Carbon Frame Chemical Damage Possibility

[Andrew R Stewart]

So I post to a few forums about bike stuff and on another (non frame specific) I replied to a concern about grease damaging carbon fiber frames. My reply was that there were no chemicals available from your local hardware store that would damage carbon fiber. This is nearly a direct quote of Craig Calfee's during a NAHBS seminar he gave a few years ago.

My question here is whether this is true. Does a person need to be concerned with the commonly available solvents, strippers, cleaners, reducers, and such that are easy to get?

many thanks. Andy. (Who doesn't ride carbon and only gets paid to work on the dang things)

[Mark Kelly]

It's not strictly true, but it's close enough.

The carbon itself is essentially inert but it is possible to swell the epoxy enough for it to weaken the bond with the carbon fibres by soaking it in the right solvent. You need to have just the right solvent for the epoxy in question and long term exposure for this to be a problem, so in practice it is unlikely to happen. If you are really interested I can bore your tits off showing you how to calculate whether a particular solvent or mix of solvents will affect epoxy but in brief, ordinary hydrocarbons and long chain esters such as found in workshop grease and oils (including synthetics) are not going to touch epoxy, the best candidates are low molecular weight, moderately polar solvents with moderate hydrogen bonding levels, like acetone.

You can demonstrate this for yourself by immersing a bit of cured epoxy in acetone for several days: the epoxy will slowly swell and soften. The effect is somewhat dependent on the crosslink density in the epoxy, which roughly correlates to Tg: it is more pronounced in low Tg epoxies. I use this to remove cured epoxy from the parts of my filament winder when I get drips. The epoxy I use requires a high temperature cycle to achieve full Tg; since the drips stay at room temperature, soaking them in acetone for a day or so softens the epoxy sufficiently for me to scrape it off. Once the parts have been through the oven this won't work (or at least I've never had enough patience to get to the stage where the epoxy was soft).

On the other hand short term exposure is not a problem: like most people I use acetone to prep carbon for bonding; I also wipe the surface of partly cured epoxy fillets with a glove finger soaked in acetone to help round the fillet. Neither of these producues any ill effect.

[Andrew R Stewart]

Mark- Thanks for the reply. Would you mind if I copy and paste it to another list? Andy.

[Mark Kelly]

No problem for me, I think TT would appreciate it if you linked back to VSalon.

You might need to edit "bore your tits off".

[crumpton]

Short nuts and bolts experiment.

Material
Mitsubishi/NCT321 MR50 G80 36+/-2% 14ply cured 90min@275f

35mm OD tube

Pic 1, Oct 6, 2015 11:30AM CST prior to acetone submersion



Pic 2, Oct 6, 2015 1:32PM CST approximately 3 hr om acetone submersion



Pic 3, Oct 9, 2015 6:48PM CST ~78hr acetone submersion



After 78hr in acetone no change in feel, tack or material thickness. its is stable and rock hard.

all measures were taken from a notched location, within +/-.5mm

i believe fully cured industrial epoxy is just that, fully cured. ymmv.

[Mark Kelly]

Nick

Thanks for the confirmation.

It would be interesting to see the result with a piece cast from say DP420.

[crumpton]

was my thoughts exactly. though it would be a rare indecent for the OP's scenario against that bonding agent. but now that i think about it, about 5 years ago i ran a similar experiment with cured tube(similar to this one) and several samples of DP420 in a coupon form against a HW store aircraft stripper(paint stripper) and it seems to have no affect. i did that test because i had heard a certain carbon frame builder used paint stripper to strip frames for repaint.

Nick

Thanks for the confirmation.

It would be interesting to see the result with a piece cast from say DP420.

[Mark Kelly]

More good information.

If the stripper was based on dichloromethane (AKA methylene choride) as was common, it's slightly further away from the ideal epoxy solvent than acetone as its polarity contribution is about half. I would have thought it close enough to damage low Tg epoxy, however.

BTW the epoxies for which I can find published solvent interaction values all seem to be coating epoxies, not structural ones. It may be that using polyfunctional amine and similar high spec hardener chemistries changes the interaction values, although I suspect that it's simply the degree of crosslinking increasing the effective molecular weight to the level where the epoxy become completely insoluble as the entropy of combination is too low.

BTW and slightly OT - I've been playing with ways of improving the strength of the butt joined part of tube to tube composite construction: see FNL 343 for illustration. I'd be interested to know what you think.

[crumpton]

snip

BTW and slightly OT - I've been playing with ways of improving the strength of the butt joined part of tube to tube composite construction: see FNL 343 for illustration. I'd be interested to know what you think.

what i see looks like it increases the surface area of the HT without increasing the diameter. thus giving more surface for the wrap? is that what you were up to? the first couple frames i built back in '03 had gussets but they were formed from the actual wrap like in a Calfee. a similar look but accomplishing a completely different thing.

this was my experiment from several years ago that i never made anything beyond a few test units. normally i keep something like this under my hat but i'm never gonna do anything with it. the tooling for the tubes lives at another facility where they were never quite able to deliver consistent quality tubes. i should get them to pellet and freight back to me. i designed the tubes so that when mitered for the ajoining tube there would be substantial overlap from the side of the mitered to say HT, BB or ST.

at any rate check out these mitered and fixtured joints. mega surface area for the actual bonding of tube to tube. i put a one of these through en14781 horizontal forces and it passed. i then rode that frame for a few thousand miles. it is still built up with 7700 but is in storage. i should get it out and take some snaps. i wished i could find the BB pic. it told the story quite well. the only wrap on the entire frame was ST to BB which was cooked prior to assembling the rest of the frame. i glued both surfaces at each joint and formed a nice fillet. absolutely amazed how strong that un-reinforced frame is. a static test of individual joints(DT to HT and DT to BB) took over 1100psi on a 2' lever to break. the joints actually remained while the tubes failed, in compression on the opposite side of load about 2-3" from the bonded joint.

true story.

[Mark Kelly]

Thanks, Nick, that's real food for thought.

I promise not to steal your ideas.