/diy/ - Do It Yourself

>>2381571
I HATE THE ANTICHRIST.

https://www.youtube.com/watch?v=wB55Gq0X-rQ

>>2382173

has anyone experience with IOMs or other RC sailing yachts?
https://www.iomclass.org/
I've looked into this a little bit and while it's niche, a good chunk of those involved are DIYing their boats.
(especially considering that a fully set up boat goes for a couple thousand bucks)

you can get hull plans for 10-30 bucks relatively easy, but then you'd need to make a styrofoam model and finally make the hull out of fiberglass or carbon yourself.
things like the bulb or rudder aren't typically done yourself (40 for the rudder, 60 for the bulb and 50-200 for the keel... god knows why it's so expensive)
the mast, sails and rigging stuff is... idk, this ranges all over the place
and lastly special equipment like the sail winch (which is basically a motor, a gearbox and a disk with 2 grooves) goes for like 250 bucks

I get that a lot of those things are small production runs and are therefor expensive, but most of them could be easily DIYed...
or at least they look like from my limited knowledge.

so any info on where to actually get some proper info
and what actually CAN be DIYed would be nice

Thoughts on LEDS for drones? I like have green/red on the front/rear to help with orientation when flying non-fpv, but what else can they be used for (practical purposes)?

>>2382807
Practical, no. Fun, potentially.

So what are my options for replacing a 540 motor with brushless while using the same mounting holes?
I know a 3650 would fit but that's almost too much motor for a boat that is just 16in long, wondering what smaller brushless would mount without modification

>>2382807
if you're chasing another drone or plane it makes it easier to see.

>>2383198
Use an inrunner motor with 540 size and you can use the same cooler tubing

>>2382173
More reason not to use DJI I guess, makes me wonder what information they get from non-DJI since they kept oddly quiet about that.

>>2383383
Hating on non- DJI doesn’t get video clicks.

Is there any way to add a barometer through a UART port?

>>2383565
Could be that, it's also possible that other drones don't transmit data that is so easily traceable.

>>2383607
Does the barometer have a UART port? (it doesn't)
Are the UART pins GPIOs capable of being enabled as the baro's interface?
ngmi

>>2383608
Mavlink is unencrypted but there are ways to encrypt the data. I think RFD900 telemetry radios should be able to encrypt data.

>>2383619
RFD900 are just generic radio modems, they don't have any in-built support for encryption. But you can ofc transmit whatever you want over them - if you encrypt the MAVLink stream on the FC/companion computer then it will be encrypted over the radio.

>>2381571
Before I dig into the fucking source code and attempt to brute force it

Is there any way to control a Taranis TX from a fucking PC? I want to make an FPV car I control via a steering wheel.

The backpack modules use SPI and should easily be controlled with a raspberry pi but can't seem to find anyone working on this.

my car has an unsealed rear diff, if I filled it up with, say, 100k, would it leak or is that so tick it'd likely stay in there?

>>2381571
Everyone talks shit about that camera being used but that is too good of a camera for that shit

>>2383864
It’s an old DSLR, it literally costs less than a fucking Mavic. What’s the problem?

>>2383755
They have a section in the manual entitled
'3.1 Setting up data encryption'

http://files.rfdesign.com.au/Files/documents/RFD900x%20Asynchronous%20User%20Manual%20V1.0.pdf

>>2383911
Neat. I wonder how many people just copy-paste the example encryption key from the manual...

Can I get a QRD on wtf is going on

I have 3 Caddx Vistas, 2 x DJI Goggles and 2 x DJI Transmitters.

Is all my equipment now shit and antiquated?

>>2384243
At most it will be slightly better, but in reality it will be worse. I don’t know what you’re worried about, huge numbers of us still use analogue.

>>2384243
https://www.youtube.com/watch?v=NK_SfyNsLdE

>>2384336
analogue is for faggots

>>2384352
>Joshua Bardwell

Found your problem.

How long will 18650s last in storage? I want to buy in bulk to get cheaper prices but won't need them all right away

>>2385185
This sort of information will be in the datasheet for whatever particular cell you’re looking at.

>>2385228
Forgot picrelated example

>>2385230
Isn't that the remaining charge after storing fully charged?
I'm interested in what permanent loss of capacity there might be when stored at storage voltage for a long time.

>>2385393
No, it’s telling you the temperature ranges at which you can store the cells (at storage voltage) if you still want to be able to charge & discharge them to at least 90%.

So if you store the cells (at storage voltage) for 18 months, they will still charge/discharge to at least 90% of their rated capacity - if the temperature during those 18 months was under 25 celcius.

In other words, you can store them for years without worrying.

>>2384395
Didn't the gen post his links before?

>>2386571
Hasn't for a long time, he's a balls deep shill now.

>>2385435
I see thanks

>>2386574
He wasn't before? Was there a line that was crossed or something?

>>2386605
It wasn’t so much a line, more the typical steady descent that so many people in his situation exhibit as they realise how much more money they can make if they start pandering to what gets clicks rather than what’s genuine.

>>2383873
No problem. If there is space and enough carry capacity old dslr is literally a perfect camera for a drone.

dad bought me a mavic 3 bros. I feel like I am failing him because I told him a bunch of bs about how I could profit from it. In reality I just wanted one

>>2383873
It should have been all made in Russia shit, instead you find an off the shelf camera.

>>2386605
it used to be a hobby then he became a fulltime youtuber which requires shilling to have a livable income.

>>2386616
Yeah, sounds about right considering the particular TX I ended up getting last year was a Jumper T-Lite since he and some others were recommending it. I think it's fine still, I still like the sticks, but I think they weren't exactly hard enough on the controller.

>2386790
Then get your 107, and make some money, retard.
Make your dad proud for once, so he doesn't feel like he just burnt 3k+ on another ""opportunity"" that went nowhere.

>>2386790
just get some nice footage and make him a movie.

>>2387094
>replying is very hard

>>2386790

Watch real estate promotional tour videos and practice filming and editing an exterior sequence. Look up a local real estate photographer with established clientele and offer your service as an "add on" for them.

If you're more entrepreneurial, you can call the realtors yourself. You should get good first so you can make a cut of your footage. It's not too hard once you get used to piloting the drone.

I used to do this for a few years prior to covid. I was a full time video editor for a local company that shut down their office in my area when shtf. I went on to do it independently and made $800 a day for a couple locations, but I also offered full package video and photo tours, but you don't need to do that to make some money.

>>2383764
Yes, the trainer connector on your radio is a serial interface, just get a serial to usb programmer, and send serial data to your radio , imitating trainer commands usually originating from the trainer tx. The software side is youts to figure out. A way to intercept joystick/wheel inputs and translate them to the serial interface commands.

>>2384243
Chink shit. If you want real open source digital fpv now, and on top cheap go OpenHD fpv . Pi3b+ or pi mini2 for air unit, a pi4 as ground unit, pi or usb cam with built in h.264 encoder, or even a usb thermal vision camera, alpha network usb3 500mW diversity network cards give you 10+ km digital long range fpv. It supports up and downlink, and can send even tx steering inputs to your flight controller.

>>2387293
OpenHD is an absolute joke compared to the DJI system. Rumours indicate this new system from Fatshark is just the same tech licensed from DJI.

There’s good reasons why the likes of FPV Blue closed up shop after DJI released their system - the open source community simply can’t come close, because they don’t have the resources to design/produce the custom silicon that is required for these systems to perform so well & shrink so small.

Analog? LOL. DJI isn't good enough for me. I need 1080p resolution and at least 60 fps (preferably higher). Latency up to 40ms is acceptable. How long do you guys think until this is possible?

>>2387518
Doesn't the current DJI system tick all these boxes?

>>2387536
V1 is 720p
V2 is 810p (wow what an upgrade)

>>2387542
Ah, cheers.

>>2387518
Just have two cameras, one for high quality recording and another low res for control

>>2387518
60 fps is garbage after you have flown 120 fps.

>>2387660
Millenial garbage brain detected

>>2387700
No, seriously. With DJI, when I switch from "low latency" 120 fps mode to "high quality" 60 fps mode I almost get motion sickness from the chop. Plus, a lot of the maneuvers is do take less than .1 second with 90-180 degrees of attitude difference, so that same maneuver gets 12 frames to cover it vs. 6 frames, which is pretty noticeable for me.

>>2387727
fly slower, faggot

Is it worth it to learn to fly 3D?
Is there any advantage other than sometimes flying inverted?

>>2387922
If you enjoy it, it's worth it. Otherwise it's not worth it.

>>2387727
Show us your flying.

>>2387966
>Show us your flying.
this general used to have excellent videos, back when we were young and reckless

>>2387293
Can I use OpenHD fpv to transmit video and ROS data from a pi4? Never heard of this before.

>>2387569
Cant an old GoPro do both simultaneously?

>>2384336
>huge numbers of us still use analogue

>>2381571
Is it worth buying the DJI FPV bundle for the goggles and transmitter, and selling the drone off? The goggles and controller aren't perma-linked to the quad, right?

>>2388339
no

https://twitter.com/RALee85/status/1526918354828673025

>>2388277
Yes

>>2388277
Yes.

>>2388636
Imagine being an engineer who worked on that realizing now that your creation is being used to take lives.

>>2388718
It isn’t really any different to cellphones or digital watches being used to trigger bombs.

>>2383198
>>2383305
I did just end up getting a 3660 motor which is the brushless equivalent to a 560 brushed, it was cheap at the time.
It's bigger than I wanted but it didn't seem to fuck with the balance which was what I was worried about. On the plus side the motor barely gets warm.

