This is some details and tested performance data for my Synergy E5S 626 with UAV conversion done to it. I don't have any good photos of the build at this point, but I will include a link to a video of it flying its maiden flight on autopilot. The helicopter is flown with a Pixhawk autopilot with custom firmware in it, based on the ArduPilot flight stack. It has a dedicated 915MHz telemetry downlink to a ground station over MavLink radio. Since I am also an Extra Class ham radio operator I equip my UAV helicopters with an analog 5.8GHz FPV feed with both audio and visual to keep track of what the helicoper is doing in flight. I've had to autorotate helicopters before due to engine failures in flight. With a UAV heli, that might be a mile or more out on a survey flight, this is impossible to do visually. The FPV makes it possible to set up the autorotation, and while there is some guesswork involved after it drops the FPV link as the heli descends I've had about 90% success rate in making autorotation landings due to mechanical failure, with no damage to the helicopter or payload.
This happens to be one of the biggest reasons I fly helicopters instead of multi-rotor "drones". You fly 'em long enough, and it's if, it's when, it will have in-flight failure. Despite what some people think about 6 and 8 motor "drones" being able to fly with a motor out, reality is that in UAV work it overloads the remaining motors and the throttle level is so high manuverability is lost and they simply burn out more ESC's/motors, or flip over and crash. Fixed wing UAV's also have a problem with stalling and crashing into stuff in an emergency landing because of their speed. Helicopters are the only UAV type that can be safely landed with no damage in the event of power loss in flight. You might hit some corn stalks or soybeans or something in the final flare, but that tends to cushion it and prevent damage.
The powertrain configuration in this heli is a Scorpion HKIII-4035-530Kv 3.4kW (4.4kW 5 sec), Castle HV120 speed controller with governor, logging and telemetry, 9.971:1 main gear ratio, tail ratio 4.5:1, Rail 626 mains, Rail 96 tail blades. It is flown on 12S power.
Tested battery configurations, depending on flight time needs, can be:
1) Single 3300 12S (147Wh) providing 14 minutes cruise flight time with 3 minutes reserve
2) Single 5000 12S (222Wh) providing 21.5 minutes cruise flight time with 3 minutes reserve
3) Twin 3300 12S (294Wh) providing 25.5 minutes cruise flight time with 3 min reserve
4) Twin 5000 12S (444Wh) providing 34 minutes cruise flight time with 3 min reserve
CG is adjustable for all battery configurations with 1.25kg payload. Note the helicopter has higher cruise flight power consumption with the twin battery configurations. Lighter payload will increase flight time accordingly. A typical MAPPIR/GoPro camera and mount is about .25kg. Power consumption goes up or down in cruise flight at about 90watts/kg of payload weight
All the above flown with a .25kg payload. A lot of different flight configurations were flown to test power consumption and determine best cruise efficiency and payload capacity. Useful payload capacity (from the single battery configurations) is 1.5kg (3.3lbs), which can consist of either additional batteries (twin 12S battery configurations) with additional payload. Or just additional hardware payload (cameras, magnetometers for mining surveys, etc), either rigid mounted or tether towed.
Most efficient cruise speed is 15.5m/s (about 33-34 mph) @ 1,600 rpm headspeed. Measured power consumption in level cruise with this helicopter is 620 watts @ 15m/s and 6.2kg (13.7lbs) takeoff weight. It had a maximum tested range of 30.6km (19 miles) with 1kg payload in that configuration, flying a combination of level cruise and climb/descent typical with following terrain elevation on autonomous survey flights. All flight testing was done on a 1.6km course that has level cruise sections, and altitude changes to measure climb/descent performance in cruise flight.
I had to modify the ESC cooling system in the heli for UAV use. The ESC does not get hot due to load. It gets hot due to duty cycle (very long running time) and the fact that it is in still air under the canopy. A person could put a cooling fan on it (small muffin fan). But this helicopter is not designed for hovering work. It is designed for cruise flight. In cruise we have plenty of air flowing over the helicopter that can be used for cooling, so why waste power off the servo rail to run a fan?
I cut a hole in the nose of the canopy as an air intake and made internal ductwork in the canopy that routes fresh air over the ESC. That lowered it's operating temp by 30-40 deg F on 30 minute flights at full payload.
The cameras used on survey helicopters are rigid-mount. Gimbals are not used or needed. Gimbals don't stand up to the wind load at the cruise speeds that helicopters are capable of, compared to multi-rotor drones. Mulit's consume increasing amount of power the faster they try to go, at a quite radical frame angle. This is not a problem with helicopters. Heli's will cruise at 30-35 mph with less than 4 degrees of forward pitch, and helicopters become more efficient in cruise than they are in hover. The wind load on a gimbal hung camera makes the servos on the gimbal shake, burns them out. Plus the rubber jello ball things that suspend most gimbals don't hold up on helicopters that can make high-G turns at the end of survey passes to come in on the next pass right on target speed for the pass. The G-forces and weight of the camera will rip the rubber jello balls right off and then your camera and gimbal is hanging underneath by the wires. Don't ask me how I know this.
The helicopter is capable of any angles a gimbal can do, and they are way smoother and more stable than multi's or fixed wings. This Synergy E5S happens to be so silky smooth it can shoot video with a solid-mounted GoPro with no rolling shutter effect in the video.
So that's most of the details of this UAV build. I'm very happy with it. This is that video I mentioned - this was flying a short figure-8 test course on autopilot on the maiden flight flown with the autopilot. I was testing stability, proper turn coordination, and PID settings in the autopilot's rate controller. The software is capable of full automatic takeoffs and landings. However, I never use that feature. We don't trust autopilots to take off and land full-size helicopters, and I don't trust 'em to do it with RC ones either. I fly in manual mode for takeoff and landings. The autopilot is only to reduce the pilot's workload in cruise flight so he/she can concentrate on running the payload or monitoring the operating parameters of the aircraft.
https://youtu.be/3INaa3Mt-3U