Excellent question and no it hasn't really been asked before. When you suspend the fuselage from the mast and move the cyclic you are right the fuselage tilts in the opposite direction. Pushing forward tilts fuselage nose up not nose down as you see helicopters do when travelling forwards, but I think the weight has moved forward despite the nose up fuselage. There is much more to it though, an old dispute concerning how gyrocopters fly. Are they weight shift or cyclic control, this has been debated for years but the general opinion now is they are cyclic control and not weight shift. This is because aerodynamics control the rotor and overpower gyroscopic forces, it works like power steering on your car. There is a cyclic change in the angle of attack of the blades and this moves the rotor around despite being fixed pitch rotors. Another fun fact is that gyroscopic forces cancel with counter rotation, the forces are equal and opposite, the is no gyroscopic effect with counter rotation if the revolving parts are Rigidly mounted to the connecting shaft. My rotor blades are not rigidly connected to the rotor shaft and so this doesn't apply but if they were massive forces would be transmitted through the rotor shaft. I explain a little bit about this in learning to fly part 4 but I think it needs to be fully explained. Sorry if I've added to the confusion, it's a complicated subject and I struggle to understand it all.

@@Ben-Dixey thank you for your reply. 'Pushing forward tilts fuselage nose up...' oh! I was expecting that it did the opposite. Any way if it's doing that then the CG would move forward. I suppose an important consideration is which point does the whole system tilt around? Presumably the CG location in the Z Axis and not the pivot of the rotor as one might at first assume. Thanks again. I'll do some more thinking

@@SunilSundar I think that's right, in the air as far as I know aircraft pitch and roll about their CG. Not the same as having the machine hung from the mast via a rope in the workshop.

@@Ben-Dixey , the best way to think of your system is that the cyclic control is acting directly on the blades rather than having a separate swash plate. Dynamically it will be the same, although you have more inertia than a swash plate. Both systems control the blade pitch with 90 degrees azimuthal lead, then rely on blade flap to produce the desired thrust vectoring. So I prefer to think of a teetering helicopter as having a servo mechanism for thrust vectoring. In other words the rotor leads the dance with the mass centre following, whereas a weight shift design would rely on pilot directly revectoring a non teetering rotor. For absolute control authority, my preference is for rigid or articulating rotors, as these don't rely on positive gee for roll and pitch moments - but that introduces cost & complexity you sensibly don't want. Obviously there is also the collective pitch issue - but lets agree to disagree on that for now. 🤔

@@martingarrish4082 Sounds good to me Martin, there is one other factor with my machine though and that is the distance from the mast pivot to the top rotor. It has to move laterally more than it tilts and I think this weight shifts machine. I can feel a difference in control force if an abrupt correction is needed. Most of the time the cyclic is easy to move, bit like an R22 but it's not when more abrupt corrections are needed. I think it can be got used to as it worked on the airscooter 2, will hopefully see in future.

Lovely comment Bob, that's really nice to hear from someone who was involved in the industry watching my video. I find the subject fascinating and hope to continue learning and making videos. Phase lag is little understood and many people think it's gyroscopic precession including some big KZfaqrs. Although related there is more going on and it needs to be explained. I'm not at a position to do that yet but I hope to understand it well enough to demonstrate what's going on.

Thanks Mark. 😊👍

Appreciate that 👍😊

😊👍

Thank you, it would be so good to achieve building a machine that is reliable. What a great toy it would then be not to have to worry about mechanical failure. I think this is something that gets overlooked. Challenge 1 build a machine that works. Challenge 2 teach yourself to fly it Challenge 3 make it work reliably Few homebuilt helicopters achieve challenge 3 and that is a goal worth perusing.

Hi Peter, yes there is a way to introduce a rubber cushioning drive. I would just have to make sure it didn't help cause a failure like the last one did. I've got a feeling that it might not be necessary but it's only a feeling. If the torque can be measured then I can tell if a cushion drive is required. Thanks

Exactly, we don't get to see the long development process of making a flying machine reliable. Its a hard task but ultimately will be very rewarding. It's amazing how much stored energy there is, enough to hold the machine in the air for any length of time I find an incredible fact. What did you fly ?

Hi Ray, you could be onto something with the teeth wear. Your explanation makes much more sense than bending teeth. If this is correct then the belts and other components didn't experience a big shock load. It looks as though all the teeth are irregular and that would correspond with tooth wear rather than a sudden drive jam. This is very good news and thanks for your superb comment. I also think your are right about the flex in the boat, in the water and in the air the boat can move with the wings, at the moment it's rigidly mounted to the trailer so flex is evident. I'll will proceed without worry. 👍😊

Thank you 😊

😊👍

😊 love Colin furze, he made a hover bike and it worked. Would love to see him attempt something else on similar lines. Hope to have contact with him one day, that would be great.

