Ad nauseum, life of composites

[pliptrot]

I am trying to imagine if it is at all possible to get any quantitative data on composite frame breakages. According to Chris Juden's recent comments, it's not a case of if, but when, such a frame will break. I have a composite frame made in 2000. I believe that this frame (a Trek 5500, allegedly made in the USA and using some secret process) is as reliable as it was possible to make a frame using last century's know-how. What I cannot ascertain if it is safe to ride or to sell: I would not choose to take money for something which is certain -or likely- to present a risk. Many critical machines now use composites in safety critical applications (eg. the Boeing 787 which has a composite airframe). What do Boeing know that the cycle industry does not?

[[XAP]Bob]

Boeing prescribe an extensive set of regular checks, as well as (no doubt) having embedded sensors. Such checks may involve Xray analysis or similar techniques which are unavailable to the typical cyclist.

As for your frame it is an unknown. If it has suffered knocks (for example when parked in a rack) then they *may* have had an effect which may or may not be visible.
If it has had any falls then the same applies. The "nice" thing about steel is that it tends to fail gracefully (i.e. get weaker, spongier, basically it tells you it's going) whereas composites tend to fail suddenly and without warning.

[TonyR]

http://www.sheldonbrown.com/rinard/EFBe ... e_test.htm

[Valbrona]

We like a man with a conscience. I think you can test ride and you can carefully inspect. Most people buying SH goods these days know the score, in as much as goods are sold on a 'as seen' basis, and of course where viewing is beforehand is not possible it is common to offer a return/refund service.

When a composite frame breaks you tend to notice it. If you haven't noticed it being broken, well it probably ain't.

[Brucey]

I think the Trek is a well made frame (should do 20000 miles or more, easy) but you know, stuff happens. The fork is the biggest worry TBH; if this fails you are into face-plant territory; the chances of the whole front of the frame coming detached (without you noticing the crack beforehand) are slim by comparison, and other frame failures are not so inherently dangerous.

I agree with Bob's comments on the whole. If you are lucky then you might hear a creak or something before a failure occurs, but when it goes it goes quickly and without much warning.

Speaking as someone who has a set of bonded carbon forks (that match a frame and can't be replaced...) which have recently been involved in a prang, I have a real concern about them. I don't think they had a big knock but realistically I can't tell.

I could have the forks X-rayed, and/or UT'd. But without 'known good' reference data for a similar component it might be meaningless, and in any event it might be more costly than new parts all round.

With steel or aluminium components they can be dye-penetrant tested and 're-lifed'; when they fail they usually crack first, in such a way as the cracks can easily be seen (or felt). Carbon just doesn't fail in the same way.

If you don't ride hard, and have not had a prang on your Trek it ought to be fit for further use.

But...if in doubt, chuck it out.

cheers

[tim-b]

Hi

There are a dozen or so non-destructive test (NDT) methods suitable for composites, and a dozen or so flaws that you would want to look for
Any one NDT method will have proven ability to pick up maybe three or four of those flaws. An airline will be able to access the necessary tech at a reasonable cost (in comparison to the cost of an aircraft) for regular checks

Assuming that you find a flaw, at what point does it become safety-critical? Which brings you back to your original question "What I cannot ascertain if it is safe to ride or to sell"

Regards
tim-b

[beardy]

I think the Trek is a well made frame (should do 20000 miles or more, easy)

Not very reassuring, the bike is 14 years old.

We could be looking at it already having done 70,000 miles if an only bike or 28,000 as a "best" bike with others for doing the donkey work.

[pliptrot]

The frame in question has been in storage most of it's life (I have a job which is far too intrusive) and I reckon it has done less than 10,000 miles. My prinicpal worry was the temporal deterioration of composite material. It is now the only "fast" (unlike the owner) bike I have and will do 10,000 miles before the end of 2015. Based on comments above (for which, many thanks) that would seem like an appropriate time to visit my favourite frame builder in Croydon and have a proper frame built to replace it, which (in the latest steel exotica) may add only 1lb to the whole thing. Apprehension when riding is very heavy. Interestingly the best feature of the Trek frame is that the steerer tube is steel - no aluminium/carbon bonds to worry about.

[mig]

is your main concern that the tubes themselves may fail ? or that the bonds between them have deteriorated?

or both of course?!

[pliptrot]

Given my experience with Araldaite, the glue is the sticky issue. I know that Duralinox frames quite often came apart under their riders due to glue failure (although the original aluminium forks were also not up to the job) and I have no objective knowledge that technology has moved on that much.

[JohnW]

I think the Trek is a well made frame (should do 20000 miles or more, easy)...........

Brucey - 20,000 miles was two years cycling for me up to a couple of years ago - now it's three years cycling.

OK, most of us will have more than one bike, but 20,000 miles is nothing is it? - for a steel frame - the oldest that I am still riding was built in 1979 and it's in use all the time..........not every ride of course but you know what I mean...........If I thought that a frame was only going to give me 20,000 miles I wouldn't buy it. 20,000 miles is two sets of wheel-rims (given the reduced mileage that I currently get from Mavics)..........20,000 miles is four sets of tyres (possibly 5 on the rear).........20,000 miles is less than the life of a bottom bracket, about half the life of a traditional build-it-yourself Campag BB..........20,000 miles is the life of 4 block/chain ensembles (5 in wet weather).

If I thought that a frame was only going to last the life of four sets of tyres, I wouldn't buy it.........would you?

My choice was always going to be steel for frames, but at "20,000 miles........." for a "well made frame............." of any material - is that right?

I had a Pennine 531 frame that I'd easily done that on, and I had a minor prang that slightly bent the top-tube. That was about 1981, and a friend took it for a hack commuting frame. He's ridden it ever since (not excessive miles by any means, but for over 30 years!) and he had it resprayed last year for further use.

Now all readers of the forum will know me as an old anachronism, and this is just my opinion, but for what its worth, if 20,000 miles is good then it's not what I call a good frame - and I'll remain an old fashioned stick-in-the-mud.

[Brucey]

I thought that remark would stir up some comment!

Trek offered those frames with a really good warranty IIRC. But arguably if they lasted a season's race use and did another as a training machine, then it would be 'job done'. If they were built to last 100K then maybe they would be built too heavy...?

Of the mainstream carbon frames I think the Trek ones are pretty good. But anything can happen to one in 14 years; if I'd had one and it had never seen winter road salt etc I might still trust it but I wouldn't be minded to buy one from someone else....

cheers

[Italia50]

i recently had a front end ding and have to conclude its 'new set of forks time'....

[Italia50]

I have a friend who is workshop mananger for quite a well known shop, his experience is that a carbon frame has a working lifespan of just 2-5yrs.After that its just fingers crossed and good (or bad) luck!!

[reohn2]

Re the CF 20K wornout/don't trust it scenario,the thing about steel or alu or any metal,they can be recycled once broken or bent.
Can CF?
In a world of diminishing resources and increasing population is CF and it's short lifespan total loss usage really such a wise choice?