It doesn't rip as hard as I would have thought however, it's definitely faster than the other Feilun ft016s I have, need something larger than a 20ft round pool to play in.

>>2389270
Show full boat

>>2389430

>>2389575
Cool boat.
Would FPV on a gimbal work on that?

>>2389634
Same problem as on a car - when you're that low down, moving at any speed, you just can't see far enough ahead.

>>2388718
chang doesn't care

>>2389782
Yeah if you really think about it probably wouldn't work very well, on a small boat especially since ripples on the water can be taller than the boat and an FPV gimbal would be alot of relative weight on top.
On a larger boat with a catamaran type hull you might be able to mount a gimbal on a small mast as it would be relatively small weight

looking to get into FPV flying. for people that have been doing this for awhile, where would you suggest i start? i like the DJI drones, they seem to take great shots and have great battery life, but the company seems awful and i wouldnt be able to use the controller or goggles with any other drone.(i believe?) i would like to practice and think the controller should be the first thing i buy, but im confused as to what they will work with. i believe in buy once cry once, so id rather not buy junk.

i have my amateur radio license and enjoy tinkering and soldiering.

guess im just looking for where to start. budget would be around 500-1000$, could obviously go lower and a bit higher. thanks for reading my stupid questions

>>2389887
If you’re starting from nothing, you should buy the DJI goggles. The simple truth is that nothing even comes close to the DJI goggles & the only reason not to get them is that you already have a lot of analogue equipment.

Do not buy a DJI controller or a DJI drone though. Buy something like the TX16S & then either build your own drones with DJI video transmitters, or buy them ready-made from Diatone, inflight, etc.

>>2390066
>inflight

iFlight. Fucking autocorrect.

>>2390069
>>2390066
thanks for the answer. i guess im confused with how i figure out what to buy together. or what is compatible? so the dji goggles will work with other equipment? im missing something here.

>>2390080
Yes, if you take a look at their website it actually explains it fairly well.

https://www.dji.com/uk/fpv?site=brandsite&from=nav

You buy the goggles, then for each drone you buy a camera bundled with an ‘air unit’. These air units just connect to battery power & optionally to a data connection from the flight controller to enable OSD data.

>>2389887
Banggood have a good selection to buy from, particularly for parts and build-it-yourself kits. It does take a month or so to get shit in though.

You're going to need parts because it's 20% flying 80% fixing.

>>2390080
>so the dji goggles will work with other equipment?
Yes, you can flick through different frequencies on them.

>>2390080
FPV equipment is more or less separate from everything else. Your video transmitter must match your video receiver: analog to analog or DJI digital to DJI digital. Same thing for your controls. Your transmitter must be paired to the receiver in the drone, the important thing is the protocol that it communicates with. Radiomaster makes multi-protocol transmitters so they can pair with different receivers. TBS Crossfire transmitter (usually an add on module) will only pair with a Crossfire receiver, FrSky ACSST will only pair with an ACSST receiver, etc.
If you're in the US check GetFPV, ReadyMadeRC, or RaceDayQuads.

I want to build an underwater FPV drone. I've tried this before and ended up tearing it all apart due to design fuckups.

Keeping the Receiver & VTX above water. The cable I'm using as a tether is 100ft of ethernet cable. Everything else is submerged in watertight container.

Two big problems, maybe someone with a big brain maybe throw some ideas down?
1. Sending signal from the receiver above water down to the flight controller. Basically the signal isn't making it down to the FC.
2. sending video signal from camera back up through the tethered cable. It's "viewable", but a lot of interference.

If the whole downfall to this is the cable, what sort of tether would work?

>>2388277
Yes

>>2390421
How deep is it?
Perhaps put the RC transmitter/receiver and the VTx transmitter in a little buoy on the surface connected via a pole with shielded wires inside (that's if the purpose is to have a discrete vehicle that can film underwater). You didn't consider just a boat with a camera on a pole to film underwater?
Also, if its analog, you can put a thing called an LC Filter between your power supply and the camera as well as the transmitter.

>>2390421
>The cable I'm using as a tether is 100ft of ethernet cable.
Just read what you wrote properly. If it's 100ft.. I once tried to transmit i2c signal over a long distance only to realise there are limits to how far you can transmit various types of protocols.
I think PWM (if that's what its using, there are various methods eg SBUS) wont work over 100ft. My initial try would be to attempt to use RS485. So a converter (MAX485 maybe) would convert the signal from the RC receiver, transmit it over RS485, then another converter converts it back on the other end.
Another approach would be to find out what ROVs are using transmit signal and copy it.

>>2390421
searched some more, try this:
https://www.instructables.com/RS485-Serial-Communication-Between-Arduino-Mega-an/
remote control PWM to arduino to RS485 to another arduino to PWM to the flight controller.

>>2390471
>>2390483
>>2390483
I did consider a boat but it's too easy. I want to have something that can dive to 30m while I sit in the kayak and FPV submarine it.

> RS485
Thanks for this, I'll look into it. I was really hoping it might just require thicker gauge instead of turning it into a side project. It might be worth just cutting the tether in half and sticking to 50ft...

>>2390538
You're trying to send what is probably 3.3V logic along 100ft of conductor. Even with a thicker gauge conductor, the receiver simply doesn't put out enough juice.

I thought of an idea. A Polaris water plane but instead of a delta wing it has a variable wing like the F-14
Would it work?

>>2391156
Scratch that, stupid idea. I would need a variable step on the bottom of the fuselage as well.

>>2390538
>I want to have something that can dive to 30m while I sit in the kayak and FPV submarine it.
why do you need a radio controller? Why not a tether?

Let say I want to build some uav that will be able to fly 50 km in good weather and carry 2kg of payload.
What knowledge do I need to have to do this? Do I need to have aerospace engineering degree?

>>2391777
>Do I need to have aerospace engineering degree?
No you don't, most things can be learned on youtube. As for how hard it is, it depends whether or not you are willing to buy a kit or even a finished plane which you can modify.
It sounds like you have a specific need rather than a desire to dable in the field..

>>2391777
>2kg of payload
You'd be surprised how big you'd need your drone to be for this.

>>2391497
>Why not a tether?
I have a tether. It's communicating to the flight controller under the water that's the problem. The signal simply won't travel through the tether.

>>2382489
I have much experience with full size sailboats and limited with RC. The issue is making something class legal, it is very easy to get competitive advantage with small changes. If you are just doing it for self then go for it

>>2391930
So use an amp. Run some duplex wire to provide power, too. That way you can just have a big lead acid battery topside.

>>2391944
Apparently CAT5 cables can only take 550MHz or something close to that. So not sure if that will work. I'll have to measure the video analog output.

No power is being transmitted through the cable, only video signal. The power is underwater with the VTX/camera.

>>2387966
Alright, let me fire up Gyroflow, I'm not posting repeats here.

>>2387966
Here you go, numbnuts, enjoy! https://youtu.be/eOUuZQB_rTk

>>2387966
Just kidding, here's a better one: https://youtu.be/iCKTBeoDQEs

>>2392049
>that video compression
i can imagine the day when lossy digital compression algorithms are replaced with lossy neural network compression algorithms, could potentially make digital fpv significantly more bandwidth efficient than it is now while also reducing latency, blowing analog out of the water
except when you zoom into the video instead of being pixellated it's like those dogs covered in eyes
imagine flying fpv for hours and then the forest starts beckoning to you
the audio could be neurally compressed too, and the whining turbines start speaking in tongues

>>2392068
That day is sooner than you think though, because YT is still encoding the HD version of that video. And it will have music.
Only the freshest material for you guys!

>>2392049
gay music didn't watch

>>2392049
Where was this filmed? Reminds me of NM or northern Arizona.

Is there a good way to join two batteries lengthwise instead of side by side? I am familiar with battery welding in packs but want a long battery instead of a fat battery.

>>2392381
You weld it in a pack then bend it into long boi. Don't weld it so you can't bend it the long way.

>>2392248
Central coast of Commiefornia.

>>2392225
Just put your wigger music or whatever on in another window. It's not rocket science, zoomer.

>>2392399
Thank you. I was considering welding tabs to each end and then having tbr tabs stick out the end and welding those together, then heat shrinking thr exposed tabs. I like your idea.

>>2392414
>Central coast
Weird. I asked this question and I live in Santa Barbara County. Where on the central coast? I'm very familiar with the coast all the way up to San Simeon and this looks like way more pine trees than most places I've been.

Underwater ROV guys signing in >>2390421

>>2391944
>>2391497
>>2390483

Turns out I was using
> 200ft of tether
> CAT5E cable

Chopped the cable in half (down to 100ft now) and got the camera working. Will try for the motor controller in the next day or two.

>>2392479
I'll give you a hint: we have a unique species of pine tree here that is different from all other pines. Now stop trying to get me to dox myself!

>>2392758
Oh. Starts with an M.

is there seriously no other decent camera drones in the market other than djews?

>>2393317
Not really, no.

It's a bit like tablets. You can buy a Surface or an Android tablet, but at the end of the day an iPad is simply better - regardless of what you think of the company.

>>2393359
How is an iPad better than a Surface?

>>2393361
In every single way?

>>2393361
The problem with all of the Surface devices is that Windows simply isn’t that good at being a tablet OS. Obviously they’re far superior if you need to run x86 desktop apps, because the iPad can’t do that at all. However as a tablet & for anything involving pen input, the iPad is light years ahead.

