Looking for a large FT or community UAV style build

[evranch]

I think I'm finally going to throw in the towel on my Maja, as even with the greatly improved power package it just is too hard to get off the ground. I'm down to about 150' of ground roll, but the ground needs to be smooth and there isn't much of that in this country. It still will not hand launch, it now just barely skims into the ground. Many others have complained about this airframe in the past on other forums as well, and there are lots of comments as to it barely missing the ground or striking on hand launch. I think there is too much weight in the fuselage and high wing loading.

So time to build something better I think! I want to reuse the power package that I bought for that stupid Maja. A 1000Kv Turnigy and 60A ESC, which will happily swing a 12x8 or 13x6 and put out about 2kg of thrust. I only want to run a single engine for efficiency.

I was looking at the Spear, but I think it doesn't have enough wing area to fly 7000mAh and ~1lb of camera. Ultimately, I'd like to fly a pair of those batteries for 14000mAh if I can. The Kraken can be converted to a single-engine pusher and might be the ticket, and they spec a 6000mAh 3s battery.

The other thing is, it doesn't have to be a pusher as this is not an FPV plane. The camera is on the belly. A heavy lift conventional aircraft could also do the job and will likely both hand launch better and rotate better from the ground, as the surfaces will be blown. I noticed a lot of FT planes seem to have no shortage of excess weight capacity in the videos...

 

[clolsonus]

I know we've chatted about this before, but you are describing a system very similar to the x-uav talon ... I'm flying an 8000mah 4 cell battery, turning a 12x5 folding prop (from memory so give or take an inch...) carrying a 1 lb-ish camera. Could easily carry more battery or more camera. I hand launch it and belly it into some fairly rough spots for landing. I'm trying to get one of our students to put together his notes and make up a howto page or video. Building the thing is super straightforward, but we can share how we integrated all the bits to come out with a pretty clean install. We did some 3d printing of brackets and a bezel for our camera cutout, streamline nose ... things like that.

I have dreams of building something similar out of foamboard, but with a wider fuselage so I can mount the camera wide-ways without needing to turn it 90 degrees ... maybe next year. I'm really intrigued by how well the Talon/Sparrow configuration flies.

If you wanted to do something more conventional out of foamboard, my other suggestions would be to start with the legacy design (basically a telemaster) and scale that up as much as you think you need to hold your payload and battery -- consider widening the fuselage if you have fat fingers like me. I have an 8' telemaster arf with more that 200 flights (almost 88 hours of air time) but wish the fuselage was an inch or two wider on each side.

There are a million ways to do these things, so no single right way ... I'm happy to share anything I've done if it helps (I work at a university so we do all our code and designs under the MIT open-source license ... we want to be able to share our work, although sometimes it's hard to find anyone interested) But there are tons of other good approaches out there too. Not trying to talk you into or out of anything here.

I was pondering this a bit today after reading some dji rumors (phantom 4 done, phantom 5 cancelled, or maybe not?) What do we have to do to get a solid fixed wing aerial mapping system with a decent amount of payload capability? Maybe it's a bit like getting into formula one racing cars ... you can't just go buy one at the local dealership, you kind of have to know what you are doing (or figure out some substantial things as you go) and you almost have to build big parts of it yourself, and then know how to setup and tune it. There is the ebee, but it's tiny and can't carry a bigger mirrorless SLR type camera with the extra good lenses.

The uav market place is weird. DJI has dominated so much that everyone is just doing quad copters. They are relatively inexpensive, and you really need zero flying skills to go out and map ... just draw a box and press go. (As long as nothing goes wrong that is ...) If the quad/drone market is collapsing (even for DJI) then what then? Will we all have to go back to the wild west of building our own? WIll new companies emerge? But fixed wing is more complex and you need to know stuff to operate them ... just like you'd need to know stuff to own and operate a racing car. It's stuff that can be learned, but we are used to DJI now ... just download the app, draw the box, click launch. Upload your pictures to the cloud, get an email back with your map ...

