| Start | Text |
|---|
| 00:00:00 | Hello, everyone, and welcome back. If you've been following along, you'll know that I'm interested
in building a robot that is capable of helping me sort coins
for my coin collection. In the last video, we were talking about my new Bambu Labs Carbon X-1 3D printer
and some of the things that it could do. |
| 00:00:17 | But in that video, I also mentioned that I
really have no real world experience using a 3D printer, and I obviously
have some pretty elaborate plans for it. But since I don't know what I don't know,
I wanted to try and print and build a few designs created by other people
in an effort to get comfortable with the limits and tolerances
of 3D printing in plastic. And I figured I'd get a few bonus points |
| 00:00:41 | if I ended up creating something as
complex as it is completely useless. So stay tuned if you'd like
to see what I came up with. So, in this video,
I'm going to attempt to print and build this seven segment display counter
created and distributed by Fhuable. This is by far the most complicated thing
I could find to build, and even the instructions say that it
is not recommended for beginners. |
| 00:01:08 | But I've never let a little thing
like lack of experience slow me down. So let's just jump
in with both feet, shall we? So I paid the $8 or whatever to download
the STL files for this project. But given the amount of time that must
have gone into the design, the clear amount of over engineering
involved, the more than 100 parts that needed to be printed,
and the 26 page instruction manual |
| 00:01:31 | that comes with it, I think
that's really an amazing deal. And again, this video is not sponsored. Overall, printing the STL
files took a long time. I completely filled each build plate with parts, and each of those took
approximately 5 hours to print. So with setup and trying not to annoy |
| 00:01:50 | my wife by printing things in our tiny
apartment when she's trying to sleep, that represents about three
days of setup and printing. But even though it took a long time to print the parts out,
the 3D printer performed flawlessly. And I didn't really have any other issues
with the actual printing of the parts, especially after solving the noob
issues I mentioned in the last video. |
| 00:02:12 | The one thing I will mention about the design is that the gears that drive
the actual segments had very important alignment marks on them that did not print
out very well, even at my printer's best resolution, which I believe is
a point two millimeter layer height. Since you need to use these marks
to properly set up the entire mechanism, these marks should have been made
clearer on the finished gear. |
| 00:02:35 | In fact, I think those calibration marks could be through holes so that they
become even clearer to the builder. Also, printing the part numbers on the actual part would have been
helpful, but I just used a sharpie marker on them before removing
them from the build plate. In my opinion, |
| 00:02:52 | 26 pages for the instruction manual isn't
quite enough to do this build justice. I think the instruction manual could have been expanded to about 50 pages in order
to make the form and function of many of the parts much more clear to everyone
beyond the original designer. For a newbie like me,
it was really pretty difficult to understand how all the pieces fit
together, how they should be aligned, |
| 00:03:15 | how they needed to actuate,
and how important it was to have every sub assembly aligned so that you could get
the correct sequence of gears set up so that the segments count
in the correct order. Now, the instruction manual is fairly
clear about how to put most of the major assemblies together,
but most of those assemblies are put together with very specific screws
called out in the appendix. |
| 00:03:37 | And I'm not sure if this is more global supply chain issues or what,
but as it turned out, those specific screws are currently
very difficult to source. It said that we needed a 2.5 millimeter by eight and said that the length
really wasn't that important. But even with that leeway,
I was unable to find the screws online, |
| 00:03:56 | not with a quick Google search, not on
ebay, and apparently not even on Amazon. I ended up purchasing screws that I
thought would work on ebay and others on Amazon, but those all
ended up being too big. These larger screws kept cracking the PLA plastic and I had to keep using Gorilla
superglue to put the pieces back together. The process was a bit frustrating, |
| 00:04:19 | but it eventually kind of worked and I
could get it to somewhat hold together. In a lot of cases, I ended up just gluing parts together that probably
should have been screwed together. And I'm not sure I'll be able to get this back apart later if ever needed,
but hopefully that won't be an issue. I built my unit with a manual hand crank,
and as you turn the crank, |
| 00:04:39 | you can see that it wants to move
through the entire count sequence. But because there was a lot of sticking between the plastic parts,
in a lot of cases the parts do not move freely enough, and that means a lot
of these segments do not actuate freely. Now, in the instruction manual,
they address that issue and ask that you use a hand reamer to ensure that all
of the parts turned freely on the small |
| 00:05:01 | nails we're supposed to be using as pivots
for many of the plastic mechanisms. I don't have a hand reamer and I didn't find one that was the right size
to purchase, but I was able to use a small Phillips screwdriver that came
with my computer repair kit, and that seemed to work pretty well until,
of course, I put the whole thing together. Now, I was particularly careful to make
sure that all of the pieces rotated |
| 00:05:25 | freely, but I think I could have been
a little more thorough with my cleaning up of the gears to ensure that all
of those rotated freely as well. Despite my best efforts,
the plastic parts were not free enough to be actuated by the plastic
springs called out for in the plans. So I tried to design even stronger springs to help snap the segments into position
with a little more oomph. |
| 00:05:47 | But as you can see in the finished product here, that didn't really
improve matters much. You need to manually push the segments
forward and back to actually see the numbers that the unit
was trying to display. But as I turn the crank here, you can see that the counter actually
does count from zero to nine. |
| 00:06:05 | But again, because the segments don't
actuate freely, I need to manually pull them forward or back so you
can see what's going on. Now, this is fine for my purposes. I mean, I was really only doing this as
a proof of concept to test the tolerances of 3D printing, and it gave me a good
baseline to work from so that I can see I need to loosen up the tolerances
a bit in my own future designs. |
| 00:06:28 | I've also seen in other 3D printed parts
that the creator often leaves room for a bearing to ensure
that the parts spin smoothly. So that's something that I should
also consider in my own design. Any parts that need to rotate or spin will
probably need either some sort of lubricant that doesn't eat plastic or
they'll need a bearing to reduce friction. All in all, I'd say this
was a successful failure. |
| 00:06:53 | I learned what I wanted to know by putting this all together, and it really
was a great learning experience. Again, I think it was a great test case for my eventual, perhaps even
more complicated robot designs. In the next video, I'll show you some of the revisions I made
to my own robot design. |
| 00:07:06 | In the meantime, maybe you'd like to check out some of my earlier videos
and see what all this is about. Please subscribe and hit the notification bell if you'd like to see
where this is all going. But until next time, that's my two cents.
Thanks for watching, everyone. |
