Events 2017
Aerotow 2017

IMAC 2017

St Asaph Fayre


Click an event for more info
Log in

I forgot my password

Latest topics
» Gary Gilmour
Thu Oct 19, 2017 7:06 pm by Allan Patrick

» Identify this aircraft
Thu Oct 12, 2017 8:00 pm by Gary M Jones

» llyn pss
Thu Sep 28, 2017 9:40 am by Tim

» Field
Tue Sep 26, 2017 1:01 am by Allan Patrick

» Cloud Hopping above RDMFC
Sat Sep 16, 2017 9:30 pm by Mark Barnes

Gate post broken

Sun Aug 27, 2017 8:42 pm by Gary M Jones

Just in case anyone is not aware, the main gate post at the field entrance is broken at its base just at ground level. ( the bolt side post) The post is only supported by the fencing attached to it. If the fencing fails then the post will fall and …

Comments: 3

RDMFC Social Media
RDMFC Facebook and RDMFC Twitter

Project Peenemünde!

View previous topic View next topic Go down

Project Peenemünde!

Post by Andy Sayle on Mon Feb 16, 2009 11:25 pm

Evening all.

Now the little rocket is almost complete (need to upload some pictures on that one!) I thought I would introduce Project Peenemünde!

For some reason, I seem to be developing a strange fascination with WW2 German hardware. Combine that with a desire to experiment with some fairly simple composite manufacturing (on a small scale!), and I came up with the idea of building a stand-off scale V2 rocket. The project is pretty simple: Design a V2 rocket to be moulded and constructed from Composite materials.

I've decided to keep it simple, and fairly small to begin with, and possibly look at scaling things up if all goes well. The exact scale I am working to is about 1:27.14 (roughly!), and this will result in a V2 with a total height of about 517mm. It will be designed to use the larger Estes motors (D class) that are 24mm diameter. The construction will be a composite skin (fibreglass reinforced with kevlar/carbon where needed) and ply internal structure and fins.

The "task" list is as follows:

1. Design "buck" in 3D CAD.
2. Design mould tooling
3. Produce "buck" using appropriate method(s)
4. Produce tooling using appropriate method(s)
5. Use tooling to mould airframe
6. Fitout airframe with non moulded parts like fins and whatnot
7. Test flight, and hope it doesn't end in disaster!
8. Tune flight performance.

At all points I'll be taking pictures and explaining what I am doing. More so I can remember for next time really Smile I'll explain the CAD stuff as well, and post some pictures. If anyone wants anything explaining in more detail, just shout! I'll occasionally slip into "techno talk" so apologies if I do!

Simple eh?

_________________
Chicks dig slop free linkages....
avatar
Andy Sayle
Club Chairman

Posts : 4721
RDMFC Bonus points : -487569809
Join date : 2008-11-16
Location : Abergele, North Wales

Back to top Go down

Re: Project Peenemünde!

Post by Andy Sayle on Mon Feb 16, 2009 11:44 pm

Right, part 1. Designing the Buck. As with most of my projects, they tend to come about whilst doodling in SolidWorks, usually whilst sat in dull meetings where I should really be paying attention Smile In this particular meeting, I downloaded a few resources on the V2 rockets, including a rather neat little cross section drawing.

It was a simple matter of inserting this drawing into a SolidWorks part as a sketch picture, and scaling it to the right size (in this case 517mm from tip to tail). That presented me with a nice outline to trace, and this gives the body profile, and fin shape.

http://www.andy-sayle.co.uk/peenemunde/01%20-%20layout%20sketch.JPG

With this sketch sized correctly, I made a few decisions about the design and construction. The model will be two piece. A nose cone, and tail/fin can. The split between the two will be roughly in the middle (and it is here that the model splits in flight to allow a parachute to be ejected for recovery). With that decided, it was time to start turning the sketch into 3D. I started with the tail cone, revolving it through 90° This was done so I can create one set of fin loacting features, and then just do a pattern to create the other 3 (saves drawing things 4 times!).

http://www.andy-sayle.co.uk/peenemunde/02%20-%20tail%20can%20revolve%20quarter.JPG

