Getting precision fit for my bike


I've been waiting to get a frame custom welded for touring, gravel, and possibly some randonneuring. At the end of last month, I rode out to Freewheel Bike in Eden Prairie to use their Precision Fit system and find out exactly what I need. 


It was the warmest day that we've had since last fall. Actually up to 60F, and sunny. Kind of windy, but still, really nice. So, of course, all of the fair weather riders were out again. It was really funny to see them all. Tights, long sleeve jerseys and short sleeve jackets, super expensive bikes, pot bellies...


I was in shorts and a t-shirt overheating. 

I thought I had a ride there and back figured out, but that fell through. It was about 17.5 miles to where I was going from my house, so it worried me a bit to ride all that way. Especially since I haven't left a more than 10 mile radius in the last three months. And it was all into a headwind.

I felt so slow, left late, and didn't really know the entire route. I thought for sure I was going to be late. The closer it got to 2:00pm, the more I wondered if maybe I should call and tell them I'd be late. And yet somehow, I ended up rolling up to the door at 2:02pm. It's actually kind of odd, but stuff like that has been happening a lot lately. Time is kind of not there, not something I have to think about. Just falls into place, without trying at all. I walk back into the kitchen as the microwave or oven timer hits zero, have conference calls that last exactly one hour and eleven minutes... Stuff like that.

How it worked

First - The Bike
They had told me to expect two to two and a half hours to go through the process. I thought that meant we'd be riding on the machine most of the time. 

Thankfully, it was mostly measuring my body and figuring out what type of fit I wanted. 

We started by talking about what I wanted while they measured my current ride to get a starting point. The level of detail and number of points looked at were pretty extensive, which made me feel pretty good about the whole process. A lot of questions about what issues I had now, what I was looking to fix, stuff like that. Probably a good 15-20 min just looking at the existing bike. Since my goal wasn't to fix things for this particular bike as much as it was to figure out the optimal geometry for the bike I want custom built. I think that was probably a bit weird compared to what other people using the system might need.

The fit I was getting was $150, you could spend $250 if you wanted full motion capture analysis. That's kind of the thing that only really serious cyclists would be into, typically ones that actually race. When you need to analyze every single motion and measure the power output and all that silly stuff. 

That's not me at all. I'm looking for a touring / rando frame that lets me ride comfortably for hours at a time with cargo, multiple days in a row. It's not the power output, it's the comfort that I'm worried about. And when you are talking about repeating a motion millions of times, even the tenths of millimeters mater. 

Second - My Body
This was one of the key things that I wanted to focus on. I've had a fair amount of injuries over the last few years, and some life long issues with the symmetry of my body. That was the real thing that I wanted to figure out. 

The issue that seemed to be the most important and what I couldn't figure out was the combo of cleat position, seat fore/aft position, and leg extension. I had actually thought that my legs were completely different lengths, and my feet different sizes. How my hips rotate and the angle between my two feet was never right. The left one would point in the direction I was facing, and the right was pointed about 30 degrees out on the right. So that was also really hard to compensate for. Bring the front of the foot in to kick the heel out, and it would change the angle of all the bones, ligaments, and muscles up into the hip would be off. So I would have tightness, numbness, and pain would run up the outside of the right leg. And to compensate for that, I'd move the left foot to where it always had pain on the quad just above the knee.

Turns out, I was mostly correct. I just didn't know how to measure it and correct it.

The feet weren't actually much different in length, but there are a lot of differences in shape. They measured and recorded the position of all the relevant outer points and also the arch height in both seated and standing positions. How my metatarsal bones are arranged is pretty different, and was even more pronounced when under pressure. Turns out that's pretty important when determining cleat position. The width and outer / inner positions of the extremes on mine were off by quite a few millimeters. Sure, that doesn't sound like much, but really, it is. Add it in with the difference in angle on my legs from hip down, and it's an even bigger issue.

My cleats were way too far forward. I had them towards the front of the ball of my foot, but they should have been further back, closer to the line through the inner / outer extremes on the feet. They were also nearly completely worn out, so I needed to buy a new pair. Oops. Should have checked that before going down there. 

There were a lot more measurements going up from there. And looking at how my injuries have affected my body. The left arm / shoulder is really off. About 5-7mm higher on the shoulder, and the arm is both longer and can't be used in the same positions as the right. 

Third - Dialing In
That's why I was there, right? Pretty much the whole point of getting fit. The machine was kind of a jig for moving every possible aspect of a bike around. Even the cranks could be lengthened and shortened by about 1.5 centimeters in either direction.

The photo's a little blurry, but this is what it looked like:


On the left, you can see the two comfortable chairs, which sat below a kind of shelf barrier with water bottles if you were getting thirsty riding. And extra chairs for your coach or if you had a friend / SO that's coming along with you. 

First they set it up with all the same angles and distances as my previous bike, and then moved to what the measurements on my body suggested should be moved. My pedals had been transferred over, and they had a bar of the same width as mine set up with dummy brakes. Even the bars and hoods were set the same as what I had.

