Product testing – or not.

I’ve been thinking about product testing these past few months as I’ve been doing a lot of work with disc brakes on road and cross bikes. The process is simple enough I suppose — come up with an idea and see how it works. The tough part relates to the time involved. In the case of disc brakes I’ve been riding them for a few months now in this snowless winter and beating the crap out of the entire system. I rode them without issue for a few months before even mentioning discs in public because I wanted to have preliminary testing done and out of the way first. But this won’t tell me how well they will last in the long term and if design changes will need to be made to make sure they are safe for a lifetime. So to that end I will be using them for the next year, give or take, before any orders I’ve taken for them come to the front of the queue. In the end they will have more than a year of hard use and abuse and tweaking before I even consider selling them to a customer.  The prototyping of the original JKS was the same way — about a year of testing and tweaking before the first customer ever threw their leg over one. The same thing with the JKS X — about a year of constant hard use and testing before a customer gets one. Being sure and getting it right takes time.

Lately I’ve read some things online about new tubes, tube shapes and fork blades being made available to builders and I wonder aloud if any of this stuff is being tested by the builder before it is offered to the pubic. I think much can be assumed by considering the design of the tubes and the material being used but only real world testing can make the builder sure everything will be OK in the long run. I’ve even heard some builders say that the tube sizes don’t matter in terms of ride quality, stiffness or strength and that the tube sizes are best chosen to match the aesthetics of modern components. I’m all about stuff looking right and agree that tube size changes the look of the bike (sometimes for the better and sometimes not IMO) but these aren’t hemlines or tie widths we are talking about — they are structures that behave a certain way based on the overall design and the materials used to realize that design and we trust our fun and our lives to these structures. Tube size and shape change the ride of a bike — period — full stop. This is not my opinion but simple structural, mechanical and mathematical fact. This can of course be a good or bad thing but one thing is for certain — changing the size/shape of the tubes will change the bike and the way it rides and lasts.

The frustrating thing about this to me is that the builders would know this if they built one and rode it. But for some reason they don’t and think that because they think the big tubes look cool with modern cranksets that that is more than enough. I couldn’t disagree more. Products should be tested first. They should be proven to be the proper stiffness for the intended rider and use, to be strong enough and tough enough (strong and tough are two very different things) and to be safe. If after all that it gives the look the builder is after then that’s gravy. The idea that this is being reversed and that bikes are being designed and built with the look being the most important thing makes me think that the person who came up with the ‘no white after Labor day’ rule should be a tube/bike designer………or maybe not.


Karin and I are in the middle of a slight revamp and remodel of our website and I’ve been putting some thoughts into words over the past few months when the mood strikes me. One of the pieces I’ve written is about my history with product testing and how it affects my product design. I had the luxury to learn proper testing protocol while working at Serotta years ago and it’s this experience that I draw on for my current product testing. This piece will be put into the site and no doubt edited a bit before it ends up in a semi-permanent spot but I will include it here now in its first draft form to give you an idea of what I’m talking about.

One closing thought……….please do not shy away from asking the builder of the bike you will trust to be safe and ride right to tell you how the product was tested before they sold you one. Bike design is not just a matter of fashion and the builder should know more than it looks cool.

Thanks for reading. The ‘testing’ article is below.



Testing —

I’m very proud of the design, materials and processes I use to make my bikes. These have all been developed in the light of things I learned over many years of building and testing of bicycles both on my own and when I ran the R&D/testing program at Serotta. I think it’s safe to say that Serotta has probably done more testing of their products than most any other small company — and I am proud to say I played a large role in that early testing.

After working at Serotta for a few years I transitioned into the role of custom builder and the next logical step was to build and test new products we wanted to offer for sale. This all sounded great but there was little info out there pertaining to acceptable safety standards for bikes and nothing out there that we’d ever seen pertaining to performance standards. So when it came time to build and test prototypes of new stuff we started with a blank slate.

One of the first products we wanted to test was a new carbon fork Serotta called the ‘F1’. It was one of the first carbon forks designed not to just be lighter but to perform better in turns and braking and of course in terms of safety. I was enlisted to help finish the fork’s design when the engineer in charge of R&D left the company before completion of the project. I worked with the composites house to help design and finish the mold used to produce it as well as test the safety, durability and performance of the fork.