FYI I’ve owned both & consider myself fairly anti Apple.

>>2393317
>is there seriously no other decent camera drones in the market other than djews?
I don't understand why not.. Could you not start off producing high end military drones, use the massive profits from defense contracts to invest in the necessary tooling to start mass producing consumer drones?

>>2393387
DJI already have complete dominance in a saturated market.

It’s no more attractive for a military drone manufacturer to try to compete with DJI in the consumer world than it is for a military communications manufacturer to try to compete with the iPhone.

>>2393429
>than it is for a military communications manufacturer to try to compete with the iPhone.
Huawei was founded by a former PLA engineer with Chinese government money in order to get rid of Chinese dependence on the west for its telecommunications.

>>2393770
This is not a good example because this is how the PLA grift works for the higher ups

>>2393387
>producing high end military drones,
>use the massive profits from defense contracts
While profits can be big they are nowhere near massive compared to easily accesible civilian products and profits made from those products.
There are several military drone makers and they hardly make profit because there are few militaries willing to spend money on drones (might change after Ukraine war), and even if they do buy them its not like every soldier needs to have one like some bullerproof vest or a rifle

>dedicated RC gasoline engines start at 300$
>big li-po battery packs can easily get to 100$ without the charger and they're a fire hazard while charging too
I didn't realize that powering a long range plane can get so expensive.
How retarded would it be to use a generic chainsaw/weed whacker 2 stroke engine to power a plane? Surely it can't be much harder than using a dedicated RC engine, right?
I'd still need a battery with decent current to start it but it doesn't really have to be much more than 1000mAh capacity as once it's running I can charge from it.

>>2394470
anon, if you don't have any experience with RC don't try to do some long range monstrosity, please. What fuckhuge batteries are you looking at that are $100? They're not really a fire hazard unless they're damaged, people have been using them for decades now. The weedwhacker motor plane thing has been done before.
https://youtu.be/mk293eyL3MM
https://youtu.be/A8Rdjo-GtGg

>>2394470
If you want to do long range, pay extra for reliable equipment, or you might end up losing it and possibly start a fire and/or kill someone. If you are looking into converting your weed whacker into an airplane, I'd fly it pretty close in.

>>2394600
>if you don't have any experience with RC don't try to do some long range monstrosity, please
I have some experience with electric 4 channel trainers, but that's it really. I have to do detailed aerial photography above a patch of forest and it would take me weeks if I relied on electric crap with 15 minutes of flight time.
>What fuckhuge batteries are you looking at that are $100?
6S, 5000 to 6000 mAh. Amazon price is around 70-80$ then I pay another 24% in VAT and 35% import tax because I live in a shithole.
>They're not really a fire hazard unless they're damaged, people have been using them for decades now
Despite this most people say you shouldn't leave them to charge unattended, especially indoors. That doesn't really put me at ease.

>>2394699
Long range is probably bad wording on my part. I need it to stay up in the air for as long as possible. A big battery would work for this but once it's dry I have to head home and charge it. I'd be limited to one flight per day. With a fuel tank I can land the plane, fill it with gasoline and send it back up.

>>2394711
Why do you want to go with 6S? Most planes fly off of 4S, that should lower the cost of the battery to the point where you can buy multiple batteries and switch them out. If your plane is light enough look at Li-Ion instead of LiPo. These will have a lot more capacity, but less amps.
https://youtu.be/Env9jD8FTZI?t=113

>>2394814
Small and light is out of the question since I have to take a DSLR camera up to take good pictures of the foliage. I also have to make something to keep the camera stable in flight so that will add even more weight.

>>2394814
>>2394821
Pretty much any large DSLR mapping plane will use 6S lithium ion, as the higher voltage is simply more efficient. We’re talking about large paralleled packs here (6S4P-6S8P) so discharge isn’t an issue - you don’t need lithium polymer for large planes.

>>2394821
> I also have to make something to keep the camera stable in flight so that will add even more weight.

I missed this part. I hope you’re not talking about trying to mount a DSLR in a gimbal to a plane, because that’s simply not realistic. There are good reasons why mapping planes use hard mounted cameras & for anything requiring gimbal stabilization you just use a multirotor.

>>2394814
>Most planes fly off of 4S
Actually most normal size planes fly off 3s, 4s is only needed on sports planes 1400mm
anything bigger maybe 6s but usually they skip over to 12s like the 3m Yak 53 my flying buddy has.
5m wingspan is exclusively gas engine territory
I have a 3m glider that flies off of 3s.

>>2394859
>We’re talking about large paralleled packs here (6S4P-6S8P) so discharge isn’t an issue
Doesn't sound like it will be much cheaper than the li-pos I found.

>>2394864
>I hope you’re not talking about trying to mount a DSLR in a gimbal to a plane
I don't know for sure, this is something I will see once I have the plane flying. Ideally I'd just make a simple rigid mount and it will work. If that gets me good pics then I roll with that. If not I will have to stabilize the camera some way. Also this isn't mapping I'm doing, I have to get aerial pictures of the foliage in hopes of spotting any signs of diseases or pests on the leaves of the trees. The camera has to be stable so I can get clear shots.
And no, multirotor isn't economically feasible. We have to survey almost 10km^2 of woods and that's a 200km drive from where I live. If I was made of money I'd get some big off the shelf multirotor, get a hotel room, spend a few weeks there and be done with things. Also we're getting fuckall funding from the university and that has to be saved for the second part which is giving the labeled pics to some machine learning/neural network company to turn those into a program that can recognize the pests from aerial shots.

>>2394878
How exactly do you propose to get close enough/detailed enough photos of foliage using a plane? Even assuming a very high resolution sensor (40-50MP) you’re simply not going to be able to fly low enough. Even if you could fly low enough, the speed would screw you over even on a very fast shutter.

>>2394881
>you’re simply not going to be able to fly low enough
On the ground I can get pictures detailed enough to tell common diseases from almost 12m away. So I think 7-9 meters above the foliage should be good enough. So idk what you're talking about not being able to fly low enough, last I checked RC planes could fly even lower than 2m above ground. Surely tree height + 7m is doable.
>the speed would screw you over even on a very fast shutter
This is one of my biggest worries but I can only see how it goes once I have the thing running. If I can't get clear shots then I'll have to bite the bullet and build an electric multirotor. And that would mean sleeping in my car for god knows how long until I get pics of the entire area. I'd like to avoid this fate if possible.

>>2388277
Yes.

>>2392488
Any progress? Any deep underwater footage you can show us?

>>2394919
>On the ground I can get pictures detailed enough to tell common diseases from almost 12m away.

Zoomed in? Or wide angle? Because when you're flying you're going to be shooting fairly wide, by necessity.

>So I think 7-9 meters above the foliage should be good enough. So idk what you're talking about not being able to fly low enough, last I checked RC planes could fly even lower than 2m above ground. Surely tree height + 7m is doable.

How consistent is the height of the canopy? Because if you're covering large areas using automated missions, you need to fly comfortably above the level of the highest point.

>This is one of my biggest worries but I can only see how it goes once I have the thing running. If I can't get clear shots then I'll have to bite the bullet and build an electric multirotor.

Honestly, having done survey work myself, I just don't see a plane working for the task you're describing. While I don't know exactly what it is you're looking for, the fact you're talking about disease/pests on leaves implies you need a lot of detail.

If you fly low/close enough &/or shoot with a long enough lens to get that level of detail, you're just going to get blurs.

If you fly high enough &/or shoot with a wide enough lens to get sharp pictures, there just isn't going to be the detail.

Something like a used Mavic 2 Pro with as many batteries as you can afford is probably the best solution. You have a lot of ground to cover, what you want is a turnkey solution that just works.

Building your own aircraft to do this will be a huge undertaking & likely won't even work in the end. With a Mavic you would be up & running after an hour of learning DroneDeploy/Pix4DMapper.

>>2394929
>Zoomed in? Or wide angle?
Zoomed in a bit.
>How consistent is the height of the canopy?
I do not know, I've yet to visit the location in person. From what I could see on google maps I'd say it's around 2-3 meters of height difference at worst.
>Something like a used Mavic 2 Pro with as many batteries as you can afford is probably the best solution
The problem is I'm not made of money. I doubt something like that would go below 700$ used.
My budget is 400$. Maybe 500$ if I'm really pushing it. I can't spend more than that. This is why I had the idea of building my own. I already have a decent camera and the 2 stroke engine so I could save money if I could make use of those.
Maybe I could try to nig rig that engine into a helicopter instead of a plane.

>>2394937
>My budget is 400$. Maybe 500$ if I'm really pushing it.

Sorry to be the bearer of bad news, but there's simply no way you can pull this off for that sort of money. Even $4000 would probably be a fair challenge to capture 10km2 of canopy with sufficient detail for ML to detect details on leaves.

Even if you somehow managed to make a junkyard weed whacker engine work in place of an electric setup to save money on batteries, that's just one expense & not even the biggest.

If you only have $400 to work with you're best option is probably to hire a Mavic for a day & cover as much area as you can as a proof of concept.

>>2394926
>Any progress? Any deep underwater footage you can show us?
I burnt out the ESC so waiting new parts. Life is keeping me busy so it's going to take a few weeks to get this shit together.

>>2394470
Go thin film PV instead of large battery. Might want a bit of CFRP if you're going for really wide wingspan for rigidity, even just a spar or three reinforcing a standard hot-wire-foam-cut wing profile would probably be enough, without any fibreglass. That kind of flier is dead slow though, and rather subject to wind conditions.

