Model MF - GB CLOSED

[Techno Trousers]

I like the idea of boxes that are as durable as possible. Your links take me to the Deskthority search for some reason, but do they happen to have an option for acid-free cardboard? Certainly overkill for this project, but if we happen to want to store some paperwork in the box along with the keyboard, it would help to preserve that (on the timescale of decades, of course). I am actually thinking about creating some paperwork to go along with the keyboards, to give a future recipient of one my MFs some context about what makes them so special.

[lot_lizard]

Your links take me to the Deskthority search for some reason

This is the new hashtag support the added to the DT forum. Anytime it encounters a # symbol, it represents as a search link with the proceeding text.

but do they happen to have an option for acid-free cardboard? Certainly overkill for this project, but if we happen to want to store some paperwork in the box along with the keyboard, it would help to preserve that (on the timescale of decades, of course).

For that we would almost have to do a box within a box (i.e.... MassDrop and the like) unfortunately. Which is certainly doable, but gets pricey quick since we need containers 2x the volume of the WhiteFox if we are going to both cushion and include space for parts. We certainly could have gotten more exotic with the box build, but we were targeting shipping and storage containers vs. ornamental pieces. Maybe next time when we have smaller profiles. The upside is we will own the box die now ($750 for a million strike guarantee), so that expense will be capitalized with this GB. Future stuff we can concentrate on the "innards". In the meantime, you could always slip the literature into a baggie.

[Techno Trousers]

Yep, great idea. I can put the documents into a mylar bag or equivalent. It's more cost effective and adds an extra layer of protection for those. Thanks for the explanation about the hashtag. I haven't been keeping up with the board software changes since I often use Tapatalk.

[lot_lizard]

The standoff spacers have been ordered (m3x4mmx4mm). As a reminder, these hold and ground the PCB to the backplate at 3 points on the SSK and 4 points on the Full-Size. The top plate is then tensioned down into the standoffs. They are the only points where the plate fasteners are attached from the top. Unfortunately, Fastenal didn't carry the size and style we needed, so had to use conventional means to procure.

IMG_0545.JPG (117.54 KiB) Viewed 5461 times

[lot_lizard]

The final version of the foam is finally put together. I realize in advance that I have WAY more interest in this aspect of our project than others, but strongly believe it is one area where we could have the greatest improvement over any existing architecture to date. Came up with several prototypes that were tested in reality (maybe 10?), and settled on this being the design.
Again, the goal is

• first consistent key feel throughout the board (remember we are closer to an F122 than a 4704
• secondly consistent audible feedback

I believe we achieve both. The core ring is made of silicone rubber that both helps seal barrel openings and stabilizes with consistent pressure at key points. There are two flavors of the ring. Those where the barrel front is supported by the overhang of the previous rows ring, and isolated (includes bottom row) barrel rings. Also, should we make our own barrels in the future this two part approach would also accommodate flat plate profiles, the silicone stabilizer rings will still fall within spec.

The voids not accounted for by the stabilizer rings will be filled with MLV (mass loaded vinyl). At least that is the goal. If MLV proves cost prohibitive, the fall back is a thicker than normal closed cell foam like neoprene. I have ordered several foams to test on the laser cutter, and will report the results.

The following are the renders to help visualize. Regardless of the material chosen for the secondary foam, this design will be the final unless issues arise. Note that the secondary (form factor specific cutout) foam will sit INSIDE of the assembly, and will not be sandwiched between the top and bottom plate on the edges.

Standard silicone stabilizer ring. Note that rectangle at the top of the barrel base stabilizes the rings barrel AND the barrel on the row above

foamSilicon2.png (27.93 KiB) Viewed 5420 times

Standard silicone ring fitted on a barrel. The example shows both barrel locks (XT and AT)

foamSiliconBarrel.png (40.42 KiB) Viewed 5420 times

Stabilizer ring for isolated barrels (barrels that do not have a row directly below them)

foamSiliconIsolated.png (27.64 KiB) Viewed 5420 times

Isolated stabilizer barrel fitted on a fictional dual lock barrel

foamSiliconIsolatedBarrel2.png (37.71 KiB) Viewed 5420 times

Standard and isolated rings fitted on stacked rows. Note the overhang of the ring on the lower row stabilizes the one above

foamSiliconDualBarrel.png (63.03 KiB) Viewed 5420 times

Closed cell foam cutout for the SSK to fill voids

foam.png (39.69 KiB) Viewed 5264 times

Closed cell foam cutout for the Full-Size to fill voids

foam.png (40.95 KiB) Viewed 5264 times

[Techno Trousers]

