Last month I detailed my plans and preparation for an attempt at a 100km flight in a blog here.
Well yesterday the weather was ideal and my preparations (nearly) complete so I went for it and achieved a flight distance of 126km in a time of 2 hr 46 min 40 sec.
I can't attach my log file here because it's too big. It's 15MB and there's a 7MB limit. Any ideas how I can share it?
I exported the KMZ (2014-01-28%2006-32-40.tlog.kmz) to Google Earth and generated the following images showing the flight path, and the altitude/speed trace with some of the vital statistics.
[Click image to enlarge]
The airframe is a Bormatec Maja with 2.2m wingspan. It’s not the most aerodynamic airframe, but its advantage is its payload capacity which allowed me to carry 21600 mAh 6P4S Li Ion batteries with ease.
I stripped out absolutely everything I didn't need including the obvious things like FPV cables in the wings, and thanks to some advice on my previous blog I also removed a few things I wouldn't have thought of such as the RC receiver case and airspeed sensor.
The advice I had from other DIYDrones users was invaluable but the biggest factor in improving the performance was the testing I did. For each variable I wanted to examine I would fly multiple laps at different settings then analyse the log files to find the best performing setup. I give more details on that in the original blog.
For the actual record attempt there were a few things I’d never done before. One was flying without the airspeed sensor (I had flown with it switched off, but never with it physically removed) and the other was the 4 cell batteries. Probably both those things were the cause of the trouble I had getting the plane to fly at the 13m/s (47km/hr) I wanted. Initially it was much too fast. Over the period of a few laps I adjusted the cruise throttle percentage, the cruise speed and the min and max speeds to much lower than I really wanted which bought the speed down (and pitch up) to where I wanted it.
Key Stats
• Distance: 126km (according to Google Earth)
• Duration: 2hr 46min 40 sec
• Airframe: Bormatec Maja
• Wingspan: 2.2 m
• AUW: 3.25 kg
• Battery capacity: 21,600 mAh
• Average Speed: 45.6 km/hr
• Top speed: 68.3 km/hr
• Minimum speed 40.5 km/hr
• Duration: 2hr 46min 40 sec
• Barometer creep: -6m
Comments
Sorry, more info here: http://www.barnardmicrosystems.com/UAV/milestones/atlantic_crossing...
I don't want to detract this effort, but do you know Maynard Hill? His plane flew 3039km with just 2.2kg fuel autonomously from Canada to Ireland.
Moglos - Congrats! Glad the Li-Ions worked so well. If you ever decide to go for the record (if what you did didn't all ready do that), let me know. I can share how to control charge temperatures to get a few more % out of it.
Just one question. How did you hold such a tight pattern over that distance? I'd be all over the place.
@ reto: It had crossed my mind that a 200km attempt might be achievable, but not in the Maja. I calculate it could get up to 165km by adding 4S3P of batteries, but that would be 4kg AUW which I think is right on the limit.
You're right the flight path was 12 WP arranged in a circle. In my testing on a rectangular circuit I could see the increase in throttle, pitch and current at each corner. I knew I needed a very smooth flight path. I thought about flying in loiter mode with a very large loiter radius, but I decided on the 12 WP circuit because that let me program a very gentle ascent for the first lap, and descent for the last lap and landing approach.
During the flight I noticed the speed fluctuated, but I thought it was just random. I didn't notice the regular pattern until I made that graph. There was a slight breeze between 3 and 8 km/hr from the north west and watching the log being replayed I can see I'm slower flying into the wind, and faster with it. But I didn't expect that, I thought with the airspeed sensor removed there would be an upwind/downwind difference in power, but not ground speed.
Congratulations!
I have never thought a Maja would ever fly so far. You really optimized it well. Your thorough and systematic testing and optimization approach bears fruit. The high capacity LiIon batteries also did a wonderful job. Great work!
Take an Albatross UAV or another high performance airframe and you will easily make 200km :-)
Your flight path looks like a circle, but is somewhat angular. I guess this is due to waypoints arranged in a circle as opposed to really flying a circle. Your flight speed oscillates constantly. I guess this might be the ground speed while flying circles due to a bit wind. Can you confirm?
@stephen: yes
@ravi: happy to, but I can't attach a file here (only when I write the original blog). If you PM me your email I'll send it to you that way.
@chris: 2.76 (I think, I haven't yet uploaded the newest version.)
@crashpilot: I did't cover the wings in the end. I didn't have a good way to test the covering as I couldn't really fly a couple of laps then apply the covering at the field and test a few more laps. I could have tested without one day, and with the next, but there would be too much random error (changes in wind, temperature etc) to recognize a small difference. And if it didn't work I would ruin the surface by taking it off. I'd really need two sets of wings to test that.
@mark: I have a dropbox (and google drive) account so I'll try that.
@eduardo: motor is Sunny Sky something like 700kVa (I'm at the office so I can't check right now) with 14 x 9 prop.
@tommy: no NCR18650B. They are 3400 mAh so I must have gotten 3600 mAh in my head from somewhere when I calculated the total mAh. It would be 3400 * 6 = 20400mAh. My plan was to land when the voltage reached 12.4V so my error didn't mean I pushed the batteries too hard.
Great work moglos :)
I was also wondering about the batteries. Did you use the new panasonic NCR18650G cells? (21600/6=3600mah)
If you send the log file, i can share it on my Dropbox account :)
Moglos: What was your propeller and motor setup like?
Those sound like it would be quite a stretch from anything I've seen. So it depends on how hard you want to make it. I daresay they're quite a bit harder than 100km for an airplane, but I'm not sure...
Well if you have already done 20, 40 or 50 sounds like an effort ;-) How about half aircraft distance, 50km for helis and 25 for multi's. Can't make it easy!
-
1
-
2
-
3
-
4
-
5
of 5 Next