With that in place, I revolved the front nose cone sketch through 360°. You'll notice I made this part hollow, but left the tail cone solid. The reasons for this are basically to make the creation of the Fin locating features easier. The nosecone also has a lip on it, which will slot into the tailcone to hold it in place up until the recovery parachute is released by the two halves of the model splitting. The mating recess for the lip to sit in will be added later.

http://www.andy-sayle.co.uk/peenemunde/03%20-%20nose%20revolve%20with%20lip.JPG

Now I moved onto the fin. I have just extruded these to a 3mm thickness. They will not be moulded (cut from 3mm liteply, or balsa instead), but I need to show them in this model, so I can create the fin location features which will be moulded in. I have also made them slighly larger than scale, mainly to try and ensure a stable flight (I could go on about centres of pressure, and static/dynamic stability and whatnot, but I need a blackboard and 3 colours of chalk for that)

http://www.andy-sayle.co.uk/peenemunde/04%20-%20fin%20extrude.JPG

More later...

_________________
Chicks dig slop free linkages....
avatar
Andy Sayle
Club Chairman

Posts : 4721
RDMFC Bonus points : -487569809
Join date : 2008-11-16
Location : Abergele, North Wales

Back to top Go down

Re: Project Peenemünde!

Post by Andy Sayle on Mon Feb 16, 2009 11:55 pm

Next up is to create the fin locating feature. I want there to be a recess to accept the fin moulded into the airframe, so that alignment is nice and easy, and strength is high. To do this in Solidworks, I have used a nifty tool called "Indent", which is basically a non-destructive boolean subtraction, with a bit of clerance thrown in for good measure! It removes material from the tail can where the fin extrusion intersects it. (fin is hidden in this screenshot)

http://www.andy-sayle.co.uk/peenemunde/05%20-%20tail%20and%20fin%20indent.JPG

Now I have the fin loacting features, I can make a circular pattern about the rocket axis to generate the missing thre quarters and three fins.

http://www.andy-sayle.co.uk/peenemunde/06%20-%20tail%20and%20fin%20pattern.JPG

Oh, one thing I forgot, was to turn the tail can into a hollow part, rather than a solid. This was done using the shell command in SolidWorks. It rather convienently follows the indented fin location feature, and gives a constant thickness wall to the part. This is why I started off with a solid part Smile

http://www.andy-sayle.co.uk/peenemunde/07%20-%20tail%20shell.JPG

Now the basic exterior 3D model is taking shape, I made a decision about how to construct this "buck" to take the tooling moulds from. I could carve it from wood, or foam. I could get it CNC milled. Instead, I am going to make use of our SLA machine at work (Stereolithography authoring machine) This is basically a 3D printer that uses epoxy resin as it's printing medium, and a UV laser as it's "printhead". All will become clear as this build continues, but for now, all you need to know is that this design won't fit on the machine in one piece! It needs splitting into 4.

So, to do that, I need to split my model into four, and make sure each part can be fitted together accurately, and so that all the relevent features that are needed on the finished model, are accurately represented. I'll go through that lot tomorrow Smile

Cheers
Andy

_________________
Chicks dig slop free linkages....
avatar
Andy Sayle
Club Chairman

Posts : 4721
RDMFC Bonus points : -487569809
Join date : 2008-11-16
Location : Abergele, North Wales

Back to top Go down

Re: Project Peenemünde!

Post by Andy Sayle on Wed Feb 18, 2009 1:11 pm

Right, so I need to split the model up into four parts to allow them to fit on our SLA machine. This is quite easy to do using the "split" command in SolidWorks. I just chose four roughly equal sections horizontally.

http://www.andy-sayle.co.uk/peenemunde/08%20-%20split%20the%20buck.JPG

With the parts split, obviously I need some features to make sure I can glue them back together once they have been made. With the parts being circular sectioned, a mating lip and recess seem easiest to produce (simple extrudes and cuts in SW). The section view below shows this quite well, and you can also see the wall thickness of the parts here is about 4mm or so. This should give the parts plenty of strength to prevent damage when taking a mould off the buck. The actual parts produced from the mould will be a lot thinner, and I'll change the model to reflect this at a later stage.

