DIY Carbon fiber repair - seatstay.

Discuss light weight issues concerning road bikes & parts.
BikeAnon
Posts: 399
Joined: Sun Mar 02, 2014 6:36 pm
Location: NY USA

by BikeAnon

Thanks, everyone.

Sib, my shopping list is in the first post. I don't think it's an exaggeration to say I have enough material leftover to do this eight more times. There is plenty of material to test and test.

I spent under $35 on material. If I were doing this again, and knew what I know now, I wouldn't bother having two different weaves. This would have come out just fine using plain-weave, or 2x2 twill. I'd pick the plain weave.

As for time.... as you mentioned, prep is where I spent the most time. Start to finish....the day it broke was Friday May 5. Eleven days later, I was riding again.

The 11 days included Saturday crying :cry: .....and agonizing over what to do next..... Sunday on YouTube and blogs, Monday ordering supplies..... a few days waiting for them to arrive. All coupled with a busy job and life.

The steps (if I did it again) with time estimates (ignoring my test work)
====DAY ONE===================
- Sanding - an hour
- first carbon wrap - an hour
- 24 hour wait (first 10 hours heat lamp)
====DAY TWO===================
- sanding.
- Coat of epoxy- 30 minutes. allow to get tacky, wait an hour
- Final covering of carbon fabric for looks and protection - an hour
- 12 hour wait.
======DAY THREE OR TWO (either would work, I'd go with day 3)========
- Light sand high spots
- Final coat of epoxy as a smooth clear coat. Maybe two coats, using the same batch of resin, spaced out from beginning of mix, to 40 minutes later (nearing the end of the mix's pot-life)
- Final heat and dry day.

I feel confident I could do a job like this again in a few hours over the course of 3 days.

I'm also confident I spent more time with the pictures and writing up the thread, than the time spent actually doing the repair......if it saves just one other frame, it's all worth it. :beerchug:

Since we're on Weight Weenies.... at least we all already own really nice scales, to help us do a good job with resin mixing. I wouldn't recommend doing this without a scale. Unless you mix big batches of resin, and have good graduated cups. My biggest batch of resin was only 30 grams (20 resin/10 hardener). Most of that got rolled out in the plastic, or squeezed out in wrapping.

I did all the work with a single sheet of 100 grit dry paper. If I wanted a cleaner line look, I'd go over it to get all ripples out with 320, then go up in grades, 500,800,1500,2000 and then a polish (Meguiers Scratch X) for glossy, or a final run with Scotchbrite pad for matte finish.


This is what five minutes with Scratch-X can do for you (another project of mine).....

Image Image

by Weenie


BikeAnon
Posts: 399
Joined: Sun Mar 02, 2014 6:36 pm
Location: NY USA

by BikeAnon

Epilogue:

The little tab on the derailleur had also snapped in "the incident". I assume this tab helps keep the chain from jumping off the pulleys if things get bouncy. I've been riding with some tape wrapped around the derailleur pulley arms for a couple weeks.

When I was working with the carbon fiber, I made a little piece to replace the tab. Today I got around to filing the piece down, and gluing it on.

Image

This is not a good "weight weenie" fix. I know I'm adding grams here. Oh well, too cheap to buy a new derailleur.

When I made the piece originally, I had thought the epoxy of the carbon fiber would allow me to attach the piece. Once it cured, the piece fell right off, as if I had used mold-release on the aluminum.

We'll see tomorrow what JB Weld does to get these pieces married together.

Image

DamonRinard
in the industry
Posts: 267
Joined: Fri Apr 29, 2011 8:32 pm
Location: Connecticut, USA

by DamonRinard

Thanks for the update. You're clearly handy with CF!

Bond failure analysis points to surface prep as a likely culprit in the disbond. If the part comes off leaving one surface clean, then adhesion was inadequate. In contrast, if the failure is in the epoxy, then epoxy was inadequate.

I'd recommend prepping the cage using a two-part chemical etch kit before re-bonding. West Systems sells one.
Damon Rinard
Engineering Manager, Road Bikes
Cycling Sports Group, Cannondale
Ex-Kestrel, ex-Velomax, ex-Trek, ex-Cervelo

BikeAnon
Posts: 399
Joined: Sun Mar 02, 2014 6:36 pm
Location: NY USA

by BikeAnon

This thread is to show what a first-timer can do without special tools.

