How Long Can a Solar Drone REALLY Fly?

Last year, I proved that it's possible for a drone to fly purely on solar power. The only problem is that drone flew for about 3 minutes before it crashed itself. Naturally, that result was not good enough for me. So, this time around, we're redesigning, rebuilding, and eventually putting the drone to the test to see how long it can actually fly for on 100% solar power. Unfortunately, the panels from the first drone did not hold up so well. And I'll show you here. This is the current state of them. And they're looking really sad.

They're pretty much all at least a little damaged or broken. And it just looks terrible. And the main reason for that is number one, these are super brittle. Even if you just bend them a little bit like that, they will crack. It's so hard to deal with. And then secondly, the double-sided foam tape did stick to the panels, but it didn't stick to the carbon fiber tubes because I didn't make these big enough. The new idea is to still use these thin carbon fiber tubes, but I've actually

redesigned the TPU sleeves that go on them. So, these have a much larger surface area for the foam, so it should stick really well. And I'm basically going to slide two sleeves onto each section to cover one panel. And then this will just stick on like that. And hopefully it holds. And that's on there. And this is looking really good. And we've got a little solar panel lollipop. Fortunately, the drone is perfect and

could fly right now. However, I did have an idea of some changes that I want to make to the design. The version one of the drone really struggled to rotate in the your direction purely because the arms were so long, meaning the inertia of the drone was very high. Because of this, I decided to trim down the arms on version two just to reduce the inertia and increase the overall agility of the drone. And from these two pieces, we have saved about 70 g of weight, which is awesome

cuz that's about 4 W of power. Next, I just had to rerun the wires through the shorter arms. You can see we've actually got quite a bit of excess. So, when I cut that off, we're going to save even more weight from the shorter arms. And now it's time to install the ESC and the flight controller on these bolts over here. It's actually flying pretty nicely. It's very locked in. There's a slight wobble, but I think I can fix that pretty

easily. So, I think it's time to take this thing outside. The good thing about the drone being smaller is it's way easier to transport now, which I always appreciate. I'm back at the flying field where I flew the solar drone for the first time. And it just shows you cannot trust wind forecast in Cape Town because it is actually pretty windy right now. And I specifically chose this morning because it was supposed to be no wind. The drone is set up and ready for

flying. As you can see, there's four batteries on here, but only one is connected. So, the other three are basically dummy batteries to simulate the weight of the solar panels. The goal this session is to make sure that the tune of the drone is perfect and also that the GPS hold is working correctly, just because I do not want to have to fly the drone potentially for hours when we actually test it. I used the tune from the V1 drone and it actually worked perfectly. So that was great. I then moved on to the GPS hole tuning and

right away it started swinging around like crazy. I actually realized that I just needed to calibrate the compass and after that it seemed to be working very well. So the drone seems to be holding its position, altitude and direction very nicely. Now is a little bit wobbly around but I think that's cuz it is very windy. But as you can see, the controller is just chilling, not doing anything, and the drone is

flying very nicely. So, I would say that it's actually ready for solar panels at this point. I've just had a look at the weather, and unfortunately, it seems like tomorrow is the only chance I'm going to have for the next 2 weeks to attempt this long solo drone flight. As you can see here, tomorrow is looking very low wind and also no clouds, which is obviously important. And every other day for the next 2 weeks has either a

lot of wind or clouds and or rain. So, that is not going to work. And that's quite unfortunate because I haven't even started putting together the new solar panel array, and it's already 1:00 p.m. So, I think it's going to be a late night, but I am going to try and get it done. The solar panel mounts are finished. And I think it actually looks pretty cool. It's kind of like a little mosaic work of art over here. Just going to peel them all off.

So, here is the plan. I have laid down and taped together three pieces of this thick white card. And on top of that, I'm going to put my solar panels in a row and then solder them together. So, I'm first going to do eight in series and then another eight and then another eight and then another eight because the idea is to have an array of 8x4 which is going to give us 32 panels in total.

