Designing and Building my own Carbon Fibre frame

Who are you (no off-topic talk please)

Moderator: Moderator Team

Posts: 143
Joined: Wed May 13, 2015 10:35 pm

by euph

Hello WeightWeenies

I am studying mechanical engineering and for my 3rd Year Individual project/Dissertation I am building a lugged carbon fire frame. I am going to make the lugs and buy some premade tubes to assemble the frame. I will have to make the chainstays myself which will be tested using the universities machines. I proposed this project to my project supervisor as a project focusing on testing the material properties of carbon fibre.

I am taking inspiration from the old Colnago C40 and Time lugged frame designs. This frame will likely be quite heavy as lugged design and premade tube wont allow much weight savings but this is my first attempt at a frame and I felt this was a design that should work well with the dissertation criteria.

I have designed the frame in Creo v3 which is the program my university uses and that I am the most experienced on.


I have based the geometry around a size 52cm Scott Addict 2016. I will be designing it with the intention of using an ISP topper as I cant source premade tube with the correct inner diameter for a conventional seatpost. I also already have a 2010 Scott Addict which uses a 30.25mm ISP. I have two toppers for this so it means fewer parts to buy.

I have 3D printed the lugs as a test to ensure I have got variouos important measurements correct and to check fitment around crucial areas like the Bottom Bracket where it may interfere with the chainrings. These 3D Printed parts will then be used as plugs to make a mould from. The headtube mould should be a simple 2 part mould however the seattube and BB lug will be 3 or more part moulds.
Once the moulds are completed I will make some inner mould shapes. I don't know the correct terminology for these? These parts will be precise as they will determine the final inner diameter of important ares of the lugs like the BB and headset as both will be pressfit.

Many things are likely to go wrong but that is why I am at university learning! I have spoken to Berk over private message and he has been helpful. I thought everyone else would be interested in my progress so I will try and post updates here. Any tips, pointers or criticisms are welcome. I am by no means a professional at this.


by Weenie

User avatar
Posts: 437
Joined: Thu Nov 01, 2012 4:17 pm

by jbaillie

Hell of an undertaking - good luck and keep pictures coming!

User avatar
Posts: 6752
Joined: Sun Jun 18, 2006 10:29 am
Location: Athens, Greece

by kgt

Good luck! Tell us more about the characteristics of the frame, carbon used, testing, etc.

Posts: 279
Joined: Tue Apr 25, 2017 11:12 am
Location: Christchurch New Zealand

by Squashednuts

Very impressive :thumbup:
I'm looking forward to seeing your progress

Posts: 276
Joined: Sat Nov 14, 2015 10:10 am

by reedplayer

would be exactly my size!;-) and design is 100% to my taste! subscribed.

Posts: 143
Joined: Wed May 13, 2015 10:35 pm

by euph

Thanks everyone.

The handling should be the same as the addict as the wheelbase, chainstay length and headtube angle are all the same.

For the layup I will produce a few different parts with varying layups so I can test them at university. But I hope that the frame will be reasonably comfortable. The stiffness should be good in the front half however the rear may not as I have designed the size of the chainstays a little small. So I will see if they are stiff enough. If not I can always increase the wall thickness.

I am cleaning up my 3D printed parts today and testing to see if I can layup resin onto them without any issues.

I plan to have moulds created by early January. But I am never too optimistic about deadlines.

Posts: 58
Joined: Sun Dec 29, 2013 3:44 pm
Location: heaven, EU

by gotgame

good luck with your project! :smartass: i would love to have this as my dissertation ( i was in mechatronics at highschool)

i am curious also from where you will get the carbon tubes and how the lugs will look. :) curious also about the molding process.
Colnago C40 custom sarroni red, dura ace 7800 - sz54
Specialized Crux cyclocross, sram force hydro - sz52
Caad12 105 - sz52
wheels and parts for sale at good prices!

in the industry
Posts: 637
Joined: Mon Mar 05, 2012 7:28 am

by kevosinn

Sounds like a really fun project! Is there a reason you are choosing lugs as the construction method? I think the word you are looking for is mandrel for your be lug area with internal figments. Keep us updated here and feel free to message me if you have any questions!

Edit- you can get plenty of stiffness out of small stays. I use small individual stays on my frames and get all the stiffness needed by optimizing layup and material. Custom carbon frames, bar/stem combos, repair and component tuning.

Posts: 2008
Joined: Tue Aug 02, 2011 9:20 pm
Location: New Zealand

by dereksmalls

Nice one!

Posts: 35
Joined: Tue Sep 10, 2013 12:47 am

by rpowell

ive always wondered what kind of stresses the lugs see. take for example fillet brazed steel frames where you basically have super high strength steel tubes bonded with low strength brass. but the brass joint rarely cracks and the frames have about the same stiffness as a welded steel frame. this makes me think that the joint isnt so critical. im curious from a FEA perspective what kind of stresses you expect the lugs to see? is it basically minimal and is covered just by making a good fit between the tubes or is there something else?

User avatar
Posts: 6752
Joined: Sun Jun 18, 2006 10:29 am
Location: Athens, Greece

by kgt


Posts: 143
Joined: Wed May 13, 2015 10:35 pm

by euph


Thanks for that. I will be sure to call them Mandrels from now on. You're knowledge of chainstays is reassuring. I plan to make multiple versions of them. Perhaps try and tune the weight rather than just overbuild for peace of mind.

With regard to the strength of lug based design I am also intrigued to see how it turns out. I need to think of methods of testing some completed lugs. For example how to support them and whether to use actual carbon tubes or use some plastic tube.

Lots of thinks to take into account when testing it.

Also. I haven't researched it yet. But are any of you familiar with FEA of carbon fibre parts? I have done some reading as to writing the layup of a part into Abaqus. But I don't know of any alternatives. Any help would be appreciated.

Posts: 3844
Joined: Thu Nov 10, 2005 5:31 pm

by BmanX

Can't wait to see how this works out. I would love to build my own frame one day but have other commitments.

Cervelo SLC 5960g/13.13 lbs

Posts: 143
Joined: Wed May 13, 2015 10:35 pm

by euph

I have some experience with carbon fibre but I don't have much experience with the moulding process which I plan to use to make the lugs. So I have been looking far a way to test the process.

I have a toy RC Helicopter which I broke a wing on and thought it would be a perfect opportunity to create a two part mould and make a carbon fibre component.

First I lay the wing in some plasticine to take a mould of one side.

Then once that was cured I removed the plasticine and cleaned the mould release.

I reapplied mould release and moulded the other side.

I split the mould apart and drilled some holes to index the mould and screw it together for the final part.

I chose to use two layers of 200 gsm normal weave with a 100 gsm layer sandwiched in between.

With new mould release applied to my mould I layed up the epoxy and the carbon.

I cracked open the mould this morning and have a new wing. There are some small voids in the thicker parts however all the carbon is nicely saturated and importantly all the mould release worked well which was an area I was worried about.

One of the issues I had read about was that the two different resin systems used would react however I think the blue PVA release agent I used should have helped that.

Much more confident about the process now and should start making the moulds for the lugs soon. ImageImageImageImageImageImageImage

User avatar
Posts: 248
Joined: Tue Jul 15, 2014 7:37 pm

by Clannagh

Very Interesting keep those pictures coming.
Good luck with this project :D

by Weenie

Post Reply
  • Similar Topics
    Last post