_Rade

I'm having the same problem. I've uploaded the craft file but it doesn't let me publish it.

@panzer1b What is the location of your base? I'm asking because it's really hard to find such flat terrain on Duna. Also from my tests I've found out that breaking ground propellers don't seem to work that well on Duna. I couldn't get the plane to go faster than 100m/s and that was at low altitudes usually below 1000m, at about 3000m plane couldn't even move fast enough to takeoff from the ground.

Structural fuselage to act the same way as the cargo bays (drag occlusion) for the small objects placed inside the fuselage. It's really annoying having to use fairings on airplanes in order to be able to shield parts like fuel cells from the aeroforces in flight.

I've updated couple of my older planes by replacing the reaction wheel engines with the new rotors and propellers. Performance provided by these new parts is amazing. While using just 2 smallest rotors to run 3 bladed contra-rotating propellers I was able to get the planes to go ~250 m/s at low altitude. Link to the craft files: VB 10 Air Racer

I'll try those panels, only problem is they are wider than the decouplers so they might not work so well.

@sgt_flyer That thing looks amazing, real shame about the in game performance. Did you try to replace every second hinge with some other part, it wont reduce the part count but it would cut the number of hinges in half and that might help. Anyways here's my new tracked rover Only 20 hinges for each thread in order to keep impact of the hinges on the performance to the minimum. Also I've redesigned the sprocket wheels, now they are similar to the sprockets on the t-34. This have considerably improved how the threads interact with the sprocket wheels. The max speed increased to ~7 m/s and it's now mostly limited by the impact tolerance of the decouplers I used for the track links. Here's the link to the craft file https://kerbalx.com/Rade/Crawler and a link to the video on Reddit

Here is my take on the tracked propulsion. By using the decouplers to make the track links I was able to cut down the number of hinges by more than a half, each track is 38 hinges and decouplers. Hinges are in the middle of the track and along with the double wheels they serve as guide for the track. Front idler wheels are mounted on a pistons which can be used to adjust the tension of the track. Whole thing works reasonably well as long don't apply to much torque to the engines, max. speed is about 5 m/s. Main problem I've had with it is that all these hinges really slow down the game. The other problem are the KSP's stretchy joints, which make it really hard to properly tension the tracks, this can cause the track to either stretch or skip on the sprocket wheel and in turn this can de-track or break the tracks. Here is the link to the craft file https://kerbalx.com/Rade/Tank and a link to the Reddit post where I have posted a video of this thing in action

My new helicopter, it's not a turboprop but it does have coaxial rotors with the working swashplate and collective, also yaw is controlled via the rotor engine rpm. Craft file H-21 Frelon

Using the new parts I've made a working drive shaft that uses universal (Cardan) joints. Best part is this thing works surprisingly well, and if you wanted you could have drive shaft that makes 90 degrees turn. Only downside is there is a power loss for every joint you add. Craft file https://kerbalx.com/Rade/Kardan

You could just attach a cubic strut to the rotor and then put 4 more cubic struts around it to hold it.

Nice to see it works, I'm glad I could help. Also if you're using 1.25m parts you might as well go for the fairing/stayputnik combo for the bearing. It's probably the one of the best bearings for the propellers, mainly because it's really strong and fairing occludes the engine parts from the drag, which makes the planes that use these propellers much faster.

@Phantomic try using some other parts that have higher thickness than thermometers, the 5-way rcs and gravioli might work. Also it might help to include short I-beam as the root part of the rotor and then attach and autostrut all parts of the rotor to the I-beam, this should make the rotor more rigid and less likely to break.

I finally did it, I made the helicopter with a working swashplate. I've used more or less the same design I've posted few days ago. First I've tried using only the rotor with the swashplate, but it can only spin about 16-17rad/s before breaking up. That's not fast enough to provide sufficient lift for the helicopter. So I've added additional coaxial rotor that just provides the lift, while the rotor with the swashplate is used to control the helicopter and provide a little bit of lift. Also the tail rotor can slide left or right via the elevon that's controlling its position. In neutral position rotor is spun just by 2 juno engines. If you want to turn right rotor is positioned so that 4 junos act on it to spin it faster. If you want to turn left rotor is positioned out of the way of juno engines, so it idles and slows down and the torque from the main engines turns the helicopter. Craft file.

Thanks for the info on that, I've never had this issue before. I've used the cubic octagonal struts as the root part for the rotor blades. I'm gonna rebuild the blades to see if this contraption will work at higher RPM.

I've finally managed to build the helicopter rotor with the working swashplate and collective. Rotor uses 2 decouplers to act as the swashplate, and then there is one rcs ball for each rotor blade captured between the decouplers. 4 elevons bellow the swashplate control its pitch and roll and this is then transferred to the rotor blades. For the collective I've used 2 elevons to push up on the rcs ball that act as the bearing for the swashplate. You can toggle these by engaging the brakes, it's not perfect but it works. Only problem is the rotor will only spin at about 7rad/s. I've tried adding more engines and using the reaction wheels but for some reason it doesn't want to spin faster. One thing that is interesting is that if you don't decouple the rotor blades the rotor will be able to spin faster, but then the swashplate and collective don't work. Link to the craft file on KerbalX.