A Thing or Two from Interbike 2016

Meld is a bicycle saddle manufacturer that makes custom saddles based on individual cyclists’ anatomies. We had a booth at the recent Interbike 2016, and it was great talking to a whole bunch of cycling folks. We thought we’d try and summarize some of the discussions and common questions asked.

Bike/Saddle fit

As part of the anatomy geometry capturing process, we have the cyclist sit on a piece of foam (also called a ‘crush box’). The foam deforms from the pressure exerted on it, and is subsequently scanned and used to create a saddle model.

The imprints captured always show an asymmetry between the left and right sides, i.e. they are not mirror images of each other. Accordingly, one of the most frequently asked questions is whether the saddle should be made asymmetric based on the imprint to provide a better fit. The answer appears to be ‘No’, for almost everyone.

  1. Everyone has a natural bias, i.e. we tend to lean a little to one side when sitting, even when our body is symmetrical. This has been the case for every imprint we’ve seen. Utilizing the asymmetry captured in the foam will aggravate/create issues when there isn’t one.
  2. Asymmetry, if present, is significantly more likely to be leg-length. Based on anecdotal evidence, 20 years’ worth of client analysis resulted in just 6 cyclists being observed to have pelvic abnormalities. Resolving leg-length differences included (a) adding a shim to the bottom of the relevant foot and (b) physical therapy before getting onto the bike.
  3. The pelvis needs to be as stable and horizontal as possible while cycling since any movement of the pelvis impacts the kinetic chain (knees, ankles, feet). A flatter saddle (side-to-side) will help keep the pelvis in a horizontal position when we aren’t sitting perfectly along its axis lengthwise. On the other hand, a saddle that’s more curved (side-to-side) can cause the pelvis to tilt to one side when we aren’t perfectly aligned.

Capturing sitting posture on foam

A question that comes up fairly often is the need to adopt the same cycling posture when sitting on the imprint foam. Since we utilize software to help shape the saddle, the question becomes the level of contribution humans should make towards the shape. Based on multiple trials, the answer quickly became ‘as little as possible’, because:

  1. Sitting on the foam captures a single, static position. When cycling, our legs move, and many of us also move forward and backwards on the saddle. The foam imprint doesn’t capture the continuous range of positions we may be in at various times of a ride, nor does it capture necessary changes to the saddle shape to better support the body while pedaling. We need to extrapolate, with additional inputs from the cyclist (e.g. via the movement fore/aft parameter), from that single static sitting position to accommodate various postures and positions. This is achieved via software.
  2. It is difficult to get people to simply sit upright on the foam. Initially the single page instruction sheet simply said, ‘Sit upright on the foam’. We now have instructions filling the entire page and it includes pictures. In order for the process to be sufficiently user-friendly, we (humans) need to do the minimum possible, and let everything else be handled by software.
  3. Anything that’s manually shaped by us, especially when we are untrained, will not look as good as that shaped by machines. If we use the foam imprint as is, we’ll end up with a saddle that looks ugly. And absolutely no one will put an ugly saddle on their bike.

Time needed to get a saddle made

Depending on how far you are from California, shipping can take anywhere between 1–3 days in the US, so 2–6 days for the imprint kit shipping to and from you. For the saddle manufacturing, it depends on the size of the queue. In the best case, we take at most 2 weeks to make each saddle, 1 week if everything goes well.

To provide a better sense of the time involved, the estimated time-to-completion is shown in the payment page (accessed via the ‘$’ icon in the dashboard). It’s definitely not 100% accurate since it’s influenced by factors sometimes outside of our control, but hopefully it helps in some way.

Can you adjust a particular dimension, say saddle length?

We started out allowing for significant user control over minute details of the saddle. Nose width and saddle length were two of them. The problem is that people weren’t sure what numbers to use. Our initial response was to start from or use the lengths corresponding to their current saddle, but there’s always the nagging feeling that they’re missing out on something if the saddle length isn’t set precisely to 27.38491 cm.

The other concern with enormous user control over saddle shape is the increased likelihood of creating a less comfortable saddle. It is difficult to get a complete picture / know all the facts necessary to create a good saddle shape (we have a blog article on this, which we call the Fog of War). As far as possible, the input parameters available impact functionality (e.g. movement fore/aft) and aesthetics (e.g. graphics), not comfort. The only parameter that affects comfort is the availability of a channel or cutout, and that’s because we cannot detect all cases where one is needed (e.g. medical/nerve damage situations).

Found this article useful? Check out other cycling-related stuff at meld3d.com/blog.