类型: IC
作者: WaS
发布时间: 2021-06-16 09:48:25
更新时间:
2021-06-16 10:06:41
原链接:
https://geekhack.org/index.php?topic=113381.0
Hey guys the past few weekends we´ve been experimenting a bit with Jesmonite and figured we´d share our findings and results with you.
Disclaimer: We hoped to be the first ones to introduce a
jesmonite project, but Brazen/Ian was a little faster, so
make sure to check out his Mason60 project as well
All disclaimers aside, let’s get going
What´s Jesmonite?
I´m going to
borrow that one straight from the horse’s mouth:
*Jesmonite
is a composite material used in fine arts, crafts, and
construction. It consists of a gypsum-based material in an
acrylic resin.*
Why Jesmonite?
One day I stumbled upon this video
https://www.youtube.com/watch?v=GGpN61XvLtU&t=10s
of Alexandre Chappel on Youtube and saw what he was doing
with Jesmonite.
At some point this came to mind while
we were building some of our boards at Teleport and I
figured, why not try it on a small scale proto for a board.
The Material is quite unique, as it can be colored with
pretty much any color (pigments), or other additives (such
as metal powder), to give it a different look each time.
Workability:
Jesmonite comes as a powder with an additional liquid
and is mixed in a 2:1 ratio, then cast into a mold to hold
its shape. After a few minutes an exothermic reaction takes
place and that stuff dries quickly, as we learned the hard
way (who reads instructions anyways, right..?)
To sum
this up: It looked different and seemed to be somewhat of a
workable material.
Initial Concerns - Weight, Haptic and Sturdiness:
As we are all somewhat used to metal boards, we were
concerned mostly about the above mentioned points. We will
go over these points later in detail, however, we did not
want to have something light (plastic like) and were
concerned that haptics would be very similar to classical
plaster.
All concerns aside, here are our first experiments in
chronological order:
-
60% board poured into a 3d-printed mold
Not gonna spend too much time on this one. A 3d printed mold for our model was a bad idea in the first place and we had no idea how fast that stuff would react. So we ended up with a puddle of half hardened jesmonite, before we could even cast. We figured to add the picture for the laughs
- Picture: Maybe read instructions next time…*
-
Testing the waters
No idea how that stuff reacts to water, so we used some of the scraps and let them soak in water for a whole week. Didn’t do much to it, but better safe than sorry
-
Numpad with silicone mold
As we needed a somewhat smaller mold, which didn’t require two halfs to be printed and we had some numpad PCBs lying around, we figured we’d keep on going with that first, before moving on to bigger casts. So we designed a 3d model of a NumPad with a rough idea of the mounting in mind, printed it and created a silicone mold out of the 3d print.
-
Picture: 3d printed Model*
-
Picture: Silicone mold of 3d model*
-
Picture: 2nd pour, this time according to the
instructions*
This turned out way better than expected and our second pour gave us the proof of concept we were looking for. We applied some clear coat finish as well, just to make sure there won’t be any changes in color over time.
[ Specified attachment is not available ] -
Picture: 2nd pour released out of the mold, sanded and
coated*
The small thing weighs in at 290 grams. The haptic is hard to describe, the best thing I can come up with is a mix of plaster and acrylic. It’s super smooth, smaller irregularities can be sanded and it’s much sturdier than expected. We mixed White and Yellow, certainly not everyones cup of tea, we however think it has a unique and pretty look to it. -
Figuring out the Mounting
As this stuff needs to be casted, we figured it would be a good idea to stick to a 1-piece design and not overcomplicate things. So we went with a simple tray mount for our NumPad. We tried several ways to get the mounting pins sturdy enough to hold a pcb + mounting plate.
We first tried to simply embed pcb spacers into the bottom of the case with glue. It worked, but we didn’t think it was sturdy enough to hold the spacers.
[ Specified attachment is not available ] -
Picture: Glued Spacers as mounting pins*
We tried adding additional material around the spacer, but that just looked clunky, and made casting it more difficult than it needed to be. We didn’t even bother gluing pins in, but here´s a picture.
[ Specified attachment is not available ] -
Picture: Additional material for the glued mounting
pins (ignore the air bubbles, this was just a quick pour
of the bottom to test different mountings)*
The other option was to cast M2 screws upside down into the bottom of the board, so only the thread would be visible and the screw head would stabilize it. Then add a PCB spacer to mount the PCB and mounting plate. This worked well, the threads are solid, don´t move a bit and everything can be mounted on them, with enough space for pcb, switches and stabs.
[ Specified attachment is not available ] -
Picture: M2 screw upside down as base for the mounting
pin*
Overall, we are happy with the Mounting, especially for a NumPad, which isn’t under heavy typing fire all the time, it works out just fine. We were experimenting with additional O-Rings, but due to little space between the switches and stabs, they would squeeze against the switch housing and we didn’t really like that. Also it seemed overkill for a NumPad, we´re going to revisit that, in case we make an actual keyboard out of this one day
-
From Experiment to a Prototype we felt was worth
sharing
After we decided on how to mount everything, we added a few design elements in the bottom of the case and agreed on a final look of the NumPad.
Again, 3d printed it, this time added some filler and sanded the 3d model to make it overall smoother, casted a mold with silicone and then poured in the Jesmonite.
[ Specified attachment is not available ] -
Picture: Hardened NumPad before sanding with brass
mounting plate*
[ Specified attachment is not available ] -
Picture: Numpad after sanding with some Cherry MX
Silent Red (they matched the red, okay
?)*
[ Specified attachment is not available ] - Picture: Numpad with some white keycaps*
-
Next Steps:*
We are not yet 100% happy with the pouring results, as with all casting materials, air bubbles build up while mixing the liquid. We are looking into different ways to reduce that. Our best bet at this point is either a vibration plate or a pressure chamber. Vacuum doesn’t seem to be a great alternative, as the stuff hardens relatively quickly and vacuuming bubbles takes quite a while. If you have other good ideas, let us know. We´ll keep you updated.
We´re still unsure whether or not we will put it up for sale in our shop. Clearly, this is also dependent on your guys’ feedback, so let us know what you think and if you’d like to have something like this on your desk and if you think it’s worth investing more time in. Colors are up for debate, pretty much anything is possible. -
PS: Still unsure of the project name, feel free to
share any ideas
*
We hope you enjoyed the read, thanks for taking your time and greetings from Germany! -
Follow us:*
https://www.instagram.com/the.teleport/
https://www.theteleport.de/en