>tfw the only RC I have is a modded Xmaxx

>>2394878
One thing you should know about a wing vs. a multirotor drone, is that the wing will get blown around by gusts of wind quite a bit more, whereas the multirotor will give you a more stable camera platform. Just something I noticed after flying both, and it is pretty noticeable.
Also, you will get in the air much more easily if you can use a GoPro for the type of photography you are talking about. Not including your transmitter and FPV equipment, It is the difference between a $200-$500, 5"-7" quad vs. $2k-$5k+ for a 10" octocopter. Plus the fact that you are putting a couple hundred dollars worth of camera into the air (that is built to crash) instead of a pricy, heavy, and delicate DSLR.
Or, just go with the best DJI SoiDrone that is in your budget for an off the shelf solution.

>>2395284
That is exactly what someone who was bullied to keep his mouth shut about what's down there would say. Someone who's smart, and knows what's good for him.

>>2395497
After you learned flying the most basic RC plane (Easy Star) driving an RC car will be unbearably boring. Also for the price of that Xmaxx you can have a fully kitted out hangar with all kinds of planes and stuff.

>>2395535
Not that anon, but for me the big benefit with cars is I can play with them when it's raining, blowing 50mph winds, etc. Can't do that with my planes or multirotors.

>>2381571
I got an Xmod years ago for christmas
It was a yellow skyline

>>2395541
>I can play with them when it's raining, blowing 50mph winds, etc. Can't do that with my planes or multirotors.
Sure you can. https://www.mgchemicals.com/products/conformal-coating/silicone-conformal-coatings/conformal-coating-waterproof/
https://youtu.be/wfZ-OnMmnMw

>>2395554
Completely missing the point. Regardless of waterproofing, it's just no fun trying to fly in 50mph winds - especially with a plane.

>>2395555
>it's just no fun trying to fly in 50mph winds - especially with a plane
Have you tried sloping with an F3B or F3J glider?

>>2395535
But where will I get big jumps in a plane? Fling dirt? Plane honestly sounds more boring. Can't race cars, can't fly it outside my front door, can't take it out on an adventure hike. So many limitations

>>2395723
When you climbed to 500m with a glider then came down whistling, finding the base of another thermal and climbing back up again, you won't have any need for jumps and flinging dirt

>>2395723
RC cars are cope for physicslets who can't into flight. Even boats are more interesting. Cars are as mundane as possible.

>>2395742
Still sounds boring af

>>2395821
>Even boats are more interesting
lol

>>2395824
Looks like a Funjet or 3D flying is more up your alley then

>>2395522
Don't rat me out just yet anon. I need to put the deadman switch in place first.

>>2395821
>Even boats are more interesting
Ok you have to draw the line somewhere, and you've drawn it too far. I may see some interest in building something like a hydroplaning vessel, but actually driving it? It's all either automatic or it's impossible to keep stable with hydroplanes so it doesn't make it any more fun to use. Maybe driving a sailboat could be interesting, for a short while.

Also any of you guys bolt BB-guns to your quads/fixedwings/boats/cars to buzz about shooting at one another? Sounds like it could be fun, if a bit expensive.

>>2396200
We regularly did dogfights at my old RC club, just simple depron deltas and we simply flew into each other. Airsoft mech is too heavy and you can't really aim it, not even in FPV, small planes just spazz around, hitting another plane is simply easier.

>>2396222
>Airsoft mech is too heavy
This depends on the parts you choose. If you go the AEG route then yes, the gearbox is heavy, the motor is heavy, the "low" capacity mag is heavy, the battery is heavy etc.
The innards of a CO2 gun will be much lighter and simpler to use.

>>2396243
Still too heavy. I did a bombing mod for my Easy Star with two modified darts and those (plus the two release servos) already add a lot of weight on the airframe, it is noticeably sluggish and much more difficult to control. A gun mods adds a lot more weight even if it is for two weak pews.
The airframes that are worth it to shoot them up are just not big enough, the ones that are big enough you don't want to shoot at.
Like I said, we went through every iteration imaginable, the easiest is to just ram each other with cheap depron deltas, the fun is the same.

>>2396200
>Also any of you guys bolt BB-guns to your quads/fixedwings/boats/cars to buzz about shooting at one another? Sounds like it could be fun, if a bit expensive.
I used to do Slope Combat where it's just destruction derby style with foam wings like that other anon said, but I've always wanted to try the Cut the Ribbon style of R/C combat too. I think it would even work with FPV drones.
There was also that laser tag system you could put on drones, but Rotor Riot did a video on it and then I knew it was retarded (like everything they do).

I read some story about a dude spending a ton of money making a big glider and then he put a Spektrum receiver in it.
now I'm fairly new to the hobby but isn't Spektrum kinda shit?

>>2396222
>>2396329
>laser tag system
How would those receivers work? Maybe cover the wings/body of the plane/quad/etc. with fluorescent paint and detect its emissions with a photodiode and optical band-pass filter? If you actually had to hit the receiver dead-on then that would be pretty difficult.

Not sure if quads or fixed-wing would be optimal. Quads have the whole "arbitrary camera angle" thing.

>>2396490
First DSM were very prone to brownouts, even a few 0.1V drop was enough for it to reboot, also it had a retarded rebinding method every boot where it needed you to turn off the radio, turn on the rx and turn on the radio for it to connect, basically rebinding every time you turned it on. A single brownout mean total loss of control.
DSM 2 improved it a bit but some newer rx are still prone to brownouts and while Spektrum implemented a new boot sequence it still takes too much time for you to be able to react.
At my field nobody trusts the system while on semi regular uses it. He regularly loses airframes over loss of control where the plane flies off in one direction, he calls it "loss of direction". It is a Spektrum DX9.
Rest of the crew uses Futaba T10, Hitec Aurora 9, and Futaba FF9 and random assortment of chinkshit TX with old FrSky modules. No brownouts and flyaways.
Spektrum has one good thing going for them and it is having satellite receivers which is a good thing for bigger aircraft and competition (carbon everything) gliders. Wish I had something similar for my Aurora.

>>2396815
>satellite receivers
What would this be?

>>2396828
Multiple physical receivers on the same model. They’re used to mitigate the issue of the airframe itself blocking line of sight between radio & receiver depending how you orient the model during flight.

>>2396815
> Spektrum has one good thing going for them and it is having satellite receivers which is a good thing for bigger aircraft and competition (carbon everything) gliders. Wish I had something similar for my Aurora.

FrSky has this too (‘redundancy receivers’), it’s not exclusively a Spektrum thing.

>>2396490
The other bad thing about Spektrum is that there’s a proven exploit that allows an attacker to hijack the air protocol & Spektrum hasn’t addressed this.

>>2396836
>Multiple physical receivers on the same model.
Ok, that's cool, redundancy is always welcome in flying machines.
Is it like a basic multiplexer based on signal level or something more elaborate?

>>2396848
All receivers have RSSI, actually all antennas have RSSI on dual receivers. If RSSI drops for a set amount of time on the active antenna it switches over to the other. Replace antenna with satellite receiver.

I have a Pavo30 3 inch cinewhoop and it's pretty awful. Shaky to fly, especially with a payload, terrible in any amount of wind, and sounds like a jet engine.
Are all cinewhoops shitty to fly or would I do better with a different frame?
I use this for slow and proximity filming.

>>2397871
I would look at your PID tune first, and try to get RPM filtering set up if your flight controller and ESC can do it.

>>2397871
Also this rate profile by UAVTech improved my flying immensely. It might take you a bit to get used to it from whatever you have now, but it works really well. Check https://theuavtech.com/presets/

>>2396329
>There was also that laser tag system you could put on drones, but Rotor Riot did a video on it and then I knew it was retarded
Some toy drones already come with such a system.

>>2397871
Ducted whoops are always going to be very loud & get buffeted around badly by wind, those are just inherent drawbacks of the big ducted design & the small (relative to the payload) props.

>>2397871
Whoops in general are shit. 5 inch is where it's at

I like drones and cars, too much of a brainlet to get into planes and copters.

>>2398865
Please stay with your drones and cars toys and let the big boys fly their planes, thank you.

>>2398865
drones are copters

Anyone ever use an FPV drone to fly bait out? I see people online doing it and invariably they're using normie DJI stuff. I have a bunch of FPV quad I've built and I'd like to try using them to fly bait out a few hundred yards into the ocean.

I was thinking of something as simple as a hook underneath. Just tilt the drone backwards to release the line.

>>2400338
that would work. Gliders sometimes use that instead of having a release servo.

>>2400334
No they are not

Underwater drone guy here.

I have the electrical components - motor and video signal working through 100ft of CAT5 cable. Electrically we are sound. I'm going to proceed with some of the structural build in the next two weeks.

>>2400535
Did you take care of the impedance matching thing, or are you just winging it? Ethernet to coax baluns are cheap.

>>2400636
Winging it. It seems to work with 100ft of cable with any signal siwtcherups

>>2400648
*without

Took my little quad out for the first time in like.... 2 years after crashing hard in 2019 (webm related....). I forgot how bad the default PID tuning on this controller was. The throttle is especially sluggish... It's gonna be a pain in the ass to get it running well...

>>2400667
PIDs have nothing to do with throttle. If the throttle is sluggish, that indicates a separate issue.

>>2382173
>cover drone in aluminum foil
>put direct LOS antenna on the bottom
>laugh at the CIA

>>2400802
>they just jam it anyway

Learn how radio works, anon.

