1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

Scratch Build – In Progress Trebuchet: Update - Construction complete!

Discussion in 'Project Logs' started by crazybob, 13 Sep 2007.

  1. radodrill

    radodrill Resident EI

    Joined:
    14 Mar 2006
    Posts:
    870
    Likes Received:
    1
    what kind of a layup sequence were you guys planning on doing?

    For the CF composite layups on these arms I'd do primarily unidirectional CF along the length of the boom as well as several plies of -45/45 plain weave fabric and a ply of 0/90 on the outside.

    Edit: and around the pivot points for the hinge pin and counterweight I'd also put some extra plies in the -60, -45, -30, 30, 45, 60, and 90 orientations.
     
    Last edited: 20 Oct 2007
  2. crazybob

    crazybob Voice of Reason

    Joined:
    21 Oct 2004
    Posts:
    1,123
    Likes Received:
    6
    We haven't worked out the pattern on the arm yet; that probably won't get done until next week or the week after. The counterweight hangers, which we hope to start Tuesday, are ten layers of plain weave on each side of a balsa core. We chose plain weave because even though all the stress is tensile, it manifests itself as bearing stress and we hope the weave will resist pull-out better than unidirectional cloth would.

    On the arm, we've considered both unidirectional and plain weave, and we also recognize that the axles will need special attention. However, in our quest for stiffness, the whole arm may be made the same thickness as the axle supports. No matter what type of cloth we use, I've considered a few layers at ±45° but can't really come up with a compelling argument for doing things that way. The longest fiber in a 45° layer will be under three inches long, and I suppose it might help torsional stiffness a bit I can't imagine it doing anything good for the strength. The 400-500 lbs of force the arm will have to support will happen when it's nearly vertical, so I think the best bet for supporting that is going to be plain weave at 0/90°.
     
  3. crazybob

    crazybob Voice of Reason

    Joined:
    21 Oct 2004
    Posts:
    1,123
    Likes Received:
    6
    Of Wood and Metal

    Well, I know it's been a while since I've posted something concrete, rather than idle chatter and theory. I plan to make up for that today, with a truly epic leap in progress and a large number of pictures. Since my last progress post, we've completed the support structure and begun work with composites.

    First, the support structure. This was given a nice protective coat of Danish Oil, and it looks quite spiffy. Unfortunately, at the moment it doesn't smell terribly spiffy, but I expect that to resolve itself within a week or so.

    [​IMG]
    The frame in all its glory.

    [​IMG]
    This shot gives you a better look at the finish of the wood, as well as the edge of the protective carpet on the bottom.

    [​IMG]
    The eight outer corners also got some carpet to keep walls and the trebuchet from damaging each other.

    [​IMG]
    Here you can really tell what's going on with the carpet-covered base. All that's left to do here is cut the holes for the stakes.

    We also shined up the axles and counterweight, and gave them protective coats of paint. The axles got clearcoated, while the weight got black rubberized paint. We also worked out a better way of carting the awkward and heavy weight around than a rope through the center hole.

    [​IMG]
    This case was just sitting around, but it turned out to be perfect for the counterweight.

    [​IMG]
    The axles will probably eventually get slots in here as well, to protect them better than the plastic bag they're in right now.

    [​IMG]
    So here's where we stand right now for everything that isn't made of carbon fiber.
     
  4. [Tom]

    [Tom] Member

    Joined:
    10 Dec 2003
    Posts:
    504
    Likes Received:
    2
    Top work Radodrill. Did you cut the wood with a router to get all the filleted edges?
     
  5. crazybob

    crazybob Voice of Reason

    Joined:
    21 Oct 2004
    Posts:
    1,123
    Likes Received:
    6
    Of Carbon and Epoxy

    So far, all my posts have been results rather than processes. Partly, this is because I'm lazy and don't like taking time out from work to take pictures. It's also partly because I figure most of you know how saws and drills and spraypaint work. However, we're now into the part of the project that might be strange an unfamiliar to many of you, so I'm going to make a better effort to post in-progress pictures as well as finished shots.

    When working with composites, it's possible to create the exact part you need, but this is very similar to making cast metal parts. Just because it's possible doesn't make it the easiest or best way, so you instead make a sheet of material and cut or machine your part from that. So yesterday, we started that process. We formed the panel that will be used for our counterweight hangers, and when it's done curing we'll cut our parts from the larger panel.

    I apologize for the quality of a few of these photos, but the composites lab doesn't have exceptional lighting and it's not really possible for us to add any.

