Resin print keyboard parts factory

User avatar
SneakyRobb
THINK

08 Aug 2020, 01:20

Hi,

I just got a resin printer and its fun. I figured I would collect various resin related things here. Such as why resin print etc. as well as my failures and learning.

The printer I bought is the elegoo mars. Its a pretty inexpensive machine.

I got this machine to make small plastic keyboard parts for beamspring, model f and blue alps. Which are my favourite keyboard switches.

Resin printing is a secondary sort of thing to all the projects that I do so I figured its in my own fashion to make a new thread for just this. I didn't want to clutter my extreme clutter threads with my trials and tribulations in producing the parts and learning how to print.

Why did I get a resin printer? User kralcifer sent me some parts on his machine for me to see the resolution and it was great so I had to get one. Here are the parts that Kralcifer sent me.
kralcifer 1.jpg
kralcifer 1.jpg (3.86 MiB) Viewed 1984 times
kralcifer 2.jpg
kralcifer 2.jpg (4.07 MiB) Viewed 1984 times
kralcifer3.jpg
kralcifer3.jpg (3.39 MiB) Viewed 1984 times

As you can see these parts are dimensionally very accurate and very smooth. This inspired me to move away from resin casting of beamspring parts, which was a major headache to just getting a resin printer.

I had posted in the Mara thread but I felt it was better to just make a new thread
viewtopic.php?p=469152#p469152

This wont be organized at all.

I want to basically make all the parts of beamspring model f and skcm blue.
That thread has my 2nd attempt at model f flippers on resin.

Here is my first model f flipper attempt. On fdm, or "normal" 3d printing attempt.
fdm compare 2.jpg
fdm compare 2.jpg (3.71 MiB) Viewed 1984 times
fdm compare 3.jpg
fdm compare 3.jpg (3.25 MiB) Viewed 1984 times
fdm compare.jpg
fdm compare.jpg (3.87 MiB) Viewed 1984 times

So you can see, regardless of dimensional accuracy that this print is very rough. The surface is not smooth. It wont really work well. This is why resin printing is cool. The surface is smooth.

Here is my 2nd/3rd attempt at a model f flipper out of resin.

The smoothness is great. I need to fix this model as it doesnt click yet, but I'm just learning resin printing now so its good.
1.jpg
1.jpg (2.6 MiB) Viewed 1984 times
1a.jpg
1a.jpg (3.85 MiB) Viewed 1984 times
2.jpg
2.jpg (2.83 MiB) Viewed 1984 times
3.jpg
3.jpg (2.6 MiB) Viewed 1984 times
4.jpg
4.jpg (2.46 MiB) Viewed 1984 times
This print actually looks great. Really happy with the smoothness.

Here is is compared to an original.
compare v1.jpg
compare v1.jpg (3.67 MiB) Viewed 1984 times
compare v1 4.jpg
compare v1 4.jpg (3.58 MiB) Viewed 1984 times
compare v1 3.jpg
compare v1 3.jpg (3.88 MiB) Viewed 1984 times
compare v1 2.jpg
compare v1 2.jpg (3.73 MiB) Viewed 1984 times

This one had a slight issue where the spring didn't sit. The original actually has a little lip that keeps the spring on. So we will adjust that!
nub on model f flipper.jpg
nub on model f flipper.jpg (3.5 MiB) Viewed 1984 times
spring attach.jpg
spring attach.jpg (4.28 MiB) Viewed 1984 times
I have thought about making capacitive prints. I can either add carbon to make it capacitive. Maybe paint it on? Will test this soon. Not sure why it wouldn't work.
paint.jpg
paint.jpg (4.03 MiB) Viewed 1984 times
paint 2.jpg
paint 2.jpg (3.61 MiB) Viewed 1984 times

For the capacitive content I can only guess but ellipse reported 18% content. This sounds reasonable. As said in the Mara thread, the model f flipper has so much carbon, you can draw with it if broken. Please for give me.
pencil.jpg
pencil.jpg (4.5 MiB) Viewed 1984 times
sad.jpg
sad.jpg (4.67 MiB) Viewed 1984 times
Will have to see about this.

