Load bearing subframe

General Composites Discussions Engineering Talk
#1

Hello all,

I’m looking for some input from guys who are engineering savvy with regards to CF.

I’d like to create a composite subframe for my motorcycle. This will need to be load bearing for a rider and passenger. I would like for it to have a foam core to add strength, but I’ve never played with cores at all.

The shape isn’t terribly complex (i’ve made way more complicated molds that turned out nice for wet-layup) and I intend to fabricate a negative mold out of fiberglass and vac-bag it. I don’t want to spend an arm and a leg on this, is blue polystyrene foam suitable? Could I make the outter shell, line it with polystyrene, and then wet-lay an inner layer, then vac-bag it to laminate it nice and tight? i’ve tried to research cores but it seems like an art that people keep secret.

Here’s a couple MSPAD (MS Paint Assisted Drawing) renderings. they’re not to scale or anything, just a digital “sketch”.

I’m having some troubles getting my photobucket pics to upload, bear with me please.



#2

Also I intend for the subframe to be “hollow” on the inside so that I can put the battery below the seat - i intend for the core to only be 1/4" thick or so.

#3

what bike is that

#4

it’s an EBR 1190rs, the big brother to my Buell 1125r. I’m basing the designs off the 1190.

#5

I’m really happy to see this post. I joined this forum this week hoping to create the exact same thing for my CBR.

The shape doesn’t seem too complicated. But how are you going to handle the mounting locations? I’m doing this for my race bike so I don’t have rear passenger pegs, but my tank hinges from the subframe. Are you going to locate that into the mold and put in a washer or something or just build it thicker in those locations and drill it out when it finished?

And my passenger mounts are threaded into the subframe, how are you handling that?

Hope you get the advice you need.

Looking forward to progress pics.

#6

It’ll be a while before you see any progress unfortunately, I’m in Afghanistan right now and won’t be able to get to this for about a month if I’m lucky.

The subframe is attached to the frame by the four fasteners (two on each side) at the front bottom of the subframe. I’m intending to bolt the CF subframe to the frame, using large washers on the front side, and the area surrounding the bolt will be an aluminum spacer/standoff rather than the regular core material.

The design for the passenger pegs is still a work in progress. I intend to use a similar setup as far as fender washers, and use some aluminum flatstock as a backing plate to distribute the load across a little larger section of the CF. Adding passenger pegs will add weight, but weight doesn’t matter when I have a passenger anyway, and they will be removeable for track-day.

I’ll put up some more MSPAD sketches when I get the passenger peg mounts a little more developed. They’ll need to withstand downward load which will translate into torsion. I don’t know if CF is suitable for the passenger peg standoffs, it will require some experimentation. Maybe some aluminum flatstock in key areas to reduce deflection.

Well my computer isn’t currently communicating with photobucket, i’ll put up a picture later on.

#7

http://s1241.photobucket.com/user/wintermetalhd/media/CF_Cross_section.png.html

#8

If you’ve not already done so, type ‘carbon fiber subframe’ into Google images and a lot of pictures come up. I’ve done similar searches before to feed my curiosity.
These pictures of OPP Racing’s subframe may be of particular use to you?? http://www.oppracing.com/image_display.php?image=56163&id=197431

#9

Nice, thanks for that link. that is helpful.

If I can get some Expancel, I’d like to make something similar, but fully enclosed (blind mold). But the cheapest package of expancel is 1,200 USD. Akzo Nobel will ship me a sample packet of it, don’t know if it’d be enough for what i need but probably not.

#10

Nice, thanks for that link. that is helpful. That really is a beautiful piece of work.

If I can get some Expancel, I’d like to make something similar, but fully enclosed (blind mold). But the cheapest package of expancel is 1,200 USD. Akzo Nobel will ship me a sample packet of it, don’t know if it’d be enough for what i need but probably not.

I’d like to learn more about how this is set up, in the mold, during layup:

I understand the flat inserts, at least they seem pretty straight-forward, but this one is a recessed cylinder and the carbon wraps from the outside, over the top, and into the inside of the cylinder. Wouldn’t that require cutting something like an “x” shape right where the cylinder goes, in order for the carbon to suck down tight to the internal sidewall of the cylinder? it seems like that could potentially get very distorted once vacuum is applied.

