Everything posted by OJTheAviator
I'm starting to learn Mandarin Chinese and just found a great joke.
First five numbers in Chinese:
(Note that the first three are lines, with 1 line for one, 2 for two, and 3 for three. It seems like it could continue, but the later numbers are unrelated.)
One day, a little kid was learning to write. He learned the first three numbers and thought "That was easy, I know how to write any number now!"
Soon after this, his dad wanted him to write down some numbers and said to the kid "Please write down ten thousand."
The kid was there for a while.
A couple years back, I got some RC parts and a couple EDFs to build a hovercraft. Didn't work, so put them in a box and left them. Until Now. Turns out, my ESCs didn't power the receiver. So, I've ordered two that actually will, plus they have more margin in terms of amperage. I've figured out a way to test the motors though, before the new ESCs get here. They work wonderfully and terrifyingly. I might try making a "jet" RC skateboard that is entirely controlled by two EDFs using differential thrust. Imagine seeing a skateboard holding nothing but a pile of batteries and wires speeding down the street at what looks like 50 MPH! (I won't do it in the road. Why don't we do it in the road? Because the thing could explode, or hit some cars, or be hit by cars.)
Maybe I'll just wait and do the hovercraft. Also, I feel like a big tail section with a tail mounted camera would look pretty great and produce a great view. Already have an action-cam that would fit the bill.
I'll probably post my results when whatever happens happens.
Man, it's hard to design 3d printed attachments to servos. Trying to make a mini robotic arm. Went the route of adapting an existing servo arm, but since the arm isn't exactly straight, the thing wobbles noticeably. Mk.2 coming up.
Servo motors (for example these 9g servos I've purchased) must have some angular acceleration, but that data doesn't seem to be listed. I'm hoping to make a little robotic arm out of 2 or 3 of them, but I'm doing some quick calculations to see if the thing can even work. The max torque listed is about 18 N/m, which would be plenty to hold the arm in any position (the torque exerted by gravity when horizontal is about .02 N/m). But, a static robotic arm is a little bit useless. So, I want to see if the servo could actually move the arm. The moment of inertia would be somewhere between 5.8E-5 kg m^2 and 28E-5 kg m^2, and if I just had the angular acceleration, the calculation would be simple to see if the little servo has enough torque. ?τ = I α
Now that I've written all this, I suppose the angular acceleration doesn't have to be constant. Maybe the thing outputs as much torque as possible, and the angular acceleration comes out of that constant torque and moment of inertia.
Maybe I'll just build something and see if it works.
Yeah, it would be A. If you get yourself into the shit with grey covering your whole windscreen and/or experience spatial disorientation, you should immediately start your scan of the 6-pack (or if you have EFIS, a scan of your EFIS display) as the only thing you can trust is your instruments. Spatial disorientation is the clue as in IMC, you have nothing what-so-ever to judge your attitude and as every single movement shifts the fluid in your inner ear, you'll get disoriented very quickly. This means that apart from your instruments, there isn't really anything else you can trust.
A good way to see the effects is by spinning blindfolded and see if you can stand straight (whilst still blindfolded). You will find it extremely difficult
I've never actually flown in IMC, but I do have an interesting simulator story. When I went to Oshkosh a few years ago, I did a lesson at their Pilot Proficiency Center (Redbird simulations sends up quite a few simulators and volunteer instructors send pilots through simulations). I choose a lesson on stick and rudder skills, and the instructor sent me through a number of challenges, everything from a crosswind landing, to a low altitude engine failure, to a plane on the runway on short final, to finally a carrier landing in a Cessna. One of the challenges he gave me was a night flight. Not quite the same as IMC, but the sim was set up to be nearly pitch black, only the airport lights visible. At first, since I was a student pilot who hadn't flown at night, he waited to give me this situation, but he decided to give it a go. I used the instruments the whole time, hardly looking at the airport until final, and ended up flying a pretty good pattern.
I suppose I can't say that it's silly to try to use sensations in IMC, since I haven't really been in that situation, but having learned everything I have in flight training, I'd definitely be trusting the instruments the whole time.