I wonder about all these things because it seems like my normally magical intuition has come up wrong on a lot of this and I'm trying to understand why. Did we all just fall into the DJI bubble way too much ... and is that a good thing or not a good thing or a meh thing? I get 4x the resolution (and higher quality sensor / detail / variation) with my home built system when compared to the cutting edge phantom 4 dji, but it took a boat load of work to get there: http://gallinazo.flightgear.org/uas/dji-phantom-4-pro-camera-vs-sony-a6000/

Anyway, whatever you decide, go for an airframe that flies well and flies easily ... you can always squeeze more performance out of v2 (sacrificing safety margin) if you want to turn the knob that direction later.

Sorry for rambling ... !

Curt.

 

[evranch]

Looking in to the Talon, I expected it to be an expensive UAV airframe from the name, but it's more of an FPV hobbyist platform at a very reasonable price! Might be an easy turn-key solution for me. It also has a much deeper wing chord than the Maja does which means it might have a better chance of getting off the ground.

I've considered making further mods to the Maja. I thought of both putting my original motor up front as an additional takeoff tractor, with a folding prop. It would be easy to mount and is so far from the pusher motor that it should deliver full performance. Very tempting just to see what happens as once it's in the air, it performs very well. But considering the beating the airframe has taken, and all the trouble it's given me, I feel like a fresh start is in order. The Maja keeps getting heavier and heavier too, with more and more hot glue injected into the EPP foam after all the wrecks.

A foamboard plane, on the other hand, uses more of a traditional airframe construction with stiff board and empty space inside. Thanks for the Legacy suggestion as I'd only ever seen it built as a twin. Looks like there is a single configuration as well and it will fly on a single power pack C - which means that the power system off the Maja should be able to fly a pretty beefy version of it. An all-foam wing that shows no signs of folding up like the Explorer wing means I could probably add a spar and carry a lot of weight. I might do a test build at 100% scale and then try to go up to around a 2m wingspan.,

I definitely would prefer a plane like the Legacy over flying wings as I usually see my Maja crabbed pretty hard into the wind when flying crosswind legs. With no yaw control, wings are fine for fun flying and FPV but don't really make sense for mapping. I've wondered before how the eBee does it, it appears to have a gimbaled main cam so maybe they bank into the wind? And how do they get such good flight times out of a plane under 2kg?

I kind of wouldn't mind this all going back to a wild west scene because it really increases the value of those few people who can build and fly. The eBee and similar solutions have always been vastly overpriced due to being a specialty item, so most fixed wing mappers I've seen are running some kind of homebuilt aircraft. Accessible drone quads have taken over the real estate/acreage aerial photo market for sure, but there are still very few solutions to cover 100+ acres. Farmers around here really want to get into using aerial imagery but their toy quads have proven useless - there is definitely a market there if I can get a reliable platform into the air. I am very hesitant to enter the market right now and embarrass myself with a plane that fails to leave the ground.

It is definitely strange how the UAV market is going, perhaps it's getting saturated with quads? GoPro is another odd one, despite their cameras being synonymous with sports footage and FPV racing they very nearly went out of business last year.

 

[clolsonus]

Looking in to the Talon, I expected it to be an expensive UAV airframe from the name, but it's more of an FPV hobbyist platform at a very reasonable price! Might be an easy turn-key solution for me. It also has a much deeper wing chord than the Maja does which means it might have a better chance of getting off the ground.

Hopefully I'm not taking away from other responses by jumping in here so much myself ... everyone else: jump in and say stuff too! So yeah, the Talon airframe was about $135 if I recall and that included free shipping from somewhere far away. You still have to provide all the electronics. It's really quite well engineered, when you go through it and see the details, it's clear that someone smart went through multiple iterations and put in a lot of thought.

The base airframe is super low wing loading and literally flies right out of my hand for a hand launch. I shared a video in a previous thread of my hand launching and even with an 8000 mah (4 cell) battery and a full sony a6000 mirrorless camera, it jumps out of my hand with a flick of the wrist. I'm launching under autopilot control (and motor powered up) but I've never had a single launch mishap <knock on wood> ! I usually wear some sort of hand/forearm protector due to the pusher prop, but when I've slo-mo'd the launch videos the prop has never been close and my arm is moving a way rapidly.

I've considered making further mods to the Maja. I thought of both putting my original motor up front as an additional takeoff tractor, with a folding prop. It would be easy to mount and is so far from the pusher motor that it should deliver full performance. Very tempting just to see what happens as once it's in the air, it performs very well. But considering the beating the airframe has taken, and all the trouble it's given me, I feel like a fresh start is in order. The Maja keeps getting heavier and heavier too, with more and more hot glue injected into the EPP foam after all the wrecks.

