I try to write these update blog posts around the middle of the month. That gives me plenty of time to make sure I don’t forget to write one of these every single month. I’m doing a bad job. We’re beginning the final week of November, and I completely forgot that I’m supposed to do this!
I was worried that I hadn’t accomplished anything except for recording two podcasts. I feel like that’s a pretty good accomplishment on its own, but it turns out that I’ve made some pretty good progress in other areas too. Not much of it was worthy of its own blog post, but there’s definitely enough to write about in this month’s update!
I’m on a podcast!
I am excited about this. I’m the new co-host on The Create/Invent Podcast!
Jeremy Cook and Max Maker have been hosting the podcast for quite a few years now. I was bummed out to hear that Max Maker was exiting the show, but I was excited when Jeremy eventually asked me to co-host the show with him.
He tells me that the search for a co-host was arduous and thorough. I sure hope he made the right choice!
So far, there are two episodes up with me in the co-host chair. There’s my introductory episode, and there’s our interview with Chad Dowdell where we talk about laser cutting and Etsy.
- The Create/Invent Podcast
- The Create/Invent Podcast on YouTube
- The Create/Invent Podcast on SoundCloud
I bought a Sony ZV-1 point-and-shoot camera
I’ve been considering this upgrade for a while. My Canon 6D is a phenomenal DSLR, and I have an assortment of nice and not-so-nice lenses for it. Newer cameras aren’t all that big of an upgrade as far as photography goes, but the Canon is severely lacking in the video department.
Canon’s color science is awesome, and the 6D can actually record beautiful 1080p video. It can’t autofocus while recording, which is a HUGE bummer when you’re a one-man operation. It also can’t output more than 480p over HDMI, so I can’t really use it with OBS Studio for live streaming.
I have a DJI Osmo Pocket, and it is an amazing vlogging camera. It is cheap. It fits comfortably in my pocket. I don’t need to carry a tripod or selfie stick when I want to talk to the camera. For what it is, it is quite good.
Podcast?! pic.twitter.com/WZKfcWGsg1— Pat Regan (@patsheadcom) November 19, 2020
I’ve been eyeballing the Sony ZV-1 ever since it came out. I told myself that I wouldn’t upgrade, because I’m not vlogging enough with the Osmo Pocket. Why should I upgrade? I can’t live stream with the Osmo Pocket, but I already have a Logitech C920 for that.
Then Jeremy invited me to co-host his podcast. The ZV-1 has lightning-fast autofocus that locks right onto your eyeballs. It can output 1080p or 4K video over HDMI with no on-screen display getting in the way, and it works great with my $12 HDMI capture device. It doesn’t require nearly as much light as my Osmo Pocket or Logitech webcam.
It is such a nice upgrade for podcasting, and it doesn’t require much more room in my pocket than the Osmo Pocket. I’m also learning that it really isn’t all that far behind my Canon 6D when it comes to dynamic range, so it will be a nice photography camera as well. The entire Sony ZV-1 with its built-in lens is smaller and weighs less than my favorite lens for my Canon, so I’m much more likely to take the Sony on the road with me.
NOTE: I need to embed a tweet here that shows off the lightning-fast autofocus!
The Sony ZV-1 is a good compromise. Its autofocus is easily on par with the Sony a6500, it has in-body optical image stabilization, but the ZV-1 weighs less than just the body of the a6500. It is half the price, half the size, and half the sensor, but the ZV-1 has a better mic.
It is a tough comparison to make, but I’m expecting the Sony ZV-1 to be a better fit for my needs than a real mirrorless camera from Sony.
- I Bought a Sony ZV-1 Vlogging Camera
- Three Months With The Osmo Pocket: The Perfect Vlogging Camera?
- I Upgraded to a Canon 6D Full-Frame DSLR
The OoberLights have leveled up!
This is super exciting. Our illustrious programmer has been hard at work on the OoberLights code for a while now. Last month, he rewrote some bits to implement a compositor. This means we are able to run multiple animations on top of each other.
That was exciting, but not something I could really show off. He was also rewriting the graphics API, so all we really did was port the old animations to the new framework. It was all working differently, but it pretty much looked exactly the same!
We now have an API for sending commands to the OoberLights via WiFi. You can send JSON-formatted configurations over HTTP to change the OoberLights animations.
All we need to do now is create a few animations and get the OoberLights to accept this JSON over USB serial, and it will be possible to implement all the functionality I had envisioned when I first came up with this idea many, many, many years ago!
