HobbyKing.com New Products Flash Sale
Reply
Thread Tools
Old Apr 04, 2014, 03:53 AM
Registered User
taurineman's Avatar
Wellington, New Zealand
Joined Sep 2009
384 Posts
Build Log
CNC Milled Plank

Again - this is a copy-n-paste from another board.

================================================== ==========

I've been doing some thinking about just how much time i (we) would typically spend building a nice slick sloper or DS ship; and how nearly everytime we end up custom fitting and custom making components and custom covering etc goes into one. It seems to me that there should be ways to reduce build-to-fly time. Now, i love building models nearly as much as i do flying, and i know there are people who are more one way than the other than i am, but since taking up dynamic soaring i find i spend much more time building and repairing than i ever did when i just did thermalling and/or front-side sloping. I dont have a lot of spare time after work, and my family etc, i have very litle time left. I want to use this time more effectively. Put simply, i want to claw back some shed-time for DSing or other things i like to and want to do.

I want to design and build a heavy and strong flying-wing/plank for strong wind dynamic soaring. The big design goal for it, is not necessarily to be the fastest, but to be a good balance of very easy to build and super strong. It needs to be a DS trainer/hack of sorts. It needs to be heavy (for wellington winds) for speed; it needs to be very strong (stand up to crashes as well or if not better than a well-built foamie); it needs to be simple (for a fast build); it needs to be cheap and expendable (dont have so much time or money invested in it, it doesnt matter if it crashes or explodes); it needs to come with all the bits which are all standardised (for easy replacement and cross model re-use) - eg no custom fitted parts.

So, i have been thinking about using modern manufacturing techniques to assist with this. Ive been thinking about doing the model design in CAM and then having it produced to a near RTF state at a relatively low cost if we could get a production run to spread the costs over. I had been thinking about 3D printing models in plastic and using snap-in glass spars. But i think it would not be heavy enough nor strong enough, and perhaps not cheap enough.

So, ive been thinking of CNC milling. I dont want to make moulds as that would mean hand-building each one from cloth and glue each time (takes too long). I want the CNC mill to produce the actual wing. This gels nicely with my previous thoughts of using a copy-router (in my iSoar build thread) from a master. This way though the master is in the digital realm and can be changed very easily, and i/someone wont have to spend time doing the copy routing.

So initially i was thinking of milling out of marine ply. And making the wing as two halves (top and bottom halves) which are then glued together. All the pockets for balancing lead, RX, batteries, LMAs, servos, tip-weight, and wire routing etc would all be routed in by the CNC router. It could also do the fin slot, and maybe even the elevon hinge gap. The CNC would also produce the PVC bay covers, and fin. So, all you would need to do is put the wiring in, glue the two halves together, give it a quick smoothing sand-down, cut out the elevons, fit out with radio gear etc and weight for balance, close bays with covers, hinge elevons, add linkages and fly. No covering required if you didnt want to.

So, the next step was to go and talk to someone who has a CNC wood mill and have a chat about viability and indicative costings and getting a road-map together to work towards my goal. I went to a place called "Maker Space" here in Wellington and Lee was very helpful. He suggested a few good things and perhaps to use blown PVC as the material to make them from. It looks nice and strong, but may have too much flex span-wise, so it would need a spar (again slot and spar can be CNC milled and glued in when glueing the two halves together). It should work out cheaper than marine ply.

I originally wanted the design to be semi-fus 60" plank like a Reaper or my iSoar Spruce-Goose. But it seems it may add to much cost and complexity to the design and starts to errode my design goals. So, it will be apure plank/flying wing like a bat or drongo, or sawtooth etc and probably 54". The trouble to overcome is getting enough weight in to balance out the solid wood/PVC in the rear part of the wing. Again Lee suggested we cut airpockets into the parts of the wing to keep rearward weight down. Im pretty sure we can do it. Flying weight may well be 3 - 3.5kg? But thats good for Welly DS.

The next step for me is to socialise this idea and see what sort of support there is for the idea, who may want to chuck in some venture capital to get to a prototype build, and ultimately to see who may actually want one or two of these wings should they get to that stage. If i can get some backers to get this moving, i can go back to Lee and move on to the CAD/CAM work, and then move onto producing a prototype or two. Should the prototypes work well enough, we could move into a small run of perhaps 10-20 of these to keep costs down.

Those who help stump up cash would take a proportional share of any profits made in on-selling the production block. Not that this is being done as a profit-makimg venture, we will be selling these at near cost, but we need to have a slim margin for future R&D and recoup initial investments.

So, Lee has indicated there is probably another 2 hours of CAD/CAM work to do to get to a point where we can mill out a prototype. We would then build and fly the prototype, and then perhaps refine it for a second prototype or maybe just go straight into the production run.