    [​IMG]
    Step one is to cut the layers of carbon cloth. The observant among you will see that despite my earlier claims of no 45° layers, that's exactly what we have going on here. Calculations showed a terrifying stress concentration around the axle- and bearing-holes with only 0° and 90° layers.

    [​IMG]
    The wood is the balsa core we're using to thicken the parts without adding more expensive carbon. We also used it as a rough cutting guide rather than measuring the layers. The excess will be trimmed off once everything is cured.

    [​IMG]
    Now comes the fun (read: tedious) part. Stacking the layers up one at a time...

    [​IMG]
    ...While saturating each layer with epoxy...

    [​IMG]
    ...And ensuring there are no dry spots, which will be very weak and could easily lead to part failure.

    [​IMG]
    In the end, you have something like this. A stack of very expensive cloth soaked in toxic chemicals.

    [​IMG]
    We're not too concerned with the cloth at the edges, because only portions with the balsa core will be used.

    [​IMG]
    Now comes the final step. To ensure proper bonding between the layers, we have to essentially clamp everything together. The easiest way to do this is by vacuum-bagging, compressing the composite against a flat aluminum plate.

    [​IMG]
    On both sides of the composite, there are layers of material designed to keep the part from sticking and to absorb excess epoxy.

    [​IMG]
    Then after the epoxy-soaked panel is dropped in place and the release layers are in place, we drop in a final layer of what is essentially sturdy fluff.

    [​IMG]
    This is designed to maintain air passages throughout the bag, so no air bubbles can be trapped in remote corners of the setup by the bag sucking down to the plate or material.

    [​IMG]
    Once the setup has been stacked, we insert a vacuum tube and seal over everything with double-sided tape and nonstick plastic. Then we turn the vacuum on and listen for leaks.

    [​IMG]
    After a few minutes, we decided there were no leaks, so we left everything to cure in peace.
     
  6. radodrill

    radodrill Resident EI

    Joined:
    14 Mar 2006
    Posts:
    870
    Likes Received:
    1
    LOL I'm not the builder :D

    Are you guys using peel-ply or release film between the composite and the breather?
     
  7. crazybob

    crazybob Voice of Reason

    Joined:
    21 Oct 2004
    Posts:
    1,123
    Likes Received:
    6
    We're using release film. The parts aren't going to be painted and there'll be no more layers added, so the rough surface left by peel-ply would just spoil the look of the carbon fiber.

    Radodrill isn't building, he's just the most active thread participant. And yeah, we rounded all the edges with a router. It didn't take too long and it made everything a lot safer and more attractive.
     
  8. radodrill

    radodrill Resident EI

    Joined:
    14 Mar 2006
    Posts:
    870
    Likes Received:
    1
    Cool. Oh and what pattern are you using for the layers?

    hehe; I'm just closely following this project since I myself have experience in composites and am also interested in trebuchets and other siege weapons.
     
  9. crazybob

    crazybob Voice of Reason

    Joined:
    21 Oct 2004
    Posts:
    1,123
    Likes Received:
    6
    Completed Carbon Parts

    Well, the first batch of carbon fiber parts has been cured, cut, and drilled, and is ready for mounting. Unfortunately, these are the weight hangers and can't be mounted until the second batch of carbon fiber parts (the arm itself) is completed. But here's some pretty carbon fiber for you to look at.

    [​IMG]
    This is what it looks like as we take it out of the bag. The bits of white stuff left on ithe edges are part of the bleeder and breather layers which remove excess epoxy and prevent air bubbles, respectively. They're still stuck on the edges because our release film was a bit too small and some epoxy got around the film and into the other layers. Fortunately, this only affected the edges, which were useless anyway as they contained no wood core.

    [​IMG]
    Here's what the parts look like after the edges were removed and everything was cut to size. The long cuts were made with a tile saw, and the ends were cut on a miter saw with a fairly vicious wood blade. The aggressive blade can cause the layers to delaminate, reducing strength, but it was our only choice for the end cuts and those areas don't matter much anyway.

    [​IMG]
    Here's the ball bearing that has to fit into the panels. The panel thickness is absolutely perfect. That's a 1/8" layer of balsa in the middle, and each side has 10 layers of carbon cloth, alternating 0° and 45°. The 45° layers barely do anything for the tensile strength, but they have a huge impact on the stress concentrations around the bearings and make the part much less likely to pull apart there.

    [​IMG]
    Both parts from this batch and one of the bearings.

    [​IMG]
    And, after a bit of careful drilling, we end up with this. We're really happy with these results and can't wait to get started on the arm.

    [​IMG]
    And a bit of macro photography for those of you who want to see how the bearing fits. The bearings are actually friction-fit into the holes, and probably won't need any additional restraint.