http://www.andy-sayle.co.uk/peenemunde/09%20-%20add%20mating%20features%20-%20flanges.JPG

In order to prevent the circular sections deforming once made, they could do with a little bit of bracing. This should help prevent it from going slightly oval. This is just a simple extrude in SolidWorks, and can be seen in the 4 pictures below (one for each segment of the model)

http://www.andy-sayle.co.uk/peenemunde/10%20-%20nose%20base.JPG
http://www.andy-sayle.co.uk/peenemunde/11%20-%20nose%20cone.JPG
http://www.andy-sayle.co.uk/peenemunde/13%20-%20mid%20section.JPG
http://www.andy-sayle.co.uk/peenemunde/14%20-%20fin%20can.JPG

Now those parts have been created in SolidWorks, they can be prepared to be built on the SLA machine. Unfortunately, the software that the SLA machine uses cannot understand SolidWorks files, so the first step is to convert them to a format that it can understand. I normally use STL files for this, and these are models that contain approximated surfaces of the original CAD geometry. It constructs these by creating lots of small surfaces the represent the CAD data. The output of this conversion looks like this:

http://www.andy-sayle.co.uk/peenemunde/16%20-%20STL%20generation.JPG
http://www.andy-sayle.co.uk/peenemunde/17%20-%20STL%20generation%202.JPG

You should be able to make out the wireframe structure that shows the approximated CAd geometry in the images above. I have control over the detail that is captured here, so if I want a small file size with not much detail (to speed up the build, and rough space envelope models) I can tell it to use large surfaces, and that a large deviation from the CAD geometry is acceptable. I can go the other way, and use lots of surfaces if I want the accuracy to be good, although it reulsts in bigger files, and longer build times.

Now I have the STL files, I have to prep them for the SLA process...

More later
afro

_________________
Chicks dig slop free linkages....
avatar
Andy Sayle
Club Chairman

Posts : 4721
RDMFC Bonus points : -487569809
Join date : 2008-11-16
Location : Abergele, North Wales

Back to top Go down

Re: Project Peenemünde!

Post by Mark Barnes on Wed Feb 18, 2009 6:21 pm

Very intresting Andy, didnt know you had it in you, he he

One question (well about 400 really but one for now) how much alteration can you do once you convert the SW drawing to the STL one is it a simple process or is it back to SW then reconvert? Im fully aware you dont make mistakes as its hardly rocket science, qust curoious

M
avatar
Mark Barnes
Club Secretary

Posts : 6501
RDMFC Bonus points : 8098
Join date : 2008-11-16
Location : Rhyl North Wales

Back to top Go down

Re: Project Peenemünde!

Post by Guest on Wed Feb 18, 2009 9:07 pm

lookin good mr s.

Guest
Guest


Back to top Go down

Re: Project Peenemünde!

Post by Andy Sayle on Wed Feb 18, 2009 10:07 pm

Evening folks.

The conversion to the STL format is done purely to allow the models to be prepared for the SLA process. It is the equivalent of taking an Excel document, and saving it as another format (eg, pdf). You still have the original file to work on if you need to, but you also have a copy in a different format. I'll actually be coming back to the Solidworks model once I have shown the SLA process. I want to use it to visualise the mould tooling, and also the internal structure/layup of the V2.

Incidentally, I can import STL files back into Solidworks if I want to, and have full access to edit the geometry if needed.

Anyway, on with the build. I've exported the STL files ready to begin the SLA process, but first a bit of background about SLA. It is quite a simple process, whereby models can be built up in a vat of epoxy based resin. We basically take a 3D CAD model and slice it up into layers. The SLA machine has a vat of UV curing resin in it, and a platform that can be lowered into it in small increments. It also has a UV laser, which scans a layer, curing where it strikes. The platform is lowered a small amount and the laser scans a layer of the part, curing the resin. It is then lowered and the next layer is scanned, curing more resinon top of the previous layer. The end result is a lump of cured resin that accurately represents the original 3D model, built up of layers of cured resin. (wikipedia has a reasonable article on it for more detailed info: http://en.wikipedia.org/wiki/Stereolithography )