Thanks for the info about the bond failure. I had sanded the aluminum with 120 grit. I don't know if the epoxy resin that I used for the carbon fiber is any good for bonding to aluminum.

This time, I used a rough file, and scratched things up a bit. Cleaned both parts with acetone.


The JB Weld seems to have done the job. A few hours in the sun, then a couple hours in the toaster oven at 120*F. It will get another 24 hours of setting before getting cleaned up.

Image

DamonRinard
in the industry
Posts: 267
Joined: Fri Apr 29, 2011 8:32 pm
Location: Connecticut, USA

by DamonRinard

Well done. With luck the JB Weld will hold.

Aluminum oxide forms in milliseconds. It's not very strong. Even a mechanically roughened (sanded) surface oxidizes quickly. In this situation, the epoxy is adhering to the oxide and not to the aluminum, which makes the oxide layer is the weak link. When it fails, the carbon part comes off (along with nearly all the epoxy) leaving the aluminum apparently clean, just like in your photo. That's what the etch kit fixes: it chemically changes the oxide. I don't know the chemistry details but in practice it makes a big difference in bond strength.

A less-successful method to reduce oxidation is to wet-sand through the epoxy just before assembling the parts to be bonded. The hope is that the liquid epoxy acts as a barrier to prevent interaction between the aluminum and air, hopefully reducing some of the oxide forming. I've found the etch kit works much better.

Let us know how it holds up in use!
Damon Rinard
Engineering Manager, Road Bikes
Cycling Sports Group, Cannondale
Ex-Kestrel, ex-Velomax, ex-Trek, ex-Cervelo

BikeAnon
Posts: 399
Joined: Sun Mar 02, 2014 6:36 pm
Location: NY USA

by BikeAnon

JB Weld seems like it has done a solid job.

I also decided to sand the seat stay repair smooth, and blend it in a little better. I taped off the paint on the seat stay, went at the epoxy with 400,600,800. Then some auto scratch-remover polish. I was going for shiny-matte. Or dull-glossy. You pick.

I threw the bike on the scale today. 15 pounds. I blame the fatter bar tape. ;) I think the replacement derailleur pulleys saved enough weight to offset the new carbon fiber. :D

Image

Image

Image
Last edited by BikeAnon on Sat Jun 03, 2017 3:38 am, edited 1 time in total.

bkw
Posts: 34
Joined: Tue Sep 09, 2014 3:04 pm

by bkw

Fantastic post, great detail!


Sent from my iPhone using Tapatalk

Shrike
Posts: 1259
Joined: Fri Jun 03, 2016 5:08 pm

by Shrike

Brilliant job BikeAnon, been reading through this after you posted in my thread earlier. Going to revisit when I'm back tonight and give having a go some thought myself :)

BikeAnon
Posts: 399
Joined: Sun Mar 02, 2014 6:36 pm
Location: NY USA

by BikeAnon

We're through the season. The repair has held up just fine.

And I weigh 260. Having a 15lb bike is actually quite ridiculous for a guy like me.

thumper88
Posts: 146
Joined: Tue Mar 13, 2012 7:27 pm

by thumper88

One thing I might do differently is use a better resin... you may get the initial strength you want with that one but perhaps not the longevity. a 5-to-1 WEST 105 is a really good choice for this kind of repair. Track record is as long as they get, and it's proven out not just for lamination but for secondary bonding (bonding to a previously-cured laminate surface) which is what you're doing here.
You may be just fine with this, but its not a lot of expense difference vs risk to go with the best resin you can lay your hands on.

BikeAnon
Posts: 399
Joined: Sun Mar 02, 2014 6:36 pm
Location: NY USA

by BikeAnon

thumper88 wrote:One thing I might do differently is use a better resin... you may get the initial strength you want with that one but perhaps not the longevity. a 5-to-1 WEST 105 is a really good choice for this kind of repair. Track record is as long as they get, and it's proven out not just for lamination but for secondary bonding (bonding to a previously-cured laminate surface) which is what you're doing here....

Thanks. Good advice. I've added your info to the first post, for anyone who might be shopping for a project like this.

P90Puma
Posts: 62
Joined: Tue Jul 25, 2017 7:00 pm

by P90Puma

This was a great thread!

BikeAnon
Posts: 399
Joined: Sun Mar 02, 2014 6:36 pm
Location: NY USA

by BikeAnon

Thanks, P90Puma.

It's held up just fine.

Post Reply
  • Similar Topics
    Replies
    Views
    Last post