|
Steve Broad
|
|
|
Group: Forum Members
Posts: 408,
Visits: 4.1K
|
+x+xSteve, maybe I missed something, but to my understanding, the boom isn't going to be thinner in the middle. The diameter never drops under 30mm. Except for design and aerodynamic decisions there is really no reason why it should be. So you could easily pull out the mandrel to the bigger side. To get something really light, you should keep the material stresses constant over the length of the part, wich will take you to a smooth wall thickness reduction towards back rotor. The tapered crosssection basically works the same way, but to my estimation doesn't completely compensate the difference in crosssection loads. The anti torsion layer however should be constant over the whole part. No, It's me misinterpreting the photo, it looked to me that the centre section was thinner than the ends but it was just an optical illusion :-) How do you get a smooth exterior finish when using a mandrel, without adding additional weight?
|
|
|
|
|
Steve Broad
|
|
|
Group: Forum Members
Posts: 408,
Visits: 4.1K
|
+xSteve, maybe I missed something, but to my understanding, the boom isn't going to be thinner in the middle. The diameter never drops under 30mm. Except for design and aerodynamic decisions there is really no reason why it should be. So you could easily pull out the mandrel to the bigger side. To get something really light, you should keep the material stresses constant over the length of the part, wich will take you to a smooth wall thickness reduction towards back rotor. The tapered crosssection basically works the same way, but to my estimation doesn't completely compensate the difference in crosssection loads. The anti torsion layer however should be constant over the whole part. No, It's me misinterpreting the photo, it looked to me that the centre section was thinner than the ends but it was just an optical illusion :-)
|
|
|
|
|
quinn
|
|
|
Group: Forum Members
Posts: 155,
Visits: 992
|
+x+xoh, i missed a bunch of posts. yes, the boom is tapered the whole length except for a bit of zero taper at the ends, so yes a mandrel would come out. hanaldo, you make good points about the bladder, indeed i would be pretty much screwed if i went through all the layup and then had a leak when inflating. a lot of risk there. sounds like prepreg seaming is no big deal, it make sense now that you explain it, but still the oven size problem. i badly want to make the sleeve work, aside from saving the few grams, it would sure look sweet having it absolutely seemless. i know saving a few grams sounds rediculous, but in this case it matters. With performance rc helicopters used for aggressive 3d flight, you want absolute minimal moments of inertia for snappy responsive cyclic inputs. removing grams at a far distance from center of gravity (out on the boom) makes a huge difference. im using titanium fasteners in the tail case to save like 3 grams and its worth it. All that being said, if the sleeve isnt gonna work, thats just how it is i guess. and like mentioned, i could end up heavier by trying to do wet layup even though theres no overlap. maybe i should consider the homemade oven with light bulbs idea for the prepreg with seams. i imagine you need to monitor this closely to maintain proper temp, or have thermostat controlling the light bulbs You could also consider putting electric elements in the back face of an alloy mould? Search the poster Bladerunner in this forum, I think he has some posts about it. No oven needed. ah, interesting. yes i can see how that would work. some heating elements, temp sensor, and a cycle to keep needed temp. ill look into that for sure. its not that im not willing to spend a couple hundred bucks for really nice molds, its just the fact that its a prototype that may need design adjustment. i guess its probably most likely that the mold design will be fine how it is, but maybe layup and thickness need to be adjusted which doesnt require new molds, so ill consider going aluminum. definately would make a really nice finished product. sounds like prepreg in a 2 part aluminum mold with either vacuum or silicone pressure its going to give the best results within my abilities. sleeve or filament winding i think has higher potential, but not within my abilities to get it to its full potential
|
|
|
|
|
Fasta
|
|
|
Group: Forum Members
Posts: 468,
Visits: 3.5K
|
+xoh, i missed a bunch of posts. yes, the boom is tapered the whole length except for a bit of zero taper at the ends, so yes a mandrel would come out. hanaldo, you make good points about the bladder, indeed i would be pretty much screwed if i went through all the layup and then had a leak when inflating. a lot of risk there. sounds like prepreg seaming is no big deal, it make sense now that you explain it, but still the oven size problem. i badly want to make the sleeve work, aside from saving the few grams, it would sure look sweet having it absolutely seemless. i know saving a few grams sounds rediculous, but in this case it matters. With performance rc helicopters used for aggressive 3d flight, you want absolute minimal moments of inertia for snappy responsive cyclic inputs. removing grams at a far distance from center of gravity (out on the boom) makes a huge difference. im using titanium fasteners in the tail case to save like 3 grams and its worth it. All that being said, if the sleeve isnt gonna work, thats just how it is i guess. and like mentioned, i could end up heavier by trying to do wet layup even though theres no overlap. maybe i should consider the homemade oven with light bulbs idea for the prepreg with seams. i imagine you need to monitor this closely to maintain proper temp, or have thermostat controlling the light bulbs You could also consider putting electric elements in the back face of an alloy mould? Search the poster Bladerunner in this forum, I think he has some posts about it. No oven needed.
|
|
|
|
|
quinn
|
|
|
Group: Forum Members
Posts: 155,
Visits: 992
|
oh, i missed a bunch of posts. yes, the boom is tapered the whole length except for a bit of zero taper at the ends, so yes a mandrel would come out.