That looks great! I really appreciate the renders helping me visualize it. I had the impression you experimented with a lot of layers at one point. Did you come to the conclusion that was overkill? This looks like it will be very straightforward to assemble, which I like! I was a little worried about the two types of stabilizing rings, but seeing the cut foam I realized that will be a perfect guide to indicate which type to use in each place.

[lot_lizard]

There is a bug in my render for the foam sheets. It is pointing at an old version of the ring. The isolated ones have a shorter width on the support silicone so the foam isn't so delicate on bottom rows of the function, arrow, nav cluster, and 10-key. I will regenerate and update the previous ones later today

I had the impression you experimented with a lot of layers at one point. Did you come to the conclusion that was overkill?

The term "layers" was a poor choice on my part. It was always something similar to this. More of a "multi-part". There is another layer than sits under the PCB itself (separating from the backplate), but it is VERY thin. Thicker than the sheets on the original F122, but something along those lines if you want to think of it that way

[fohat]

Will the foam come "assembled" or will we be inserting the core rings into the main sheet ourselves?

If assembled, will it be glued tight or just sitting in place?

That is light-years beyond my art foam and leather punch approach!

[lot_lizard]

Will the foam come "assembled" or will we be inserting the core rings into the main sheet ourselves? If assembled, will it be glued tight or just sitting in place?

It might take me a few initial passes to get the settings just right on my cutter, but the goal is to have the silicone rings be added to each barrel before hand (a little tedious) and have just enough tension on the barrel to remain in place when inverted.


So for assembly:

• silicone rings are added to the barrels ahead of time
• top plate is inverted
• foam sheet is placed over the top
• pre-assembled barrel/ring is added
• then business as usual

The foam is thicker then the silicone rings with a much lower shore durometer value, but won't be thick enough that is makes assembly difficult. I initially considered having one side of the foam "unit" have adhesive to hold all of this together and coming preassembled, but we would need a service to do this for us (pricey because it is so artistic with small quantities).

After A LOT of testing, I am confident in it all. The silicone I settled on is a 40a shore value (so think something similar to Dr. Schol's Gel Inserts, and ~1mm thick before applying pressure. The foams I am looking at would be ~20a shore value when compressed to ~.5mm. That is the tricky part in this, getting the foam density vs. thickness when compressed to ~.5mm to be approx half of the stabilizing rings. I have ordered a few more samples now that I found foam I like through testing. While the cutter is busy cutting out these silicone rings, we will pick the exact foam to be used and order it off in bulk. Silicone rolls have been ordered btw...

[Hypersphere]

I assume you mean "silicone" rather than "silicon" (although they both contain silicon).

[lot_lizard]

I assume you mean "silicone" rather than "silicon" (although they both contain silicon).

haha... wow... yes, and thanks for catching. I updated the references (there were many). Also, the foam renders have been updated to reflect the shorter isolated barrel cross section I mentioned earlier.


About the only thing left where we could improve would be the gaps in between the barrel bases themselves. Remember they are ~4.1mm tall, and this foam would sit ABOVE that. So to prevent debris even future, we could fill those base voids. I am going to save that for another day I think though. Unless we are waiting on something at the end of all of this and I have down time. Just have to make sure whatever was selected there was at least 4mm thick and wasn't conductive in any way.

