In this video, we're going to be using
3D printing, 3D scanning, and 3D CAD modeling to repair broken stuff and just
generally solve a bunch of problems. And even if you don't have these exact
products or these exact problems, stick around, because along the way, you're
going to learn about half a dozen take home lessons on how I, an amateur with no
formal engineering training whatsoever, approach different design challenges to get great results. And, of course, how you can do the same. Let's get after it. So, throughout this video, I'm going
to be using both CAD and 3D scanning to generate models suitable for 3D printing. And I'll be doing that using
this latest greatest 3D scanner from Revopoint, the Miraco. Now, Revopoint sent this over and
they agreed to sponsor this project video, but what they didn't do is
send over a proper carrying case for it, like so many of other 3D
scanners that I've tested in the past. And that's a bit of a problem,
because the whole beauty of this 3D scanner is that it's fully portable
and requires no computer to operate. And our problem is then compounded by
the fact that in addition to all of these beautiful lenses up front, this
thing features a big, beautiful, and very vulnerable touchscreen in the back. So before we even dive into the rest
of the video and take this thing out into the wild, let's just go ahead and
whip up a quick carrying case for it. Shall we? The first thing I did was search
for a model of the Miraco on both Thangs.com and GrabCAD. And although this product wasn't even
released at the time, wouldn't you know it, someone had already created and
uploaded a free CAD model for me to use. Now to be safe, I took some
measurements of my own just to be sure, but here we have the first real
teachable moment when it comes to designing solutions to your problems. Don't reinvent the wheel. If you're like me and you actually
enjoy designing new things, then it can be really, really tempting to
just jump into CAD and start creating. But unless you're literally doing
it as a learning exercise, there's really no sense in spending your time
redoing the work that others have done. So before you start any design challenge,
I really recommend searching places like Thangs.com, which aggregates
all the different printable models from most online repositories, as well as GrabCAD, which has a lot
of the more technical engineering models of products and the like. With the model and my own measurements
in hand, I actually reapplied rule number one when it came to
actually creating a solution. See, instead of designing my own case from
scratch, I started off with this design for a parametric rugged box, which allows
you to customize the dimensions in your own specifications right in Fusion 360. I altered the width and
depth to fit the Miraco. Then figured out the ideal ratio of
lid to base so that the product would sit nicely, but still be easy enough
to pull out without inverting the case. I then added 5mm to each
measurement and hit print. To finish up the case, I just assembled
the box and cut up some leftover foam from packaging materials. Look, I'm not going to go so far as
call it another teachable moment, advocating that you hoard foam and
bubble wrap, but it definitely makes sense to keep some of this type of
foam around for projects like these. With that, we have our handy
carrying case that I can just throw into my backpack without fear of
scratching this beautiful 3D scanner. And while I might just reprint the base
with some kind of handle in the future, I think we're ready to get started. So, what should we fix next? So, recently I connected with a
local subscriber named Ron who does a ton of volunteer work for Yad
Sarah - a non-profit organization here in Israel that lends wheelchairs,
walkers, cribs, and all other types of medical equipment to people in need. Now for years I've been looking for a
way to volunteer here in my city that both leverages my current skills but
also challenges me to grow in areas where I want to learn and develop,
and this came along and it was the perfect opportunity to do just that. With literally thousands and thousands
of wheelchairs, not to mention all kinds of other medical equipment, Yad Sarah serves something
like 350,000 people a year. And that means a lot of small, broken
pieces that are difficult or impossible to replace from the manufacturer,
but pretty easy to 3D print. For this reason, Yad Sarah
actually already owns a bunch of 3D printers, but before Ron came
along, they didn't have anyone who actually knew how to operate them. And until I came along, they didn't
have anyone to actually model up the replacement pieces they need. So yeah, this was a perfect opportunity
to help an organization doing great and very important work, while expanding my
own CAD skills and taking advantage of my army of 3D printers and 3D scanners
and filament that are just sitting here. I started out by trying to scan
some black plastic parts which as you'll see throughout this video
meant that I definitely needed to use ATTBLIME sublimating spray. The Miraco doesn't claim to be able
to scan dark objects and while you can do feature or marker tracking
with these sticker thingies, I've always found it best to just
spray any remotely dark objects. After getting the hang of the scanner
and all the different settings, I managed to get some good scans of the
external geometry, but as with any 3D scanner, it's really hard to actually
get the scanner to see deep inside to the part to perfectly recreate it. I checked with Revopoint on
this, and they concurred. This brings us to teachable
moment number two. The truth of the matter is that, at least
for functional, mechanical parts, unlike say, a bust or a piece of art, 3D scanning
