Something I’ve learned is that editing short clips (social media videos that are usually limited to 60 seconds) and inserting screenshots to get your point across effectively is hard. And after I’ve played around with a few different methods in attempts to improve that, I think I’ve significantly improved it.
Last week, when I was working on editing a clip, I decided to experiment with screenshots; this time around, I tried an “on-screen look.” Instead of just slapping an image full of small text, I used a couple of really cool features in Premiere to animate, blur, and highlight the text to bring focus to the actual important part.
Here are examples of a video that has editing and no editing of the screenshot:
In the second example, I added a bunch of extra effects to get across the focus of the screenshot. One of those changes that I had never used before was the “Basic 3D” effect; it adds that tilting/pan animation, which really emphasizes the “on-screen” effect.
Some more minor changes were the blurs. In the background, I added a stock photo of the topic (in this case, a girl crying) and some blur to blend it into the background. Another blur that I added was around the headline. A simple “Gaussian Blur” with an inverted mask did the trick. The highlighting effect was also done by simply keyframing two orange blocks with the opacity set to “Multiply,” which blended the colors, instead of just turning down the opacity for a transparent look, which would work just not as clean looking.
Most of this I learned from a handy tutorial video by the channel Storysium. Which brings me to my next point:
It’s easier than ever to start Video Editing
The great thing about video editing is that almost everyone needs to do it. Thus, there are hundreds of thousands of tutorials on basically any way to edit. And all for free on YouTube. Most of my learning has been from just watching shorts, or other videos in which I see someone do a really cool effect, and then I look up or try to replicate that myself. I myself am still learning stuff, and I think I can still improve many aspects of my editing, but with the free use of YouTube and many people willing to share their presets for small prices, it’s a quick process to get started with video editing today.
I spent about three weeks working on 3D printing an RC plane. I had initially searched for a somewhat easy RC plane to print and assemble as I’ve never had any experience with RC vehicles in general. I eventually found a good model that was both free and looked simple enough for me to be able to print. The model I ended up with was the Piper J-3 Cub by 3DLabPrint; 3DLabPrint is a Czech Republic-based company that designs 3D printable RC plane models. They have a bunch of options to choose from when it comes to RC planes.
It took me about three weeks from starting printing to assembling electronics and crashing flying; I went with Polymaker LW-PLA, as they had a summer sale when I purchased; you can get a roll of plastic (and support me) by clicking here.
3DLabPrint recommends their plastic, but the shipping would have been pricey as they are located in an overseas country. But there is a difference between PolyLight (3DLabPrint brand) and PolyLite (Polymaker brand). Polymaker prefoams the material, which makes it easier to print as it has very similar properties to regular PLA, and thus, I was able to print it at my stock PLA slicer profile. 3DLabPrint’s LW-PLA is not prefoamed and requires more tuning to compensate for expansion and other variables when it comes to printing.
For the first couple of parts I printed, I hadn’t gotten the printer’s slicing profile correctly; for example, I accidentally added two bottom layers and two top layers, which I don’t need for this model as specified in the instructions. 3DLabPrint provides extensive instructions on how to assemble and prep your printer before printing. But after some trial and error, I got the slicer settings down and printed the rest of the parts. It took me about a week to print and assemble all the details, but I didn’t purchase the electronics till later in the following weeks.
In total, I spent about 68 hours of just printing. That also includes any mistake prints and or extra parts, or multiple of the same parts, and other materials printed like the landing gear and tires. All the pieces were printed on my heavily modded Ender 3; I believe a stock Ender 3 (or similar-priced printer) should be able to print this material just fine. I would advise printing in 1.15 extrusion multiplier (or flow) in your slicer and printing a bit lower temps, at about 195°C for the hotend.
Close-up of Finished PartEnder 3 Printing RC Plane
If you don’t have a 3D printer, I recommend a Sovol SV06 for a cheap option or a Bambu Lab P1P if you want to go for a more quality and hassle-free brand.
At first, I ordered the wrong parts; the ESC was made for drones and hadn’t come with a 5v line that was standard for RC planes. I returned it and got the correct one. (List of my configuration below)
The only problem now is that I have no idea how to fly an RC plane. I watched a couple of videos to check I got the directions right and propellor was working correctly, (thanks to the folks in PW3D’s discord channel, I could ask noobie questions and get informative answers.)
I tried to fly and crashed it almost immediately, which I believe is due to user error and not a design flaw. At the moment, I need to repair it and attempt another take-off again. I can say I got it in the air (6-9 feet at least) with 75% throttle but had not correctly lowered the throttle, and it came straight down, cracking the wing assembly and a couple of other parts of the main body.
Crashed on Aug 19th, 2023
That’s been my experience so far on 3D printing an Airplane. If you have any questions or thoughts, leave them in the comments box below, and I’ll get back to you.
*Disclaimer: Most products linked are affiliate links and will support me if purchased through them.
I recently wanted to learn more about Fusion 360 3D designing. To further my 3D modeling experience, I started another project, this time to redesign one of my existing 3D printers from scratch.
It started with looking at the Prusa Mini+ and how the X gantry worked. The Prusa Mini is an FDM 3D printer by Prusa, a German company. It has a similar style to how my Printrbot Simple Metal is oriented. One significant difference is that the X gantry moves along the rods instead of the rods moving along a centered Z gantry. For example:
As shown in the figure, the Prusa Mini+ is designed so that only what is needed is moving, the Fans (part cooling, hotend cooling), Extruder, Hotend, and Z sensor. One plus of having the Prusa Mini Design is that there’s very minimal weight and thus less mass movement. I only have to carry the needed weight instead of a significantly heavier X arm on the Printrbot. This isn’t necessarily a downside. 90% of the Printrbot is machined, so it’s highly durable and can “fall off a table and go right back to printing” – Brook Drumm (CEO of Printrbot).
