3d models for components of XR4ti and Scorpio

[brokencase]

Hey,
So the the soft for it is... Early. EXSCAN its called, its not bad, and they ARE getting better for feature recognition. If I have a very detailed item, it works great, if its a tube, or flat piece, its pretty... Bad. At that point I use markers, and sometimes add another feature rich item beside it, and just delete it in post process. If you'd like, you can contact me directly, and I can share my screen and scan an item live.

I don't think I need a demo. But thanks for the offer. I think I have a pretty good idea.

I have a Microscribe CMM arm. But I find that I have not used it very much. But this may change.
To be honest I've gotten pretty good at reverse engineering with just a set of calipers or from vector translation from photographs.

For the case of the Merkur key above...
Took Ed's top view picture and read it in a paint program and cleaned it up so that it was a greyscale image with good contrast.

I then read that image into Inkscape and have it trace the bitmaps into an SVG. Convert SVG in DXF.

Read the DXF into my favorite DXF editor and choose the best side profile, delete the other side and then mirror it to get a nice symmetric profile.

Import the DXF into OpenScad and extrude it into 3d. Shrink it in XY and then extrude it again but slightly taller.

Run the "Round Anything" library on each extrusion, then run hull() which essentially is like wrapping the two extrusions with saran wrap.

And then that is it. Little touch ups to add the bumps and cutouts. Scale it as needed.

I guess the point is that most items are easily modeled with either linear or rotational extrusions, so even with the Microscribe I'd rather pick a few points and get a curve an then extrude that to get the full surface and hence a cleaner model...as opposed to a 3d scanner where you end up with a somewhat rough tessellated surface

[brokencase]

My garage doors are frozen shut so, without my micrometer, new pics added to link; the last two show the edge of a credit card aligning the key edge with the ruler's edge.

https://photos.app.goo.gl/EG5San9WYwdidw1p9

YMMV

Those last pictures will do it. Thanks again.

[Nmstec]

Hey,
So the the soft for it is... Early. EXSCAN its called, its not bad, and they ARE getting better for feature recognition. If I have a very detailed item, it works great, if its a tube, or flat piece, its pretty... Bad. At that point I use markers, and sometimes add another feature rich item beside it, and just delete it in post process. If you'd like, you can contact me directly, and I can share my screen and scan an item live.

I don't think I need a demo. But thanks for the offer. I think I have a pretty good idea.

I have a Microscribe CMM arm. But I find that I have not used it very much. But this may change.
To be honest I've gotten pretty good at reverse engineering with just a set of calipers or from vector translation from photographs.

For the case of the Merkur key above...
Took Ed's top view picture and read it in a paint program and cleaned it up so that it was a greyscale image with good contrast.

I then read that image into Inkscape and have it trace the bitmaps into an SVG. Convert SVG in DXF.

Read the DXF into my favorite DXF editor and choose the best side profile, delete the other side and then mirror it to get a nice symmetric profile.

Import the DXF into OpenScad and extrude it into 3d. Shrink it in XY and then extrude it again but slightly taller.

Run the "Round Anything" library on each extrusion, then run hull() which essentially is like wrapping the two extrusions with saran wrap.

And then that is it. Little touch ups to add the bumps and cutouts. Scale it as needed.

I guess the point is that most items are easily modeled with either linear or rotational extrusions, so even with the Microscribe I'd rather pick a few points and get a curve an then extrude that to get the full surface and hence a cleaner model...as opposed to a 3d scanner where you end up with a somewhat rough tessellated surface

That's kinda what I still do for flat smaller items, but I use a printer scanner. In my lucky case my combination saves in a perfect scale, so load it up in illustrator, trace it out, and stick it into the plasmacutter/3d printer!

[brokencase]

That's kinda what I still do for flat smaller items, but I use a printer scanner. In my lucky case my combination saves in a perfect scale, so load it up in illustrator, trace it out, and stick it into the plasmacutter/3d printer!

Yeah, don't get me wrong, I'd love to have the laser scanner just for the speed aspect. Maybe I will get one someday.

But I am amazed at how good I have become with the modeling. I'm a software guy at heart and OpenScad has opened up the world to me. OpenScad used to be pretty slow to render, but they have made some new options that makes it much faster.

I'll give you a good example where I don't think a laser scanner would work out...
https://www.thingiverse.com/thing:5457921

I don't believe a laser scanner could reliably detect the overhang in the thin perimeter groove of that item.

[Nmstec]

That's kinda what I still do for flat smaller items, but I use a printer scanner. In my lucky case my combination saves in a perfect scale, so load it up in illustrator, trace it out, and stick it into the plasmacutter/3d printer!

Yeah, don't get me wrong, I'd love to have the laser scanner just for the speed aspect. Maybe I will get one someday.

But I am amazed at how good I have become with the modeling. I'm a software guy at heart and OpenScad has opened up the world to me. OpenScad used to be pretty slow to render, but they have made some new options that makes it much faster.

