Making fossil replicas with 3d printing

[Misha]

Making replicas of my fossils has always been something very interesting to me, but for the longest time I wasn't sure about how it should be done, casting although it could preserve a lot of detail can be pricy if you want to make replicas of multiple different fossils and it is something that I am just very unfamiliar with. From that point my attention turned to 3d printing as this technology can allow us to replicate objects fairly inexpensively, but when I started to looking at 3d scanning for actually making the models I found that scanners were not all too cheap either ranging from about 500 dollars into the thousands. This is when I discovered photogrammetry, which allows you to make 3d models of objects with just a number of regular photographs taken from different angles, no need for expensive tools to scan the object. I found a free program called 3DF Zephyr that allows me to upload up to 50 pictures of a certain item and construct a 3d model from them.

Yesterday I installed the program and immediately went to find one of my fossils that would be good to start off with, I wanted something where I wouldn't need to scan it from the bottom and something with good surface that would work well with the software. I thought that my Flexicalymene ouzergui from Morocco should be perfect for this.

I read that you should take pictures so that every desired point of the object is seen from at least 3 different angles and so I took the bug and started doing that, I used 3 different elevations for the camera and rotated the bug taken pictures from many different angles, I ended up with a total of 45 photos.

I then uploaded all of these directly into the software and it used all but 1 photo. It took a few hours to construct point clouds of a bunch of different points on the bug and some time after to reconstruct a mesh from that. What I ended up with looked pretty good and I was really happy with it for my first attempt, especially since I have heard that it may take several for you to get the model looking correct but nonetheless there were some issues with weird floating geometry around the bug. To address that I first selected everything disconnected from the model and removed that, then I took a lasso tool to select and remove all of the blobs attached to the bug that I didn't need. Here is what the model looked like before and after this:

Now I saved the mesh as an OBJ and brought it into a different program to fix all the holes in it, make it solid instead of just the hollow shell I got from the scan and smooth over certain parts.

This step was very successful and the model looked great, now it was time to print it. I took it into the next program to scale it properly and slice it for printing, I scales it roughly to the same size of the original bug and added all the needed parts to print it. The first print got disconnected and failed but then I made some slight changes to the model and leveled out the print bed, after that it was very late and I went to sleep but woke up today to a perfect copy of my Trilobite, it had some supports holding it up which were a bit reluctant to disconnect but after a bit of cleanup it looked great.

Here is the finished print alongside the original:

This has been a really fun thing to try out and now I want to do it with some more fossils, I think the next one I am going to try will be a small crinoid and then a more challenging model like a brachiopod where I will need to do some processing with photos as I will be taking pictures from all sides of the object.

I encourage anyone interested to try it for themselves, if you want any help or additional information as to how I did certain things here I would be happy to discuss that with you.

Thank you for looking,

Misha

[Nchapman]

Very cool Misha!  What printer are you using?  Materials have been expanding rapidly in the last few years which makes this an interesting topic.  What slicing software are you using?  Slic3r, Repetier, manufacture specific?  3d printing is something I'm very passionate about and truly enjoy so if you every have any question don't hesitate to reach out to me.  I've built almost every opensource printer out there and have been the technical editor on numerous 3d printing books/papers.

[Misha]

10 hours ago, Nchapman said:

Very cool Misha!  What printer are you using?  Materials have been expanding rapidly in the last few years which makes this an interesting topic.  What slicing software are you using?  Slic3r, Repetier, manufacture specific?  3d printing is something I'm very passionate about and truly enjoy so if you every have any question don't hesitate to reach out to me.  I've built almost every opensource printer out there and have been the technical editor on numerous 3d printing books/papers.

I have a few printers, the one I used for this was my SOVOL SV01, and the slicer is nothing special I just used Cura. I haven't done much more of this 3d scanning yet as I haven't had time but would like to continue with some other pieces, hopefully soon. I am also looking to try and expand into trying either SLA or DLP printers, I want to try making model recreations of extinct organisms and I think having one to print those would be nice to really capture the details.

[Nchapman]

I have not used the SOVOL SV01.  What's your thoughts on it?  It looks like the board is still using the older Atmega 2560 (Similar to the Arduino Mega/Ramps 1.4).  If your ever looking to upgrade your controller I would recommend the Smoothieboard V2.  Arthur and the community over there is fantastic.  Cura is a great slicing program.  Getting into the DLP/SLA printers is a great move.  It would allow for higher details and complex shapes while printing.  Just watch out as there is a lot of cheap printers out there.  Check out either the Prusa SL1S or or any of the Formlabs printers.

[rodrex]

Do you think it's easy to lasso out individual parts and print them individually and then put them back together?

 

I have some large cretaceous fish that are mostly 2D, but Id like to convert into a 3d replica, like the ones you see Triebold Paleontology doing for the Kansas Chalk fish.

 

I also have multiple specimens of the same species that I want to use to complete the whole skeleton.

 

thoughts?

 

thanks 

 

Rodney 

[jpc]

This would be a challenge with a surface scanner or photogrammetry because these technologies do not see the parts of your squished fish skull that are underneath other pieces.  It would work better with CT scanning where the CT scan sees into the specimen.  You would have to use a program like 3DSlicer to manipulate the CT data.  I just downloaded it yesterday and am hoping to find time to learn it.  I have some volunteers in the museum where I work who are learning it.