I have some indicative costings here (and they are loose): About $450 to get all the CAM/CAD work done and a prototype(s) built. All going to plan, the production block models could be around the $150 mark.
taurineman is online now Find More Posts by taurineman
Reply With Quote
Sign up now
to remove ads between posts
Old Apr 04, 2014, 03:53 AM
Registered User
taurineman's Avatar
Wellington, New Zealand
Joined Sep 2009
384 Posts
Heres what i've been working on in Sketch-up. Like a predator bat, but with backwards sweep. Hopefully to allow the more of the nose to hang out in front of the CG to counter its shear weight. It has a PW-51 @9%
taurineman is online now Find More Posts by taurineman
Last edited by taurineman; Apr 04, 2014 at 04:01 AM.
Reply With Quote
Old Apr 04, 2014, 03:54 AM
Registered User
taurineman's Avatar
Wellington, New Zealand
Joined Sep 2009
384 Posts
Small update: i have spent some time with the CAD expert and i have also given him the layout of all the pockets, wire and spar slots. Its looking like it will come out at about 1440mm wing span to keep the max thickness in the centre under 30mm. I have speced a very generous lead bay in the nose for balancing, as i suspect it will need a fair bit.

Spars will be two 20mm ply spars on the edges.

I have not heard back from them in over a week. So, i think i will need to chase them up.
taurineman is online now Find More Posts by taurineman
Reply With Quote
Old Apr 04, 2014, 03:56 AM
Registered User
taurineman's Avatar
Wellington, New Zealand
Joined Sep 2009
384 Posts
We're just ironing out issues in CAD model at the moment. Wire tunnels popping out where they werent expected to etc. We're close.

Multiboard surface when sanded should be reasonable - it will need hand-sanding tho. The sanded finish is unknown at this point, but almost certainly wont be mirror-like... Again, the goal is not to produce a super-shiney super-slippery record-setter - rather something which will be more workhorse than stallion...

To get the prototype(s) out, im looking at somewhere around $500 mark. Im actually expecting it to be a bit more now as the complexity of the design is going up as we uncover issues in it.
taurineman is online now Find More Posts by taurineman
Reply With Quote
Old Apr 04, 2014, 03:56 AM
Registered User
taurineman's Avatar
Wellington, New Zealand
Joined Sep 2009
384 Posts
Apologies for the slowness, but we've had to revist the terms of engagement with Makerspace, as the project was taking up more much design time that expected. Finally, i have negotiated $300 for the design and tool-pathing work, and materials are extra. So thats pretty close to what i was expecting, so we're still on track. Big thanks to Makerspace for making this work!

So i was again down there today working with the design guy (Steve) and he's done a superb job with the 3d model - now with all the pockets, recessed cover area, spar tracks, wire tunnels etc. We've nutted through the manufacturing logistics etc too. It has to be done a little different to how i wanted, but it should end up good. I wanted all the machining to be done upfront and then the two halves glued together; but we have to do a two-stage machining process: mill wire tunnels and spar slots, then glue together, wait till cured, then mill the wing shape into the glued block.

We need to do it that way to avoid milling feathered edges, which wont come out accurate as they flex under the milling head and you end up with the wing profile getting distorted. Seems there is no way to avoid the revises two-stage process.

So by next week we will be in a position to mill two prototypes: one in multiboard and one in solid wood. I will give the wood one to Jono to gear up and evaluate, while i will do the multiboard one.
taurineman is online now Find More Posts by taurineman
Reply With Quote
Old Apr 04, 2014, 03:57 AM
Registered User
taurineman's Avatar
Wellington, New Zealand
Joined Sep 2009
384 Posts
Small update: i went in to check the final design and from that we made a couple minor tweaks: re-aligned the servo pockets to right angles to the hinge-line; made the spar slots 18mm as that is the standard thickness of the pine.

I heard back today, and he has completed the changes. I just need to go in and give them the final thumbs-up before he commits to setting up the tooling-paths.

As for the materials, i was going to go with 1 multiboard prototype and 1 birch ply prototype, but they have a new material called Corelight, which is a composite of bambo and PVC and should be stronger and more rigid than multiboard alone (which is 100% PVC). Its also nicer to work with and cheaper.

http://www.psp.co.nz...s/PVC/Corelight

I need to eyeball some myself and get a feel for it. But it looks promising.

Makerspace are pretty worried that the ply will splinter and rip out chunks when being milled leading to a rough finish. So, maybe i can get them to mill a sample? I will enquire when im there later in the week.
taurineman is online now Find More Posts by taurineman
Reply With Quote
Old Apr 04, 2014, 03:57 AM
Registered User
taurineman's Avatar
Wellington, New Zealand
Joined Sep 2009
384 Posts
I got a sample strip of the corelight and a same sized sample of the multiboard for testing. I clamped one end to the bench and used the digital fish scales to pull the other end of the strip up 10mm and measure the pull weight.

The Corelight takes 4.4kg to deflect it up 10mm.
The Multiboard takes 3.5kgs to deflect it up 10mm.

Both peices were about 6cm x 29cm
The multiboard was 15mm and the corelight was 16mm thick.

So adjusting for the thinner board (its 6.6% thinner), i added 6.6% extra to the multiboard to get 3.7kg - to estimate truly equal thickness boards.

