How do I make a Star Trek replicator?

DavidR8

Scrap maker
Administrator
Moderator
Premium Member
Maybe you could also do something like that with a touch probe in the mill spindle and a DRO to take most of the measurements. Then make a 3D design in your favorite CAD program from that.
I supposedly have the ability to do that with my CNC router and its touch probe.
 

Brent H

Ultra Member
@Susquatch - in the full “AutoCad” you do this: take picture, convert to a PDF, import it into AutoCad. After that scale the picture to the correct measurements. It is just a matter of tracing the photo and then checking measurements to confirm fitment.

It works great for converting drawings or, as @YYCHM mentions, drawing parts to scale from a good quality pic. There is some distortion with camera pics but it is easy enough to extrapolate and correct measurement/perspective errors.
 

Matt-Aburg

Ultra Member
I want a star trek replicator. To scan stuff, accurately enough, and then 3D print it. What is required? what is the process? What software is needed? How much for various options? Does it take forever? Can I measure and model it in CAD faster than the actual process of scanning, cleaning, printing?

See this holder for hex shaped bits? I want more of them. How can I scan this with some sort of scanner? Then model it up and print it with a 3-D printer.? If I can do that then I have a 2023 version of a Star Trek replicator.

Somebody on the floor must be doing this. Tell us about your scanner in the process and how much the scanner will cost.
For that clip holder, I would scan then only use it as reference geometry. It looks like one holder could be pattern copied. This probably would not warrant the cost of a scanner to do in this case. On more complicated objects, you could scan, clean up, and print. The scanners range in price from a couple hundred to 100,000$. Of course, you get what you pay for.. If you want a good starter scanner, I would recommend either a Peel or Einscan scanner. These are not metrology grade, or certified scanners. They would do the job though. You will still need to clean up and redraw the part if it is prismatic. For organic, you can auto-surface it. For 3D printing you can leave it as polygon mesh. Here is in my opinion of a starter scanners.. I would avoid Creality lizard or anything that costs 2-300$.. you really get what you pay for.


or made in Canada by Creaform, for the consumer market

 

Susquatch

Ultra Member
Administrator
Moderator
Premium Member
So you had your head scanned for a crown? Your wife must be very understanding.

Ya, I guess I stepped on my own dic* on that one.

Seriously. They scanned my head and then did a 3D print of my head on a post. Reminded me of a Greek sculpture. Trouble is that the sculpture looked better than I did! Prolly cuz it had no fur......
 

Matt-Aburg

Ultra Member
Seriously. They scanned my head and then did a 3D print of my head on a post. Reminded me of a Greek sculpture. Trouble is that the sculpture looked better than I did! Prolly cuz it had no fur......
not sure which way to go on this diversion.. either they were trying get you to shave,,,,,,, or amazing the tech was available that long ago, had to be re-invented after someone got around the military patent..
 

Susquatch

Ultra Member
Administrator
Moderator
Premium Member
or amazing the tech was available that long ago, had to be re-invented after someone got around the military patent..

Lots of stuff has been around for a long time. Just way beyond the financial means of an average guy.

I think my head was printed on a 3D printer using ABS filament. That was around 20 years ago. The printer was several hundred thousand dollars. Now you can buy one for several hundred. Roughly a thousandth of the cost back then. I do not recall any military patents on this technology. I seem to recall most of this 3D printing stuff was an outcome of the work done at one of the Fraunhoffer Institutes in Germany. But I'm not certain that's where it started. I did go to Germany around the turn if the century to evaluate early 3D printing of metal. It wasn't ready for application yet. I don't know if it ever evolved or not.

However, I did regularly review Military technology for industrial applications.
 

mbond

Well-Known Member
Premium Member
Lots of stuff has been around for a long time. Just way beyond the financial means of an average guy.

I think my head was printed on a 3D printer using ABS filament. That was around 20 years ago. The printer was several hundred thousand dollars. Now you can buy one for several hundred. Roughly a thousandth of the cost back then. I do not recall any military patents on this technology. I seem to recall most of this 3D printing stuff was an outcome of the work done at one of the Fraunhoffer Institutes in Germany. But I'm not certain that's where it started. I did go to Germany around the turn if the century to evaluate early 3D printing of metal. It wasn't ready for application yet. I don't know if it ever evolved or not.

However, I did regularly review Military technology for industrial applications.
For those who may be interested. Reliable 'scanning' relies on a method called computed tomography. The math behind this has been known for 200+ years, but the technology only became practical in the 1970's for medical imaging. My doctors were very excited to show me a 3D rendering of my tumor from their multi-million dollar toys. X-ray based CT is by far the most common, but visible light is increasingly common. It is necessarily less precise because of the increased wavelengths involved, but also much easier and safer because you don't need a magnetron to generate x-rays, x-ray detectors or risk x-raying someone by mistake

In any event, several pictures are taken from different angles / perspectives, and then combined to deduce the x, y, z coordinates of edges / vertices from inverse matrix operations. The biggest challenges are when the exact position of the camera isn't known (or isn't very accurate), or when the lighting is inadequate or directional (shadows). I'm skipping many details obviously

To think about why a single 'point and click' solution won't work, think about the example of a UFO. From the point of view of a single person on the ground, there is a point of light in the sky that appears to 'hover'. But the source of the light could easily be traveling towards or away at hundreds of miles per hour. Then when the point of light suddenly veers to the left of right, it does not exhibit 'instant acceleration' - just a turn where the speed it already has is now in a direction that is apparent to that one person.

Anyways, once the outline of an object is in something like Autocad, it can be sent to a 3D printer - additive manufacturing device. Plasmas arc designs work well with many metals, but can be _very_ expensive too. I think that there is a lot of work yet to be done in this area
 

Matt-Aburg

Ultra Member
In any event, several pictures are taken from different angles / perspectives, and then combined to deduce the x, y, z coordinates of edges / vertices from inverse matrix operations. The biggest challenges are when the exact position of the camera isn't known (or isn't very accurate), or when the lighting is inadequate or directional (shadows). I'm skipping many details obviously
You are referring to structured light, with feature recognition. As in the video of Einstar, it loosed tracking often. Laser is way better because it uses markers that define space related to the camera. I doesn't need spray to lighten dark surfaces and can work in practically any lighting condition. Unfortunately, the cost of a start laser scanner is over 10 K.
 

mbond

Well-Known Member
Premium Member
Ambient light has a lot of restrictions. Emitted 'light' as in lasers in the visible band, x-rays etc. are clearly better. 'shadows' happen for those sources too - where the object has a complex shape and rays are blocked or can't reflect to the detectors. All of this is expensive i think
 
Top