    The design of our arm has been completely changed from the original drawings. The dimensions will still be the same, but in order to make construction easier the arm will just be two parallel panels, with a hollow, open space in the middle. I suppose this will decrease drag a bit, but that's just an extra benefit. We spent several hours trying to plan how we would cut the cloth and wrap it around a thin double-tapered core and we couldn't come up with anything, so we revisited the design entirely. At some point I'll try to get a new 3D model up, but we might actually get the parts done before the model - we hope to start cutting the cloth for the arm on Tuesday.

    We've also gotten a very slippery polyester fabric for the sling, and some 50-lb test braided ice fishing line for the connecting lines, so hopefully we'll be seeing some progress there in the next few weeks. And finally, I think I have a clever idea for the release mechanism. It's a secret for the moment, but it should release fairly smoothly and manage to stay out of the path of the moving parts.
     
    Last edited: 5 Nov 2007
  10. radodrill

    radodrill Resident EI

    Joined:
    14 Mar 2006
    Posts:
    870
    Likes Received:
    1
    Nice looking CF work. In all sincerity though the 45deg weaves aren't as useless as you might think for the tensile stress; since the loads can be dispersed between the layers and the weave also helps to more effectively transfer the tensile loads from the 45deg fibers to the -45deg fibers, granted not as good for the tensile loads as the 0deg is.
     
  11. ssR

    ssR Carbon God

    Joined:
    11 Aug 2006
    Posts:
    607
    Likes Received:
    5
    Very nice work crazybob. You used pretty much the same techniques i use while making some parts in my project. And you use the same carbon type i got access to (among some others :p). Its 195gr/m^2 plain weave isn't it?

    About the skin thickness, isn't that a bit of an overkill? i mean a sandwich with a total of 20 layers of carbon and if thats the carbon i think it is then each layer is creating a thickness of 0.28mm so thats like 5.5mm of carbon. Maybe i missed it somewhere but how much does that chunk of iron weights?

    Also i noticed that you have some pinholes in your skin which is a pity cause overall it looks very good. You could avoid it by using a finish layer of thin fiberglass or adding some aerosil to the resin in the first layer

    Either way looks very good so far, waiting to see the finished product and maybe a small test video :)
     
  12. crazybob

    crazybob Voice of Reason

    Joined:
    21 Oct 2004
    Posts:
    1,123
    Likes Received:
    6
    Thanks. You know, I'm actually not sure about the material weight - we ran calculations based on the strength and stiffness, but I don't think we ever paid attention to the density. I'll let you know when the next batch comes in. And yeah, it's plain weave.

    The skin thickness isn't overkill at all. Optimization is a big part of the project, since this is an engineering design assignment, so we've done the calculations. The weight is 23 lbs and the way it falls, it subjects everything to almost 1000 lbs of force for an instant during firing. With that kind of thickness, the tensile strength is overkill, but the resistance to the bearings tearing out is the real limiting factor, and our thickness was chosen with that stress in mind.

    Finally, you're right, there are some pinholes. However, that's mostly an aesthetic thing, as the epoxy in the holes of the weave on the outside layer isn't really doing anything. I had been thinking the fiberglass layer might dull the appearance of the carbon fiber, but I had forgotten that the only times I've added that layer in the past, it was on parts that weren't vacuum-bagged. Looking at your results, I think we'll be putting that layer on when we do the arm.

    Again, thanks for your comments. It's nice to see some new interest in the project.
     
    Last edited: 5 Nov 2007
  13. radodrill

    radodrill Resident EI

    Joined:
    14 Mar 2006
    Posts:
    870
    Likes Received:
    1
    If you use a very fine weave, lightweight fiberglass skin then it'll barely be visible and still show off the CF nicely :D
     
  14. RaptorLord

    RaptorLord New Member

    Joined:
    6 Aug 2002
    Posts:
    156
    Likes Received:
    0
    I wouldn't change a thing - that arm looks like my kind of Oreo cookie! :D

    nice work!
     
  15. crazybob

    crazybob Voice of Reason

    Joined:
    21 Oct 2004
    Posts:
    1,123
    Likes Received:
    6
    Thanks! Although that's one crispy Oreo...

    Today we're going to start cutting cloth for the arm, and I think if we have a bit of extra time I'll throw together a small sample with a few different layups to compare the aesthetics.
     
  16. crazybob

    crazybob Voice of Reason

    Joined:
    21 Oct 2004
    Posts:
    1,123
    Likes Received:
    6
    Invisible Progress

    Well, it's been a long time since I last posted, so I figured I'd give you guys an update. We've still been working on the project, but at a greatly reduced rate due to other schoolwork.