We are in exciting times! We’re close to having a workable feature set for the OoberLights. We’re about to order some test PCBs for the redesigned power stages. If those power stages work well, we shouldn’t be far from ordering the final prototypes with three rings instead of two, and those prototypes are likely to be identical to the production parts!
I have a door marked pirate
Look at that door, dude. See that door there? The one marked “Pirate”? You think a pirate lives in there?
Look at that door marked pirate. Do you think a pirate lives in there?! pic.twitter.com/fOboSJdDmA— Pat Regan (@patsheadcom) November 16, 2020
After recording my first episode of The Create/Invent Podcast, I noticed that the various doors in my home office are boring. They’re plain, old, ordinary wood, and they have no signage!
I decided that the quickest and easiest way to remedy that would be to 3D-print a sign reading “PRIVATE.” It is the first time I’ve done a multicolor print on my Prusa MK3S. It went smoothly. The bottom 80% of the print ran with white filament, then the printer prompted me to load different filament, and the job finished up with black filament. No problems.
In practice, though, this isn’t working so well. That shallow depth of field on my Sony ZV-1 means my sign is completely out of focus on the podcast. No one will ever wonder if a pirate lives in my closet!
- I’m Ordering a New Prusa 3D Printer!
- Eight Hours With My New Prusa MK3S 3D Printer
- I Bought a Sony ZV-1 Vlogging Camera
I upgraded my drone’s Crossfire firmware and locked them at 150 Hz
TBS Crossfire is the hardware and protocol of the 900 MHz control link I use to fly my FPV freestyle drones. It is one of those things that I set up and forget about. TBS releases firmware updates pretty regularly, but I don’t usually upgrade unless I’m excited about a new feature.
The last exciting update I skipped added support for CrossfireShot. This is a wired protocol used to communicate between my controller and the Crossfire radio transmitter. TBS bragged about it reducing latency by 25%, but latency between the controller and RF transmitter is already tiny. I believe this saves about 0.5 ms out of 10 to 26 ms. A nice upgrade, but not a big deal.
Depending on your signal quality, Crossfire runs at 50 or 150 frames per second. One of the more recent updates allows you to lock Crossfire at 150 Hz at the expense of range. I’ve finally hit a point where I’m not fixing broken quads every month, so I figured it was time for a Crossfire update, and I’m really glad I finally did it!
I’ve only flown three batteries, and I haven’t done a back-to-back comparison between 50 Hz and 150 Hz yet, but I think it makes a huge difference. The difference isn’t where I expected.
I feel like I have leveled up my throttle control, and by a lot! I was doing a bit of proximity flying under one of our favorite trees, and I can smoothly catch my quad after S-turning a branch, and I can catch it just about right under the branch. I didn’t feel as confident in this before.
I’ve already been flying 150 Hz most of the time, but Crossfire was constantly dropping down to 50 Hz. It wasn’t exactly predictable, so I was always flying with variable latency. If I’m to believe the graphs on ExpressLRS’s GitHub repository, 50 Hz has a latency of 26 ms, and 150 Hz has a latency of 10.5 ms.
This will deserve its own blog post after I fly and test more.
I fixed my AmazonBasics backpack
I have two drone backpacks: a big, honking, expensive bag from ThinkTank that fits everything I could possibly need when I go out flying, and a small, cheap AmazonBasics DSLR backpack for when I want to pack light.
There’s no room for a 5” miniquad inside the AmazonBasics bag, so I wind up strapping it to the elastic cord running along the outside of the backpack. My quad we is fairly weighty, and when I’m bouncing around on my Gotway Tesla unicycle, I’m putting a lot of stress on the loops that hold the elastic cord in place.
Six of these finished #3dprinting. I realized I needed 4 more, so I queued up more. Then I realized they're too fat to fit. I adjusted the model, and I just started printing 6 more.— Pat Regan (@patsheadcom) October 23, 2020
I need 10.
I am a dopefish. pic.twitter.com/w5guTgvbDK
This is a little like the $15 tripod I fixed in September. Why am I working so hard to repair a $25 backpack? It seemed like an easier problem when I started.
The design was simple. I just didn’t anticipate how difficult it would be to fit my little doodads into the tiny loops on the backpack. The first set didn’t fit, so I had to scale things down.
The upper fabric loops were in rough shape. I used some E-6000 adhesive to shore those up and keep them attached to the strain-relieving 3D-printed part. I figured the others would all just sit in place, but I couldn’t make the lips on top and bottom of the prints wide enough to hold them in place. I wound up gluing all of them!
If I had known glue would be involved in every single loop, I wouldn’t have 3D-printed anything. I could have just cut up some pieces of soft scrap plastic to glue into the problematic loops.