The first eight panels are finished and it actually took me a lot quicker than I thought which is great. What's really cool is that the panels actually still generate a voltage even when they're face down on the table like this. So you can check that everything's okay. So the first two we've got 0.42, then two in series, we got 0.92, 1.4, 1.8, 2.4,

2.75, 3, and 3.5. So that's how I know that everything's good. and I can move on to the next one. I am finally done with all these panels and that was exhausting, but it's looking really good. It looks kind of massive at this point. So, I'm a little concerned about the wind, but

everything's connected and the voltage adds up nicely. So, all I need to do next is build the frame and then stick the frame on here very carefully. It's time to stick this down. I'm a little bit stressed, but I'm sure I can do it fine. And that is looking pretty good. And there we go.

Oh, I do not want to drop that. And that is the Soda drone ready for flying. For anybody who's looking into actually making projects like I do, I know the equipment and machinery I use can be really expensive, but fortunately today's sponsor PCBW might be able to help you out with that. This is a prototype model of a new drone that I'm working on. And if anybody can guess in the comments what this is going to end up being, I'll be very impressed. But

for this one, I actually got this carbon fiber frame manufactured by PCB for me. And the process was really simple. After drawing up the design, I just uploaded to their website, selected my preferences, and after that, when you make your order, it'll essentially just show up at your door. And it's honestly really high quality stuff. and they do a bunch of different things like 3D printing, CNCing, and later this year, we actually want to play around with some metal 3D printing. So, that's going

to be a lot of fun. If you want to try out their services, you can use the link in my description below, and you'll get a $5 coupon off your first order. I'm back at the flying field, and I've got the drone there. It's got its cover on it at the moment. And the weather is pretty good. I think we're definitely not going to have any clouds, which is obviously really important. And the wind is very low. For those of you who don't know, Cape Town is a very windy city, especially in summer. So, this is pretty

much as good as it gets. So, over here, I've got a power meter and I've got a bank of resistors. You might remember this from my previous video. And what this allows me to do is change the resistance applied to the load so I can pull a specific amount of power. So for this I can actually see how much power the solar panel is able to generate. And by increasing this one by one I can up the power output. And by the time we get to about 3 or four resistors is when I

know that it'll be ready to actually take off. We're at 20.6 6 volts, which means they're getting it's getting full sun and we're on one resistor. And you can see it's pulling 24 watt. And if we go to high end here, the voltage will basically just collapse. And then we know we've gone too far. So I'm going to up it to two resistors. And now we're at 48 watts, which is great. And it hasn't collapsed. Now let's go three resistors.

70 watts already. Okay, that's quite impressive. 69. And our voltage is still fine. Okay. Four resistors. 88 watt. 89 watt. Okay. So, it's actually holding there, which our hover volt our hover power is probably about 70 W. So, in theory, that would be fine, but we want a bit more headroom for a bit of gusty wind or strong movement. So, let's drop it to five resistors. I'm guessing it won't be

able to handle this. And okay, well, it is handling it. It's on 100 watts now, but our voltage is quite low. So, this is kind of pushing it, but it's still okay, actually. If I go to six, surely it can't handle six. Okay, it's handling six. The voltage is dropping a lot, but it is pushing 110 watt. So, I underestimated the strength of the sun, I guess. This thing is so fragile.

I'm always so nervous it's going to break. That's why we really need as little wind as possible. No batteries, pier solar power. I don't know where I would hide them, but a lot of people thought the last video was fake. So, I must say there is something really cool every time I see this thing take off without any batteries on board.

GPS hold wasn't as good as I'm used to. And I just checked now and it's only picking about 13 satellites up, which is a lot lower than normal cuz it normally gets about 20 and it sometimes can get up to about 26. So that's probably why the hold isn't great. Hoping for the best that the GPS hold would work as is, I took off again to see how long it could fly for. And then this happened. Oops.