[lot_lizard]

If anyone was curious, this is a very scaled down (60dpi) version of the SVG that is being used to cut the stabilizer rings out of the silicone sheeting. The line width has been increased so that it is visible. The spoiler would be the actual hairlines at .001 inch (needed to tell the laser to cut instead of engrave), but it is hardly an interesting picture to the naked eye at this resolution. The engraver has a 36x24" cutting bed. This utilizes all of that space with a .5 margin all the way around. Total would be 1491 rings per silicone sheet (1092 standard, and 399 isolated with extra stabilization bar around the AT barrel lock). That should take the engraver a while

siliconeRings.png (1.56 MiB) Viewed 5306 times

Spoiler:

siliconeRings.png (722.64 KiB) Viewed 5306 times

[E TwentyNine]

If anyone was curious, this is a very scaled down (60dpi) version of the SVG that is being used to cut the stabilizer rings out of the silicone sheeting.

Is there a meaningful reason the pattern shifts near the bottom of the graphic?

[lot_lizard]

Is there a meaningful reason the pattern shifts near the bottom of the graphic?

The bottom 8.5 rows are the isolated version of the ring. So barrels that don't have a row underneath them that need to be further stabilized. The standard barrel rings have an overhang if you go back and look at the renders that actually supports the front of the barrels on the row above it. These "isolated" barrels wouldn't have that benefit and needed to be slightly altered

[Techno Trousers]

Wow, that's a LOT of rings. But I guess you're going to need somewhere near 10,000 total, right?

Are you planning to send a ring for every potential barrel location, or will you poll everyone as to how many they will need for their chosen layout? I guess the main differences would be in Windows keys or no, and split space bar or no.

[lot_lizard]

Wow, that's a LOT of rings. But I guess you're going to need somewhere near 10,000 total, right?

Are you planning to send a ring for every potential barrel location, or will you poll everyone as to how many they will need for their chosen layout? I guess the main differences would be in Windows keys or no, and split space bar or no.

Over 15k of them will be needed (I have several orders for myself and friends that weren't included in our totals). Shouldn't take more than 4 days, and it can be left "unattended" (have a webcam in the shop). The cutter goes through the 30a shore rubber pretty cleanly, so there shouldn't be a need for multiple passes either. The foam itself will go exceptionally quick.


Everyone will receive enough rings to account for every possible barrel location with a few spares as well. It is much easier for me to just throw in extras than to spend my time counting exact quantities. In the end, we will cut approximately 8 square yards of silicone sheeting. Just waiting for it to arrive (there was a 1-2 week backorder almost everywhere I looked). We are technically operating under a business license, so we had access to distributors that don't sell directly to the public (though our quantities were tiny). I explained that we need this as a "sample size" to prove a concept

On the 3d printing front, we are 1/3 through with the black spacers (reminder that all the whites are already complete). Moving along reasonably well

[micrex22]

There are MINOR adjustments really across the board to be honest (aside from the core cluster). Arrow/nav/10-key are too far left, Esc too far right, and only the slightest adjustment to the function row. Again, please don't think I shifted things all over creation (these are .1-.3mm bumps).

I've actually noticed this as well, and it has always bothered me (particularly with the arrow cluster being too far left). So I'm actually quite glad it's being adjusted...

Well guess what! It appears earlier Model Ms had a different orentation for the barrels (not sure if this is due to the barrel mold or the chassis itself since both are slightly different from later revisions). Check it out, pay attention to the arrow cluster and numpad, you'll see how the chassis is shifted quite dramatically from both:

Early Model M (orientation too far to the left):


Later Model M (orientation too far to the right):

[alienman82]

removed.

[lot_lizard]

forgive me for not reading this entire thread, but I haven't been on DT in over a month, is there an expected shipping schedule?

As of yet, nothing hard... I am waiting to have the latest prototype plates in hand to validate the complete package. Once we know this, we would order the PCB, controllers, and plates (which turns around in a couple of weeks). Most all other parts are either ordered, or in hand.

[E TwentyNine]

Well guess what! It appears earlier Model Ms had a different orentation for the barrels (not sure if this is due to the barrel mold or the chassis itself since both are slightly different from later revisions).

I don't know that you can pin this to early/late models, factory variation, or something else.

There's a number of factors - barrel variations relative to each other, how the barrel plate sits on the backplate, where the mounting holes are in the backplate, where the mounting towers are in the case, and where the cutouts for the keys are in the case as a whole and relative to each other.