What I needed to do was design a test rig that would put the fork through its paces in a safe, controllable and repeatable manner. This meant taking it out and riding into curbs until it broke was off the menu (thank goodness) so I constructed a testing table to do the dirty work. The test table held the fork by the steerer and pushed back and up on the front hub with a pneumatic cylinder with preset and adjustable load. This allowed me to test three things — the deflection when subjected a given load (stiffness), the ultimate strength of the fork (by turning up the load until the fork failed) and the fatigue resistance of the fork (testing at a low load but with a very high number of reps until failure). All good things of course but without some numbers to compare things to the results would mean little. So we set benchmarks by testing forks that had been shown to have low failure rates out in the real world. I tested Serotta’s traditional steel fork as well as carbon forks from most every major manufacturer on the planet. We learned a lot by doing this and it set a consistent and high standard for Serotta’s new F1 fork.

The F1 did pretty well right out of the box in its first iteration. We tweaked the design and it got better and better and got to the point where it was stronger and tougher than even the best steel forks. One of the things we ended up changing was the adhesive used to bond the fork tips into the blades. I’d wondered how hot it got in a trunk of a car and after placing a thermometer in a black car trunk and seeing how hot it got (up to 150°!) I developed the ‘black car trunk test’. I worried that the temperature would get so high that the adhesive would breakdown so I made an oven type device that was placed over the test table so that I could run the fork test at 150° F. Sure enough the adhesive softened and we saw failures where the fork tips were bonded into the blades. A change of adhesive kept those riders who left their bike in the trunk of their car all day so they could ride right after work safe. It was a very good lesson learned. In the end we tested and broke over 160 F1 prototypes before a single one was sent out the door and the real world failure rate was extremely low. Even after the fork was introduced I would pull one out of the bin every few weeks and put it in the test rig and run it until it failed just so we could be sure that something didn’t slip or change over time that would affect the end results.

There is of course more than one type of testing that is important and while safety is paramount it’s a moot point if the fork doesn’t perform as it should. Being a racer and having ridden more different bikes than I could count gave me a good basis to compare different versions of the F1 to one another and to other forks on the market. So I rode all different versions of that fork — some were wonderful and others not so much. We tweaked and fine tuned and I’m proud to say came up with one of the best riding and handling forks of its day and I’d even put it up against the forks of today.

In time the testing program greatly expanded from forks to frames, framesets and stems. I made a modular testing system so I could do all of it on the same test plate. I ran tests similar to what was done with the forks — we bought all kinds of frames (different brands, models and materials) and broke them. After they failed they were cut up, measured and analyzed. I cataloged the results and then set about making sure Serotta’s products surpassed the level of its competitors. This was not all that hard frankly once the standards were set and you could watch the frame fail in front of your eyes on the table. Make the tube butts a bit shorter or longer, change the shape of a lug, braze or weld it differently…all these little things changed how the bikes lasted in the fatigue tests and in time we were able to set a new standard of strength, toughness, durability and safety. The simple changes we made to improve the product would never have been made had we not done the testing that had shown the weak links — it’s hard to improve things unless you know the areas that could use improvement.

The other aspect of testing was performance testing. There are no hard and fast numbers when it comes to performance testing — no clear pass/fail in most cases. My job was to come up with new designs and take them out and ride them to see how they felt to me and then tweak them and reevaluate. It seems like a never-ending cycle but in time certain things became clearer and the focus narrowed. Some things were very straight forward — like — does the Ti road bike ride better with or without a chainstay bridge? It would save the company real time and money if the bike worked as well or better without that little piece of tube welded in between the chainstays so it was worth testing. Build the bike without the bridge and ride it for a few days and then tear it down, weld in the bridge, build it back up again and ride the same roads and hills I did a few days prior. It was much better with the bridge. Other things were less clear. Steering geometry was one of the less clear things and frankly there is a lot of personal taste and preference here. With something like the chainstay bridge it was obvious and I doubt you could find anyone who’d prefer it without the bridge but determining what head angle, fork rake and trail we wanted to use was less black and white. In the end it was decided to make bikes that pleased us first and foremost and then let the chips fall where they may.