>>2398869
...ok

>>2400667
I remember this webm. Still made me laugh hard, especially that ending after all that drunken flight

>>2400404
Quadcopters are literally twice the copter your little h*licopter can ever be

>>2400338
I might try doing this in angle mode (assuming you fly acro mode). It could make the hook dropping maneuver a lot easier because it would self level. Especially over open water.

>>2400404
>quadcopters aren't copters

>>2401085
Nope

>>2401085
>LiveLeak
>the chink actually makes it

I feel cheated.

>>2401337
Well you should pay more attention, the way I see it they cut right before it goes bad. After clearing the river he sets the van in forward and didn't bottom out the steering wheel for the turn. So unless that van has a unbelievable sharp turning radius he just clipped his right front wheel over the rivers edge or could have even tipped the whole van into the water give the engine's weight.

>>2401485
Do you seriously think that after the video cuts, the driver proceeded to just yolo straight forwards off the edge? Are you fucking stupid?

>>2401485
>he sets the van in forward
[Citation needed]
The way he plays with the shifter after the camera guy closes in makes it impossible to tell where the reverse is. It could be right - back or right - front but it could also be far left - front.
Point is he could be switching back in reverse instead of going forwards.
>didn't bottom out the steering wheel for the turn
You can see him do 3/4ths of a turn so I can assume the video cuts before he does the full 1.5 turns.

So I do air balance for work and part of my job is going around to all these fucking duct boxes in the ceiling and getting the info off of a sticker.

Sometimes these things are in INCREDIBLY annoying places, mainly because of shit like the tile under it being blocked.

Is there a fairly small (as small as physically possible) that I could reasonably zip around inside a ceiling and take pictures capable of making out numbers on a 6" x 4" sticker with numbers that are 1/8-1/4" large?

>>2401776
Realistically, no. Even with a caged setup & a lot of practice, you'd get it stuck every day.

>>2401485
>Trying to be clever this hard
Sad that you try so hard to look smart when you are so fucking retarded

>>2401776
You want to fly a drone in the ceiling? Lol, lmao even
If you really want something RC then your most realistic option would be a small RC car that you place on top of the ducts and then drive it to the labels. Still, this assumes a whole bunch of things like the top of the ducts being open space and the ducts themselves being rectangular, not round, etc.

>>2402023
Yah no sadly the ducts are flex most of the time so aren't really soild. Also they go up and down insanely.

Yah I honestly have no idea for a solution I just want to make my job easier and I'll pay for it you know.

Now the ones that are in open air in the "industrial" buildings a drone would be 100% doable and helpful though

>>2401776
Long endoscope? Long endoscope with pneumatic muscles to control it?

>>2401776
I vote for action camera mounted to an aluminum painter's extension pole. You'll probably need to mount a light on there too.

>>2402465
No ability to bend around and get where it needs to be

>>2381571
How hard would it be to take something like Pic related, and turn it into a little motorized cart I can have follow me? Not expecting it to be easy, but I have a budget of 800 bucks, and more then enough free time. The environment i plan on using it in is a hotel, again, not expecting it to be simple by any means, just wondering if someone could point me in the right direction.

>>2401776
You might be able to use a little micro tank track rovers that people make. Mount the camera on a servo so it can look up when you need to see up. They don't weigh much, so driving on the ceiling tile wouldn't be an issue.

>>2403414
you could but an aruco or april tag marker on your back and have it track you visually. 800 bucks is pushing it but my approach would be camera to raspberry pi running ros connected to a flight controller running ardurover controlling two motors. Going round corners might be an issue.

>>2403585
Do you have some good links describing the basic processes of connecting the controllers to the motors?

>>2403414
in a hotel? with customers? public? autonomous vehicle?
zero.
its a cool project, have fun, do not fuck yourself for life with some liability bullshit

>>2403605
https://ardupilot.org/rover/docs/rover-motor-and-servo-connections.html

You want to start off with a commonly used rc car model and only after that works, tackle the cart. You will have to learn robotics engineering, it should take months and months, just keep at it and unless you're dumb, it will eventually kinda work.

>>2403418
ive built this one to pull a string behind it when driving in my ceiling, when it reaches the opposite side where i want to pull a cable to, i remove the string from the tank and tie a cable to it, pulling it with the string through to the other side, i modified a roughly $15 micro tank with a new motor controller, mini RX using sbus and a micro drone fpv camera. https://www.youtube.com/watch?v=psQBiMXgT4c

>>2404053
CUTE!

https://youtu.be/lKDSt6SAeuM
Put your drones to use other than making 5 second webms for a dead general

Which osd can display altitude?
I have a microminim osd and Ive tried several different GUIs and firmware, none of them can display the current altitude from a Mavlink message.
I can get an arduino to read altitude using Mavlink so this must be possible.
I can compile and modify this firmware:
https://github.com/ShikOfTheRa/scarab-osd

I doesn't seem able to deal with altitude though.
Should I just switch to a different flight controller with integrated osd?

>>2404159
>flying during a FAA NOTAM
the second you take off the FAA will immediately confiscate all of Bardwell's quads and super duper ban FPV again.

>>2404053
GIBE

>>2404384
>living in a 3rd world authoritarian shithole
Shig

Are solar powered drones a meme?

>>2381571
I build and bash around quad opters for fun.
I've 3d printed an eclipson wing and I am excited to fly my first wing (I found a simulator so I don't die)

Very hype, should have painted the canopy before mounting the camera but owell.

>>2405965
Very nice.
Can you provide the 3D files? I think I'd like to try making one too.

>>2405909
A glider can work, multicopters nevah
>>2405965
That 3D printed?

>>2405965
Nice! Get a pair of folding props and hub for it

Got my Zohd Drift up and already want another plane. Thus drift isn't great in the wind and it's not very fast, so I'm wanting something better. I'm also still not particularly experienced so something stable that can reliably auto launch would be nice as well.
I was looking st the Skyhunter Racing but someone I flew with this weekend said the Mini and Micro Skyhunters are shit.

>>2406206
Yea 40 or so pieces

>>2406192
https://www.eclipson-airplanes.com/modela

I did the free one without landing gear, will probably buy their bush plane honestly, it prints like magic. I used the polymaker lw-pla instead of "true" lw-pla, it's still a lot lighter, but also more rigid, figured that would compensate adding an FPV camera and shit.

>>2406227
What would you recommend? This is a 1m wingspan, iunno weight, under 700g.

I got an APC 7x4 with a 42mm hub...but just doesn't feel like it's going to produce enough trust, this 9x4 feels "safer" for a first flight.

Coming from drones is wyld when you don't have to worry as much about this.

>>2406364
>Coming from drones is wyld when you don't have to worry as much about this.

Quite the contrary, I would say that prop choice makes far more of a difference to a multirotor than it does to a plane. You would never do a change as drastic as 7x4 to 9x4 on a multi.

Is it worth it? Seller said it crashed and the camera isn't stable needs adjusting, it flies.
I also have a work shop and 3 d printers so I can print and fit parts

>>2406961
No, not even remotely close to worth it.

I just got interested in making a RC car that can be 3D printed and adjustable so I can more or less swap shells that might have different lengths for wheels and such

Where are the good resources/channels for options and setups?

>>2406961
>buying pozzed and letter agency backdoored DJIs
ngmi

>>2390421
I've thought about this idea a lot.
I think you should build a base station buoy/boat with your wireless transmission link connected to your underwater exploration vehicle with a fiber optic fiber cable. You can get long fiber cables with, none, single, double, or triple parallel steel wire from aliexpress for a reasonable price. Then use some fiber to ethernet converters and connect them to something like a rhaspberry pi, jetson nano, or other mini computer that can interface with your video cameras.
So you either drag the bouy along with your undersea vehicle, or the bouy also has a winch and solar array to power the minisub over the cable, and to wind it back if it malfunctions. The buoy station could even serve as a satellite transceiver to starlink, and you could control and transmit over the internet an undersea vehicle an ocean away.
Otherwise, have a capable undeswater vehicle with a long fiber optic cable to a ground based tether. Just have a fiber optic cable connecting the vehicle's control system and camera system to a laptop.

>>2407167
Hey, some update

I've been piecing together all the pieces and this is the current build. I'll post a few more after this. Nothing is stuck together yet. The motors are still on order. I tested the tether/cable and it sends both video signal from the FPV camera to the VTX, and also signal from the RTX to the ESCs. The only trick is powering both the FPV camera and the RTX close to the source (aka don't transmit current through the CAT5E cable).

Theoretically, the above water box could be in a bouy with a few km range with this setup. I'll have figure out how to reel the tether/cable in so it's not floating on the surface, but that comes well after this is built. The problem with fibre optic is you can't transmit power through it. You need thick gauge and the $ starts adding in here.

>>2407409
The underwater box is glass with a plastic seal top. Will this work? Who the fuck knows. It has x3 6S lipo batteries, x3 ESCs for the motors. The FPV camera is taped to the front with a separate 6V power supply.

>>2407411
The overwater box has VTX and RTX. The FPV sends a signal to the VTX which I can see through my goggles, and the RTX sends signals down to the ESCs. Technically the plastic box is a waterproof diving box which could be shoved into a buoy.

>>2407414
Last photo here is what the future build is. The motors are on order. Two at the back for forward thrust and left/right motion. A propeller at the top will force the sub down. This means the top prop must always be on for the sub to be underwater, i.e. if the battery dies the sub will auto-float.

The underwater box will sit on a grate. I'll move and reposition the LED and Go-Pro on the grate as necessary.