For my blue alps project I wanted to make some alps parts. This somewhat turned out. Mostly due to my learning/ignorance I made some mistakes but here is a part I failed at the first time. You can see the resolution vs the original though. Very fine! The part here is not correctly formed but shows the power of resin printing.
resolution.jpg
resolution.jpg (4.39 MiB) Viewed 1984 times
I used this plastic box lined with 405nm UV lights to cure the prints. Hi SneakyRobb here with the power of resin printing!
uv light curing box.jpg
uv light curing box.jpg (3.57 MiB) Viewed 1984 times
Now we wash the prototype in our patented Isopropyl jar with part holder! Don't mind the missed mustard seed...
wash jar 3.jpg
wash jar 3.jpg (4 MiB) Viewed 1984 times
This print was meant to have 3 parts. Due to my inexperience with supports only 1 worked.
oops 1.jpg
oops 1.jpg (46.36 KiB) Viewed 1984 times
So I clipped the working one of supports. My failures of the others were due to supports. On a normal printer supports hold up over hangs. On the resin printer, everything prints up-side-down. So supports have things hang off them. Totally different! You can see the slider supports with no slider.
bin for clip.jpg
bin for clip.jpg (4.25 MiB) Viewed 1984 times
And what did we get? Now the slider I admit doesnt move that well. But. It fits. and the switch does click!!! For such a novice as myself, I can see the power and I hope you see it to.
alps compare.jpg
alps compare.jpg (3.58 MiB) Viewed 1984 times
alps compare 2.jpg
alps compare 2.jpg (3.2 MiB) Viewed 1984 times
alps compare 3.jpg
alps compare 3.jpg (4.53 MiB) Viewed 1984 times
fit 1.jpg
fit 1.jpg (3.59 MiB) Viewed 1984 times
fit 2.jpg
fit 2.jpg (3.61 MiB) Viewed 1984 times
fit 3.jpg
fit 3.jpg (3.69 MiB) Viewed 1984 times


I would like to make a tutorial here at some point soon, but I hope you can see the power of this technology through my clumsy first steps. The resolution and smoothness are absurd and might very well allow us to reach keyboard levels not previously possible.

User avatar
SneakyRobb
THINK

09 Aug 2020, 01:37

Hi,

So I wanted to make some more stuff today so I did. Great.
maximum keyboard.JPG
maximum keyboard.JPG (111.91 KiB) Viewed 1887 times
This is what I wanted to make. So if you are print resin, it prints up-side-down. Like above. You can see how the supports are different

I had wanted to buy a nexus slider from a user but it didn't happen due to sadness. So I will just make one.
robb nexus v1.JPG
robb nexus v1.JPG (38.02 KiB) Viewed 1887 times

Anyway here is my cat Freddie as a kitten.
cat 1.JPG
cat 1.JPG (3.7 MiB) Viewed 1887 times

So I used elegoo translucent resin. Here after 90+ mins is the build plate.
YES 1.jpg
YES 1.jpg (4.06 MiB) Viewed 1887 times
yes 2.jpg
yes 2.jpg (3.96 MiB) Viewed 1887 times
yes 3.jpg
yes 3.jpg (4.29 MiB) Viewed 1887 times
I need to clean them but here are some of the parts pre-cure
top1.jpg
top1.jpg (3.67 MiB) Viewed 1887 times
robb nexus.jpg
robb nexus.jpg (3.69 MiB) Viewed 1887 times
skcm slider.jpg
skcm slider.jpg (3.81 MiB) Viewed 1887 times
I really wanted to make the skcm slider and nexus so I did.

I will wait now for curing but happy so far.

User avatar
matt3o
-[°_°]-

09 Aug 2020, 08:18

you really need to get better resin though :)

gianni

11 Aug 2020, 22:58

Brilliant, you have my attention :-)

User avatar
SneakyRobb
THINK

12 Aug 2020, 02:46

matt3o wrote:
09 Aug 2020, 08:18
you really need to get better resin though :)
Aw. Any suggestions!? This one worked pretty well.
gianni wrote:
11 Aug 2020, 22:58
Brilliant, you have my attention :-)
Hey thanks.