Oh, as an IR(R) holder (UK-only IR-lite, basically full-IR privileges with one major difference, min vis for take-offs/landings is 1500m), you'll quickly find that your body is telling you to do one thing but your eyes (i.e. instruments) are telling you to do the opposite. It is something you get used to but still, your instincts will always want you to trust your senses, which is something you should not do
Anyway, I personally found IR training to be quite easy, apart from NDB/DME approaches using an ADF or doing holds over a NDB with an ADF. However, once we switched into an aircraft with a RMI (automatically aligns the ADF with your magnetic heading), it's no harder than using a VOR so it became quite easy.
My new favorite news source: AP news, Oddities section. Following are some highlights.
Animal shelter says "world's worst cat" is up for adoption
“We thought she was sick. Turns out she’s just a jerk.”
A tongue-in-cheek profile of the foul-tempered feline says her dislikes include “dogs, children, the Dixie Chicks, Disney movies, Christmas and last but NOT least ... HUGS.” It says she likes lurking, pretending to be sick and “staring into your soul until you feel as if you may never be cheerful again ...”
"She's single and ready to be socially awkward with a socially awkward human who understands personal space."
“I’m looking at her right now, and she’s rolling around in her little bed, looking all sweet and cute, but the minute you try to rub her, she slaps you. We thought she was in pain and took her to the vet and he said: ‘No, this cat is just a jerk’,” Lowery told the newspaper.
One person wrote on the shelter's Facebook page that they hoped Perdita finds “the goth home of her dreams.” The shelter replied, “Us too ... like soon.”
Cops: Waze app directs casino-bound drivers into wilderness
2 elephants escape from circus in Russian city
I wonder if it's possible to make a computer inside a ti-84 plus? Of course, I don't mean a regular PC, but I know the Raspberry Pi Compute Module would physically fit inside. Maybe it's possible to design a custom PCB that would fit inside a ti-84 and incorporate the Pi, use the original keyboard (at least appear to externally, maybe a new push buttons inside), and add a color LCD where the old B/W display was? Also would need a battery able to supply enough power. This would be an awesome project, but I know nothing about PCB design, and not a lot about circuit design, so it would require a lot of time invested. Also would require a broken ti-84.
I hacked an Echo Dot to be more private!
Well, hack is a strong phrase, it was all physical with soldering, no software or firmware changes. And, if Amazon still listens while the mic is "muted", I'm no more private than before. Also, any phrases I say to it will still be treated the same.
My issue was that the mic button is annoying to use from across the room. The whole point of a smart speaker is it only needs talked to. No need or reason to walk over. So, having to get up and unmute to change the music is annoying. I'd rather not have a giant corporation constantly listening though. So, muting was preferable to not muting, but again, that defeats much of the utility of a smart speaker.
What if I could unmute from my desk, while the Echo is on a stand across the room?
I decided to take it apart because I wanted see how it worked, especially to see if the button physically disconnects the mics and if there could be inserted physical disconnect switches. After deciding that the button is likely disconnecting via software/firmware, and that disabling the mics would be more trouble than it's worth, I thought I might try to add another mic button that could be extended to anywhere in the room.
The top board in the Echo seems to contain most of the logic, and also has the buttons for actions like volume and mute. The issue is that the buttons are SMD. I do have a hot air reflow station, but I wanted to preserve the original button in place, not remove it and move elsewhere. So, I used a multimeter to discover one part of the button goes to ground, and the other connects to one of two tiny dots of solder on the surface of the board, right by the button. So, after many minutes of attempting to solder a dot smaller than a pencil tip and two strands of copper wire twisted together, using the tiniest tip I have, the connection finally stuck. Then, I led the uninsulated wire across the board, insulating with Kapton tape as I went. Switching to insulating wire, I directed it up the side of the chassis (which is later covered by the cloth and plastic outer shell). Nearly done, I reattached a ribbon cable from the main board to the power board, and plugged the Echo back in. It didn't work. Turns out, I had connected the ribbon cable incorrectly. Finally, it worked! Finished up by adding a sort of port to the outside of the echo for button use.
Note the lighted mute button:
-WIKIBEAR: Hi guys, I'm Wikibeaar!