Pool noodle will decorate the top pipes to create this buoyancy. I may yet drill holes into the lower pipes to get the buoyancy level and right. Weights can be added to the grate as necessary too.

That's it. Thoughts?

>>2407417
Helps when I include the photo

>>2407081
I would recommend you buy the chassis and then alter it as you need. Attached is my ground bastard RC car. The chassis and motor were bought, the rest is build (clearly). The chassis actually takes a lot of abuse as you drive it around, I don't think 3D components would take it well. You could easily print a shell and switch it in and out over it.

My own recommendation is that if you FPV it, elevate the camera (as in pic). The things are so low to the ground it makes driving it difficult.

>>2388277
Reddit moment

>>2381571
I wanna suck the dick of all engineers in itt

>>2405965
>How to absolutely guarantee that you will never find your plane after a crash

>>2407542
Are you a woman?

>>2407409
>>2407411
>>2407414
>>2407418
I love this monster, please keep us updated.

>>2395821
>RC cars are cope for physicslets
Not if the car has only 1-2 wheels, nerd

>>2407566
>/diy/
>woman

so am i gonna get fucked in the ass for putting a cellular transceiver on an rc plane or does nobody care?

>>2408932
Why would anybody care? There have been cellular controlled UAVs for many years. The drones at my new job are 4G+satcom.

>>2395821
I dunno , I had a shot of a friends 3s Typhon at a quarry launching it up cliffs and spraying rooster tails off the tops of sand dunes, I liked it so much I just ordered a 6s. Cars can be fun (if you get rid of the spektrum shit)

What's the best flying camera for me these days? Autel, DJI, or someone else?

>>2409382
DJI.

While you may not like the potential privacy implications, the simple truth is that they make the best consumer grade flying cameras - by far.

>>2409692
Should I save up for a mini 3 or can I get away with a mini 2?

>>2409382
I put a gopro and storm32 gimbal on a diy drone, the results were surprisingly good.

>>2410296
The only major improvements with the Mini 3 are vertical video (the camera physically rotates) & compatibility with the new controller with the built-in screen (which costs extra).

>>2410496
It's not just video quality that makes DJI camera drones so good, it's the integration with the software. You have zero control over your GoPro while it's airborne, you can't even swap between video & photo modes, let alone make changes to exposure, do object tracking, etc. I know this is /diy/ but there are some things you simply can't /diy/.

>>2388718
Chinese engineers don't have high emotional functions.

>>2410528
>You have zero control over your GoPro while it's airborne
Aktchually.. with my present hardware I can set it up to turn on and start recording while in flight.
So far, the auto-exposure has given good results. The limitation seems to be a single pilot's ability to fly and control the camera simultaneously (admittedly I'm fairly new to flying). I don't know that more bells and whistles would do anything for me. I use the gimbal to pitch the camera and control yaw by pointing the drone towards what I want to film.
Tbh I feel that for truly good results you would need two operators, one flying the drone and the other one controlling the gimbal and camera.
As for capabilities such as touching an area on the screen to change the exposure (like on a mobile phone), touching an object and have the drone follow it while avoiding obstacles etc I agree with you, if you can diy that, you might as well start a drone company.

>>2409019
what is your job?

>>2410528
No need to have control of a GoPro in flight, I have found that they adapt exposure really well (although sometimes not as fast as you would like), and you can just leave 4k/5k recording on and pull stills from that later. The stabilization and horizon leveling is just as good as a gimbal too, you only lose out to a gimbal if you are wanting to pan your shots.
Basically what I'm saying is that I will fight my 5" freestyle quad against anyone's DJI onions-machine, full contact, to the death or ground, whatever comes first.
Post throwaway email if you accept my challenge.

Took my little quad out today, got the throttle issue fixed and the rates & pids working pretty good. Found a little bird nest with some eggs, so that was nice too

To fix the throttle problem, where I just wasn't getting enough response under certain situations, I had to disable the anti-grav mode. Does that make sense?

>>2410970
Don’t want to dox myself/the company this soon, but it’s a drone/landing infrastructure outfit that does a lot of BVLOS stuff.

>>2410838
In my experience dual op is definitely a big help (it’s how I did most of my commercial AP work), however there are actually certain things where automated flight modes simply produce better results. Nine times out of ten, vision based object tracking with something like the Mini 2 is smoother than a camera operator trying to track the same object using a pair of controller sticks on a bigger & more expensive setup.

>>2411020
I’ve done it all - 5” quads with hard mounted GoPros, DIY gimbal copters with just pan/tilt control, small single op DJI stuff like the Mini 2 & big dual op DJI stuff like the Inspire range.

A 5” quad with a hard mounted GoPro is great for one specific type of filming, but it’s absolutely useless for everything else.

A DIY gimbal copter with no ability to toggle between video/photo, let alone to use vision based flight modes, is a very hard sell compared to a DJI, if your main concern is actually the filming. Of course if what you actually want is a fun DIY project & you’re not too concerned about the practicality/results, then that’s a different story - which is exactly what the OP copypasta says.

Has anyone here made any obstacles or things to practice flying around? The place I go to is just a flat empty field, except for some trees along the edge. I've heard about hula hoops and pool noodles, but any other ideas?

>>2411467
I would say pvc pipe with some 3d printed joiners would let you build some pretty cool stuff while still being mobile.
Other than that, have you found all the little gaps to go through on the trees? How about multiple gaps that you can fly through in series? Are there any gaps that seem too small to fly through? You probably could actually go through them, just bring your preferred method of drone-tree extraction with you.

>>2411654
Well, the other side of the trees is the parking lot. Even when there aren't any cars over there, I'd prefer to not fly over hard surfaces.. I might see if I can put something together using PVC and pool noodles, or PVC brackets and hula hoops...
I have to do some soldering first though...

>Try to launch ZOHD Talon 250G for the first time
>Rolls hard right and crashes
>repeat a couple times while playing with stick/flap levels while throwing
>airframe shredded
>salvage all internal components
>Buy second ZOHD Talon 250G kit to
>transplant components
>check CG, check servo configs, etc
>go for first launch on new airframe
>Nosedives hard on launch
>Breaks wing support immediately
Am I just retarded? is this airframe just a bitch to fly?
Suggestions for a new airframe to transplant my existing electronics into?
Current build: Matek 411WSE, ES900RX, foxeer micro, matek mq8 GPS, lumenier 1k mah 2s battery

>>2412313
These small 2S planes need full throttle & a decent throw to launch successfully. Remember also you'll have no control authority until the air speed over the wings/control surfaces is high enough - so if your launch is too weak & you throw it skew to the right, it's just going to go to the right regardless of what input you provide on the sticks.

Triple check your control surfaces move the correct way, both from stick input & from physically moving the plane while in a stabilized flight mode, then try launching in that stabilized flight mode - if your mechanical trims are way off you'll never have time to realise & react before you crash, but a stabilized flight mode will compensate & get you airborne.

>>2412408
Ive been launching in manual vs acro in an attempt to validate that stabilization is not the source of crashes. I could be throwing it softly, thats a valid critique. Should I just be yeeting this thing as hard as I can without effecting the release point/angle?

>>2412421
If you’re new to fixed wing, launching in a stabilized flight mode (assuming everything is configured correctly) will give you a much greater chance of getting it in the air successfully. Triple check everything is configured/calibrated right then trust in the flight controller.

As long as you’re throwing straight & not inadvertently twisting/flicking your wrist, you can’t really launch too hard.

New to this stuff, is there any reason to use 2 blade propellers? Some people on the internet say they're more efficient, but by how much?

>>2412749
There are so many nuances to prop choice that it’s not as simple as saying ‘two blade props are x% more efficient than three blade props’. Depending how you fly, you may find that three blade props are actually more efficient for your specific use case.

Just try a bunch of different props & see which ones you like the most. Unless you’re whole project is about min/maxing efficiency, focusing on number of blades is kinda pointless.

>>2412749
Depends on the context but for most flying stuff the 2 blade is more efficient because you will be spinning it very quickly.
Props with more blades are usually more at home with motors that spin slower but provide lots of torque. This is usually not the case, most BLDCs used in flight applications have fuckall torque but can easily reach 20-30k RPM.

>>2390421
>>2407418
>>2407417
>>2407414
>>2407411
>>2407409
Underwater guy signing in. Need some insight here.

My motors arrived. However due to the complexities of the model, I need to be able to power the RTX with a separate power source than that of the ESC.

The left of attached works. This is when a battery powers the ESC and the ESC then powers the motor and the channel of the RTX.

The right is what I need, the battery powers the ESC, the ESC powers the motor, but a separate power source powers the RTX and the signal is sent directly to the ESC. But this doesn't work. I've tried to trick it by sending the 5V of the ESC to a second RTX and powering the first RTX separately, but that doesn't work either. Anyone have any ideas?

>>2413247

Ground needs to be shared between components. You can't transmit a signal if there's no return path for it. Connect the ground of the ESC and 5V supply and it should work normally.

>>2413254
>Connect the ground of the ESC and 5V supply and it should work normally.
Will this work through 100ft cable?

>>2413316
Your basic approach here is completely wrong - you shouldn't be trying to transmit RX outputs, ESC signals, FPV camera feeds, etc. over 100ft of cable.

>>2413444
>Your basic approach here is completely wrong - you shouldn't be trying to transmit RX outputs, ESC signals, FPV camera feeds, etc. over 100ft of cable.

Another ROV enthusiast here - how do you work out the FPV camera in this case? Or other signals anyway?

>>2413516
Most ROV setups that I've seen (both /diy/ & commercial) use some sort of ethernet based 'tether'.