Okay I figured I would use an actual camera to get some nice photos. Still learn.

Here is a compare between an alps switch opener made with my FDM printer vs the resin one.

In retrospect clear resin was not the best choice for photo taking, hard to see the details.

Tried to make this visible but the resin one is actually ultra smooth!
fdm and resin compare 2.JPG
fdm and resin compare 2.JPG (2.96 MiB) Viewed 1747 times
fdm and resin compare.JPG
fdm and resin compare.JPG (2.7 MiB) Viewed 1747 times

I had the alps switch shell opening facing the top of the build plate as it printed, this was not optimal printing technique as the opening was slightly weird. As well the lower housing clips are not that bendy, they are easy to brake off.
opening error.JPG
opening error.JPG (2.89 MiB) Viewed 1747 times

Here are just various items in my current experiment box.
various items.JPG
various items.JPG (2.88 MiB) Viewed 1747 times
Here I will compare the stems for blue alps.
resin stem.JPG
resin stem.JPG (3.06 MiB) Viewed 1747 times
stems compare.JPG
stems compare.JPG (2.84 MiB) Viewed 1747 times
stems compare 2.JPG
stems compare 2.JPG (3.26 MiB) Viewed 1747 times
The resin stem actually works perfectly and no joke feels super smooth and excellent. Works perfectly! Really surprised. Printed it just to test before printing a 4x scale one.

The model f flipper is almost ready, but having slight issues trying to get the spring to stay on
items 1.JPG
items 1.JPG (3.14 MiB) Viewed 1747 times
The nexus slider came out great. As mentioned the original nexus slider had a spine down one side. I was not sure why this was there. When I tried it in the switch though I think I figured out out.

Regardless it looks great and I inserted into this random cherry mx keycap multiple times and no breaks or anything. Very excite.
new nexus 1.JPG
new nexus 1.JPG (3 MiB) Viewed 1747 times
new nexus 2.JPG
new nexus 2.JPG (2.8 MiB) Viewed 1747 times
new nexus 3.JPG
new nexus 3.JPG (3.04 MiB) Viewed 1747 times

Here you can see what happened to the click leaf. Some how the slider bent it slightly. It still clicks but not very well. I'm not 100% that this was the cause but I don't know why it would be bent otherwise. Perhaps the bottom of the keycap somehow compressed the clickleaf down. That spine might prevent this. I will add this to the next attempt. Don't remember if posted it before so here is the nexus spine and the bent click leaf.
nexus click leaf sad.JPG
nexus click leaf sad.JPG (2.83 MiB) Viewed 1747 times
nexus spine.JPG
nexus spine.JPG (26.3 KiB) Viewed 1747 times

gianni

13 Aug 2020, 11:44

For your next project, I'd like to suggest printing IBM logos (for keyboards or old computers), like you already did in one colour.

It would be nice to replace some old IBM logos with a modern one, printed with the letters in relief (or carved), in two colours if it's supported.


viewtopic.php?t=18905

https://drive.google.com/drive/folders/ ... sp=sharing
Last edited by gianni on 13 Aug 2020, 11:57, edited 1 time in total.

gianni

13 Aug 2020, 11:45

Will you make available the 3d model for the model f flippers? What file format are you using?


Another idea: printing key blockers for the model f (stem plus flat surface, level with the keyboard plane). This is useful since most users remap the model f at and xt, so removing the unused keycaps maybe nice and useful

User avatar
sharktastica

13 Aug 2020, 13:45

Awesome stuff! I've been eyeing up the Elegoo Mars for a while now since it is decently well priced.

User avatar
SneakyRobb
THINK

13 Aug 2020, 16:04

sharktastica wrote:
13 Aug 2020, 13:45
Awesome stuff! I've been eyeing up the Elegoo Mars for a while now since it is decently well priced.
Hi, its a great printer. Recommend getting it or the Saturn whenever it comes out.
gianni wrote:
13 Aug 2020, 11:45
For your next project, I'd like to suggest printing IBM logos (for keyboards or old computers), like you already did in one colour.