-MYSELF: Hi Wikibear! Why don't you tell the kids about Winnie-the-Pooh?
-WIKIBEAR: Winnie-the-Pooh, also called Pooh Bear and Pooh, is a fictional anthropomorphic teddy bear created by English author A. A. Milne.
Do you know another English thing? Canada! And Elon Musk holds Canadian citizenship and is polluting space right now, hooray! Today at around 9 EST, Musk plans to launch 60 satellites at once, greatly increasing the number of objects in Earth orbit. If this number grows too large, we won't be able to get to space again! Yayyyyy! Hooray for bad foresight and sustainability!!
-MYSELF: Wikibear, is that really what kids should hear?
-WIKIBEAR: You can't silence the truth! Yay!
[Though, glad to hear they are attempting a fairing recovery. I actually don't hate SpaceX, but I felt like writing a Wikibear script.]
Just tuned into the Times Square livestream. Imagine by John Lennon was incredible to hear just as a new year and decade begins. I always have in the back of my mind at these moments how these decades, years, even days, hours, minutes, and seconds are simply constructs. But, what do we have that we didn't make up, that we didn't imagine into existence? Even the arbitrary nature of New Years can remind me of the power we have to change, to imagine a new decade, and then to act on that notion. I'll leave this song here. I don't know of any better for this moment.
Happy new year everyone!
Well, I'm trying to do some stop motion. Unfortunately, my DSLR doesn't support live view, and seems to not be able to be controlled by Windows 10 at all. So, I've taped a webcam to the viewfinder, also taking pictures with DSLR, that way I can see what I'm doing, and also have it in not-potato quality.
This is rather pleasing:
I couldn't tell if you meant "for" to mean "the use of" or "the origin of". Schnoz answered the first one.
As for how I got the reputation: not quite sure myself. I suppose I just follow a bunch of people and many of them follow me, so my occasional but steady status updates get reactions. And, ever so often, I venture to the open forum, instead of staying in the rivers of status updates. So, it's been slow but rather steady growth to 1000.
- Show next comments 12 more
No more satellite images today. This is because I've learned a very useful fact from trying to get images last night: don't bother standing out in the rain and cold just to get static-y satellite images, it's not worth it. At all. I only seem to get good images if I can fiddle with the antenna, and have the laptop on the ground next to me. Laptops don't agree with puddles, and antenna fiddling doesn't agree with laptop holding. Also, radio dongles and displays don't agree with rain. With all this disagreement and wetness, and the inability to get images free from static, I'd rather do something else. Such as sitting inside and playing a game, or watching James May create ridiculous sized train sets, or composing needlessly long and wordy status updates on a particular tech forum.
@Levent was asking how I've been getting the satellite photos, and my reply for the status update was getting both too long and maybe too interesting to keep as a reply, so I'm making a new update.
Quick and simple explanation of the concept:
There are three NOAA satellites (NOAA 15, 18, and 19) that broadcast images down using a radio. These three aren't the only satellites that do this, but they are the easiest to receive. Using a radio that hooks up to a computer through USB (a Software Defined Radio), and an antenna, I can tune to any frequency, including the one that the satellites use. I can use a program to turn the radio signals into audio, and then use another program to turn that audio data into an image. Honestly not certain how the data is encoded by the satellite or decoded by the second program, but that's roughly how it works.
More detailed explanation of how I did it:
I forgot that The Thought Emporium was the place I first saw this, but after watching that video again, I'm pretty sure they introduced me to the concept. I essentially followed this guide from RTL-SDR.com, though with a few changes. I used the Nooelec NESDR SMArt SDR. It was recommended somewhere (I forget where) and the price is pretty good. I think it's comparable in price and quality to something like the RTL-SDR.