The HomePlug standard (aka 'Powerline networking', those plug-in adapters that let you use your home's mains power wiring as network cables) is a common /diy/ approach.

https://www.ardusub.com/introduction/hardware-options/required-hardware/tether.html

>>2413444
>>2413516
>>2413554
I actually got it electrically working, both FPV and motors through 100ft of CAT5 today. The tip on connecting the ground worked really well, thank you. There is no difference in either the video signal or motor signal in transferring it 100ft vs. 2".

Pic related. Everything in the blue box is in the ROV below water.

To answer the other question, why am I doing it this way?
1. I have a lot of quad drone parts lying around waiting to be used (spare RTX, VTX, ESCs, etc.).
2. Theoretically, because the RTX and VTX are transmitting above water wirelessly to my transmitter and goggles respectively, I can now send the ROV a few km out and he has a depth of 100ft.

>>2384243

Fatshark made new goggles and teamed up with Caddx aka Walksnail to get a Digital system in it based on DJI. Its gonna be 1080p and better FPS than DJI. And its a smaller form factor without the stupid DJI radio link.

So we get dual antenna Digital VTX without the unnecessary DJI radio link that the vista has. As long as they fix all the bugs before release it should be a really great digital system for FPV freestyle pilots.

Its nicer than DJI stuff because the goggles are made by fastshark with OLED screens. So the goggle form factor is nice and small like the old fatsharks. Not big like DJI goggles.

>>2384336
>huge numbers of us still use analogue.

I been waiting for something like this Walksnail before I got a digital system. I didnt like a bunch of stuff about the DJI.

The new Walksnail Digital system is gonna be 1080p 60fps I guess right now DJI is only 720p on their goggles but I am pretty sure the V3 will come out with 1080p also.

This thing has singlehandedly ruined my enjoyment of my 5" quads...

>>2388636
he cute

>>2395555
except it sounds like it would be extremely fun

>>2414847
It’s fun for about 10 seconds, then it just gets old.

>>2414847
It's really not

Charged up all of my batteries, fixed a few weak solder joints, and drove to the park.... only to find out there's a giant soccer camp taking up the whole place...

>>2415516
What, you're not going to attempt to sell some "drone footage" to those over protective parents?

>>2415600
Good idea... but this is the quality of my camera

>>2415605
>The Virgin GoPro vs. the Chad Comfy Scan-lines

I have a pair of Echine 800D goggles. Should I just get a better pair of analog goggles or shell out the money for dji? I would also need an hd drone... So I'm looking at around 800 dollars for it all... Or 400 dollars for some descent analogue goggles? Is it really worth it for digital?

>>2415659
DJI is gonna come out with version 3 soon so you can get some of the old ones second hand. If you already have an analog drone you can put an HD camera and system inside of it. You dont need to get a whole new drone.

Right now I am working on making my flight controller and VTX stack shorter so that I can fit the new Walksnail Digital system into my drone.

I can lower my stack by removing some plastic stand offs and replacing them with rubber stack dampeners. Might be able to get an extra 4mm and then I should be able to fit the new VTX in my stack.

>>2415662
Check for interference between the USB plugs and connectors, they can get a little tight when you are chasing millimeters...

>>2414597
Knowing fatshark the hd goggles will have tiny little lenses and feel like a McDonald's toy.

>>2415659
A better pair of analog goggles isn’t going to improve the image quality, if that’s what you’re hoping for. In fact the smaller binocular style Fatshark goggles are actually a *worse* visual experience than the bulkier single screen box style goggles like the EV800D.

I started with Fatsharks, but then ‘upgraded’ to the EV800DM & it’s so much better.

It's seeming like the best bet is to just go dji.

>>2415915
Is the signal on the 800d goggles good enough for an fpv wing? It has a 400mw camera.

>>2415934
Forgot pic

Okay I have a few things I need advice on.

I came across the DJI/Ryze Tello which has a hacking aspect per se, mainly just computer programming stuff for it, like object detection, using the usb port to power things, etc..

I may end up buying this, but wanted to know if there is other drones similar to this functionality to it?

I first got into FPV when pandemic started, but barely flew my drone because I ended up going too small for what I wanted, and I didn't really like have the huge fpv goggles (some highly recommended eachine boxy ones)

and I am wanting to know if there is a non fpv option of building my own drone but view it with my smartphone? I thought runcam came out with a wifi digital version or something for this, but its been a year or so since I looked into it, and just looking for advice on this.

If you're wondering why, its mainly to reduce bulkiness of stuff I have to carry without having to buy fat sharks or something.

>>2415934
>>2415935
Yes, the receivers in the EV800D are absolutely fine as long as you use decent antennas. The stock omni antenna is probably fine (it's hard to fuck up an omni) but the stock patch antenna is genuinely useless & should be thrown away.

I've done over 1km with my EV800D & a wing with its VTX accidentally set to only 25mW, because I was using a decent directional antenna.

>>2415936
>is other drones similar to this functionality to it?

Not really, because this is all about software rather than hardware. Churning out drones is simple & cheap, but developing software to enable them to do complex tasks like object detection is very expensive. Unless you want to sink vast amounts of time into reinventing the wheel setting up OpenCV/ROS with Ardupilot, you just need to get the DJI.

>if there is a non fpv option of building my own drone but view it with my smartphone?

You can buy 5.8GHz analogue video receivers that connect to your smartphone via USB, but the high latency they have makes them unsuitable for most FPV flying.

You're not going to get a wifi video stream beyond about 100ft, so that's really not an option.

If you're worried about bulk, consider a monitor instead of goggles.

>>2415969
One more question. I ordered an r9m receiver for the wing.. Will I need to solder it on or will it just clip into a wire? Only asking because the wing says it requires no soldering but doesn't come with a receiver.

>>2413984

What VTX/RTX parts and controller are you using there?

>>2415969
>If you're worried about bulk, consider a monitor instead of goggles.
True, I remember watching this video here about this
https://youtu.be/nm0-lWEut3c

Maybe I can find one for a few bucks at a thrift store, or just buy one.

>>2381571

upgraded my m1 to 3 bladed system

>>2415975
Depends what specific R9 receiver you actually ordered. Some of the R9 receivers have servo connectors, some are solder pads.

The R9M is a transmitter module, not a receiver.

>>2416022
Looks like it'll need to be soldered. Hopefully it's not hard since I've never soldered anything.

>>2416028
Practice on something else first. You don't want that to be your first ever attempt at soldering, it's *small* (picrelated).

>>2416039
Wtf do I test it on. I don't have circuit boards laying around

>>2415908

The screens are smaller than the DJI goggles that are huge. But its because the Lens's are better. That is how fatshark is able to have a small form factor goggle. By having BETTER optics. Inside the goggles it looks the same as DJI and the FOV is pretty nice.

I would only buy the DJI V3 system if they came out with Live Audio. Then they get my money. Otherwise I will go for the Fatshark Digital system cause its lighter.

>>2415933

DJI v3 might be good but we dont know. So far we do know that they will be OLED screens just like the Fatshark ones have. So that means the screen technology is the same. Also they will both do 1080p so that means one of them is not better than the other one for resolution. Fatshark will have smaller and lighter form factor over the DJI and they have OSD working. We are pretty sure that DJI is gonna give us a working OSD overlay but we dont know 100% yet.

Really the DJI needs to have something better than the Walksnail or its not as good. DJI will have radio link and probably heavier than the Walksnail stuff. So unless DJI puts a Microphone on the new Air Unit and makes them really smaller I dont see any reason to buy DJI over Walksnail.

>>2416093
What is your opinion of hdzero?

>>2416100
I like that its fixed low latency. But the penetration ability of the system is very bad. I dont like the way the break up looks. Its almost like you are flying Analog. If I am going to a Digital System I wanna be using the DJI quality stuff because the picture is more clear.

I think the best use of HDzero is if you are a racer and you only fly in fields and you wont ever have anything blocking your signal. Or if your a freestyle pilot and you live out west where there are no trees and you fly in wide open spaces all the time.

So personally I am not interested in HD zero because they way I fly I have lots of obstructions to my signal. DJI based digital systems will give me a clearer picture. Also I do freestyle so I dont need the super low latency like a racer needs. Sure it would be nice but most of the time the DJI stuff is good enough.

>>2416079
You can buy soldering practice boards. They’re a helluva lot cheaper to screw up on than ruining a receiver.

HD zero for FPV using goggles is actually about 25ms. With the DJI systems in low latency mode they are very close to that. And if you have a clear signal you will have really low latency. But if you fly behind something like trees or whatever the latency goes up but your picture stays clear. With the HD zero system your latency stays the same but you get rainbow colored breakup.

>>2416118
HDzero seems a lot better now with the 1 wat vtx.

>>2416127

Bardwell tested the 1 watt VTX in a parking garage and the DJI system did much better.

https://www.youtube.com/watch?v=v0gbLQM97Fs

^^ the above is a long video. It compares HD zero to Analog but he throws in a test of DJI. With the HD zero system he had to land half way through the flight because the video was so bad. He could not complete the test with it.

But when he flew with the DJI system it was very clear and it shows how much better the DJI based gear is. Their stuff is superior.

>>2416138
>>2416146
Never said it was better than dji but why would you be flying in a parking garage?

It's seeming like the best bet is to just go dji.

>>2416248
Oy vey

>>2416184

Its the best way to best penetration. That is why hes flying in a parking garage. I fly from inside my house sometimes in the winter and I have a big metal garage in the way too. Along with some trees and cars. So yea Penetration is important.