It would be nice to replace some old IBM logos with a modern one, printed with the letters in relief (or carved), in two colours if it's supported.


viewtopic.php?t=18905

https://drive.google.com/drive/folders/ ... sp=sharing

------------------------
Will you make available the 3d model for the model f flippers? What file format are you using?


Another idea: printing key blockers for the model f (stem plus flat surface, level with the keyboard plane). This is useful since most users remap the model f at and xt, so removing the unused keycaps maybe nice and useful
Hi, these are good points. I have made some of the ibm logos before and sent to some people in FDM. Does anyone have the actual dimensions of the original oval? The ones I printed were too big and I don't have a model m to measure the oval size.

I will release the flipper file when It is complete and working. I am using fusion360 to make the items.

The blank model f barrel is an interesting idea! I may have to do that yes.

As well to fully illustrate what I mean with that spine on the nexus slider. Ignore this is the wrong side of the stem and visualize the otherside. You can see how the bottom of the keycap overhangs the alps slider width. I think somehow that is pinching the clickleaf weirdly. You can see how the spine would somewhat prevent that by making the keycap flush with the slider.
keycap spine.JPG
keycap spine.JPG (599.45 KiB) Viewed 1612 times

User avatar
SneakyRobb
THINK

15 Aug 2020, 00:43

Hi,

So still on this nexus slider.

Here are some comparison measures I made
nexus slider 4.JPG
nexus slider 4.JPG (37.88 KiB) Viewed 1532 times
nexus lel 3.JPG
nexus lel 3.JPG (38.71 KiB) Viewed 1532 times
So I tried to update both with the vertical height and the spine.
nexus redone 1.JPG
nexus redone 1.JPG (42.71 KiB) Viewed 1532 times
Looks pretty good lets compare
nexus compare 1.JPG
nexus compare 1.JPG (127.66 KiB) Viewed 1532 times
nexus slider.JPG
nexus slider.JPG (71.98 KiB) Viewed 1532 times
I will try to print this tomorrow. One thing to note is that the nexus original designers seemed to have omitted the vertical part of the switch that goes past the alps "slits."

I am not sure why they did this. Maybe due to mold costs it was simpler. When the slider is fully up it would seem this would at least partly help the dust incursion into the switch. No matter. I have made models in both configuration.
no wiper control.JPG
no wiper control.JPG (44.45 KiB) Viewed 1532 times
yes wiper control.JPG
yes wiper control.JPG (43.63 KiB) Viewed 1532 times
alps compare 3.JPG
alps compare 3.JPG (66.71 KiB) Viewed 1532 times

gianni

15 Aug 2020, 07:02

Talking about something not strictly related to your current problem, I'd like to say that stem tolerances are a strange beast, do you remember box switches? They where off by a smaller margin than what is (nowdays) considered an acceptable tolerance for 3d printers (0.1mm). Probably to play safe you should buy a cruciform to check the safety of each switch mount, to avoid breaking keycaps.

Did I miss something or you didn't actually measure the keycaps with a digital caliper? I would surely do it.

User avatar
matt3o
-[°_°]-

15 Aug 2020, 16:59

SneakyRobb wrote:
12 Aug 2020, 02:46
matt3o wrote:
09 Aug 2020, 08:18
you really need to get better resin though :)
Aw. Any suggestions!? This one worked pretty well.

you may try with ABS-like resins or technical resins usually called "engineering" or "tough". I used FunToDo Industrial blend with some degree of success. Harzab Model is also decent. Let me know if you need more specifics.

gianni

25 Aug 2020, 12:50

If you need a model for the key blocker, probably you can use this stl file as a base to work on (well obviously in your version the top will be square not rounded).

https://www.thingiverse.com/thing:1849140
$$Clipboard01.jpg
$$Clipboard01.jpg (2 KiB) Viewed 1194 times

User avatar
anthonymak

26 Aug 2020, 04:50

any update on the electric paint?

gianni

26 Aug 2020, 07:47

https://www.blackmagic3d.com/Conductive-p/grphn-pla.htm

Conductive filament will surely have a longer life than paint.

