Making of carbon fiber suspension link process

General Composites Discussions Engineering Talk
#1

Hello ,
The past month i am working on a specific project.
I am trying to recreate a link of my bike’s suspension. The reasons for this are the following:

  • It is relatively small piece so i dont have huge waste of materials between attempts
  • Although it is a small part i believe i can learn a lot from it - how to form a specific shape for example replicating/ creating indents in a piece, finishing and combining it with metallic parts and … it is also funny …SO dont judge this effort by numbers or
    if it worth replicating something like this , because the only reason for this is knowledge from problems that i will face… So i have some photos i would like to share with you and a BIG list of problems i am facing right now … Any but really any comment is more than welcome.
    And here are the problems that i am facing right now

(1) I am doing this using prepreg. The mold of the piece is made of the aluminium original in room temperature, then i put the prepreg in the oven at 130 C which is inside the mold and the the mold expands about 3 mm in length which makes my final product to be 3 mm longer than the original and the diameter of the circles around 0.5mm bigger , also because the mold is stretching more than the Carbon Fiber the C.B pulls the one circle destroying it shape in a somewhat like egg … we talk about maybe 1 mm but it is enough to destroy my piece and the fitting of the components. I have calculated the degree of thermal expansion and planning to make a piece which is a bit smaller (about 3 mm) and then lay up the Cf of normal dimensions so when the piece will expand it will not destroy the shape of the holes and it will go to the dimension of the original…but this is just theory.
So is there anyway to kill thermal expansion ? Is there any material like silicone which doesnt expand???

(2) Another problem i have come across is that the prepreg’s resin is for some reason
sticking to the silicone mold -not everywhere just in 2-3 (different) places each time and when i try to detach it …i detach some pieces of the mold as well . - Should i use some kind of release agent ? and if yes which one is suited for 130 C temp and on silicone surface ? But why is that happening ? Silicone should be ok shouldnt it?
Note: i have used 2 type of silicones the white one is normal 2 parts silicone the red one is able to withstand 300 C and so it seems so… but it sticking 10 times more to the epoxy and because it is very easy to get teared it is almost useless .

(3) If you noticed in the hole of the part there is an indent line where the metallic stopper ring goes into not allowing the bearing to go out …this line must be absolutely straight and fine along the circle , and i am almost there but i dont know about how much it will last… and the same goes for the bearing is it right to seat it inside carbon fiber directly? maybe i should use some kind of metallic ring ( to which i have absolutely no access to)

#2

interesting project and approach.

Why did you decide to try and use a core? Did you do any computer modeling and simulation to determine the stress points and help determine the ply schedule? What is your ply schedule?

The epoxy shouldn’t stick to the silicone. Maybe the silicone isn’t cured properly? Which silicone are you using?

You can get rid of thermal expansion by not using heat cured epoxy. This part looks like it could be done in an infusion, thus eliminating the need for heat curing and expansion. Then you can post cure.

Are you trying to directly copy the piece from the bike? You probably could make this piece and not use the same shape as the original. You could also make this piece and then machine the holes out after it’s cured, then you don’t have to deal with cutting and stuffing as much. it looks like you could even probably make the carbon piece and then machine it after ward to final shape.

I think you you might be right about needing some metalic parts, for the bearing.

good start, best way to learn

#3

[QUOTE=sammymatik;56448]interesting project and approach.

Why did you decide to try and use a core? Did you do any computer modeling and simulation to determine the stress points and help determine the ply schedule? What is your ply schedule?

No i did not do any computer modeling. I dont have a program to do it nor i have the knowledge to use it although i am civil engineer our programs are just for civil works.
I know / guess the stress points though .
The ply schedule i have is that all (8 layers) the fibers on the main body to lay along the length of the part except 2 pieces which is 45 degrees ani one which is 90 . - that is for the solid one. The only way (i think) to determine the stress criteria of the piece is to brake the original and she its limits … another way would be to analyze in a program the aluminum one and take the limits from there…but it is not that easy for me. The biggest problem that i have is that the part( the way i do it) is made of two rings which are connected to the main body… so in stress maybe the rings will detach from the main body - of course at the process i chop off some fibers and i connect the with them with the the main body…but still this is not Pro…