Sometimes I'll attack things on multiple fronts to see which one ends up being most productive. You could keep working on the Maja and start building up a talon on the side. And there are other options (the believer is a twin and well spoken of.) I have an old skywalker (fat foam glider t-tail) and that actually is a pretty good work horse as well ... not nearly as well engineered as the Talon though, nor does it fly as nice as the talon.

A foamboard plane, on the other hand, uses more of a traditional airframe construction with stiff board and empty space inside. Thanks for the Legacy suggestion as I'd only ever seen it built as a twin. Looks like there is a single configuration as well and it will fly on a single power pack C - which means that the power system off the Maja should be able to fly a pretty beefy version of it. An all-foam wing that shows no signs of folding up like the Explorer wing means I could probably add a spar and carry a lot of weight. I might do a test build at 100% scale and then try to go up to around a 2m wingspan.,

I definitely would prefer a plane like the Legacy over flying wings as I usually see my Maja crabbed pretty hard into the wind when flying crosswind legs. With no yaw control, wings are fine for fun flying and FPV but don't really make sense for mapping. I've wondered before how the eBee does it, it appears to have a gimbaled main cam so maybe they bank into the wind? And how do they get such good flight times out of a plane under 2kg?

The yaw angle is entirely dependent on the wind speed/direction and the aircraft speed and direction. At the same airspeed, a flying wing crabs exactly the same angle as a traditional airplane. Maybe with a traditional aircraft you could go into a full on slip to try to straighten the airplane out relative to the ground, but now you are in a high drag inefficient configuration and I don't think I'd want to fly like that.

You could put a big camera on some sort of gimble, but that's a lot of engineering and cost (especially if you are budgeting in ways to try to avoid purchasing a second motor/esc for your next system ... a gimble that can swing a 1lb camera around smoothly isn't going to be cheap or easy or small.

Probably someone knows something I don't know here, but I prefer to fly a plane like a plane. I want a rudder so the airplane tracks straight, but I don't use it to try to anti-crab in the wind or keep the wings level(er) in a turn. Prove me wrong though ... I like hearing perspectives I hadn't considered yet.

I kind of wouldn't mind this all going back to a wild west scene because it really increases the value of those few people who can build and fly. The eBee and similar solutions have always been vastly overpriced due to being a specialty item, so most fixed wing mappers I've seen are running some kind of homebuilt aircraft. Accessible drone quads have taken over the real estate/acreage aerial photo market for sure, but there are still very few solutions to cover 100+ acres. Farmers around here really want to get into using aerial imagery but their toy quads have proven useless - there is definitely a market there if I can get a reliable platform into the air. I am very hesitant to enter the market right now and embarrass myself with a plane that fails to leave the ground.

It is definitely strange how the UAV market is going, perhaps it's getting saturated with quads? GoPro is another odd one, despite their cameras being synonymous with sports footage and FPV racing they very nearly went out of business last year.

There was so so much hype. There's an article in the current model aviation magazine that talks about how this crazy hyped up dji drone thing took the knees out of the traditional RC hobby for a few years too. Only now in the past year or so has the RC model industry started to bounce back and gain traction with new innovations.

The flite test community has been rock solid this whole time though ... always a bright spot and always a new cool project going on here at this forum.

We still don't know if DJI is working on a phantom 5, meanwhile the phantom 4 is all but discontinued and going out of stock all over. (according to rumors anyway ...) That would make sense if the drone hype market bubble burst ... we wouldn't necessarily hear about it the moment it happened because the media and the market have a vested interest in keeping the hype train running ... but if no one is buying DJI's any more and many of the people that did buy them have them on their shelves collecting dust, then it makes sense that DJI isn't hurrying to build 1 million phantom 5's that might never sell. What do I know though, I'm just totally speculating here. But we might be looking at a market and industry in shambles for a few years and it might be really hard to get a descent pro-sumer drone for a few years without building it yourself .... well new anyway ... there will always be ebay.