There are also conductive resins that can be used for sla printing

https://www.3dresyns.com/pages/conducti ... lectronics

User avatar
SneakyRobb
THINK

27 Aug 2020, 22:37

Hi,

Yes the electric paint seems to work. I am trying to finalize the flipper model though as my previous ones do not click/flip. Can't say for sure if the paint will work while on the clicking flipper inside the assembled keyboard module.

Here is the current untested version
flipper newer.zip
(17.75 KiB) Downloaded 10 times
Took a while to get the bottom and the spring retainer better. Will print hopefully Saturday and get actual clicks. The bottom is much more like the original now.

Conductive filament will last longer and I have been using it to make beamspring capacitive elements. For the beamspring though the capacitive element doesn't really do anything. It gets yanked up and down sometimes. It's kind of captive.The FDM printer doesn't have the best tolerances but it doesn't matter as long as the bottom is flat.

This is a major problem for the model f flipper though as the flipper unlike the beamspring is a working component. It pivots and rubs. The FDM with it's often very scratchy sides makes for a hard time printing accurate things. This is doubly bad for the flipper which is a small flat element. Very difficult to make the fine detail working edges of the model f flipper with a 0.2mm nozzle shooting the stuff.

Regardless it is a good idea.

This is why the paint is currently more attractive.

Old
flipper old.JPG
flipper old.JPG (45.81 KiB) Viewed 1025 times
New
flipper newer.zip
(17.75 KiB) Downloaded 10 times
Attachments
flipper redo 1.JPG
flipper redo 1.JPG (51.29 KiB) Viewed 1025 times

User avatar
SneakyRobb
THINK

27 Aug 2020, 22:39

gianni wrote:
25 Aug 2020, 12:50
If you need a model for the key blocker, probably you can use this stl file as a base to work on (well obviously in your version the top will be square not rounded).

https://www.thingiverse.com/thing:1849140
IBM_Model_M_Typewriter_Keycaps_for_German_edition.zip

$$Clipboard01.jpg
Hi,

This is a good idea I might take it!. I will probably saw some xt barrels in half to get interior measurements later. I do find the blank pad idea attractive. I think it will also keep the original tension/pressure of the model f sandwhich nicely preserved.

gianni

27 Aug 2020, 22:53

Sorry, I got lost in your words and I didn't understand why conductive filament doesn't work in the model f flipper. You won't use it because it is worse (for the details), than the standard filament?

gianni

27 Aug 2020, 22:56

Furthermore there is an old post that shows two different kind of flippers. Maybe the squared one isn't from IBM and was used for repairs?!?

viewtopic.php?t=9715

If your flipper doesn't click, maybe change the angle of the spring? This surely modifies the buckling in a substantial manner.

User avatar
SneakyRobb
THINK

27 Aug 2020, 23:42

gianni wrote:
27 Aug 2020, 22:53
Sorry, I got lost in your words and I didn't understand why conductive filament doesn't work in the model f flipper. You won't use it because it is worse (for the details), than the standard filament?
Hi, conductive filament will work electrically. It just is hard to make a working flipper out of because the flipper needs to be smooth. FDM doesn't produce the smoothest finish. The model F flipper is a much more functional item vs the beamspring capacitive element. The little ears on it rotate and rub everywhere.

The model F is a huge improvement over the beamspring and one part that gets a lot of function is the flipper. So it is not that easy to make a basic flipper copy. It has tolerances and a function that need to be replicated closely.
gianni wrote:
27 Aug 2020, 22:56
Furthermore there is an old post that shows two different kind of flippers. Maybe the squared one isn't from IBM and was used for repairs?!?

viewtopic.php?t=9715

If your flipper doesn't click, maybe change the angle of the spring? This surely modifies the buckling in a substantial manner.
I have been trying to fix
flipper with spring.JPG
flipper with spring.JPG (37.38 KiB) Viewed 984 times

I am not sure about why IBM had different flippers. I do know that they basically work interchangeably.