To prepare the STL files for building on the Machine, I use a bit of software called 3D lightyear. This allows me to open an empty build platform, bring in the STL files I want to build, and arrange them on the platform. The screenshots show the maximum envelope for building, and the platform at the bottom too. I have arranged everything in a particular fashion, to ensure the parts are built to the best possible standard (Screenshot is corrupt, I'll sort that out later....)

http://www.andy-sayle.co.uk/peenemunde/18%20-%20Empty%20build%20platform.JPG

Once the parts have been arranged and orientated, I need to add some supports. These are primarily there to anchor the parts to the platform. Because the VAT of resin stays still, we lower the platform into the resin bit by bit. Each time it is lowered, a fresh coat of uncured resin is applied to the surface, and then cured by the UV laser, to build the next layer on the model. It effectively builds it from the platform upwards. Obviously the parts would be free to drift around in the vat if they are not tied down to something, so the supports do just that. They are arranged according to a load of rules, and it takes a fair bit of time to get them right.

http://www.andy-sayle.co.uk/peenemunde/20%20-%20part%20supports.JPG

You can see from the screenshot, the supports are like a lattice structure. This provides adeqaute support to the parts during the build. Once the part is complete, the supports just break off nice and easily. You'll see that when I build the parts Smile

http://www.andy-sayle.co.uk/peenemunde/21%20-%20editing%20part%20supports.JPG

Once the supports are designed, we have slice the models up, so the SLA machine knows what to get the UV laser to scan, and cure. This is done automatically by the software, and our machine works with a layer thickness of 0.1mm.

http://www.andy-sayle.co.uk/peenemunde/22%20-%20slicing.JPG

With the slicing complete it is normal to check the output files, to make sure there are no errors, or unsupported bits of model that could result in a failed build. This is done by stepping through each slice layer in turn, and inspecting the results. It is quite quick once you know what to look for Smile The screenshot shows a typical layer halfway through the parts. The different colours on the screen represent where the laser will scan, and with what power/speed dot size etc.

http://www.andy-sayle.co.uk/peenemunde/23%20-%20review%20slicing.JPG

That completes the SLA preparation, and the next step is to get a bit of machine time to build it! The build process is not amazingly fast. This particular build will take around 30 hours or so, according to the build simulation output. Fingers crossed I can get some machine time over the weekend though Smile

The whole process is brilliant for rapid prototyping, as I can go from a CAD design on my PC one afternoon, to having a built part the next day (for small stuff). The finished parts are perfectly usable, depending on the exact type of resin used. The stuff I will be using is almost clear, and mostly epoxy based. It is designed to be quite close to nylon in its physical properties.

More later Smile

_________________
Chicks dig slop free linkages....
avatar
Andy Sayle
Club Chairman

Posts : 4721
RDMFC Bonus points : -487569809
Join date : 2008-11-16
Location : Abergele, North Wales

Back to top Go down

Re: Project Peenemünde!

Post by Andy Sayle on Mon Feb 23, 2009 11:15 am

Morning all,

The SLA machine was busy over the weekend with normal work stuff, so I didn't get chance to build the parts I wanted. Fingers crossed this weekend coming will be free though! In the meantime, I have done a bit more CAD work on the next stage of the process.

The parts designed for the SLA process tend to have quite a thick walls (heavy!) and are in more parts than I shall build the rocket in, so I have modified the CAD model to now represent the actual manufactured parts (I have saved a copy of the model in it's "SLA ready" status, for future reference). The first thing to do was to join the parts back together. The body will be made in two pieces, a nose cone, and tail can. To show this in SolidWorks, I just joined the two nose SLA parts backtogether, and removed the mating features I added. I also thinned the wall down to 0.75mm, which is a rough estimate of the laid up composite wall thickness I shall be looking to acheive (possibly less!). I have left the mating flange on the nose, as you can see.

http://www.andy-sayle.co.uk/peenemunde/24%20-%20Join%20Nose%20cone.JPG
http://www.andy-sayle.co.uk/peenemunde/25%20-%20mating%20flange.JPG