So just to summarize, I have 2 Talons flying and one partially assembled and I'm still sold on the design personally. Not trying to talk you into or out of anything of course, everyone has their own use case and preferences ... and airplanes are always purpose built to get the best performance. But should you decide to take a closer look at the talon, I'm happy to share any details about our builds and electronics integrations. (And there are numerous build threads and videos online as well.) I think it flies amazing. If you have any doubts, spend $18 on a FT sparrow kit, spend 2 hours putting it together and installing the RC gear, and the go fly it. I think it will blow you away for how well it flies for it's size, even in substantial wind. Scale that up a couple 100 percent and you have the talon.

 

[evranch]

Watching some videos today it seems that everyone's Talon just leaps into the air, that's really what I'm looking for. Some guys running over 10AH as well, and there's a 150km 5 hour flight with a custom 18650 pack... obviously it can carry some weight. One thing in common though, almost everyone is running a 4s setup.

Whether you're trying to steer me towards the Talon or not, I definitely value your opinion as you appear to do this for a job and your resulting imagery is very high quality. So if you say "fly the Talon" I'm more likely to listen to you than poke around looking at stuff that might work. It's a proven platform for $135. The only other thing I may do first is build a Legacy to play with just because I like building planes anyways and the community could really use an accessible, plan-buildable UAV platform.

I'm talking about pulling the power system from the Maja not so much due to low budget as for a couple of reasons:
1. Finality to the Maja project. The power system is in a new plane, the Maja is hanging on the wall, no more tinkering with it. Done.
2. Shipping and parts availability in Canada SUCKS. I reuse a lot, because that means I can build today rather than a month or two from now. That quad build I posted? I ordered all the parts in March and only received them on Friday. Those parts came from GetFPV in the USA, not even on the slow boat!
3. I'm a farmer and while not poor, I'm cheap like all of us are. I hate to see anything go to waste, even if it's only $60 of motor and ESC, they are basically brand new.

Just your suggestion to spend $18 on the Sparrow kit drives the point home, since to you in the USA it's a casual buy and here it's more like "Spend $18 on the plane, $30 on shipping, $50 on customs fees, and then wait approximately a month which is the worst part". Might as well just buy the full size Talon then ! I love the FT planes as I keep a stock of foamboard and 9g servos here and if the urge strikes me to build a design I like, I just build it right now.

This is another reason I would like to build my mapping airframe, so if I lose it in a crash I can build another in a day or two rather than wait up to two months for the boat to bring me one.

The one valuable thing I'd really like to carry forward is my collection of large 3s batteries, as they are hard to come by in Canada. So it'll be disappointing if I need to go to 4s to get enough wattage for reliable takeoffs. (Or maybe series them up for 6s? Heh heh heh.)

After your explanation I realize I overthought the crab thing, I was too busy building my quad today and not thinking. I was thinking from the ground and not the plane, of course you don't have to use rudder to angle into the wind, you just fly a heading that gives the proper track. I should know this, but watching it go overhead it sure looks like it's in a sideslip due to the low airspeed:wind ratio. Duh. After reading up on it, the eBee gimballed cam is for taking rapid angled shots for better point cloud resolution. Neat idea.

Still, I don't think I'll do a wing. Having a real rudder gives a lot more versatility in choosing a landing strip and gives properly coordinated turns. I found when I got the auto-rudder properly tuned up in Arduplane its turnarounds were a lot sharper than when it was making washy uncoordinated turns, also it didn't lose altitude or surge the throttle.

I suspect cheaper drones are really eating into the prosumer market. The main reason to buy DJI IMO is their good cameras and high-def video link, for inspection or other tasks that require a good live view. A lot of consumers don't need this though, they just want an aerial selfie from a Chinese drone, or a fun flyer like the tiny whoops or toy drones. Previously if you wanted a quadcopter you pretty much had to buy DJI or build from scratch, now you can buy 250 racing quads prebuilt for ~$200.

Myself I decided to build a quad from Mavic parts rather than buy a Mavic, because I was really impressed by their power systems and folding props, but didn't need a lot of the features (HD video, gimballed cam) or weaknesses (folding arms, proprietary batteries), or to pay $2500CAD for something that will likely get regularly crashed on the job. Real DJI parts are amazingly cheap on Ebay which likely means there are a ton of dead or crashed ones out there!