The same thing was done with tube diameters, wall thicknesses, and shapes. We tried different brands and alloys and custom mixes of tubes and I got to do a good bit of the testing. I loved this part of the job. It meant going home with different bikes made in my size and riding them hard to see how they felt. At first it was hard to tell the difference in certain things but in time one develops a feel that they can trust. In a way I think it might be like wine tasting. With time and attention one really starts to recognize and appreciate the differences.

I learned much during that intense time.  I didn’t realize then that I was doing work that wasn’t being done at other small companies — in fact I figured they all did this — but I was wrong. Very few did anything more than build something and ride it around for a bit before selling it to the public………and some didn’t even do that — they would just build it and sell it and the first time they were ridden was by paying customers. The wonderful thing it did for me was to teach me how to design both frames and frame components, and build in a way I would never have done without going through this long process. It also let me ride many bikes in many different configurations and it was through this process that I was able to develop a feel for how small changes can make profound differences in the end product. I was fortunate to be given a big task, have it well funded and then let loose to get the job done and learn as much as I could. It was a lot of fun and the knowledge and experience I gained at that time helps shape the bikes I build today.

This entry was posted in Bike, Musings, Process.  

Share this Article:

4 responses to “Product testing – or not.”

  1. Jon K says:


    Great post(s). I’ve read about your testing experience at Serotta in the past. I’m curious if you do any bench testing today. Is a year of hard braking on a test frame enough to satisfy you? Are there companies out there that could run one of your disc forks through a million cycles on a test rig and see what happens?

    On the other hand, is the hand wringing over discs on the road misplaced? Its not like we haven’t had discs off road for more than a dozen years. Are the forces involved that different?



  2. kirks says:

    Hey there,

    Thanks for the good questions.

    Yes I do a good bit of bench testing. In the case of the disc braked fork I’ve been doing static load and deflection tests to compare blades before and after the mounting tab is installed. I also do load tests that checks the deflection along the blade so I can measure how the flex is distributed over the blade. The value in this is finding out if the blade is ‘hinging’ at any point. It’s a pretty sure bet that if you get an area of concentrated flex (hinging) that you are much more likely to have a problem there.

    I also do static load failure tests. In effect I hold the fork and add weight to it until it fails and then I study the failure mode and location and make changes if necessary. All fatigue testing takes place on the road after the previous tests have shown me I’m in the ballpark. For the disc brake I’m doing long repeated down hills on washboard dirt while dragging the brake to give the fork cyclical spiking loads. These will often break something else like handlebars or wheels and are low on the fun scale but do a very good job of stressing the fork in a short period of time.

    And yes – I think the forces involved with disc brakes are predictable and well known. There is a lot of concern in the carbon fork crowd about disc brakes because unless the layup of the fork is done properly to deal with the loads in the middle of one blade then there can be issues. Steel forks are a bit of a different animal and are much better at dealing with combined stresses………….and as long as the disc tab is designed to spread the load and not cause a hinging point then it’s not as complicated as it might seem. So I think we should be concerned with the design to be sure that it is safe but I don’t think that the task is all that difficult with steel.

    I hope that answers your questions. Be well and thanks again for reading and your comments.


  3. SteveP says:

    Thank you for this post. I like that you are tackling road disc brakes in a thoughtful and deliberate way. It inspires confidence. I’m almost certain that I’ll be opting for disc brakes when my order comes to the top of your queue.



  4. Dwight Kellams says:

    My vote for all time best blog posting. Don’t know how you have the time and money resources to do all you do but keep up the great work. Wish you, Dave Wages, and Kelly Bedford and all your expertise were still in one place at Serotta but am glad you and Karin are happy where you are. Tried like crazy to get my buddy from Evansville, IN, to stop by your and Karl’s shops while his daughter was living in Montana and place order for a custom frame. Sadly his daughter has moved to Colorado and he is probably going to buy an off the shelf carbon frame; but I tried.

Leave a Reply

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