I have also joined the two fin can parts together, and thinned the wall down, as per the nose.

http://www.andy-sayle.co.uk/peenemunde/26%20-%20fin%20can%20join%20and%20thin.JPG

The mating flange (blue highlight in previous picture above) is used to hold the two halves of the rocket together during the flight. The rocket seperates here at apogee, and this allows the receovery parachute to be deployed. Obviously it requires an internal feature in the tail can of the rocket. This can be quite tricky when using a female mould to produce the part as the feature is on the wrong side of the composite layup (the non-tool side), and therefore difficult to layup accurately (it has no tool there to keep the surface accurate. To get around this manufacturing issue, I shall be laying up the fin can in the female mould as normal, and then bond a mating ring inside the fin can. This will be moulded so the internal surface is smooth and round, and a perfect fit on the male flange on the nose cone.

http://www.andy-sayle.co.uk/peenemunde/27%20-%20mating%20ring.JPG

The next picture shows this as it would be assembled, with the nose cone inserted. The nose cone is on the left, the tail can on the right. The blue section is the mating ring, and this is bonded to the tail can. This type of construction has the added benefit of greatly strengthening the body of the rocket

http://www.andy-sayle.co.uk/peenemunde/28%20-%20Mated%20body.JPG

That's it for now. The next job is to model the mould tooling. More later Smile

Cheers
andy

_________________
Chicks dig slop free linkages....
avatar
Andy Sayle
Club Chairman

Posts : 4721
RDMFC Bonus points : -487569809
Join date : 2008-11-16
Location : Abergele, North Wales

Back to top Go down

Re: Project Peenemünde!

Post by Rich on Mon Feb 23, 2009 7:32 pm

I personally think we could all do with a comprehensive report on this subject.
avatar
Rich
Treasurer/Membership Secretary

Posts : 4638
RDMFC Bonus points : 5494
Join date : 2008-11-18
Location : 53.3200°N, 3.4726°W

http://www.futaba-rc.com/

Back to top Go down

Re: Project Peenemünde!

Post by Andy Sayle on Wed Feb 25, 2009 8:53 pm

Ha haaa! You should be glad I am not giving you a step by step SolidWorks tutorial on how I modelled it, instead of the little insights I am giving Smile

So long as people find it interesting, I'm happy to write as much as I need to (usually whilst sat in dull meetings at work Shocked )

Cheers
Andy

PS. interesting Avatar Rich, I'm guessing that is powered by something a bit more Hefty than the little Estes motors afro

_________________
Chicks dig slop free linkages....
avatar
Andy Sayle
Club Chairman

Posts : 4721
RDMFC Bonus points : -487569809
Join date : 2008-11-16
Location : Abergele, North Wales

Back to top Go down

Re: Project Peenemünde!

Post by Rich on Wed Feb 25, 2009 9:12 pm

Just throwing these things in to see if your awake, I like the idea of rockets I'll have a go when the field is a bit dryer, have you done that secret project of ours yet..!!!
avatar
Rich
Treasurer/Membership Secretary

Posts : 4638
RDMFC Bonus points : 5494
Join date : 2008-11-18
Location : 53.3200°N, 3.4726°W

http://www.futaba-rc.com/

Back to top Go down

Re: Project Peenemünde!

Post by Andy Sayle on Thu Feb 26, 2009 8:47 am

Dooo eeeeeeettt! They are dead easy and quick to build, the small ones are not expensive to run, and best of all, there is fire and smoke involved!

If you want to cut down the costs, you are welcome to use my launching gear (as is everyone) plus I can sell you the motors for a bit cheaper than the shops Smile

Cheers
Andy

_________________
Chicks dig slop free linkages....
avatar
Andy Sayle
Club Chairman

Posts : 4721
RDMFC Bonus points : -487569809
Join date : 2008-11-16
Location : Abergele, North Wales

Back to top Go down

Re: Project Peenemünde!

Post by Sponsored content


Sponsored content


Back to top Go down

View previous topic View next topic Back to top

- Similar topics

 
Permissions in this forum:
You cannot reply to topics in this forum