👍 thank you

Hi Dk, Steel box would work and it wouldn't be too heavy. I think the seam could be sanded off with an air file, it would solve the potential issue with needle rollers indenting into the material. The boat is coming together and I'm pleased so far, pretty fun seeing a wing on, its going to be quite a ride I think.

Hi Tyger 😊 interesting comparison. A lot can be gained from the study of models. There are some great videos on YT, the birth of the bell helicopter where Arthur Young built some fantastic models. Your 700 series must be a beast but I suspect there will be a size difference in the bearing, but if enough of the 700 series sprag bearings were used it would power my helicopter. It's actually quite a low torque application being right on the engine, high speed but low torque. I was amazed at the small size of the driveshaft on an MD500. Watching Cletus McFarland and heavy D's videos.

Hi, a mechanical fuse does make sense as long as it only breaks during an extreme event like a what I experienced, worth considering. A couple people in the comments suggested that the teeth are not bent but worn to give the illusion of bent teeth. I'm not sure which is right and I can't tell visually. Things are getting exciting with the boat, a different challenge but there's a nice bit of crossover Knowledge that can be applied. All good fun that's for sure. A steel sleeve inside the aluminium pulley might work. 👍

Good observation, I think you're right.

No misery at all, I've loved every minute of this entire project, hopefully many more years of enjoyment to come.

Thank you 👍

@@Ben-Dixey you are welcome my boss

Titanium is a fantastic material choice, never made anything from titanium and it's very expensive but it's the right choice. It's only the cost that would prevent me using titanium.

A 6 degree wash out would provide more lift for sure just harder to make the blades. The blade route is definitely stalled and creating drag. Leather would be the material used before modern clutch material ?

@@Ben-Dixey Leather? Absolutely. I havd it for bike brakes ,unreplaced since 2003. It was early clutcn and brake material. Crazy glue it on. Acetone takes it off.

Six degrees sounds a little excessive.

Thanks for the educated comment, I was hoping it was belt wear but having measured the teeth where the belt runs and measured the teeth where it does not, it seems there is only 0.2mm difference. I don't think this is enough to give the illusion of bent teeth. The material was 6082 t6. I'm looking into changing to a chain drive instead.

@@Ben-Dixey okay, I designed ultralight planes for a 20 years period (search YT for fresh breeze flying car, I designed and mainly built it), and I prefer belt drive over chain. A chain usually has a lower power2weight ratio. And since it is stiffer, peak loads in your drive train will be higher. You may have benefits from something like a hardy clutch in your drive train, it may add some elasticity and also reduce peak loads. And check belt lubrication. And if you want to add a freewheel, try to integrate it into your biggest pulley, with a big diameter, since torque is always critical. Cheers, Stefan

@@StefanWode Cheers Stefan, well done for designing/building a flying car, I'm sure it was difficult to keep the weight and strength at acceptable levels. Seems to work well from the videos. Is it still in use ? I take your points about chains. With the belts I'm using a slightly smaller than recommended drive pulley size. Lowest advisable was 22 teeth, I'm running 20 teeth. This was changed to increase lift but I don't know what effect that would have on belt life. The tighter radius would increase the amount of bending. On the bottom belt it's also being reversed more times which is also decreasing the life of the belt. I Would also like to increase the ratio again to perhaps 7.5:1 currently running 7:1. I can't do this with the belts without making the large pulleys even bigger. Currently 350mm diameter and expensive to make. Chains therefore look an attractive option but as you say will be a less forgiving drive with regard to torque variations. Some motorcycle drive sprockets have a rubberised insert. If I can find these at the right size that would be ideal.

Hi, your suggestion would simplify things and I think would work. Have you seen YT videos on the helitrike ? They abandoned the idea after some dynamic roll overs but I think they went the wrong way about testing the machine. It's a lot of work to change it but worth considering and a couple other people would like to see me go down this route.

Interesting comment, I have a friend who encountered the same thing. Long driveshaft to prop, I think his solution was to have a flexible drive coupling. Is some sort of flex within a driveshaft not the answer?

Your belts are as bad as a torsion driveshaft . And as you have a frame to hold the axles and pulleys in place that also has a strong influence. Do not think your frame I so stiff that it does not impact tv. It does!

You should at some point measure your tv at an easy to access location. I made a simple strain gauge to adc to small arduino like cpu and then Bluetooth module to send over the air the measured tv. I did measure that at 3khz, and recorded that on a pc. I could do run ups from starting the engine all the way to 2700rpm to see all the resonance and they were there! A 4 cylinder engine will produce a fundamental excitation at 2x and a bit 4 and 8 x. Anything related to imbalance will showup at 1x.