I'd be interested to see what the sample set was for "case fitment" for the MF assembly. When reassembling an AT/XT F, the assembly can be shifted in the case before everything is buttoned up, not so much with an M.

I'm wondering if it might be of use for the mounting holes on this backplate to be oval (long horiz axis) so the assembly could be shifted slightly in the case and allow for case variations.

[lot_lizard]

I'm wondering if it might be of use for the mounting holes on this backplate to be oval (long horiz axis) so the assembly could be shifted slightly in the case and allow for case variations.

We actually don't use the back plate oval holes at all. Those pegs in the bottom shell have always been the weak link in the ruggedness of the Ms design, and often break during shipping. The new design is a tighter fit, and our oval holes have been made slightly bigger so they don't interfere.


I can't say I have tried every case version that ever exposed by any stretch, but the assembly has been tested in 5 different models of the SSK, and 5 of the full-size (along with early and late versions of the 1401).

[E TwentyNine]

I'm wondering if it might be of use for the mounting holes on this backplate to be oval (long horiz axis) so the assembly could be shifted slightly in the case and allow for case variations.

We actually don't use the back plate oval holes at all. Those pegs in the bottom shell have always been the weak link in the ruggedness of the Ms design, and often break during shipping. The new design is a tighter fit, and our oval holes have been made slightly bigger so they don't interfere.

Is there a pic of that (somewhere in this thread)?

[lot_lizard]

Is there a pic of that (somewhere in this thread)?

Three dimensions are basically identical to the stainless steel plate prototype in the first post. The oval cutouts now have an extra .5mm all the way around I believe. All of the weight would now rest against the front of the lower shell. I will try to take better pics when I get the final protoype plates in

[tentator]

Omg i just read now about that nifty silicone ring to attach to the barrel and I really like that idea!! Would really like to be able to test them znd "hear" the difference!! Are they planned to be shipped in this round? Or just an idea for the future?? I'd really like to beta test them!!

[andrewjoy]

Have you seen any of the treated top plate yet ?

[lot_lizard]

Are they planned to be shipped in this round? Or just an idea for the future??

They will be delivered with the rest of the order... Happening this go round. I have only tested on a small scale, but it is very effective.

Have you seen any of the treated top plate yet ?

I have seen the treatments, but not the treatment on our plates. They are producing the plates this week I believe, but I am unfortunately traveling through the weekend. I will try to visit them early/middle next week. Work getting in the way of our fun again

[andrewjoy]

Work getting in the way of our fun again

It always does .

Cannot wait to see both plates in that it will looks so amazing ! , Will look like the top of a beamspring plate , but with no rust.

Oh wait my beamspring has no rust .

[seebart]

Oh wait my beamspring has no rust .

Show-off!

[lot_lizard]

Had a few days of work and play time with baby girl, but have some nice updates.

• 3D spacer prints are ALMOST done. Another day or two max
• Received LITERALLY 19 samples of foam from my local foam producer to play with. We have already settled on the 30a shore silicone rubber for the barrel "ring stabilizers", but have been conflicted about the rest of the foam layer as to the material and density. Along with several samples of neoprene and the like, they included silicone foam (not rubber). This stuff is incredible. Many call it is "silicone sponge". The laser cutter absolutely OWNS 10a silicone foam (low density), and reading the spec sheet... it really is something special. The pricing is not cheap, but we accounted for up to this per square yard (thank god the laser cutter has a 36" bed). Things get dramatically less at standard roll widths.
• For our stabilizer rings on silicone 30a shore rubber, I think I am going to have a roll die made. Since we have proven it would work for flat plates as well... I would need to eat about 200 total on the run versus just using the laser to power through it, but we will own the die for future needs. We can make this up easy on [to be determined]
• Received all the fasteners from Fastenal, but still waiting on the Torx variants

[Techno Trousers]

Ooh, exciting update! Does your last sentence mean that Torx is coming but not delivered, or that you're still awaiting an answer on it?

Also, I like the sound of "silicone foam." What's the expected durability of that when kept under pressure in our keyboards?