Woo. >> Definitely a low wind device. Yo, the leg is broken. >> Oh no. >> You have a spare? >> No. >> Wow. It really needs like no wind at all. Hey. >> Yeah. >> How long was that flight? Can you check? >> That's only 3 minutes now. 4 minutes. So it was 2 and 1/2. >> Okay. So, in a perfect world with zero wind, the drone could in theory fly indefinitely while there's enough sun in the sky to power it. But we don't live

in a perfect world, and I think it's basically going to be impossible for me to get as little wind as I want in the outdoors where there is also sun. So, I'm going to have to make a few adjustments to the drone to hopefully make it able to fly in at least very small amounts of wind. And one of those things is a little battery. So, I've got this little setup here. It's literally five little small lithium ion cells in series. And these are basically there to

supplement the power from the solar panels when they can't deliver enough. And when I look back at the log, there was a very clear moment right before it crashed where the voltage from the solar panels just plummeted. And that means that it was overloaded. And what this will do is basically take over when that happens and make sure that the drone doesn't crash. And this is how it works. Firstly, these little things are dodes and they basically are one-way valves. So they will only allow current to flow that way and not back that way, which is very important. In normal operation, the

drone will work like it has been where the solar panel just provides a voltage to the ESC and powers the ESC. Now, what it'll also do with its excess voltage is charge the battery through this stream here. And this battery will be charged up to this voltage minus the drop across here. Now when the solar panel drops too low with its output voltage, the idea is that the battery will then take over from these leads until the solar panel

comes back up to its proper voltage and then it'll take over again. I have roughly soldered together my circuit and it actually seems to be working perfectly which is a very nice surprise. And this is the setup. So over here we've got the power supply which is acting like the solar panel supplying 20.4 volts. And over here is our little circuit. And then here is our battery. And this is the battery voltage. Now if

we turn on the solar panels, you can see it's now recharging the battery at 11 W. And you'll be able to see the battery voltage here is climbing. So that is charging up nicely. But what happens when we actually connect up the system to the ESC's, which are these over here? Now, when we plug it in, you'll be able to see that the system is supplying 83 W to the ESC's. And now, if we turn the solar panels off,

you can see it's still supplying 60, 59, 58, 57, and we turn the solar panels back on. And you can see it's jumped up again now and the batteries are recharging again. I was a bit worried that this tiny little battery wasn't going to have enough power output to handle the drone if the solar panels cut out. So I switched out to this battery which is quite a lot bigger and has a high C rating. So this should definitely handle it. And over here is our circuit.

And all I've added here is a little BMS. And this is basically to make sure that the battery doesn't get overcharged. And it also balances out the cells in the battery. At the moment, the panels at the ends are super wobbly, as you can see there. And that's really bad cuz as soon as they start flattering like that in the wind, it destabilizes the whole drone. So, I'm actually going to change out the center spar here to a thicker one. And that should make it a lot stiffer and means that it will not be

able to twist as much. On top of that, I'm also going to try and lower the panels a bit just because having them so high up here really raises the center of gravity of the drone. So, I think if I lower them down a bit, that'll also make it a little bit more stable in the wind. If you're concerned about the panels blocking air flow to the propellers, don't stress too much because I did run multiple simulations in a new version of Air Shaper to see how the air flow interacts with the panel being above the propeller and the results were actually

pretty good. Even with the panels quite close to the propellers, there wasn't a major drop in thrust or efficiency. To further increase the wind resistance of the drone while maintaining enough power to fly, as well as matching the voltage from my battery, I decided to remove four of the panels from the grid. It's time to give this another shot. But this time, we have come all the way to Stalenbos. And the reason for that is Stalenbos is generally a bit less windy than Cape Town, so it should be a bit calmer. And we've also got this farm to

fly at which has a bunch of trees here so that should block out the wind later hopefully. Right now it's super calm so it can definitely fly in this. However, the wind always picks up in the afternoon and the drone is not ready to take off yet. So as soon as there's enough sun to take off, we're going to do that and hopefully we can start flying and get some stable flight. While we're waiting for the sun to get a bit stronger, as you can see here, I've now mounted the GPS on top because the