Another great thing about the Prusa Mini+ is that Prusa manufactures everything in-house and 3D prints most of the parts for their printers. I have found that to be useful as they also release all of their step files which I can use to modify the designs to my liking. Same with the Printrbot, even with it being shut down a couple of years ago. At BrookDrumm.com, there are uploaded .F3D files for every model he’s produced and even some that haven’t been manufactured,
Prusa factory tour showing off Prusa’s print farm.
I started by studying both printer designs in Fusion, how they worked, what made them different, and how I could combine both to create the best of both worlds. Next, I began designing my first iteration of my “Printrbot with the Prusa Mini style”PrintrMini.” The first one was very rough. A ton of wasted printed space, an inefficient motor mount with weight not being distributed properly, and a couple of more minor issues: At this point, I’m hoping for one of my last iterations of the design.
After some feedback on a Discord server, I redesigned both the toolhead and the X-Z gantry mount. This time to be as compact as possible with as even weight distribution as I could manage and utilize many of the original parts. This was V2:
(Above: I figured out how to render in Fusion, so I tried my hand at it. V2 of my design)
I also bought an E3D V6 Hotend since the original hotend (Ubis 13s) was rather old and had some flaws compared to the newer V6; since the V6 hotend and the Ubis hotend have very similar designs, the original extruder could mount it, which a massive bonus with not having to order another extruder assembly.
I decided to go for a very similar style in V2 to the original X/Z gantry attachment. I utilized the same motor mount and lead screw to make it more compact and utilize as many original parts as possible.
At the moment, I’m working on V3; version 3 would be a machined X/Z gantry with an almost identical design to the original Printrbot billet but without the bearings for the X gantry:
(Above: V3 of my X/Z gantry mount)
I didn’t change the toolhead mount in V3, as I’m still finishing up some smaller details on the mount. But since it’s 3D printed, I can always print a new design if my current model has flaws. The new X/Z mount would utilize press-fit bearings and press-fit 8mm rods. Using a machined mount would provide several benefits, one being that it’s aluminum and would be super strong compared to my plastic 3D prints; another is that it would still utilize the aesthetics from the original Printrbot with it being machined metal; it is called Printrbot Simple Metal after all.
I’m still getting a quote from my brother (CovenantMFG) for that at this time, so v3 may or may not happen. If not, ill probably figure out a way to 3D print it, most likely in Carbon Fiber infused filament for strength and temperature resilience.
So far, this experience of learning fusion and changing my design over and over for more compact and practical parts has been delightful.
A big plus in this process has Fusion’s student plan. The student plan allows me access to almost every feature for free as long as I’m a student. Previously I had been using the Personal license, which included limited features, such as only up to 10 design files at a time and only being able to import .step files for converting 3D models.
Recently I bought a used Canon SL1 camera; it’s an old but great camera to use as a webcam or any general-use camera taking. One benefit is that it works with a lot of software using the USB port to capture a live video feed from the camera. When using the USB port it does have a bit of delay between the camera and the live feed, but it isn’t super noticeable in recordings.
I also discovered it works with Octolapse, a plugin for taking seamless-looking time-lapses of 3D prints. Following a tutorial by the creator of Octolapse (FormerLurker), I started setting up my camera with my Octoprint server.
The tutorial starts with the initial installation of the main program to interface with the camera. gPhoto. gPhoto supports a wide range of cameras, including my camera, which can File transfer, live view, and capture photos or videos.
What is gPhoto?
gPhoto is a set of software applications and libraries for use in digital photography.
gPhoto Wikipedia
With the installation, I generally followed the provided steps. Updated any updates that my Ubuntu server needed, updated Octoprint. When installing gPhoto, I had to use the “alternative instructions” since I didn’t run my Octoprint server on a Raspberry Pi, the structure of Linux flavors and different device configurations can make it differentiate the process configuring scripts and general directory locations.
Next is editing the “sudoers” file. It is similar to the standard setup, but instead of the regular path, I used the provided command to find the install directory where gPhoto was installed. I also used the output inside the line of code I needed to insert into the last line on the file.
After that, I can test gPhoto with a snapshot command; surprisingly, it worked! I heard the audible click, and it saved a picture to my server. Once that is set up, the next step is to create a script for the plugin to run in Linux that holds a picture, deletes it from the original SD card, and saves it to where the snapshots need to be stored to later process the whole timelapse video.
The script included with the tutorial works out of the box for me, thanks to FormerLurker’s documentation.
After that, I add it to the plugin’s configuration. Which also went well. I continued the guide with “Configuring a DSLR camera,” After which I started a print, and the final result after rendering is below:
Final result. (Actual quality of the printed cube was terrible due to a lack of tuning at the time)
One issue after rendering, which I could only fix by uploading it to YouTube, was the encoding and format of the rendered video. VLC media player says its renders at 4K, meaning 4000 pixels across. For some reason, my computer could not play 4K in the regular Windows media player or even my other video players without rendering it at 1/16’s quality. This may be due to the lack of support in my computer’s GPU (Intel Arc A370M Laptop GPU), but I’m unsure. But I can change that in Octolapse’s settings when it renders the final video. The F-stop was also not set correctly, which I can adjust in the camera’s menu.