I glued a quad 1000ohm strain gauge bought on AliExpress in the driveshaft at 45 degrees to capture the torque and that works like a charm . Those are only a few euros.

What I did in the end is to tune the engine flywheel to make sure the tv happens at an rpm regime that is not used while flying. 1600 rpm is low power and unusual in flight. You also have its 2x harmonic at 3200rpm but lycoming do not go over 2700

Cheers John, my centrifugal clutch disengages at 2500 rpm, so driveshaft and rotor rpm would have to drop to this level before it became free wheeling. Is this the case with yours then ? Did you consider a dog clutch ? I'm not sure which is right on my teeth, bent or worn. Material was 6082t6

@Ben-Dixey My centrifugal clutch is a hybrid between a centrifugal clutch, sprag clutch, and roller cam clutch. It initially engages under centrifugal force acting against springs, then friction between shoes and drum allow it to rotate partially acting against other springs. The more torque applied the greater the friction. When torque is disengaged the shoes travel in the same direction as the drum to a small degree allowing the engine to decouple if the engine was to seize. It only needs to decouple immediately if the engine seizes or if you shut down the engine. If the engine loses power but is still maintaining rpm it does not need to decouple completely or quickly. Honestly a sprag clutch is completely unnecessary for what you are doing, especially this early on. If you have a stable aircraft and want to venture higher look into it, but for now just hovering in ground effect it's not necessary. I wouldn't be surprised if the clutch you are using doesn't slip a little also, I'd do away with the centrifugal clutch also. Pull starting it might be difficult, though you could just crank the rotor from the seat. Install a decompression valve for easy start up. Ditch electric starter, clutch, make it as light as possible.

@@chippyjohn1 Ah, clever designed centrifugal clutch there by the sounds of it. 👍 That's correct I don't need one for what I'm doing, I would just feel happier with one, if the engine seized rotor inertia would destroy everything. Not planning on another drive shaft failure but it would be safer with an overrun. If the engine cut out it would also allow what little inertia there is to cushion the landing.

Yes it is quite a large span, aspect ratio is still quite low at 4 but I was worried about the wing loading. 28kg/m2 is what I've gone for, I'm still doubtful if there is enough wing area for the weight with two people. Hopefully the effect of the ground will prevail. I would love to reduce the span of the wings, increasing the chord is possible but not so easy. Adding ailerons would help. The dihedral was just to add clearance to the water and is detrimental to ground effect.

Have the free run on the engine ? Yes it could be designed into the clutch and would work just fine. I decided I wanted to have it inside the pulley so that if the driveshaft jammed it wouldn't affect the rotors. Having the freewheel on the engine wouldn't isolate the driveshaft from the rotors.

@@Ben-Dixey indeed thats a must the drive shaft is scetchy in your disign. One way bearings would work with good press fitting and locktite. Just make sure both rotors keeps free running losing one is as bad as losing 2 . I would never used belt system as a simple gearbox would be less heavy and more secure. And i think less work aswell. How ever i like your project and respect mate Few people would open a can off worms as building a helicopter. Respect.

Your concerns are of course valid but this all depends on the amount of torque being applied. Normal operation is only 55lbft, very low and very under-stressed for the size of the box. You may be still be right but it has to be proved. If you took the load on each roller it's only 2.4kg, you could push that load with your finger. Is that going to indent into the tube? I can work out the twist on the tube but I haven't yet, I can also test it with a torque wrench, early indications suggest there won't be any twisting of significance. Having said all this there is another option called a ball spline, I'm going to investigate this option. Precision ground all steel, can take the torque and rpm but I don't know the price or weight.

I don't know how many rollers you have, but I think you are calculating the force over all the rollers evenly in a linear plane, however it will encounter torsion. Just like the square drive on a ratchet, the force/torque is applied to the corners. If you look at a socket that has been used on a rattle gun you will see the corners are loose. If you have four rollers that are 40mm each in length the actual contact area might only be 4mm. Considering the needle roller is round the actual contact area of that is a hair line. 0.4mm of contact per roller, or 1.6mm per circumference.

At 75nm of torque, calculated at the corner radius that is 145kg per 1.6mm2 or 885mpa for one set of 4 rollers. I'd be looking at 100+ sets of rollers, that's a lot. The rollers at the beginning will experience most of the pressure though. I talk to much, best of luck, just try.

@@chippyjohn1 There are 280 rollers in total but only half will be under load and as you say only the extreme corners and the first few will be taking load. I'll make an insert to test the idea under repeated torque loads and see what happens.

@Ben-Dixey what's the bore and stroke of the engine on the helicopter? Just curious.