panels were actually blocking the signal last time. So, I'm going to fly it just on the battery now and make sure it's all stable while we're still waiting for the sun to get a bit stronger. On the first flight, I actually got super stressed because the drone kept on rotating without me giving an input. And after looking at it closely, I saw that one of the motors had rotated on the arm. But luckily, I had some super glue on hand. So, I actually glued the motor mount to the arm, and it seemed to fix the problem perfectly. Still on the

battery now, but it is on GPS hold. And it's not perfect, but I think it's good enough for us to actually try. Now, with a lot of uncertainty, it was finally time to launch the drone and see exactly how long it can fly for. Seems to be holding pretty nicely. I'm not touching the controller. However, I am very cautious that it might drift off because it's also got very limited amount of power. So, if it needs to fight the wind or something, it won't do a great job of that. But, it's doing pretty well. I

think we're flying for about 3 minutes now. So, we're over 20 minutes now and it's still flying, but there is a very light cloud in front of the sun, which I'm sure, as you would could have guessed, is not a good thing. And the wind is picking up a bit. So, we got our two arch nemesis really not helping us now,

but it is still flying at least. Unfortunately, at this point, the GPS hold stopped working, which means I basically had to fly the drone manually all the time to make sure that it didn't crash. >> I am very sure that without the battery to supplement that high power usage when it wobbles or the sun dips, >> Yeah. >> it would have not made it this far cuz that's every time that it had to land. Previously, didn't really crash, but it had to land was when the panels dropped

out. So now the battery is actually taking over when the panels do drop out and allowing it to stay in the air. >> All right, 1 hour. >> Yay. So, we just passed the 1 hour mark and

for me I count that as a successful flying solar drone. As far as I know, this is the longest solar powered drone flight. So, that is very exciting and definitely a victory in my book. Now, how long it will fly, I'm not sure because we've got a lot of factors here. We've got clouds, we got the wind, we've got my mental strength and my coordination. So, we'll have to wait and see. But for now, I'm really happy with

the 1 hour mark. >> That's pretty nice. That makes me feel a lot better cuz now >> Yeah. All the electronics settle down as well. That >> they could have cooled down or something. But look at that stable. >> Oh, luxury. >> You go have a nap. Maybe not. Maybe not enough. The GPS hold working was very short-lived. So I had to carry on flying it fully manually which was very

mentally draining. >> Okay, there we have 2 hours. Very cool. That's a very exciting milestone. Now I can relax even more. But we've come this far. I'm definitely not going to land it yet. So, either the wind is going to take it down or the sun is going to take it down. One of those two options.

GPS hold seems to be working again for a bit. So, I got a bit of a break on my fingers and we're at 2 hours and about 47 minutes in flight. So, that is pretty impressive. All right, there we have 3 hours 32.

So, we've just passed 3 hours and 32 minutes, which means this has flown longer than the endurance world record drones, meaning this is the longest electric multi-rotor drone flight ever, which is very exciting. And I guess it was all worth the effort. Now, we just need to see how much longer it'll fly for. It's still going strong, so who knows how long. It's a bit sketchy. Oh,

come on. Go. >> At this point, it was just a waiting game. But the wind did start picking up as well, and the flying started to get a little bit sketchy. So, while we still had a successful flight in hand without a crash, I decided to land the drone. The drone has landed and we have done over 5 hours of flight time, all on solar power, which is incredible. I honestly didn't think it would go that long. I mean, I knew in theory it could,

but I thought the wind or something would get in the way, but we were super lucky with the weather. We had a little bit of wind and a few close calls, but eventually it pulled through and made it. So, I am very, very happy with that result. And a big thanks to Stefan and his uh uncle for letting me use their farm here. That was a huge help. If you enjoyed this video, please don't forget to subscribe. And also in the comments down below, if you have any suggestions or ideas or things you want me to try,

then please let me know because I really enjoy reading those. And who knows, maybe one day I will build one of your