We also used my Brooks Team Pro saddle since I knew I liked that. I've actually been riding a different synthetic saddle all winter with almost exactly the same measurements, and it's even more comfortable because it resists and doesn't conform to my body. The Brooks has worn in to fit my broken position and uneven weight distribution. It's a thing that tends to just compound problems. The dimples for my sit bones have gotten deeper depending on angle, and that's led them to keep getting deeper, leading to even more uneven distribution... Classic vicious cycle. But I do know that most of the dimensions on it are already perfect for my anatomy, so there was no need to go through the yellow saddles in the pic.

The wear of my saddle did actually change the measurements for seat angle. Because there wasn't a sure center line, and because a tilt of 1-2 degrees left / right would tilt the front / back position, I ended up being off quite a bit at the start while also thinking it was at the same angle I had it. Since I didn't know it was off, it made me interpret the rest of the distances differently.

We fixed it pretty quickly, but even later trying to dial things in more, it was apparent that the tool to measure couldn't deal with my particular saddle well. If you have a Brooks, it can move 2-5 degrees trying to measure if it tilts. Probably not an issue for most of the people using the system, who are probably using lightweight totally synthetic components that are 100% symmetric.  

I stepped up onto the pedals in a similar way to getting onto a trainer or rollers (taller, pedal first without a platform to stand on) and then started pedalling slowly while they fired up a small notebook PC on a bar extending from the place the handlebars were bolted on.

That was the thing that really bummed me out. The system couldn't connect to the computer. So I wasn't able to actually see things like my power output per leg. We spent a fair amount of time trying to get it going, but it was on Windoze 8.1 so that makes me think it's kind of scary. Especially since the GUI launched with nothing but a task bar and only had "About" in the help menu. It surprised me that a system that cost so much time and money to develop would have such a horrible UI. Especially if it's a product that is being marketed towards bicycle shops, who are often not all that technically savvy. But I'm a programmer who really tries to focus on UI / usability of the tools I build. 

Regardless of the netbook working or not, we were still able to get a lot done.


 Not a lot of difference, but the few that are larger offer a lot of insight. 

When I showed up at the shop, The issues I had were starting to show up from the 17.3 mile ride to them. So over the course of talking, the pain from the ride was gone, and I was relaxed from the chair and just resting / talking. I think that actually helped a lot. So even though I haven't really ridden enough in six months to remember where the pain happens exactly it was really easy to identify. 


What worked for me?

  • The ride out there wearing me down actually helped me focus on what I really needed. Originally I had thought I would get a ride out, and I'm actually glad I didn't. Could have dealt without the tailwind, but honestly, the distance and the fact I rode there really helped make sure that it was accurate to my state while on the road.
  • The level of care measuring the angles on everything and the understanding/explanation of how the movements would work. And also how to check these things myself in the future when I do things like clean my bike to the point of removing parts, installing new components, replacing cleats... The communication there was really nice.

What would I change?

  • I'd offer a version that doesn't use the computer but just focused on the human / jig measurements. We didn't use the computer at all. The result was still good, but it felt like that part was lacking.
  • The jig should have two adjustable items for the handlebars. One to set top tube height, and the other (inside of it) would set the height of the stem. That would have the part that moves forward or back. This way, you're not just stretching the front of the bike. And if you are being fit for an existing bike, you can set a fixed point for that forward / back length for the headset and seat tube. It can still all be 90 degree angles, but you could use some really simple trig to get the proper stem length and angle. And actually probably calculate that based on the max height of the fork that's in the frame. If I was trying to perfect an existing frame, I'd want that a lot. Having the saddle starting position fixed that way too would end up helping people just replace components rather than deciding that things are totally unfixable, and also tell the shops exactly what they need to find for a part. 
  • Resistance on the cog you spin. Maybe it's part of the others tests with the computer, but it would be really nice to feel it under different uses. Since I'm thinking touring, I would have liked to feel what the system would have felt like hauling 60 pounds up a 3% grade, stuff like that.  

Do I have any ideas?

  •  I'd really love to see an open source / instructables way of making something like this. The real thing that mattered was the measurements, and the degree of precision that they could be made with. But it didn't really require a computer at all. It was just being able to move things precisely and use the right tools to measure. 
  • It would be really awesome if this became something that could be accessible enough so that custom framebuilders could have them in their own shops. I had to ride 35 miles to get these details, and actually on the way back 8 blocks from my home stopped and chatted with the person that will be welding the frame. If I could have actually been on a machine even without a computer at all talking to him? 

    That would have been far more useful. He's going to be bending the metal, he's going to know 100% what's possible and what's not. I went and gave $150 to another company* to help me communicate with the builder. It feels a bit like that becomes a game of telephone. It seems like that should have been something that was more of the process of working with him to get the minutia right.

    I think that you could probably make a pretty decent analog version pretty cheaply if you know how to weld, so an open source plan to build them might help a lot of builders and riders.  


Overall, I'm pretty happy. It sucked having second headwind on the way back, but even though they were more tired my legs felt less pain. I know a whole lot more about how to think about my fit and work with it. As my riding progresses and I look to get more kinds of bikes, I will probably use it more. It looks like a really great system, even when it's not 100% working. 

* Not dissing on Freewheel here at all.They're all really awesome, and I have dozens of friends that work(ed) for them. Just thinking that if it would be nice if the process of buying a four figure custom frame an "all in one stop" experience