other change in the design which would be too subtle to detect.

Why do the springs have slightly different angles between the variants.

I don't know.

3. Floss mod
We discussed how you have seen the "floss mod" method of model f spring ping dampen.

wiki/Dental_floss_mod

As well as how you had experimented with similar sound dampening. Could you elaborate on whatat methods did you try? Fibrous material like floss, foam plugs, grease etc?
We did assemble an entire keyboard with foam rubber strips inserted in the bucklingspring. It was tested and found note to be as desirable as the unmodified, louder version.

Were there any attempts to silence beamsprings? If not, how would you do it now? My personal thought would be putting thin foam on-top of the capacitive plate so when it goes up it cant smack as much.

As you suggest, adding foam to the top of the flyplate or to the striking surface on the housing. On the key release, it may be possible to create a super flat circuit board and lap the flyplates super flat creating an air cushion to reduce the sound as the flyplate hits the circuit board.

4.
We can see in
US Patent 3699296
That you had designed a circuit closing buckling spring. This is very cool. I think you briefly mentioned on the phone about a double flip action or similar prototype.

Can you elaborate on that?As I recall, there were two problems with contacts using the lateral force of a buckled spring. One, a soldered connection to one end of the spring reduced the feel and a pressure connection was a second contact that might jeopardize reliability. The second problem was that he switch was teasable on release. In other words, the contact can open before the spring snaps straight.

Was there ever at attempt at having a conductive surface inside a model f housing to have a double action switch
The first thought for Keyboard F was a flipper plate that pivoted on a center circuit pad and when the spring buckled the flipper transferred from one sense pad to another sense pad. The surface areas were too small and the transfer feature was of little value so the design reverted to the current sense pad arrangement.
5.
This is one part of user HaaTa's excellent keyswitch testing data force data.

https://chart-studio.plotly.com/~haata/68

For the IBM switch section we can see that non-angled stem beamsprings like in my IBM 5110 unit are seemingly different in force from angled stem beamsprings. Such as in the 3278. To me the 5110 switches are lighter.

The switches look basically identical to the eye. Do you know why the more common angled stems have a different force curve? Is the beam itself different?

I really don't know, but I am pretty sure that the inner workings of the keymodules are the same. This data seems to be a single module and the differences might be normal variations from key to key.

[Scarpia]

This is so much fun, fascinating to hear about the beamspring and Model F designs from the actual inventor. What an absolute treat!

As a daily user of a (reproduction) Model F keyboard I thank you for the truly inspired design. Even after more than a year on my desk it still brings a giddy smile on my face every time I sit down to write or code.

[devo343]

First Off thanks for starting one of the best made keyboards ever made. Ive never used a model F but i have used both a Unicomp M and a 1993 M.

My question is when you first started designing the switch did anyone ever doubt what you were doing?.

[rpl]

Hi, Mr Harris,

It's something of an honour to chat with the guy whose genius saved a couple of generations of office workers from RSI's, towhit the 'B' and 'F' mechanisms.

Going back to the 70's - 90's, I was a data entry operator. Temp'ing between other jobs/careers through agencies (and occasionally on my own) I logged maybe a quarter-billion keystrokes, quite a bit of which were on s/34 and s/38 terminals ¹.

The tactility of IBM's beamspring and buckling-spring boards meant being able to pull entire shifts, not just reasonably well, but almost entirely in-the-zone ("turbo mode") without strain. (Not even counting the grandeur of a massive chunk of steel, I think I liked the B models better for data entry ².)

Sorry, if some of my queries aren't in your wheelhouse : earlier posters in this thread covered lots of the technicalia.

How much of the decision process regarding keystroke ergonomics (things like force vs displacement, down-up hysteresis) came from a clean slate, and how much based on earlier machines' development.

In that vein, manufacturing costs aside, in what way did the target demographics and computer state-of-the-art influence your designs ?

Anything that you think should have made it into those boards, but didn't ? or the other way'round. Not limited to your own area.

I (vaguely) recall the 5251 keys had a bit of an horizontal-plane wiggle to them. If (on a new, unused keyboard) that even existed, was it deliberate ? ³

Were beamspring keyboards speed governed ? I think I read where they waited until the solenoid was re-armed before accepting the next keystroke. I certainly didn't notice any inhibiting factor, though I did approach them with more articulation than the (linear keyswitch) boards from other mfr's.

In regards the solenoid, who was it said "We have to make these things louder !!!" ? (which probably confuses anybody under 60, and even makes me smile, though I commonly used the feature for its intended purpose).

Thanks again,

Pat

--------

¹ as well as the 5251 and F122(?), also occasional dabbling on Selectrics, 029's, 3741's, 3278's... possibly one of the Displaywriters... as well as the linear offerings of pretty much everybody else.

² I think that was due to the square-spherical keycaps giving two axes of positioning feedback, as opposed to the cylindrical type's one. Though, the F keycaps were certainly more comfortable on fingers at rest.

³ I actually liked it : gives the fingers something to do while waiting for the system to catch up, instead of locking frozen like a gargoyle.

[Dick Harris]

Hi Dick,

Was there ever a reverse dye sub legend as far as you know?

Like you dye the entire keycap except for the legend? Maybe the legend has a sticker covering it during dying

Robb

Robb, Reverse is interesting. It is new to me. We never discovered a way to sublime white characters and reverse sublimation printing could theoretically do this. The way we did sublimation was to print the characters on a piece of paper and "iron" them onto the keybutton. Reversing the process would require inventing a way to uniformly dye the keybutton everywhere except for the desired character. A challenging problem, but it would provide any color characters as long as a blank keybutton can be molded that color.