I have been changing the bottom of the flipper. Heavily targeting this point.
PIVOOOT.jpg
PIVOOOT.jpg (75.72 KiB) Viewed 984 times
That is the key working part of the flipper so hopefully that can work soon.

Ill add the current nexus slider files here as well I suppose.




nexus sliders.zip
(205.09 KiB) Downloaded 8 times

gianni

28 Aug 2020, 00:45

Personally I'd make the area circled in red taller and longer. Then you can probably cut it away bit by bit until it works, using an exacto knife and sandpaper.

In simple words, the points in red and the center of the spring must not be aligned. The red points must be a little bit on the left with respect to the center of the spring, but not too much or the buckling will be too difficult. Looking at the photos it seems that the distance between the center of the springs and the pivot point (circled in red), is smaller in your print when compared to the original)

Considering that the spring applies a force of 60 grams and that deformation can play a role, in a second moment, you can make the two bridges thicker.

gianni

28 Aug 2020, 04:43

I couldn't figure it out in my phone, but it seems that ellipse flippers have less detail/ resolution than yours, and that the red parts are quite smoothed out. Sincerely that makes sense, and with a simpler model you could make a flipper whose base is made of two different surfaces, with a small intermediate offset/ step between them. Instead you and ibm added a lot of complications that aren't really necessary.

User avatar
SneakyRobb
THINK

29 Aug 2020, 01:17

gianni wrote:
28 Aug 2020, 04:43
I couldn't figure it out in my phone, but it seems that ellipse flippers have less detail/ resolution than yours, and that the red parts are quite smoothed out. Sincerely that makes sense, and with a simpler model you could make a flipper whose base is made of two different surfaces, with a small intermediate offset/ step between them. Instead you and ibm added a lot of complications that aren't really necessary.

Hi, your points we well met and correct!

I have made a mathematical 3d model of the flipper. It is not a production sort of model. The model I made has harsh sharp edges that would be bad for an actual mold that is machined out of metal. If we were to use the model to make molds those harsh edges would have to be smoothed yes.

So yes, the ellipse and OG flippers would be smoothed and optimized for injection molding! I will test that specific model though and post the results though about that red area height.

gianni

29 Aug 2020, 12:17

$$hammer.jpg
$$hammer.jpg (24.87 KiB) Viewed 805 times
Right now I don't have a flipper at hand, anyway in my opinion the only facts that count in this preliminary phase are:

- the distance between the center of rotation identified by the yellow arrow and the center of the spring

- the angle between the axis of the spring "at rest" and the green plane

- when the red plane is in contact with the traces of the pcb, verify if the yellow point is still on the pcb (verify if this happens also with an original flipper).

If these measures respect the original flipper, in my opinion it should work like the original one.

If I had to take these measures, I would literally tape the flipper on the photocopier's glass and then scan it. Taking a picture results in many more errors. I've seen that you have alredy broken a flipper in half, so you can scan it in the center plane and in the side plane :-)

Later I'll go look if I have some spare flippers and barrels and try to take some measures.

User avatar
SneakyRobb
THINK

30 Aug 2020, 23:17

gianni wrote:
29 Aug 2020, 12:17
$$hammer.jpg

Right now I don't have a flipper at hand, anyway in my opinion the only facts that count in this preliminary phase are:

- the distance between the center of rotation identified by the yellow arrow and the center of the spring

- the angle between the axis of the spring "at rest" and the green plane

- when the red plane is in contact with the traces of the pcb, verify if the yellow point is still on the pcb (verify if this happens also with an original flipper).

If these measures respect the original flipper, in my opinion it should work like the original one.

If I had to take these measures, I would literally tape the flipper on the photocopier's glass and then scan it. Taking a picture results in many more errors. I've seen that you have alredy broken a flipper in half, so you can scan it in the center plane and in the side plane :-)

Later I'll go look if I have some spare flippers and barrels and try to take some measures.
When I lay it flat It feels like there is resistance between a sheet of paper when trying to slide it in. I tried using a glass plate and some feeler gauges and there doesn't seem to be a thinness where there is zero resistance. I think the yellow point is always in contact. Which would make senses as a pivot point.