    The smallest and most important part we've been working on is the release mechanism. We were stumped for a while, because the release needs to hold ten pounds of force and be fairly secure and resistant to accidental firing, but also has to release very quickly with only a light touch on the trigger. Additionally, all the parts of the release have to stay clear of the other moving parts.

    However, I discovered that a small modification to a mousetrap would make it work flawlessly for what we needed. Of course, we don't want to stick an ugly mousetrap into our trebuchet, but I managed to recreate the mechanism out of heavy steel wire. I think it will probably end up mounted using screw-eyes. You'll get pictures once it's mounted, because until then it's very difficult for a picture to convey exactly how the release works.

    On a more obvious and practical note, we've also been working on the arm. We've finalized our design, and even had a first attempt at making it, but ran into a few snags.

    Preimpregnated carbon fiber (pre-preg) is fantastic stuff. It's already soaked in half-cured resin, so instead of dealing with itchy, tempermental cloth and mixing epoxy, it's more like working with thin cardboard. You just cut a piece to shape, pull the backing paper off both sides, and add it to your part. Once you've got the layering all correct, you bag it up just like normal and stick it in an oven at around 350 degrees F overnight. It's much faster, cleaner, and more accurate than the plain cloth we used for our first part.

    We had decided to make all of our parts from plain cloth because we couldn't find an oven big enough to cure our arm. However, last week we learned that one of the clubs on campus has an oven just barely big enough for our arm. Eager to escape the mess of mixing epoxy, we decided to use pre-preg. Yesterday, we built the whole thing. After about 3 hours of work cutting and assembling, we carried it over to the club office and stuck it in the oven. That's where it all went wrong.

    The person who helped us get set up in the oven probably hadn't used it before. There was an aluminum plate in the bottom, covering the heating elements. He removed this, thinking it would reduce the heat too much. I wasn't close enough to see into the oven at that point, and if I had been I would have stopped him and kept the plate in - it was there to distribute heat evenly, and had it been in place I would be posting pictures of an arm right now.

    In the absence of the plate, our part was suspended just a few inches over the bare heating element. As a result of this, our bag material melted, which both removed the vacuum compressing the layers and melted plastic directly to our carbon fiber. By the time we recieved a phone call and went to fix the problem, the curing had gone completely wrong. The resin was mostly cured, to the point that rebagging the same part and starting the heating over would have had no effect. Unfortunately, due to the loss of vacuum, it was cured badly and the arm was much too weak to be used.

    We won't have an opportunity to try again until January when school starts up again after the holidays. It remains to be seen whether we will try again with pre-preg or just revert to using plain cloth and mixing epoxy. As much as I would like to avoid messing the the epoxy, I'm not confident that this oven will be able to heat our part evenly enough to work, even with the distribution plate in place.
     
  17. mvagusta

    mvagusta Did a skid that went for two weeks.

    Joined:
    24 Dec 2006
    Posts:
    4,639
    Likes Received:
    523
    That blows :waah: I say go with using plain cloth & mixing epoxy. Your first arm looked great, the last thing you want is to take another risk in the oven. A smaller part wouldn't be risky, but we know that you can make a big arm using plain weave and it comes out perfect. So what happened to the first arm? You say you ran into a few snags, so does this mean the first arm can't be used or something??
    If you have to make another arm, use one of ssr's tips. If you go with the fibreglass final layer, then as radodrill said it will look beautiful. Not that your first arm didn't look excellent!
     
  18. crazybob

    crazybob Voice of Reason

    Joined:
    21 Oct 2004
    Posts:
    1,123
    Likes Received:
    6
    Thanks. The first piece, the one I posted pictures of a while back, wasn't actually an arm - it was the vertical piece that connects the arm to the counterweight. It came out perfectly and we will be using it. The part we're working on now is the arm itself, and will be almost four feet long, so it's a bit more of a challenge. I think we'll most likely use plain cloth, just like on the first part, but this time we probably will include the thin layer of fiberglass to get a smoother finish.
     
  19. NZ_mod_man

    NZ_mod_man New Member

    Joined:
    26 Aug 2007
    Posts:
    241
    Likes Received:
    0
    Dude thats's awsome! Love it!

    Keep it up! :thumb:
     
  20. Lupo_IM

    Lupo_IM ItaliaModding staff

    Joined:
    30 Jul 2007
    Posts:
    72
    Likes Received:
    0
    wow!!! great work with carbon!!! it looks very nice!!!
     

Share This Page