 

[clolsonus]

Watching some videos today it seems that everyone's Talon just leaps into the air, that's really what I'm looking for. Some guys running over 10AH as well, and there's a 150km 5 hour flight with a custom 18650 pack... obviously it can carry some weight. One thing in common though, almost everyone is running a 4s setup.

4S is what made the most sense to us too, but we were starting from scratch. There is probably some combination of motor and prop that would scream under 3S ... I once ran across a 2S motor that screamed ... not sure where it derived it's magic from, but probably there was some sort of smart person and math involved. We have some ultrastick 25e's here that are much smaller/lighter and we run them on 3S ... and they don't feel overpowered ... so I don't know. If you did jump to 4S it might make your Maja a different bird too?

Whether you're trying to steer me towards the Talon or not, I definitely value your opinion as you appear to do this for a job and your resulting imagery is very high quality. So if you say "fly the Talon" I'm more likely to listen to you than poke around looking at stuff that might work. It's a proven platform for $135. The only other thing I may do first is build a Legacy to play with just because I like building planes anyways and the community could really use an accessible, plan-buildable UAV platform.

I think it would be cool to work up a basic design that is primarily foamboard and quick to build that is somewhere in the 2m wing span range and 4-5kg all up weight. It would require some engineering around the wing spar/attach point and the main fuselage/payload box probably needs to be beefed up. Maybe control surfaces and hinges would also need some extra engineering. If you are putting $1000 camera in there, you want an airframe that is going to last more than 2-3 flights under full load. I'm not sure where the line is where foam board starts to not be the best choice and beefier construction is required. It's not just about scaling up, it's about doing the extra engineering so the scaled up version doesn't fall apart after 2 minutes of flying. (No offense to some of the big foam board projects I've seen ... they look like tons of fun.)

I'm talking about pulling the power system from the Maja not so much due to low budget as for a couple of reasons:
1. Finality to the Maja project. The power system is in a new plane, the Maja is hanging on the wall, no more tinkering with it. Done.
2. Shipping and parts availability in Canada SUCKS. I reuse a lot, because that means I can build today rather than a month or two from now. That quad build I posted? I ordered all the parts in March and only received them on Friday. Those parts came from GetFPV in the USA, not even on the slow boat!
3. I'm a farmer and while not poor, I'm cheap like all of us are. I hate to see anything go to waste, even if it's only $60 of motor and ESC, they are basically brand new.

Just your suggestion to spend $18 on the Sparrow kit drives the point home, since to you in the USA it's a casual buy and here it's more like "Spend $18 on the plane, $30 on shipping, $50 on customs fees, and then wait approximately a month which is the worst part". Might as well just buy the full size Talon then ! I love the FT planes as I keep a stock of foamboard and 9g servos here and if the urge strikes me to build a design I like, I just build it right now.

Good point ... the sparrow should be a really quick scratch build if you have a couple sheets of foamboard laying around.

This is another reason I would like to build my mapping airframe, so if I lose it in a crash I can build another in a day or two rather than wait up to two months for the boat to bring me one.

I have a way of turning 4 hour projects into 40 (400?) hours spread out over 6 months ... my problem is I sometimes stray from the instructions and the defaults ... too many of my own ideas pinging around in my head.

Still, I don't think I'll do a wing. Having a real rudder gives a lot more versatility in choosing a landing strip and gives properly coordinated turns. I found when I got the auto-rudder properly tuned up in Arduplane its turnarounds were a lot sharper than when it was making washy uncoordinated turns, also it didn't lose altitude or surge the throttle.

I would advise against building a big flying wing for mapping unless you can keep the wing loading very light and keep it hand launchable. I went down this path in a previous life and our original 6lb prototype eventually became 12 lbs with payload and battery and even after sizing up the wing area a bit, we still had something we could barely launch. We had a guy engineer a launcher, and then my partner redid his own launcher later, and it was still barely launchable ... lots of mishaps and frustrations. In the end we felt like we did a lot of good work and made a pretty cool airplane that spec'd out really nice, but was a white knuckler to operate. That was when I decided I'd give up 10-20% performance for something that was easy to launch, land, fly, and manage.

