Not-So Skilled Labor

Not-So Skilled Labor

In all the time I’ve been visiting bike companies, I’ve never once handled pre-pregnated (better known as pre-preg) carbon fiber while in its still-useful state. I’ve handled a variety of materials in an inert state, well past their useful layup phase. So while I knew that pre-preg was sticky, all I had to go on was what I’d been told. Put another way, it’s the difference between theory and practice.

While I was Allied Cycle Works, I was informed that I would layup a tube. This was news to me. And from the way they were talking it sounded like the tube was destined for my bike. I was going to lay up a tube for the bike I was going to review. I was going to build part of a carbon fiber bike. What reassured me is that if I did a bad job I was the only person who could get hurt.

As reassurances go, it was of limited value.

Allied’s head of engineering, Sam Pickman, walked me through the process of laying up the sheets of carbon fiber on a mandrel. The plastic mandrel, which had the shape of the top tube, only very slightly smaller, was covered in a mylar bladder to keep the pre-preg carbon from sticking to the mandrel. Of course, there’s a flip side to being less sticky than the mandrel. It’s not very sticky, at all, so when you go to stick carbon fiber on the mylar bladder, it is something other than easy. But you manage.

We started near the back of the tube, on the drive side, with a ply that didn’t go all the way around. The layup schedule showed how the sheet was placed, which is to say what orientation, as well as the location. But figuring out exactly where that is on the mandrel would have been impossible without some guide marks that had been drawn on the mandrel. At the very front of the mandrel, where no carbon would be laid, the letters, T, B, D and N had been written to reflect top, bottom, drive side and non-drive side, to make sure that workers could tell the proper orientation of the tube and the layers on it. When you picked up the mandrel, it was easy enough to tell, as it tapered in diameter from front to back and it flattened toward the rear; there was also a slight downward curve front to back; these are all features I’ve come to expect to see in top tubes. However, lay that thing down on its side and it tends to roll over onto the top surface; as a result, there were plenty of times I nearly laid a layer on the bottom of the mandrel, rather than the side.

I didn’t have my reading glasses with me, so the process of removing the backing from the pre-preg wasn’t super easy. The surface on which I worked was non-stick, so the approach is to place the pre-preg sticky side down and then take a razor and score the backing at a corner until it comes up. You’ve got to do this at a corner into which the fibers run, not against them because if you try to pull the backing off across the sheet you’ll simply pull the fibers apart. There were plenty of times when I laid the sheet down, went to score a corner and then picked it up to double check that I hadn’t lost track of which corner when I flipped it over.

Once that first layer is down, laying additional plies is much easier. Pre-preg loves to stick to pre-preg. Think of how tape will stick to itself. It’s not quite that sticky, but it’s close. It also comes with that level risk. You’ve got some level of ability to line up the sheet before pressing it down, but once two plies have adhered to one another you’re pretty much done. So getting it right the first time is terribly important.


How difficult it is to lay a ply is influenced by the orientation of the fibers. The long, rectangular strips of 90s that are used to combat bending forces are really easy to lay down. However, there were sheets of 0s and 18s that loved to just instantly drape across the tube and because they wrapped three-quarters of the way around I had to work to make sure that I got the positioning right near the middle of the ply and then gradually work out from there, the way you would with a bumper sticker. One more than one occasion as I gently held a corner aloft with my left hand to keep part of a ply from sticking as I smoothed with my right hand, I damaged a corner of the sheet by pulling the corner apart. I was shocked because I didn’t think I’d had any tension on it.

Another challenge is that when you smooth the ply you must do it in line with the orientation of the fibers, not across them because you’ll risk pulling the fibers apart the way I did in those corners.

There were two dozen-ish plies in the top tube. I was too busy laying them to really count. The plies went down in a pattern that was almost like a palindrome in that the mirror image of the first ply I laid was the last ply I laid. The plies in the middle were long, rectangular strips on the top and bottom of the top tube, arguably the most symmetrical pieces in the top tube.

Sam was by my side for nearly every step of work, congratulating me on what I did well in addition to coaching me on difficulties to watch for. On more than one occasion he reassured me that minor imperfections in my technique and orientation weren’t the end of the world. It gave me the sense that the work in Asia isn’t always as skilled as I think it is.

I was relieved however when we got to the end and he let me know that the front triangle of the Alfa All Road that I would be reviewing was already built. The top tube that I’d just sweated over was going in the trash. My reaction was equal parts relief and disappointment. The relief was obvious, no? Was it really a good idea for anyone, me included, to be riding the first carbon fiber tube I’d ever laid up? Probably not. But I was also a bit disappointed because I realized I would have been very curious to see that tube molded into a finished piece, if only to see how my imperfections played out.

After Sam slid the mandrel out of the bladder and then twisted the mylar bladder out of the tube in a process much like tightening the twist tie on a loaf of bread, he looked at it and pronounced, “Not bad. It’s really pretty good.” Then he looked at the clock. I’d taken 45 minutes. Then he told me, “I’m not quite sure you’re ready for the production floor yet. The ladies in there do a top tube in about seven minutes.”

I’ve brazed steel joints with silver. While laying up carbon fiber didn’t have the danger factor of working with molten metal, performing layup was every bit as difficult, every bit as dependent on experience and technique, every bit as nerve-wracking as brazing steel.

, , , , , ,


  1. Jeremiah

    Thanks for the great read. It’s really interesting to hear your perspective on the skill necessary to do the carbon-fiber lay up. Speaking as a total outsider to bike manufacturing, this was an enjoyable piece.

  2. Scott G.

    In Italian factories there are pre preg strips of the La Gazzetta dello Sport added,
    interleaved with the carbon for damping, cheaper than flax.

    1. Author

      I’d like to see photos of that. It seems inconceivable as it would allow the layers on either side of the newspaper to shear away from each other. I mean, that’s an idea that seems dumber than building a bike out of rubber.

  3. TomInAlbany

    My 66 year old mother-in-law does lay-ups for a carbon fiber plant in way-upstate-NY. And, she’s respected in the company. She knows what she’s doing and is caring enough to do it right.

    She never walked my through it, though. I’m going to have to ask her to do that now.

    Thanks, Padraig!

    1. Author

      So that last layer is Inegra and yes it helps with bar impact, but it also dampens vibration to help give a smoother ride.

    1. Author

      I was thinking of you as I was going through this. I can’t imagine how you teach this.

  4. Michael Bell

    Very Mike Rowe of you, but then he would’ve worked in an all too easy joke about the “pre-preg”.

    I’d assume all the pieces are precut, maybe laser cut? Are they labeled or just presented in the order they are to be used?

    Interesting that it seems to be all women on the production line. A local medical device manufacturer (stents, etc.) uses all women to make their products, because of the similarity to handicrafts like sewing.

    1. Author

      They use a gigantic plotter with a blade, rather than a laser cutter or a large die-cast plate.

      Yes, the production line is all women. One person said every time they hired a man to work the production line they decided they were an engineer and could figure out a better way to do things. That was not quite what they were looking for.

  5. RSN

    I took a steel frame building class at Brew Bikes in North Carolina last fall that included a TIG welding component. Never was I more relieved to hear that Steve, who has welded thousands of frames, would be welding mine too. TIG welding is hard, and steel tubes are surprisingly delicate.

Leave a Reply

Your email address will not be published. Required fields are marked *