hanaldo, you make good points about the bladder, indeed i would be pretty much screwed if i went through all the layup and then had a leak when inflating. a lot of risk there. sounds like prepreg seaming is no big deal, it make sense now that you explain it, but still the oven size problem.
i badly want to make the sleeve work, aside from saving the few grams, it would sure look sweet having it absolutely seemless. i know saving a few grams sounds rediculous, but in this case it matters. With performance rc helicopters used for aggressive 3d flight, you want absolute minimal moments of inertia for snappy responsive cyclic inputs. removing grams at a far distance from center of gravity (out on the boom) makes a huge difference. im using titanium fasteners in the tail case to save like 3 grams and its worth it. All that being said, if the sleeve isnt gonna work, thats just how it is i guess. and like mentioned, i could end up heavier by trying to do wet layup even though theres no overlap. maybe i should consider the homemade oven with light bulbs idea for the prepreg with seams. i imagine you need to monitor this closely to maintain proper temp, or have thermostat controlling the light bulbs
fasta, roll wrapped 45 degree tube would absolutely work to directly replace the tube im using now to solve the torsional strength issue, but i would still need the boom supports. im taking it a step further by making a tapered boom to eliminate the boom supports. solves the torsional strength issue and also reduced weight and much nicer look.
sounds like the expanding silicone is a good replacement for the bladder, but then baking is required and im back to needing a custom built oven
|
|
|
|
|
quinn
|
|
|
Group: Forum Members
Posts: 155,
Visits: 992
|
+xYou are overcomplicating things by a lot. First: a 45° biax cloth (or 45° cut weave) wrapped around the core with just 10mm overlap will give you exactly the same stiffness than a similar thickness. I would just make it on a positive mould/mandrel/plug. If you got a cnc, you could easily machine if from an aluminium piece much smaller than it would be needed for the mould. Than laminate the sleeve over it, compress it by wrapping it wich tape and remove the plug after the resin is cured. Filament winding would give you the ultimate result however. Maybe you could make the aluminium mandrel, and speak to someone with an winding machine (someone here has got an x-wonder, if I remember right). This should not be too expensive. Yes, machining an aluminum positive would definitely be affordable. It's like 1/3 the amount I would need for machining 2 negative halves. I do like the idea of doing it that way, but a couple questions. The boom goes down to very little taper for the last 300mm or so, zero taper for 50mm. What kind of force is it gonna take to get it out of the mandrel? I know they do actual tubes this way but use a hydrolic press to get the mandrel out. I guess I can machine some kind of apparatus that the base of the boom sits against and use a long bolt to pull the mandrel into it as it's tightened or something like that. Other question, when pulling sleeve over mandrel and compressing with tape, what kind of surface finish could I expect from this? I'm assuming I would need some sanding and clear coats?
|
|
|
|
|
oekmont
|
|
|
Group: Forum Members
Posts: 550,
Visits: 27K
|
Steve, maybe I missed something, but to my understanding, the boom isn't going to be thinner in the middle. The diameter never drops under 30mm. Except for design and aerodynamic decisions there is really no reason why it should be. So you could easily pull out the mandrel to the bigger side.
To get something really light, you should keep the material stresses constant over the length of the part, wich will take you to a smooth wall thickness reduction towards back rotor. The tapered crosssection basically works the same way, but to my estimation doesn't completely compensate the difference in crosssection loads. The anti torsion layer however should be constant over the whole part.
|
|
|
|
|
Fasta
|
|
|
Group: Forum Members
Posts: 468,
Visits: 3.5K
|
Seems the simplest method if you were happy to use a bought pre preg tube would be to have the tube supplier make your tube to your laminate specs with the 45 deg torsional layers you want? Or you could supply a pre preg tube maker with your own custom alloy mandrel? One good method for internal pressure in moulds is simply a solid silicone block. With your female moulds, layup a mock laminate from fibreglass or other like sheet wax to simulate the correct thickness plus a bit of extra clearance, close the moulds and pour cast a silicone block. Then you can do a pre preg laminate with overlapped split join and insert the silicone block when you close the moulds. No bags, no vacuum needed, the silicone expands a lot with heat. Of course this will only work with pre preg. $200 of alloy is very cheap and high quality mould compared to the effort of trying to make a composite split moulds that are of lesser quality and durability IMO.
|
|
|
|
|
Steve Broad
|
|
|
Group: Forum Members
Posts: 408,
Visits: 4.1K
|
+xIt is 30mm in diameter at one end, and 40x48 at the other. In between the shape should continuously change diameter, because that is the smartest way to do it. Should go off the mandrel quite easily. With the bladder technique you will likely not get that very last bit of fibre content you would achieve with positive methods or prepreg. This easily negates the advantage of getting rid of a small overlap. So how do you pull the mandrel out if the ends are bigger than the middle? I obviously don't understand how mandrels work :-)
|
|
|
|
|
oekmont
|
|
|
Group: Forum Members
Posts: 550,
Visits: 27K
|
It is 30mm in diameter at one end, and 40x48 at the other. In between the shape should continuously change diameter, because that is the smartest way to do it. Should go off the mandrel quite easily.
With the bladder technique you will likely not get that very last bit of fibre content you would achieve with positive methods or prepreg. This easily negates the advantage of getting rid of a small overlap.
|
|
|
|