So with that in mind the Corelight is about 20% stronger/rigid than the Multiboard. So, i think we'll go with Corelight instead of multiboard.

Corelight peice was 141.3g
Multiboard peice was 147.8g

but given the multiboard is 6% thinner, that would be an equivalent of 156.6g if they were identical thickness.

so whats that corelight is about 10% lighter give or take all things being equal - so 10% less dense, but 20% more flex-resistance. Bamboo is good stuff!

So, one proto will be in Corelight. The other will be in birch ply.
taurineman is online now Find More Posts by taurineman
Reply With Quote
Old Apr 04, 2014, 03:58 AM
Registered User
taurineman's Avatar
Wellington, New Zealand
Joined Sep 2009
384 Posts
Here's some screen grabs from the final Solidworks prototype model.
taurineman is online now Find More Posts by taurineman
Reply With Quote
Old Apr 04, 2014, 04:19 AM
ETS....Energy retention system
timbuck's Avatar
gold coast australia
Joined Aug 2008
5,342 Posts
Sorry jumped the gun..
timbuck is offline Find More Posts by timbuck
RCG Plus Member
Latest blog entry: Gizmo 36 pics
Reply With Quote
Old Apr 04, 2014, 04:25 AM
Registered User
taurineman's Avatar
Wellington, New Zealand
Joined Sep 2009
384 Posts
Quote:
Originally Posted by timbuck View Post
The only worry I have is stability through twisting , warping etc.
My hope is that ply will stabilise the wood nicely. The corelight board will be naturally extremely stable.

Quote:
Originally Posted by timbuck View Post
One of my best mates is in the ply industry I'll ask him for advise and get back to you
Cheers Im opting for non-marine grade Birch ply. Again, these are prototypes to see how they machine and finish, and ultimately fly.

Quote:
Originally Posted by timbuck View Post
$150 is great , but very cheep.
It may well be more expensive than this. Time will tell.

Quote:
Originally Posted by timbuck View Post
Have you looked in to paulonia ??
yep - that's what my Spruce Goose is hand-crafted from. But its bagged in Glass to give it split/splinter resistance.
taurineman is online now Find More Posts by taurineman
Reply With Quote
Old Apr 04, 2014, 10:24 AM
Objects in mirror are losing
1000MPH's Avatar
USA, NV, Reno
Joined May 2003
3,566 Posts
Very cool project. I've been rolling around a similar idea in my head for a while but using a proven planform such as a combat wing or licensing the JW design. I'm not much of an original thinker.
I wish we lived closer, I would be willing to invest.

My only suggestions to your design are that the elevons look a bit small and I don't feel the servos need to be 90 to the hingeline to go fast. I'd bring them inboard enough so you don't need to use extensions. Are you spec'ing std size?

I'd love to see this become a "must have".
1000MPH is offline Find More Posts by 1000MPH
Reply With Quote
Old Apr 04, 2014, 12:31 PM
DS Addict
ALEX HEWSON's Avatar
Christchurch, New Zealand
Joined Jul 2005
5,187 Posts
We all need to live in Kernville. Turn Kernville into a DS village
ALEX HEWSON is offline Find More Posts by ALEX HEWSON
RCG Plus Member
Reply With Quote
Old Apr 05, 2014, 02:36 AM
Registered User
taurineman's Avatar
Wellington, New Zealand
Joined Sep 2009
384 Posts
The prototype is speced for 12mm mini servos like the HS-5085mg or KST type alu cased servos. I could move them inboard a bit to get a bigger servo in, but the goal for this is to be a hack/workhorse with a rather modest speed range < 200MPH, so I feel the smaller servos will suffice and it means I can drive the elevon from the center to minimise twisting of the elevon.
taurineman is online now Find More Posts by taurineman
Reply With Quote
Old Apr 05, 2014, 03:49 AM
DS Addict
ALEX HEWSON's Avatar
Christchurch, New Zealand
Joined Jul 2005
5,187 Posts
Nice work Pete. Just got to hope the smaller gear trains hold up to landings/arrivals
ALEX HEWSON is offline Find More Posts by ALEX HEWSON
RCG Plus Member
Reply With Quote
Old Apr 05, 2014, 06:30 AM
Arrarrar!
josh18's Avatar
Australia, NSW, Wagga Wagga
Joined Jan 2010
3,286 Posts
Nice! This is a really interesting idea, cant wait to see more!
josh18 is offline Find More Posts by josh18
Reply With Quote
Reply


Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Build Log CNC-Multitool hot wire cutting and milling machine build chiloschista CAD/CAM 70 Yesterday 01:15 PM
New Product CNC Milling Services nexty CAD/CAM 9 May 21, 2014 06:00 AM
For Sale 4 Axis 20x24 CNC Router/Engrave/Milling w/ CAD/CAM SW -Turn-key ngr1 Non R/C Items (FS/W) 2 Apr 17, 2014 04:50 PM
For Sale 2x Armattan CNC 355 Full Milled Carbon Fiber Prototype bobepine Aircraft - Electric - Multirotor (FS/W) 12 Feb 16, 2014 04:14 AM