Technical FAQ: how to find cracks in carbon frames

Published March 3, 2020 10:14AM

Dear Lennard,
An x-ray technician recently told me that fluorescent penetrant inspection could work to inspect carbon fiber parts for defects. Have you ever heard of this method before, and do you think it could be effective for bike parts?
—Chris

Dear Chris,
Good question! Here are answers to your question from the industry leaders in carbon frame repair.
― Lennard

From Calfee Design:
“It can help to see the very fine cracks on damaged frames. But we use a 10X loupe for that.”
Craig Calfee
Calfee Design, Inc.

From Ruckus Composites:
“We have been using fluorescent penetrant dye for about 8 years out here. Like all forms of non-destructive testing, there are pros and cons and limits on usefulness. Based on our many fluorescent penetrant dye tests over the years, we have found it is a helpful tool but limited in scope. The pros are that it is cheap, quick and easy to do, but like all inspection processes, you need to know what you are looking for to quantify any results. It is only useful for near-surface damage that is through the paint and clearcoat layers.

There are also many types of dyes out there. Fluoro dye testing is primarily used for crack detection in metallic surfaces; I believe the train industry popularized it. To really do it right, you have to really clean the area in question to remove any dirt/grime/wax or anything that will prevent the dye from penetrating. Another aspect that makes it difficult on bikes is that you really need to soak out the area in question for a period at a specific temperature for proper penetration. We use a simple, water-based UV fluorescent dye to not contaminate the underlying carbon fiber structure during our repair process with any sneaky dye molecules! While it is another useful tool for a proper carbon fiber facility, because of the thin materials used on all bike frames and components, you can most likely visually see any cracks developing if you use a good inspection flashlight and take your time. But UV dyes make for some very compelling images!

Dear Lennard,
For a long time, I have been whining about the lousy skewers that Rocky Mounts are using on their roof racks. How after a month they no longer lock, and, unless if they’re lubed regularly, they get really stiff and don’t clamp properly. Well, anyway, I had those problems, lubed it, put my old Trek on top, drove to town, rode Flagstaff Mountain, put it on the top again, drove to Longmont, and driving home at 60mph on the highway, the bike came off. Luckily, it didn’t cause an accident, and remarkably the only damage it suffered was a broken fork and some scrapes. There is no other visible structural damage.

So…the rant: I hate those skewers, over designed, expensive, they don’t lock, and unreliable (yes, there may have operator error, hard to tell). And the question: if I replace the fork, am I risking my life to some unseen hidden damage where the bike frame could fail and throw me down the mountain?
—Derek

Dear Derek,
Bummer! Glad you asked. Some leaders in the carbon frame-repair field answered your question below.
― Lennard

From Broken Carbon:
“Any time we come across a situation like this, we automatically recommend a replacement frame. Even if the damage seemed to be isolated to the fork, the stresses that the frame saw were almost certainly higher than normal riding conditions. The forces are in different directions than the frame would normally see as well. It’s just not worth the risk of something happening down the line.”
Brady Kappius
Founder, Broken Carbon

From Calfee Design:
“We do this kind of inspection quite a bit. We clean the frame and look closely for cracks. A coin tap test is very effective. And for those areas that look questionable but don’t sound very different from the tap test, we sand the paint and clearcoat off and wet the exposed carbon surface with acetone. You can quickly see where acetone stays wet in a crack as it evaporates. Similar to the flouro-dye test but without the flashy colors. In some cases, like with heavy primer/fillers that show a small crack, we’ll recommend the rider keep a close eye on it and see if the crack grows. A tiny mark is placed at the end of the crack with a razor blade. 90% of the time, it’s a paint crack that doesn’t grow. 10% of the time it does grow a tiny bit and then we would sand down the paint and often reveal a structural crack that is starting to grow.

Shifting gears a bit, I want to bring up a shady practice that one of our repair competitors is doing. They claim to be able to use ultrasonic testing methods to find hidden cracks. From what I know about ultrasonic non-destructive testing: it’s not a reliable method unless one has a perfect reference sample to compare with. And that’s only with sophisticated (expensive) equipment.

Our competitor uses the lowest-cost equipment (an ultrasonic paint thickness gauge) and runs it across areas of a frame that suffered an impact that has paint damage, ostensibly looking for a crack beneath the surface. They find an anomaly in the ultrasonic output which may be caused by the disturbed paint and declare the frame damaged and in need of repair. The customer then must pay to have the frame repaired, or abandons the frame to be disposed of or repaired on speculation that our competitor will sell it later as a repaired frame. Furthermore, competitor claims to be certified by the ASNT Level II. A quick check on ASNT.org and you can see they are not listed. Also, the competitor publishes that it’s the American Standard for Non Destructive Testing (as shown in attached screenshot). But ASNT is actually the American Society of Non Destructive Testing. A convenient mistake?

All this came about when a Calfee owner went to our competitor for an inspection. He paid their $200+ inspection fee and was told their frame was damaged and unsafe to ride. Not feeling confident of that evaluation, he sent us the frame. We inspected it and found no damage beyond some chipped paint. I’ve got photos proving all of this if you are interested. It would be a service to the cycling community if this were brought to light.”
Craig Calfee
Founder, Calfee Design, Inc.

From Ruckus Composites:
“Great question and a surprisingly common scenario. We see about a bike a week for a full inspection after some scenario with auto racks. I usually look at it from a projectile motion viewpoint. Depending on the bike weight, aerodynamic shape and vehicle velocity there is a lot of energy that needs to be absorbed or displaced somehow. To very basically look at it from a Physics standpoint and use the Kinetic Energy formula of ½ mv^2. (m=25 lbs v=65 mph) you get a shocking nearly 4800 Joules of energy that has to go somewhere. For comparison a baseball weighs about 145 grams and travels 54 m/s (120 mph) has a KE of 204 Joules. A 0.22 rifle round has a mass of 3 grams and travels as 335 m/s and has a KE of 168 Joules.

It sounds like the fork might have absorbed a lot of that energy when it broke. From my experience, the frame is likely damaged as well. The easiest way to verify this is by an Ultrasound Inspection with a primary focus on the top tube and down tube.”
Shawn Small
Owner/Engineer, Ruckus Composites

Dear Lennard,
I have an e-bike, and it has to be leaning on a wall perfectly still when I turn it on, or I get an error code. Holding it up isn’t good enough. Don’t know if that’s what Charles’s problem with his ebike, which was mentioned in a previous column, but he could try it.
—Tamar