#4

The reason why i used a core is that the part with the core 50 % lighter and ‘‘feels’’
as strong - have not tested it though… In general i like combining cores with CF …

The thing with machining is that i am trying to create something which by birth will need the least machining possible … also i dont have a very extensive range of gear that would allow me to have good results via machining the shape , machining the holes , machining the indent etc…

I am using Wacker 4470 silicone… i red at the brochure that i could apply some graphite powder in order to help the mold to detach . Have you ever heard of this?
Graphite powder can be made from the graphite of a pencil yes?
Thank you …

#5

Not sure about the graphite powder. I know people use it for filler and I would imagine it might help for dimensional stability, not sure if it helps for release. Though silicon should self release, I"ve never applied release to silicone unless you’re pouring silicone on silicone. You should be able to buy graphite powder, I"ve used it before for filler.

The tech data sheet for the silicone says that it’s specifically for molding and has good thermal stability. So I think you got the right stuff. To me, it sounds like there is an issue with cure. If the silicone’s cure was inhibited by contamination through oils or latex, or if the silicone wasn’t mixed fully, it can have areas where it didn’t fully cure and would feel a bit tacky and never develop the full properties. We’ve had that happen at work more than once… wasn’t me pouring. You just have to be very clean, use no latex, and double cup. It could be something else but usually silicone is great.

I hear you on the computer modeling stuff. I do a little but am more of a shop person in that respect. It just seems like the core might make things a bit more complicated in the style of mold that you’re using. You are vacuum bagging the parts as well right? It just seems like this part could be made without core, as a more solid piece. I think you need the thickness for the bearings but, it might also be possible to bond on a metal bit for the bearing after curing. There’s probably more than a few ways to make this part. You’re going for it, which is a good place to start.

As for the machining and trimming. I mean you can make the part, and then finish it with a dremel for cutting and sand paper to clean it up. Nothing overly complex. It’s just that it’s often easier to make a part that has the holes cut out after the part is cured or trimmed to final size rather than making a net molded part. Again, it depends on the part and what you feel comfortable with.

o

#6

I am not vacuum bugging … reason for this is that i know exactly each void of the mould after the 2 parts of the molds have closed and i put just the number of plies needed - i use prepreg the resin is just the amount needed for the material that i use…what i do though is applying presure with the bolts that i have on the picture “mould 3 full set”

#7

I see, looks pretty close from the finished parts.

another thing to be aware of is the metal in contact with the carbon. You’ll need to put some fiberglass between them to prevent the metal from galvanic corrosion.

Also, do these bearing get changed at any interval? Rather a snap ring you could bond them in?

#8

Oups…yes you are right…the carbon fiber It generates corrosion even when it is a finished surface? I mean what will be really in contact with the metallic bearing is the resin of the prepreg right?
i was with the impression that the corrosion happens when for example you embbed a metal inside carbon fiber not if it touches it… is it like that?

#9

Maybe instead of glass i could use a layer of kevlar?? Whould there be any benefits from that? I dont like glass…

#10

the layer has no structural value, just a super thin layer of glass is what is usually used to buffer carbon and metal to avoid galvanic corrosion of the metal. You could use kevlar as well. Kevlar is a little trickier to work with and you’ll see it. Glass is transparent and in a very light weight fabric would basically be invisible. And glass is cheaper.

#11

Ok…understood. thank you
Maybe i could dk this ay the same time i lay up the prepreg??? So the glass will have as resin the one tgat is inside the prepreg,and not after bdcause i dont know how good tge bonding will be…

#12

The sacrificial ply should be part of the ply schedule and laid up only the area that the metal would come in contact with the carbon. Some parts will have a sacrificial ply to protect from any surface abrasions in the parts normal use. It’s just a buffer layer either between the part surface and whatever it would come into contact with. In this case between the metal and the carbon.

#13

New process …use of aluminum epoxy Vac Cast…Hope for less thermal expansion according to datasheet … Used silicone rings at the holes though in order to be able to remove the part from mold…