From a safety point of view, would it be possible to use magnets on a rotor that isn't directly coupled to the blades? Or is direct control over the movement important? I believe my ceiling fan uses this principle. Means that if I stick my hand in the blades, the motor can keep going as normal, and I don't do damage.
I like this idea, however the speed of the fans even at 50rpm still makes it dangerous. Suggest some sort of mesh over the fans, chicken wire comes to mind
The fans are slow enough that even if you could reach them, they would likely cut, but very unlikely to break or remove digits. They are also covered with a cage and deep inside the exhaust cowling. They are reasonably high and deep enough that no kid could get an arm in that far without standing on something and a deliberate effort to do so. When i have it on display, it will have bollards and tape around it to keep people at a safe distance with warning signs. The engines will also NOT be started without my personal supervision and eyes-on.
just go something like what aXis linked. I can 3D print both rubber (TPU) belts and pulleys for you to your specifications if you need! TPU belts are very strong and light - I used them with great success in my stepper-driven DSLR pan/tilt device. You could also have them without teeth and use a tensioner, providing a way for it to slip in the event of an unexpected stall and protect both kid-fingers and your cheap geared motor
+1 to the belts and pulleys and you can then drive all three of the one motor. Something like mount the motor centrally and have three separate belts/pulleys off of it. I imagine it would be easier (/cheaper?) to source smaller belts and pulleys than attempting to make up a serpentine belt system or controlling individual motors. Spin up one of the motors and check how it decelerates when power is removed; they have a reasonable armature rpm and don't go to a dead stop. Although depending on pulley sizes and as the 'turbines' are 600mm dia it could still be too fast, hopefully a belt system would slip. Anyway, see what you think.
It's very hard to run those brushless motors at low speed. Even low KV motors are hundreds of RPM per volt. Not to mention, brushless ESC's have a startup interlock so you have to program a special sequence in the arduino. Simple brushed gearmotors will be perfectly adequate for this job, preferably with a magnetic coupling for safety.
to be clear I was only suggesting pulleys and very short belts as an easier method of coupling the motor to the fan than coming up with a direct drive adapter of some kind, with the added bonus of gearing it down a little at the same time
This is what I would try first. The fan might need to be close and support reasonable static pressure but then the only danger is in the inertia built up in the blades themselves.
'air coupling' isn't a solid idea: Given the size of the fans he wants to turn, at a minimum he'll need something like five or six 120mm fans to get enough airflow going - keep in mind resistance, tolerance around the fan, tolerance between the blades. The fans will cost way more than a simple cheap geared motor It will require bearings on the fan hub instead of just a sleeve with some grease, as would be adequate otherwise it assumes cutting holes in the case to fit fans is compatible with the design and aesthetics Fitting a geared motor surely has to be less work Multiply the work and cost by three fans...
I'm not really try to argue about which is the correct way to skin the cart but I don't think airflow over the entire metal blade will be needed given the required speed (I.e. 1 rotation per second or so). Particularly if the fan can be mounted close to the blades and can supply a bit of static pressure. A smaller, fast running fan would produce more thrust over a smaller area and might actually be better. It wont have to act over the entirety of the surface area of the metal blades in order to make them turn. As long as the fan can draw air from somewhere external to the duct. Happy to be proven wrong here though. There's a reason I said it's what I'd try first and not "This would be 100% perfect guaranteed". Most people on here would have fans knocking around and could test this quite easily. Literally hold the fan close to the blades and see what happens. Indeed it does but motors and the required belts or coupling would potentially have the same issue. Maybe if everyone's time is free (and even then. fans are a few bucks on ebay) but mounting and coupling the fan is going to be way easier IMO.
right I see what you mean now, a fan right up against the fan blades would have a decent mechanical advantage I was picturing fans, say, around the circumference of the engine casing and relying on the average flow of air across the whole fan still would be a yucky aesthetic, but maybe could be mitigated with a matt black mesh grille over the fan?
Great suggestions everyone and thank you. I have lots of options to think about. Some won't work because of the way the engine nacelle and it's framing and formers are set up. With lots of changes, other options may work, but it's a lot of extra work. I think the geared motors are the simplest for now. I can set up a slip-notch coupling that engages only one way, so once the motor stops, the fan can continue to spin and disengage, slowing down naturally. I'll keep you posted. Baz.
Just use large stepper motors, maybe with a 1 to 6 ratio belt drive. Low torque so fingers don't get sliced, can be finely controlled from zero to max speed. Max speed can be controlled simply with a program inside an arduino, you could then integrate it with other special effects as well. Using steppers would allow you to overcome inertia with a low ramp up.
Steppers are horrible for inertial loads due to cogging and lack of feedback. You need an encoder to have any confidence you are driving correctly. They also suck at high speed, although not so much of an issue here. Keep it simple, brushed DC and some kind of compliant coupling for safety.
aXis with the belt step down that will be fine. Cogging is fine, and so is lack of feedback, you just want them to spin slowly with low torque.