is great for getting reference geometry, but it's not a replacement for CAD skills. That's not only because of that interior
geometry, which you have no way of scanning, but also because things like
debris, uneven sublimating spray, or damage to the original item, mean that
you're never going to get perfect edges. And that's what you want when you're
dealing with functional mechanical parts with small tolerances. With that understanding in mind, I
imported my first reference models into RevoScan's beta app, and I was delighted
at how quickly and easily I could then clean up the mesh, fill the holes,
and do all that jazz right in the app. I then exported it from RevoScan as an
STL, and imported it into my CAD software of choice, which is Onshape, where I
then used it as a reference for creating my own model, basically from scratch. And that turned out perfectly, apparently,
because Yad Sarah immediately sent me more parts to model up for them. At this point, you're probably thinking
to yourself, it probably would have just been faster to just use a set of digital
calipers and measure things manually. And you might be right. Depending on the complexity of the part
and how proficient you are with your 3D scanner, it might be faster to do so. Later, we're going to look at some
projects where it just wouldn't have made sense to try and take
manual measurements when I can quickly and easily use a 3D scanner. But for now, teachable moment
number three is consider the best way to collect measurements. That might be 3D scanning, it might
be digital calipers, and it's worth taking a quick second to think
about it before you actually start. Up next, let's switch gears from fixing
broken parts to engineering new ones. I use this Elgato Stream Deck Plus for
live broadcasts and automations when I'm recording here in the studio, so I
wanted a way to carry it back and forth in my bag without it getting damaged. Now to do that, I removed the bulky
stand that it comes with, turning it into basically a flat device
that I could then, you guessed it, whip up a parametric case for, just
like I did with the Miraco scanner. But this meant that I no longer had
a stand on which to keep it upright, and while I toyed with the idea of
using it out of the case, I didn't want to strain the cable, and it just
wasn't comfortable to use lying flat. I could've packed the original stand
with it and taken it with me every time, but that takes two screws to attach
and detach every single time, and I figured there has to be a better way. To abide by teachable lesson number one,
I first checked online to see if anyone had any models or CAD files that could
help me out, but the only ones available were ones of the full stream deck,
not the dimensions without the stand. And while I could have modeled this
up on my own, it has all these kinds of irregular angles and small details
that would be a little bit tricky and time consuming to measure, so I whipped out the Miraco, slathered
the stream deck in sublimating spray, and scanned the bottom of it. Now my model didn't turn out perfect, but
I don't need it to be for my purposes. I then took it through the RevoScan
app, which once again, once I actually learned how to use it, because there
were no instruction manuals available when I did all this work and testing,
made it really easy to get a working mesh that I could use as a reference. I then ported that over into Onshape,
where it was as simple as creating my ideal base, then using a boolean
modifier to cut one out of the other. I also cleaned up the cutouts for
the screws since my 3D scan wasn't that great around that area since
I scanned it with the screws in. I also tried to use the thicken operation
on the STL to give myself a little bit of extra wiggle room, but it didn't
work properly since Onshape is really not suited to working with meshes. But fortunately, you can use a
built in feature of your slicer, X-Y hole compensation, to expand holes. And I just did that, compensating
with, I think it was half a millimeter. Now, my initial prototypes
worked phenomenally well. They held the stream deck firmly,
while allowing me to quickly yank it out for travel whenever I wanted. So based on that success, I decided
that I wanted to print it out in ABS and add in some weights. So, I used the built-in negative
volume modifier right within my slicer and I added a simple box and
then I put a pause in my G-Code. When the printer then paused, I squirted
in a bunch of glue to prevent rattling, then tossed in a bunch of rusty old
screwdriver bits and bobs that I've just been saving for such an occasion. Again, I'm not going to advocate hoarding
in this video, but old pieces of scrap metal are another one of those things
that it definitely makes sense to collect in a bin for projects just like these. So once all that was done, I went to
put in the stream deck and realized that I'd this time forgotten to
set the X-Y hole compensation, so the stream deck didn't fit at all. And I figured that if I'd forgotten to
do this in the slicer, someone else would likely forget too when I upload the model. So. I bucked up and scaled my scan up by
1.5%, rolling back in the design process so that I didn't have to redo things. And this is another teachable moment. I know that a lot of people like to
use Blender and it definitely has its place, especially when modifying meshes. But point number four is that when
designing prototypes, it's best to use non-destructive workflows. I can't imagine what a pain in the butt
it would have been to redo this if I had done it in a software that didn't allow
me to roll back in the timeline, insert a step in that point in the timeline. and then roll it forward as
if nothing had ever happened. Anyways, here is the finished result. Check out how it even has the little