I decided to go with a V-Dipole antenna tuned to about 137.5 MHz. I copied this design almost exactly:Spoiler
I couldn't find 1/8" aluminum rods in the local home improvement store, so I went with 1/4" and it's been working great. The "Choc block" terminals were also a little tricky to find, but I found a set under the name "Lugs". Turns out, they are very similar to a terminal ground bar, except a terminal ground bar is like lots of lugs welded together. I probably could have cut up a terminal ground bar to make it work. I figured the antenna should be somewhat elevated, so I used some assorted bits of metal and hardware lying around to fasten it to the top of a 10' wooden pole. Chopped the pole into three roughly equal sections to store it more easily, and used tall corner brackets to make a sort of friction socket on the ends of the rods
The guide suggested using the program SDR#, but I had already used a program called CubicSDR and was a little familiar with it. Plus, CubicSDR is available on both Windows and Linux, and while my desktop runs Windows, my laptop runs Ubuntu Linux. All SDR# would need to do is to turn the radio data into audio, which CubicSDR also does. In fact, I probably could have used any SDR software. So, I went with CubicSDR. If I were to do this from the beginning in Ubuntu, I might go with Gqrx SDR, which is in the Ubuntu app store and is easier to install.
The only program I know of that decodes the audio data into an image is WXtoImg. Unfortunately, the dev seems to have abandoned the project, and the original website is no more. Thankfully, https://wxtoimgrestored.xyz/ exists, so I was able to get the program after all.
I also needed a way to run the audio from CubicSDR to WXtoImg. While you could run a physical cable from your mic jack to headphone jack, I didn't have both on my laptop, and that's not an elegant solution whatsoever. Windows was simple. A program called VB-Cable installs drivers so that you can set your audio output as the input for another program. Linux, not so simple. There seems to be a handful of options, of varying levels of complexity, none of which being particularly intuitive to me. Ended up using PulseAudio, which is built into Linux (as far as I know). In order to get PulseAudio working, I had to run in the command line
pacmd load-module module-null-sink sink_name=Virtual_Sink sink_properties=device.description=Virtual_Sink
which sets things up, then I set the audio output to "Virtual_Sink". At this point, every sound that would be coming out the speaker is now going into the PulseAudio input. Then, I set the "pulse" output as my input for WXtoImg. The first couple times I tried getting data from the satellites, I set something up incorrectly with the audio, and just got a black screen instead of an image. Turns out, WXtoImg has a helpful audio level display at the bottom right corner, so once I got the volume between 50% and 90%, things started to work.
My first image was rather fuzzy with noise, but that was because I was testing to see what position works best for the antenna. Still testing, but I think it's good as long as the antenna is horizontal (like in the design picture), and the antenna is rotated to point at a certain angle to RF noise sources. Not sure what that angle is, but I've found that there is just more noise when the antenna is pointed in certain directions, and I just rotate it until the noise goes down. Also, seems like my laptop emits a non-negligible amount of RF noise, so it seems to help when I use the SDR on a USB extension cord, and keep my computer a few feet from the antenna.
Now I'd like to get more images, but the NOAA 15, 18, and 19 satellites pass by mostly in the early to mid morning and late afternoon, getting into the night. Last couple attempts at night haven't worked out; they've just been giving me static. I'd like to try again to see if it was my setup, since there should be some UV data at the very least.
Congrats to anyone who read through to the end, this has been a longer post than I expected! I've been really enjoying this, and it's been fun sharing images and methods. I'll give one or two more images:
I burnt my RTLSDR trying to capture GSM packets (due my shitty antenna shorting out), I kinda hoped to achieve something like this but with ADSB packets because IIRC NOAA satellites are in geostationary orbit over NA and only satellites over middle east spy and shitty tv satellites. Its really cool to see you made it work, Justin's video kinda got me worried that I was over my head (he had a lot more problems due to his antenna design).
Only if Santa was real and got me a HackRF One so that I would find some willpower to make my own legit antenna. Oh well.
I've heard that even TV "bunny ears" antenna can work for NOAA satellite reception. Might be worth a try. Also, I'm pretty sure NOAA 15, NOAA 18, and NOAA 19 are in polar orbits, so they should pass over the middle east at some point. Still very new to this though.
Maybe I'll try to recieve some Russian images, I think they can be higher quality.
Satellite thermal data! More images below.
Ignore any big differences that look like bands, that's just static. The first big static line was because of my computer being too close to the antenna, the second is still a mystery to me.
Other fun images...
Sea surface temperature: