DIY Carbon Fiber Bike Saddle v1

I got down this rabbit hole from listening to the very nerdy Escape Collective podcasts that explored body position and how TT saddles may allow for better hip rotation even on a road bike. The Wove saddles seemed to be popping up on every pro gravel privateer's bikes, and I was intrigued. Their CEO did a podcast with Escape and everything he said made sense to me. The one issue was that these saddles sell for a mind boggling $500. I do not believe this price is unfair for a high quality product made in the USA, however I am a college student and unfortunately don't have this kind of money lying around. 

I have always enjoyed making things myself, and for whatever reason I had seen some youtube videos from Easy Composites about making carbon fiber parts. It didn't seem particularly challenging to do so from a 3d printed mold, and there the idea was planted. 

Design:

I began modeling my ideal saddle, taking deisgn inspiration from the Wove Mags saddle, Fizik Aeris, and a few other TT/Tri saddles. I had been riding an old ISM Adamo from my girlfriend's family's bike closet that was great, but also unbelievably heavy which was part of the motivation for making a carbon saddle. I had some troubles designing an organically curved saddle pad in Fusion360, my CAD software of choice (because it’s free, of course), but I eventually settled on a design I liked. I modeled these as super thick with the right top shape for the pads. I duplicated these, offset them to create the right pad thickness I desired, and then cut the final pads into their shape by removing the offset section. For the mold, I took copies of these pads and connected them to create a solid shape for which to lay the carbon fiber into. This used the underside shape of the pads to create a perfectly shaped shell.

Materials:

I printed the molds in two parts that I connected with alignment pins on UGA's makerspace printers out of PLA. I used Vaseline and acetone mixed as a release agent as I read somewhere online that it worked, and I didn't want to buy another thing for my first project that I truthfully didn't expect to be successful. The carbon and epoxy I used came straight from amazon in a small kit for less than $30 if I remember correctly. I also bought some generic steel saddle rails- these are surpisingly hard to source. Vacuum bagging materials were peel ply, breather cloth, and generic home vacuum bags like you'd use for clothes. Tools were mixing cups, paint brushes, and dremel.

Creation:

This was the very first time I ever worked with carbon fiber. Probably not the brightest idea as a saddle failure sounds rather unpleasant, but I've always had slightly questionable judgement. We (myself and Logan, who is always up for these questionable projects) coated the molds with release agent, cut the carbon fiber alternating between 0/90 degree orientation and 45 degree. I believe we used 5 layers and its been plenty stiff and strong enough, but more probably would've been better. We went layer by layer, adding epoxy, the carbon sheet, wetting out, and then going to the next. We finally added peel ply and breather and then vacuum bagged it. We put the vacuum port in the middle of the shell which was a bad idea. It distorted and wrinkled the carbon a bit, but this was only visible on the bottom so not the end of the world.

I did not consider the interaction between the saddle shell and rails- this was a major oversight. Refusing to be deferred, I decided to cut the rails to match the shell as best as I could using my newly acquired dremel. Not wanting to hold it while using a dangerously fast spinning wheel inches away, I decided to tape the rails on the pillars outside my apartment. Because thats smart. It was a process to get it to somewhat fit the saddle's shape, but eventually I got it decent enough with the tools I had. I then considered the stress concentration of the relatively small rail contact patch on the carbon shell. This seemed like a recipe for cracking the shell with the point load. To combat this, I modeled a larger encompassing section where the rails would attach that I would fill with epoxy and chopped carbon around the rails. Designing the mold for this was a little tedious as I tried to make it as self aligning as I could, and it took a couple revisions to come to something I was happy with. After the rails were put into the mold, we stuffed the mold sections with as much carbon and epoxy as we could using a pick. Very janky process to say the least. I didn’t consider the removal from the mold and ended up having to submerge them in boiling water to be able to peel away the PLA from the carbon/epoxy sections. When done, it fit the shell far better than simply attaching the rails would’ve and I was much less worried about it cracking from the point stress.

To join the two parts we now had, we put lots of epoxy on the contact areas between the two pieces, laid the rails onto the shell, and then covered the places we could with one or two more sheets of carbon fiber to reinforce their connection, especially if the bike was lifted by the saddle. I didn’t realize until after, but the rails were most definitely not perfectly straight vs the shell. I can’t notice it when riding though so no big deal.

Pads:

I wanted to create 3d printed pads like all the big companies are doing nowadays. I didn’t have access to a printer that could print in TPU, so I sent it to someone online to print for me with instructions for 30% gyroid infill with no walls. No walls would show the infill patters as done in professionally manufactured saddles, and 30% was the infill percentage recommended by ChatGPT. When they came in they were absolutely rock solid- I was very upset. Apparently the guy I sent it to thought I was using these for the saddle shell itself so printed them at like 85%? Not sure how he came to that conclusion. He then sent over some more proper ones after more waiting, and these were also disappointing. Even from when I got them, they were starting to tear themselves apart because of the no walls I requested and some suspected carelessness. Oh well, I just wanted to get it done. 

After clear coating the shell, I bonded on the pads using 3M marine weld that I found at walmart. Would recommend, it held up super well.

After a month or so, though, these pads were completely shredding. This coincided well with when I bought my own printer, so I then printed new pads with solid outer walls which have proven much harder wearing.

The finished weight of the saddle with the first pads was 200g even. For comparison, the ISM I was riding before this was around 450g, so the weight savings are pretty significant. 

Riding: 

Well, it feels and rides like a saddle which is something I was worried about. I made sure to get pictures of it before to show off and talk about in internship interviews (a side motivation of this project) because I was sure it would break immediately. Fortunately it did not and has proven reliable over the past handful of months at this point. 

The first pads were comfortable and maybe a bit soft until they self destructed, and the newer ones with walls I printed (2-3 walls, 35% double line infill that I came to from testing small sample pieces) are far too firm. I’m working on another saddle when I have time that’ll be somewhere in the middle of the two. Maybe two walls and 35% single infill? Still testing and playing around with settings.

Conclusion:

Not sure if I would necessarily recommend this to others, especially as a first carbon fiber project, but it’s worked out well for me so far. I have honestly forgotten to be worried about it for better or worse, but thankfully it’s just on my road bike so there’s less room for failure vs gravel or mtb. I’m exploring different shapes for this next one and will try to make carbon fiber rails to make it a bit lighter. Also experimenting with 3d print infill percentages to dial in the comfort a bit better.