half millimeter dimples for the triangular rubber feet at the bottom. Super cool and this has made it so much
easier for me to take the stream deck back and forth, and I have one of these
bases both here at the studio and at home. Mission accomplished. Alright, that's a lot of small indoor
type things, but as I mentioned before, this is the first 3D scanner
I've ever heard of that allows you to scan things on the go without needing
a computer or even a power outlet. So, what do you say we take this
thing out into the real world? My in-laws just bought a brand
new apartment, and like most new apartments in Israel, one of the
rooms has these three large, ugly, irregular holes near the ceiling. And so, my father-in-law asked me
if I could design something to cover them, you know, make them a little
bit more aesthetically pleasing. This was a bit tricky for a few reasons. First of all, these holes are covered
in plaster, meaning that the circles are in no way regular shapes, which
we're gonna come back to later. Third, one of them has an AC
line going through it, which needs to be accounted for. And the other, only part
of the circle is showing. This is one of those cases where taking
measurements would have just taken way, way longer than just 3D scanning. So I headed over there with the
Miraco in hand, and I took some really, really quick scans. With those scans in Onshape, it was
as simple as just generating some lids based on the reference geometry. Which, I made with a nice chamfered
edge, each with its own lip of various different sizes, so that it would kinda
get stuck in there on the plaster. For the irregular shaped one,
I simply traced the scan in the sketch, and I extruded it as is. But for some reason, probably laziness,
I decided that instead of actually using the reference geometry as a boolean,
like I did in other parts of this video, and cutting away material, like I did
with the stream deck, I would just draw perfect circles and extrude them. I figured that, worst case, I'd
have to chip away a little bit of plaster, but that it would
ultimately lead to a tighter fit. Yeah, I was wrong. After printing these monstrosities
out on the equally monstrous Comgrow T-500, the covers didn't fit at all. And I had to do a lot of hammering at
the plaster to get them to slide in. Not only did I have to chip away all
that plaster so that I wouldn't have to reprint the parts, but it also took me
a whole lot longer to clean up my mess. See, my diameters were right, I'd
just chosen to actually ignore the reference geometry, which was stupid
because I had it right there in my mesh. And that's teachable moment number five. Do it right the first time. Sometimes it's tempting to cut corners
in designing a solution, either because Someone won't see it, or a piece of
design doesn't seem to be functional. And don't get me wrong, if you're
confident that certain geometry isn't important, for example, if it's there
simply to facilitate injection molding or some other type of manufacturing, don't waste time on it for the sake
of thoroughness, but when it comes to things like how a part fits or mates
with something else, cutting corners and promising yourself that you'll figure it
out or modify it later is just gonna lead to redesigning, reprinting, and reworking. Plus, you'll get a much better sense
of pride over your finished project and a greater sense of satisfaction if you
know that the job is actually done right. Alright, for our final fix, we are
headed out into the parking garage where a tragedy recently occurred. My wife and I picked up a Tesla Model
Y recently, and although we love it, the turning radius and the camera-based
parking assistant both really suck. That's a problem because our parking
garage is extremely tricky to get out of, and because we opted for the
fancy 21 inch UberTurbine wheels. Long story short, I don't want
to talk about it, but we've been curbing the wheels at least once
a month ever since we got it. Now, I initially tried some
of those cheapy peel-and-stick pinstripes, but those things flew
off literally in the first drive. Upon further research, I did discover
that there are these modular replaceable covers by MagBak, which allow you
to protect the wheel with something a bit sturdier and better adhered. And look, I'm all for supporting small
businesses and rewarding people who come up with innovative products and all that. But as someone who owns 3D printers,
I literally cannot justify spending 250 dollars plus shipping for
a bunch of pieces of plastic. Plus, since I'm 3D printing these
on my own, not injection molding them, I can make my own changes and
improvements, for example, making my own two-tone versions if I want. So, instead, I set about creating
my own version of this design, once again using the Miraco to
gather my reference geometry. This was a much more difficult type
of design due to the curvature of the parts, and I knew from the get
go that printing it wasn't going to be easy on an FDM machine. Now, although the Miraco gave me a
very accurate and useful 3D scan to work with, I literally spent many
hours across multiple days sketching, modeling, and then re sketching this
design, and then deleting everything and starting all over again. I just couldn't get an initial
prototype to where I felt it was going to be even remotely close. I tried splitting the model with the
3D mesh, the boolean operators, but because Onshape is not that great with
mesh models, nothing was really working. And then, because Blender isn't
great for parametric designs that need to fit together, I really
didn't want to switch over to there. So I just kept redesigning over and over
and over, and then deleting because I wasn't happy with what was coming out. To be completely honest with you guys,
I actually considered completely giving up on this model and then chalking it