Not 100% sure, but I think you could be right about abrasion, makes much more sense. Not something I thought of, been trying to find examples of pulley wear to compare. 👍

​@Ben-Dixey If the belt teeth are loose in the pulley valleys it is wear, if it is tighter it is bent. Looks bent to me.

No it can't. Only to be used for hovering and perhaps some slow forward speed.

Thanks for the comment, yes I am worried about that, I'm not sure if it will be a problem or not. I might be able to do some testing on this. A rough calculation suggests the load on each roller is very small but it's not that simple as the load isn't evenly distributed across the roller. I'd like to work because it's a very light weight solution. There is another option which I'm looking into but I think it will be heavier and definitely more expensive. Galvanic corrosion would need an electrolyte ?

2m

@@Ben-Dixey obrigado, tank you

What brand of outboard boat engines do you have in your country ? Mercury, Yamaha, Suzuki, Honda? All the 2 stroke ones might be suitable.

@@Ben-Dixey شكرا لك

@@Ben-Dixey هل ينفع محرك سياره ٢سرندل او ثلاثه سرندل

لقد رأيت تعليقاتك في بعض الفيدوهات انك تنوي صنع الشفرات من الخشب في النموذج الثاني هل الخشب قوي ام انه سينكسر

@@tahaali4219 Wood can be used for blades but if using that material it needs to be reinforced with a metal spar screwed to the wood. I'm not keen on using wood anymore but Kaman helicopters still use wooden blades.

To have fun is the main objective.

What's the final objective of your comment ? Let us mortal souls know how to use the capital letter button ?

You aren't forced to watch it. People waste hours watching sports, I much prefer watching Ben.

Do you say the same to people who take part in extreme sports? TT Motor bike racers, Base jumpers, off piste skiing ?

😊 thank you!

@@Ben-Dixey no, thank you. It's people like you that make youtube a wonderful place to live in the dreams of others, not to mention your ingenuity, conviction, and insanity for tackling a difficult engineering project.

Hi. I measured the teeth width at the peak with a vernier and they only deviate 0.2mm from where the belt runs to where it does not. Sadly I think this means that the teeth are in fact bent. I would have much rather it was wear but I don't think 0.2mm would be enough wear to give that illusion. Thanks for the comment.

@@Ben-Dixey Chain sprockets have a much wider pitch than these belt pulleys, and those only need 0,5 mm wear to give a totally bent appearance: The human eye generally is sensitive to asymetries. I wish I could paste a pic into this message. But here is what you can do to check: The pitch is known. On a new or undamaged pulley the space between the teeth should be pitch minus tooth thickness, agreed? If the teeth are bent, that distance remains unchanged because the entire teeth has moved. If it is wear, that distance has increased because one side wore off and the other side is still in its original shape and position. Check if that gap has changed... The thing is, I have a VERY hard time believing that toothed belts actually can bend the pulleys' teeth. I have never ever seen such a thing and I have treated that kind of beltdrives like an ogre. Much more likely the belt gets stripped than the teeth of the pulleys bent. Even on shockloads.

@@rigididiot I tried measuring the distance between the teeth, difficult to measure with any consistency. However worn teeth makes more sense to me. I took the belt spacer off and looked down the length of the teeth and you can't see any deviation along the length. You would have thought if the teeth were bent you would see a wave where the belts were not running. I'm leaning towards tooth wear now, if I was to cut a slice of the pulley I could put it up on the profile projector, trace the tooth profile and the reverse it. That should be able to identify where the tooth is distorted. I probably won't be using this pulley anymore so wouldn't mind sacrificing it to try and find the answer.

@@Ben-Dixey There should be ways to slice the pulley and look at the metal grain to see if it was deformed or worn, but I am not familiar with that kind of metallurgy. I know metallurgy labs can do that kind of thing, but no idea if it can be done in a DIY environment. Anyway, it is in general important to retrieve the real reason for that deformation, because engineering based on things that did not happen, won't help you solve the problem. As for your sprag clutch, I am not sure if one-way roller bearings are the way to go. Those things deteriorate over time, and the type you showed, well,they are only double the size of what is in my RC helicopters, and those things only transfer maybe 3 hp.... I would tend to think that a ratchet type AR clutch would be a much safer option. I also still have my reservations on the square section drive shaft, despite the linear roller bearings. I would look into the solutions car designers use for drive shafts.

@@rigididiot That's right, it is important to know. Lots to consider and testing to do before decisions can be made. I have reservations too about the driveshaft and sprag clutch ideas, I think these can be tested before going forward though. You can get a driveshaft that Volvo make that uses a ball spline, this would solve my problem. Way over engineered and heavy for what I need. Thomson make a ball spline that's suitable, my guess is it wont be less than £500. Making my own would be challenging.