From an engineering standpoint, if the below picture is a representation of a tube-built subframe, would the bottom design offer more strength than the top two? If downward load is applied at the top left, then chords A and C would be in tension and chords B and D would be in compression. However, the compression wouldn’t be a straight line but would have a little curve to it right? I’ve got to go to school for this stuff, someday.

#11

Hi,

your project looks great. I just want to add a small comment to your sub-frame.

You pointed out correctly that if you put your load at the tip, the top will be in tension and the bottom in compression. However, if your transverse links connecting both surfaces are normal to them, they won’t carry any direct stress from your applied load and will only provide stiffness. My suggestion for you is to create a triangular trusses structure and try to predict the load distribution along each node and link of your frame.

Something like this:

#12

calculating load on chords or strands is something i’d like to learn how to do. but that will probably have to wait a few years until I can go to school. unless you can recommend a book or website that would teach me how to analyze that kind of stuff (which i’d love to learn).

I know this is dependent largely upon design, but if that was a solid triangle, wouldn’t it be stronger (talking load being applied at one downward angle) than a truss system of the same weight? that was my thought behind making it a solid piece.

#13

When analyzing trusses you idealize them as only carrying load in their axial direction. This means that if you have two trusses at a 90° angle, there won’t be any load any load transfer between the two. So you need your trusses to be at a lower angle between themselves so that you create a stiffer structure with well distributed loads.

That’s right. A solid triangle is stronger, but the whole point of engineering is making your structure as light as possible. And if you fill the core with foam you have to realize that foam is not as strong as carbon fiber and it is probably only used to provide stiffness, not strength.

#14

i’m under the impression that the trusses would have to weigh more than the solid pieces in order to achieve the same rigidity. my intention for the foam core is to increase cross-sectional area, reducing the bending moment occurring in the subframe when load (a passenger) is applied.

the trussed subframes are beautiful, it just seems like the solid triangle would give me a better “juice vs squeeze”. I’m thinking about doing a combination of both. compound designs like that usually seem to have more rigidity than a standalone piece.

I’ll be doing some experimenting before i make the final piece, so I’m sure I’ll figure something out.

I plan on laying up some 3k 5.7 oz as that’s what I have handy (bought some second quality at around half cost). I’ve never made a structural/load-bearing component, so I have no idea how many layers i should layup. Any guidance??

#15

if i wanted to use some thicker carbon to add bulk (say 12 oz), will biaxial or plain weave “stretch” and conform to all the complex shapes that 2x2 twill will? the twill is super easy to work with, i just wonder about the other types.

#16

This website does a pretty good job breaking down how to calculate trusses.
http://www.slideshare.net/fazirahabdulghafar/calculating-truss-forces

#17

Thanks foss!

#18

I wouldn’t view this as a truss problem since you aren’t really going to make trusses (are you?.. you shouldn’t). Think of it more like a cantilever beam problem with a variable cross section as a function of length (I’m not sure, but there may not be many standard solutions to this available). There are two major things that affect bending stress (applied axial stresses aren’t really a huge deal, but shouldn’t be ignored).

the applied moment - this depends only on how much weight and how far away you want to apply it (in reference to the mounting points), so you don’t have much control over this… Since you aren’t going to be using FEA or anything fancy like that, you’ll want to make yourself and your rider pretty fat.

Part geometry - This part that you have a little more flexibility with. Cross sectional area is important, but what is more important is the area moment of inertia (if you increase cross sectional area then you will increase the area moment of inertia, but there are ways of maximizing area moment of inertia by making the part bigger in the correct directions). Suppose you have a long thin piece of aluminum… if you bend it in the thin dimension it will deform easily, but if you bend it in the larger dimension it will be much more difficult.

This stuff is just basic beam bending problems. It’s not specific to carbon fiber… carbon fiber actually muddies the problem a bit, but the basic concept still applies. That’s all I’ve got time to write right now, but do some google searches on bending stress and calculation of area moment of inertia.