I suspect cheaper drones are really eating into the prosumer market. The main reason to buy DJI IMO is their good cameras and high-def video link, for inspection or other tasks that require a good live view. A lot of consumers don't need this though, they just want an aerial selfie from a Chinese drone, or a fun flyer like the tiny whoops or toy drones. Previously if you wanted a quadcopter you pretty much had to buy DJI or build from scratch, now you can buy 250 racing quads prebuilt for ~$200.

Myself I decided to build a quad from Mavic parts rather than buy a Mavic, because I was really impressed by their power systems and folding props, but didn't need a lot of the features (HD video, gimballed cam) or weaknesses (folding arms, proprietary batteries), or to pay $2500CAD for something that will likely get regularly crashed on the job. Real DJI parts are amazingly cheap on Ebay which likely means there are a ton of dead or crashed ones out there!

One of these days i need to buy/build myself a quad, but still haven't run out of fixed wing ideas and fun ... maybe some day.

 

[evranch]

Been busy with other projects, but I did pick up some more foamboard in the city and plan to try out a Sparrow since it looks like a super easy build. Also going to do a single-engine Legacy. Of course I need time to build both of these and it's the busy season now! They were all out of white so it will be a black Sparrow. They are never all out of foamboard... spring is here and everyone is building FT planes?

I really want to see that HD foamboard from the truck drop video become available, that would make building larger planes a lot more practical. I suppose foam can be double or triple-laminated but at some point it's getting to be a waste of time.

Thanks for the advice on the wing, I absolutely don't want to require a launcher since it's one more thing to carry around.

I'm unfortunately thinking of doing the dual-motor setup for the Maja in the short term (one tractor, one pusher, shut one down once I'm in the air) since it is just so useful for fencing projects. As it is, with the pusher motor, I can only launch and land in my bin yard and then just have to fly to the location I'm mapping and back. Luckily I'm only working on the home quarters this spring, and it's only a couple minutes' flight to get to the target areas.

 

[clolsonus]

Good luck with the scratch builds! Last week I did a tiny bit of work on my own scratch build project (a 50" wing span foam board fairchild f-24.) This is just a simple design, but will be my first foam scratch build. I drew up the wing in inkscape and laser cut it out. I definitely need to do some design tweaks before I build v2 because a couple things didn't quite fit like I had hoped. (Learn as you go I guess.) In the end it took me maybe an hour to put the wing together after spending an hour figuring out the laser cutter in the maker lab. (This is without any servos or control linkages.) Then when it came to putting it together, there were lots of places I could have bungled up the assembly because there was no Josh build video to work from!

All this got me along two sort-of conflicting thought tracks:

1. I still have this picture in my head of an ultra simple UAV foam board design (that probably isn't all that simple but we start out with good intentions.) I am imagining something in the 1800-2m wing span range that woul fly at around 3 kg all up weight give or take. I'd love to work something out that was so simple, that most of the parts could just cut out of foamboard by hand with a ruler and a couple quick measurements. I'm picturing a few engineered pieces to get a strong wing center and spar, and maybe a fuselage box to securely hold a big battery and a pound of camera and survive a couple dozen rough landings over a season. My picture is of something squarish, but with just enough shape to make it look like an airplane.

2. My scratch project is exactly 50" wing span and I'm starting to feel like if I went any bigger, it might be worth leaning back towards more traditional build techniques. I envy the guys that can spend 20-60 hours a week cranking out airplane builds and doing all this fun stuff, but for someone like me with a day job and family and pets and grass and not enough energy for anything any more ... the quick easy projects are attractive.

So I ponder all these things for a while and then look back over at my Talon which basically does all the stuff already. Why am I thinking about designing something out of lower quality materials that will probably have more drag and be less robust and capable. Time/cost wise probably won't save me anything because the talon starts out so cheap and is a very quick build. Strategically it makes no sense to design/build something from scratch. The only reason I would have right now is that I wanted to do it ... which I might (!) but I'm not sure. My personal plan is to work through this 50" scratch build project first and see what I think about things after that is in the air. I'll need to decide if it's worth iterating and doing a v2 to clean up the prototype design glitches, or just move on to something else, and if I still think foamboard is cool for medium size airplanes, or if more traditional build techniques would work better for me. Or maybe I should just stick with the FT kits and HK arfs until I retire and can spend 40 hours a week on my hobbies.