I have modified my flipper model to have the point and the flat part inline.
flipper inline.JPG
flipper inline.JPG (28.75 KiB) Viewed 731 times
The previous flipper I printed does click but the spring doesn't stay on well. I am very surprised at the very specific dimensions of the model f flipper.
Attachments
Flipper flatter.zip
(33.88 KiB) Downloaded 10 times

User avatar
SneakyRobb
THINK

01 Sep 2020, 18:31

gianni wrote:
25 Aug 2020, 12:50
If you need a model for the key blocker, probably you can use this stl file as a base to work on (well obviously in your version the top will be square not rounded).

https://www.thingiverse.com/thing:1849140
IBM_Model_M_Typewriter_Keycaps_for_German_edition.zip

$$Clipboard01.jpg
Hi,

Slightly different and I haven't printed it yet but I made these 2 files that hopefully will fit in order to make blank barrels.

Robb
Attachments
blank model f barrels.zip
(30.83 KiB) Downloaded 6 times
blank xt peg.JPG
blank xt peg.JPG (26.66 KiB) Viewed 649 times
AT F122 blank.JPG
AT F122 blank.JPG (33.4 KiB) Viewed 649 times

gianni

02 Sep 2020, 06:35

SneakyRobb wrote:
01 Sep 2020, 18:31
gianni wrote:
25 Aug 2020, 12:50
If you need a model for the key blocker, probably you can use this stl file as a base to work on (well obviously in your version the top will be square not rounded).

https://www.thingiverse.com/thing:1849140
IBM_Model_M_Typewriter_Keycaps_for_German_edition.zip

$$Clipboard01.jpg
Hi,

Slightly different and I haven't printed it yet but I made these 2 files that hopefully will fit in order to make blank barrels.

Robb
That's very useful and nice but it has a different function. Your project requires to remove the barrel, while (I think), most people could be interested in something that replaces the keycap when trying new layouts with the model f at / xt. Obviously when the layout is tested many people will use your keyblocker.

User avatar
SneakyRobb
THINK

02 Sep 2020, 18:02

gianni wrote:
02 Sep 2020, 06:35
SneakyRobb wrote:
01 Sep 2020, 18:31
gianni wrote:
25 Aug 2020, 12:50
If you need a model for the key blocker, probably you can use this stl file as a base to work on (well obviously in your version the top will be square not rounded).

https://www.thingiverse.com/thing:1849140
IBM_Model_M_Typewriter_Keycaps_for_German_edition.zip

$$Clipboard01.jpg
Hi,

Slightly different and I haven't printed it yet but I made these 2 files that hopefully will fit in order to make blank barrels.

Robb
That's very useful and nice but it has a different function. Your project requires to remove the barrel, while (I think), most people could be interested in something that replaces the keycap when trying new layouts with the model f at / xt. Obviously when the layout is tested many people will use your keyblocker.
Hi, I think I get you. I suppose such an insert could be made. It would still stick up a bit as it would both need to tense the spring to raise the capacitive flipper but also not be able to press it down. Although maybe the tensed spring sits below the barrel anyway so it could be flush... I will investigate this.

gianni

02 Sep 2020, 19:39

Hello, thanks for the follow up and for your patience. I would proceed this way. Convert the STL that I posted to DWG (probably you can do it with the freeware blender).
Then with your fusion software move the top of the key towards the bottom. Proceed with the lowering until the top of the key (the flat part), rests on the top of the barrell. At this point you have the exact measure that prevents the key from actuating and from moving.

The stem of the new key must have the same length of the inside of the barrel, plus about 1mm or a little more to allow the little nibs of the stem to effectively "clip" the inside of the barrel.

gianni

03 Sep 2020, 09:59

About the flipper, could it be that it doesn't work perfectly because the spring is not seated correctly on the little stem? When the spring buckles, it applies a "moment" (not sure if the translation is correct), and if it is not seated firmly, the buckling doesn't translate in a "moment" acting on the flipper.

Post Reply

Return to “Workshop”