I won’t have to buy a new backpack every couple years, which is good, but it isn’t like this backpack would have been trashed. It just wouldn’t be of use for carrying miniquads any longer!
My old MakerFarm i3 3D printer has a new home!
I bought my 8” MakerFarm i3 printer way back in 2014. These awesome, overbuilt, overpowered 3D printers were sold in a kit. I bought an already assembled machine on Craigslist for about $100 less than the price of a kit. There were some extra tools, filament, and odds and ends in the box.
It was poorly assembled. The Y-axis carriage fell apart during the first week. That was about the only thing I actually had to fix. The wiring was terrifying and embarrassing, but I didn’t want to rerun all the wires unless there was a problem. There was never a problem.
My old #3dprinter came home for a visit so I could help my neighbor get it back up and printing. I made such a huge mess trying to clear enough room for the printer and a laptop on my work bench.— Pat Regan (@patsheadcom) October 22, 2020
The mess is already clearing itself, and we printed a test part successfully! pic.twitter.com/rrOP6BIgAv
Until the connector for the heated bed on the RAMPS board melted. When that happened, the question was whether I should make the easy repair or upgrade to a Prusa MK3S.
The Prusa MK3S has been amazing. My old printer beats the Prusa MK3S in a few ways. That old extruder is overpowered and nearly unjammable, and all its stepper motors are over-sized and much more powerful than the Prusa MK3S.
The real Prusa machine wins, though. I rarely even bother to open Octoprint now. I just check the button to immediately begin the print in PrusaSlicer, and more than 9 times out of 10, everything just works. That’s amazing and awesome.
My MakerFarm printer lives across the street now. My neighbor had it repaired in no time. He brought it back to my house a couple of weeks ago so I could help him verify that everything was working and show him how to get a print out of it.
It was nice to see the old machine complete one last print job in my office!
- I Bought a 3D Printer: I Have No Idea What I’m Doing
- Control Multiple 3D Printers Using A Single Raspberry Pi and Octoprint
- I’m Ordering a New Prusa 3D Printer!
- Eight Hours With My New Prusa MK3S 3D Printer
I’m reorganizing my drone battery-charging station
This is so minor yet so exciting. I’ve been meaning to do this for years, but it just never happened!
Brian and I both chose the IKEA Hemler for our drone part storage. We chose it because it is made of sheet metal. If our LiPo batteries manage to ignite in there, it isn’t going to set the drawers ablaze. We both charge our batteries in a Bat-Safe now, so I haven’t been giving much consideration to battery storage.
What I’ve done so far is clear out the top two drawers. They used to have random spare parts, but I don’t do any drone repair at the charging station. I decided to dedicate the top two drawers to LiPo storage.
The third and fourth drawers have already been dedicated to propeller storage. There are still some random things in the bottom two drawers, but I’m working on figuring out what might need to go in there.
My workbench has storage for all my replacement parts that might need to be soldered in place. The battery-charging station is now dedicated to things that might go into or out of my drone bag. Batteries come out of the bag when I get home. Before I leave, I can refresh my props and fill my bag with freshly charged batteries.
This is going to be my first CNC woodworking project. It will sit atop my IKEA Hemler. I'm going to put the Bat Safe on drawer slides, and my big ThinkTank drone bag will rest on top.— Pat Regan (@patsheadcom) September 25, 2019
This will make it easy to swap batteries between my bag, the charger, and Hemler. pic.twitter.com/sqQ3zfSAuj
I started designing a sort of rack and drawer system to mount on top of the IKEA Hemler. I’d like to be able to store my giant ThinkTank bag on top of this setup. I want to be able to open the bag, then shuffle batteries between the charger, drawer, and bag as necessary while keeping everything stacked in one place.
It’ll be nice to have all my important drone storage fitting within about 12” of wall space. It will also be nice when I can store, charge, and then pack my batteries in my ThinkTank backpack without having to hunch over!
I’ve accomplished more this month than I expected, but I’m finding myself disappointed that I haven’t touch my network engineer’s toolkit project in more than a month. I might have to see about moving that project closer to the front of the list for December.
I’m really late getting this post published this month. I’m already into the weekend, and when the weekend is over, it will be December. I expect December will be a busy month, so who knows what I’ll get accomplished? Maybe a lot!
Am I doing a good job learning to podcast? Did I upgrade to the right camera? Am I doing a good job organizing my office? Am I doing it all wrong?! Let me know in the comments, or stop by the Butter, What Discord server to chat with me about it!