>>2416342
Hdzero seems to work pretty good in this guy's house https://www.youtube.com/watch?v=0eUO2LLU2TM&t=641s

>>2416248
If you are ok with the large goggles and you want to use the DJI radio then yea DJI is good. But the Walksnail is the same as DJI technology, just without the radio link and in a smaller form factor.

DJI is a great company and if they can give us a better solution with their new Air unit than the Walksnail then I will buy DJI. I plan on waiting till both are out and then choosing. I already know the Walksnail cannot do Live audio and the fatsharks dont even have an Audio port on them. So that means the VTX does not have a microphone either. But I heard that somebody hacked one of the old DJI air units and they got it to work with live audio.

So if the new air unit from DJI is small and lighter and has a Microphone on it then I think they might be doing live audio because the goggles already have Audio. If you buy the DJI FPV drone it has Live Audio to the goggles. So that means they are thinking about it for FPV.

If DJI does not come out with better air units with Audio I will be disappointed and I will just buy the Walksnail and use it till somebody finally adds audio.

>>2416349
They should make the cameras cheaper. Dgi bnf are so funking expensive.

I dont buy Bind and Fly's. I build all my own betaflight quads. That way you know how to repair them when something happens. And something always happens. Just the other day I crashed in a tree and one of my motor wires popped free of the solder joint. I had to solder it back on to get back up in the air. Being able to repair on your own saves loads of money.

The flying lawnmower is real. And my dad almost crashed it on takeoff.

The flying lawnmower is real. And my dad almost crashed it on takeoff.

https://www.youtube.com/watch?v=e6avmvfH4nk

>>2415976
RTX is a FS-iA6B, basic. Only going to use 3 channels.

VTX is an AKK FX2 40cm. Again just something simple I have lying around from my quads.

>>2416387
I remember using the flying lawnmower video as an example of the kind of suboptimal designs that small scale allowed for when I was imparting some UAVs courses...
Good times.

>>2416387
Sounds like Johnny Dep is filming this video lol

This guy is like every jewish stereotype imaginable. I saw a video of him saying he doesn't understand why hdzero is going open source because "why would anyone work for free". He literally couldn't imagine doing something without getting something in return.

I dont understand why HDzero seems to have so many fanboys when the DJI gear is clearly better.

Also the Digital system that DJI built is here to stay in our hobby. They are tinkering around deep in the code to give us digital FPV and now more than just DJI can do it. Other companies can now start using the same chips that DJI does to give us Digital systems. And they will all be DJI quality because its all the same type of technology.

>>2416856
Why does the DJI Jew get so defensive about HDZero?

>>2416859

HDzero is too close to analog. The breakup is bad.

Evan Turner is one of the fastest racing pilots on our hobby and he says that he can race with digital. The latency is not a problem for him as long as signal is clear.

So the thing that HDzero is supposed to be best at (low latency) does not even matter.

Here is another Bardwell video because he is the best. lol

Hes with Evan testing the DJI goggle latency and shows that he can race using that system.

https://www.youtube.com/watch?v=c-5HAXhas6k

>>2416861
Didn't even answer the question. Just more shilling

>>2416856
>Other companies can now start using the same chips that DJI does

Only to a certain extent. DJI have the resources to pay for custom silicon, no other company in the FPV scene can do that. This is the main reason given by the FPV Blue when they dropped their project after the release of the DJI platform.

Has anyone ever built an actual rc submarine? I figure you could have long antennae for gps, vtx and rc. You should connect a servo to a piston to control a ballast. That way you could set it up to loiter several meters below the surface in a set point. You could also get it to 'terrain follow' in clear shallow water. It would be way more elegant than having a tethered ROV.

>>2416898
>Only to a certain extent. DJI have the resources to pay for custom silicon, no other company in the FPV scene can do that

Wrong. The Walksnail VTX that is coming out is gonna be using the same silicon that DJI is using. You dont understand what is going on yet. People so say that DJI paid to make some special chips dont have the full story. Those chips are owned by Leadcore and DJI just had them tweaked a bit. So the reality we are in now means that other people can buy the same chips DJI is using from Leadcore.

The people making DJI's chips are called Leadcore and other companies can now use the same chip technology as DJI. That is why I am saying that this "DJI" tech is here to stay but its not even made by DJI its made by Leadcore.

>>2416947
Or it just means that Fatshark are paying a license fee to DJI.

>>2416947
DJI is a so good. Make pilot very happy.

>>2416978

lmao. Ok yes China is doing all this. People are working deep in the code of the cellular hardware to change stuff for Digital FPV use. And yes. This does have military implications. We have seen recently how drones are being used in Ukraine...

We wont know everything that is going on... But they are giving us low latency HD digital systems so I dont care I will just use the new technology.

>>2416980
Nice try, Chingchong Bingbong. DJI lost me the big race. Look at this horrible digital breakup I got.

>>2416982
DJI does not have rainbow brakeup shill.

>>2417014
How am I a shill? I fly analog and it's crystal clear. I'm not shilling any system. Once again the DJI Jew can't take any sort of criticism.

>>2417021

I have flown analog too and it was fine but the fact remains that the DJI stuff is better. And the future of FPV will use the same chips that DJI does. New people only buy analog when they are poor.

Sorry if you are poor and cant afford good digital gear. But I will enjoy it.

>>2417115
The point stands that DJI has terrible breakup as seen here >>2416982
I don't get that with crystal clear HDZero
https://www.youtube.com/watch?v=7bE45uR9Nfw

Whatever weirdo.

>>2415969
Is this antenna good?

>>2417476
Yes, but only if you point it in the right direction. The problem with directional antennas on goggles is that the vast majority of people end up pointing them in completely the wrong direction (usually straight at the ground).

>>2381571

>>2417554
Fuck drones, I want an fpv wing so bad!!! I can't wait for payday!!

>>2417554
>Jesus Fest 2022

Apparently this year it wasn't just Sunday morning that they had 'worship', the Saturday night entertainment was 'worship music'. I get that they're mostly/all religious, but this is taking it a bit far.

>>2417592

I was expecting it to be a lot of religious talk and whatnot but it was very tame. Even on Sunday it wasn't noticable after 10am.

There wasn't a ton of worship music time that I noticed. I think it was a couple of songs they played each night but I wouldn't know because I can ignore that music the same way I'd ignore one of those top 100 radio stations and they weren't playing it loud.

>>2417592
>>2417600
They're religious?
does that mean I can crash my FT build into demons to exorcist them?

>>2382173
welp, this suggests that they can get some sort of data from your diy builds. not sure if their flight mapping or real-time identification is the same in quality (they spoke about integrating the dji software into theirs, so interpretation of 5.8ghz would yield way more telemetry data) but it just confirms it knows if any drone is atleast *flying*.

>>2417918
forgot screenshot
ALTHOUGH they do go out of their way to separate "drone flights" and "flights with unique IDs" so that's promising

>>2382173
Make sure you purchase your DJI digital fpv system goy

>>2382173
Whelp, maybe I'll just buy a new GPU and play FS2020 in vr instead.

>>2417931
let's be honest though, how many flights are you gonna be doing over active concerts/sporting events that have this service?

>>2417955
If your drone's not in your line of site, then you've broken faa regulation.

>>2383619
>>2383755
You could probably also use something like a ubiquiti rocket modem on either end and establish a TCP/IP link for it with a VPN or similar. Do any FC/Software stacks support control over IP?

>>2417961
and? has nothing to do with drone shield if your not flying a DJI registered in your name

>>2417964
No they're going to force drone manufacturers to have a chip in the drones. You literally don't even know what's going on

>>2417967
still don't see how this affects anyone not flying DJI shit but okay.

>>2417973
Well since most of you fags suck DJI's cock and buy all their FPV gear, it'll affect a lot of you.

>>2417967
>>2417918
>laughs in /diy/

This sort of thing is never going to be enforceable across the whole /diy/ scene, the boomers at their model flying clubs won’t have it. Civil aviation authorities already have a track record of making allowances for boomers & that’s not going to change. My local model flying club doesn’t even allow models with autopilots/flight controllers, they’re sure as hell not going to start putting active tracking transponders in their glow powered balsa models.

>>2418121
>they’re sure as hell not going to start putting active tracking transponders in their glow powered balsa models.
They're going to put them and enforce them. Ask them about their stance on mandatory covid "vaccine"

>>2418139
The existing regulations we have are almost exclusively never enforced, what makes you think any new regulations are going to be any different?

I'm so confused about receivers. I bought an r9m module for my taranis q x7. I'm been looking at some bnf that don't come with a receiver. My question is how is would it be to just solder an r9 receiver in? And which r9 receiver should I buy? There's so many of them.

>>2417979
what? this board hates DJI. it's packed with coping poorfags.

>>2418341
You want the R9 MM OTA. Like I already said though, these receivers are tiny so aren't a great introduction to soldering.

Note that FrSky have two different air protocols - their legacy ACCST protocol & their newer ACCESS protocol.

Your QX7 &/or your R9M module may not be modern enough to do ACCESS. However it doesn't really matter, because all R9M modules & the R9 MM OTA receiver are backwards compatible with ACCST.

The only reason to buy anything other than the R9 MM OTA is if you need >4 old-fashioned servo outputs. You would only ever need this if you were building a big fixed wing model *without* a flight controller (eg you're just directly connecting servos/ESCs to the receiver).

>>2418495
The r9m is the access version. I hate rc. Why can't this shit just be simple.