I'll give you a good example where I don't think a laser scanner would work out...
https://www.thingiverse.com/thing:5457921

I don't believe a laser scanner could reliably detect the overhang in the thin perimeter groove of that item.

Actually that piece isn't bad, it may miss a few details but a bit of help in solidworks (my preference ) would do the trick. It's the really flat or symmetric parts it sucks with badly.

I am a hardware guy at heart, and hate 3d modeling! but we all do what we have to!

[brokencase]

So I suspect you often must "clean up" the laser scans afterwards in SolidWorks?

[brokencase]

Anyhow..

Moving on with the Scorpio Flashlight key...

I've decided to make it work with a 2032 lithium battery. This works well with standard white light LEDs. The duration should be pretty good with this arrangement also.

So it won't be like the original unit...other than the original outside form.

My plan is a dove tailed cover...on the push button side, that is about 25 mm wide. The cover will have a hole for the push button which will be a separate piece.

I'm trying to make it as few parts as possible and easy to assemble. Hopefully only three parts. In other words, print the parts and bend the LED leads and you are there.

This may entail building into the model a formed spring on the push button part.

Such a simple thing and much imagination required...

[brokencase]

All done. Now need to print it. Not sure if I'll do it in FDM or UV resin.
I made a square hole for the key shaft to go in. My plan was to cut one of the spare keys such that there is a square section of plastic from the prior key fob surrounding the metal portion that can be JB welded into the square hole.
.
.

body1.jpg (70.66 KiB) Viewed 11905 times

body2.jpg (76.31 KiB) Viewed 11905 times

body3.jpg (88.86 KiB) Viewed 11905 times

[brokencase]

Logo and cover. The cover has a hole that one of the LED leads slides into and then becomes the push button contact to the battery.

A similar slot for the other LED lead is in the main body. The cover dovetails into the body. I'll drill a hole for a single screw to hold the cover in place.

Merkur logo gets printed in white plastic and glued into the oval in the main body.
.
.

cap1.jpg (82.42 KiB) Viewed 11905 times

logo.jpg (54.74 KiB) Viewed 11905 times

[Ed Lijewski]

Re a spring: you'll want to butt-test whichever one you use; i.e., with the key(s) in a pocket a weaker spring might allow the light to illuminate while a stiffer one won't.

YMMV

[brokencase]

There's a little "block" connecting the "button" to the "cover". I can change it's size to affect the "flex".

My first FDM print of the body did not go so well. I tried using the fine setting in the ABS profile in the IdeaMaker software which I have never used before. It needs some tuning. Support was melded into the part and not easily removable.

I think I will change the orientation and switch back to a lower res and lay it down flat to minimize the support and try again. At the same time I will try the UV printer.

[brokencase]

Second FDM print is functional but cosmetically not very good. This item's detail is going to require the UV resin printer.

The problem with the UV resin printer is that the plastics are brittle. I have found that I have to mix a small amount of the flexible resin to the normal brittle resin to get something that is more "ABS like". Which is what I will have to do here.

Sure would like to get this working on FDM.

[brokencase]

BTW, Found another Merkur related 3d model - Ford Sierra/Merkur lower dash fader cover
https://www.printables.com/model/330567

[brokencase]

Well - I got sick the past week or so. Things have cleared up (no Covid thankfully) But yesterday after two attempts I was able to get a good result on the UV printer (AnyCubic Photon Mono, BTW)

The problem that you run into with the UV printer is you have to avoid exposing a large cross-section of the model to the exposure window at the bottom of the vat. Some resins are viscous like gear oil. If you have large area then there is a strong suction effect as the build plate moves up. This is often strong enough to rip part (either partially or fully) off of the build plate. The solution is to used a ton of heavy support and to tilt the parts.
My error is I often overlook setting the heavy support option in the software or I don't tilt the part enough, but I am getting better at it.

So here is how the parts were ultimately laid out on the build plate in Chitubox software:
.
.

chitu.jpg (304.03 KiB) Viewed 11621 times

.
.
..and here is the final part after it was printed, support removed, and then painted with some VHT black satin epoxy paint.
The battery/light/button assembly works perfectly. I did have to place a small cardboard ring around the top of the battery. I used the self adhesive ones that are used in 18650 battery packs.

Ahyhow - It is much brighter than the original arrangement.

What I have yet to do is bandsaw the metallic portion from one of my copied keys and JB weld it into the new fob. Still not sure if there is enough bite in the hole of this fob and if the plastic/epoxy bond will be strong enough. Stay tuned.
.
.

painted.jpg (530.2 KiB) Viewed 11621 times

[Ed Lijewski]

What does your sacrificial key look like?

YMMV
.
.
.

20230106_131125_copy_1209x1209.jpg (228.52 KiB) Viewed 11585 times