up to just this particular fix being beyond my current level of CAD skills. But then I remembered a really,
really valuable lesson that I actually learned in the incredible learning
CAD series by Michael Laws over at Teaching Tech, and it's a lesson
that is so valuable that I think it definitely is worth repeating here. Get to your prototype
as quickly as possible. See, oftentimes in the comfort of
our chair with our digital calipers in hand and our warm cup of coffee, it's very comfortable or tempting to
try to get to your first prototype and get it to be as perfect as possible. Especially if you want to save
on filament in the process of printing and reprinting them. But no matter how skilled of a
designer for designs of any level of complexity or which need to interface
with parts out in the real world. It's just not realistic to
get it right the first time. Now, I've found this to be true with
pretty much each and every design that I did for Yad Sarah, and it's especially
true when you're creating something from scratch, simply based on a sketch
or an idea, even if you have a 3D scan. Of course, don't confuse this
with teachable lesson number 5. We don't want to cut corners on
our designs or leave things off. But if we do, particularly when we
know that we're going to need multiple rounds of revisions, it's really, really
important to just move from CAD to the prototype as quickly as possible. Because ultimately, the way to get to the
best result isn't by perfecting the design miraculously in a vacuum that is your CAD. It's by accelerating the cycle speed
between prototype and revision. In other words, more revisions faster
is going to lead to more feedback from fitting the part in the real
world, which is going to lead to less frustration and ultimately a
better, more perfect design faster. Anyways, with this new mindset,
I went about printing out small pieces of prototypes and matching
them up to the wheel itself. And here's the funny thing. Despite hours of back and forth,
deleting and redoing my first prototype, was actually pretty darn close. This I guess is really just a
testament to where 3D scanning can save you a ton of time and effort. Because I can't imagine having to get
that exact shape of all these different curves through trial and error alone. Especially since it would be so
hard to try and get a profile of the inner part between the spokes. So I was honestly blown away by
how close my first prototype was. Actually clipped on, it just broke in
certain areas once I clipped it on. So, from there, I went about refining
and improving my design throughout six different iterations of prototypes,
getting closer and closer each time, until I finally had a model that fits and works. In fact, it fits so well that it's
actually really hard to get off the wheel, and I don't even think I'm
going to need to use double sided tape to actually install them. So, as we speak, I'm printing out a bunch
more of these, and I'm actually debating which material and which color I want. So I actually decided against
doing the two-tone ones because of the long print times and the huge
amounts of purge material necessary to switch every single layer. But I'm still not sure if I should
print them in black or white. On the one hand, the white would look
like those Porsche Taycan wheels, which are really, really cool. And I then don't have to worry
about perfectly matching this shade of matte black on the wheels. But, on the other hand, black
will hide the fact that there are wheel covers on there at all. But then I'd probably need to paint
them with the exact matching black or at least sand down the material so
that it perfectly matches the wheels and I'll have that matte finish. Plus, I'm really not sure if I
should go with ASA or PETG, so I'm experimenting before I print them
all out and cover the entire car. Anyways, enough rambling. I would love if you guys commented
below and let me know whether I should go with black or white, and I'll keep
you updated with just a quick YouTube short showing the final result. So, do make sure that you're subscribed. By the way, annoyingly, while I was
working on all this, the sellers on AliExpress did manage to beat
me to the punch, and they're now selling their own version of the 21
inch UberTurbine covers for like a 1/6 of the price of the original. But I do still think it's really, really
cool that I now have my own design that I can print replacements for as needed
whenever I break one of these or curb my wheels, so still a worthwhile project. So there you have it, a handful
of different repairs and problems all solved using 3D printing. Plus a bunch of what, at least to me, have
been pretty valuable lessons in how you can best approach using a 3D printer and
3D scanner to solve your own problems. If you learned something, please make
sure that you're one of the 15 percent of viewers who are actually subscribed. Yikes. And leave me a like so I know
to make more content like it. Also, if you want to pick up one
of these brand new, super awesome Revopoint Miraco scanners, I'll
leave a link in the description. You can check them out and they should be
on sale by the time this video goes live. I've really been enjoying
working with the Miraco. It's definitely going to
become my go-to 3D scanner. And I have a lot of 3D scanners, but the
fact that I can take this one portable and mobile means that really all the other
ones are kind of useless at this point. Hey, thanks for sticking around to the
end, and a special thanks to my Patreon and YouTube members, and an extra extra
special thanks to our Nylon and Peek members, 2 Krazy Ketos, Chip Cox, Amir
Chen, Kris Miller, and Don Arledge. You guys rock. Anyways, that's all for this week, but
I'll see all of you on The Next Layer.
