Tension Drop Meets Marginal Gains
Nothing beats riding in newly arrived Fall weather. In the north hemisphere, those start now! Zipping through scenery colored with seasonal change, feeling the lingering warmth of Summer, such a magical experience—an invigorating and restorative tonic that reaches the core of our human experience. Frankly, despite my attempt, it’s well beyond words.

© New Yorker—cover 9/23/19, by JJ Sempe
Much of cycling is this way and doesn’t get better with analysis. Still, I find it irresistible to consider all the variables from physics to aesthetics, after the ride. Maybe I’m just trying to keep the endorphins coming!
On my list of leisure pursuits are podcasts and essays by well informed bicycle experts who do their best to demystify as well as entertain. We benefit from a fine selection of such voices these days. Near the top of anyone’s list…should be Josh Poertner’s “Marginal Gains” podcast. Last month (Aug/2019) #006 included some interesting discussion—“What’s going on when the dish changes the first time a tire is mounted on a new wheel?” Find it about 18:00 in.
This topic does not get the attention it deserves. Thankfully, Ryan Morse poses the question. Josh’s answer only touches lightly on the issue, it’s covered in more detail here.

Ryan “diablo” Morse. Photo by Courtney Paige.
Summary
Bicycle wheels lose a measurable amount of tension when a tire is mounted and inflated. Historically, this amount was insubstantial and no dedicated counter measures were needed. But today, owing to two recent trends, we have perfect storm that spells trouble and confusion.
- Tubeless clincher tires requiring a tight bead fit to the rim.
- Rim profiles ill equipped to manage the pressure of tight beads and inflation.
Outcome
Some rims shrink way too much with tight tires and can lose 50% or more of their tension and asymmetric ones (all rears and disk fronts) become uncentered in the frame.“Tension drop” is worrisome and there is little consensus about what to do once a bad rim design is in the workshop.
Hats off to Josh’s podcast for hosting the topic. However, it isn’t enough to suggest builders finish only with the target tire mounted and inflated. That means:
- Extra time for the build.
- Bare wheels are over-tensioned and undished (centered).
- Everything must be reconsidered if tire or inflation pressure changes. Good grief.
Yes, for a team mechanic—there is no choice. For the rest of us, this is too much to ask. Rims and tires should never interact this way (30-80% tension drops and associated dishing changes). It is unnecessary, sabotages the tensioned structure, offers no benefit, and will historically look pretty damn stupid. It introduces liability issues that could take decades to heal. If you can think of a similar mechanical nightmare from cycling’s past, please advise.
This is not just a product of unripe, emerging technology with which we must be patient. This is reckless design with inadequate engineering—driven by ignorance and lust for the illusion of innovation. Our industry is not the worst for this sort of issue but it’s extremely embarrassing after decades of relative stability and dependability in wheel design. I am certain this would not have been tolerated by many wheel builders of the past. I can name a few who must be rolling eyes (or in their graves) over such nonsense.
Josh concludes the section by praising wheel industry players who have great tools and insight such that they can work miracles. I am seeing quite the opposite; just glad not to daily deal with the issues that Ryan and other contemporary builders must.
Suggestions
So what’s a conscientious builder to do? I see several options:
- Keep track of worst offenders—combinations of tires and rims that produce giant tension drops. This list is ever-changing and must be gleaned first hand, from blogs and forums, from FB and Instagram. For many of these you must refuse to build.
- Navigate this dangerous landscape carefully. Disclose enough to all concerned so you don’t end up holding a ticking grenade and no one suffers regrettable surprises. For each customer, there is a best decision.
- Wheels for massed start events should be a free of these issues. For competition, it pays to steer clear of “cutting edge” gear.
- Rejoice in two trends that ameliorate “tension drop” trouble.
- Disk brakes make wheel centering a much smaller issue.
- The trend to lower tire pressures (overdue, thanks Jan…) also diminishes tension drop.
- Whatever your leverage or profile in our industry, please question this dangerous, confusing, and unnecessary situation. Sooner it passes, the better!
Our situation is better than it might be and, as ever, a sound examination of the science helps in managing the problem. Please share your own observations and solutions!










According to folks at DT, going above 120kgf on a bare wheels, can crack hub flanges when a flat happens and tension increases quickly. Locking nipples, such as Sapim Securlock or DT Prolock seem to help prevent low tension spokes from unwinding.
1. “folks at DT” and “bare wheels” are very general. No details, no specs, so hard to apply. 2. Yes, it is important to be mindful of high tensions (over 100kgf), where you apply them. Tension brings increasing strength and fatigue life from 0 towards 100 but can become a liability depending on many variables in particular wheels. 3. Locking nipples work well but I prefer a thread compound that resists unwinding (at zero tension) but lowers turning resistance under high load (near finished tension). Spoke Prep, FIX, and a few other materials have this objective.
I have read elsewhere that hub manufacturers need to take into account spoke tension so that their bearing races are not pulled too far apart once there is tension on the hub flanges. If they do this and then the wheel is detensioned once a tire is inflated on the wheel, does this again negatively affect the bearing races (causing too much friction)? Are there known hub manufacturers who take both these forces into account? Is there a particular point of inflation where the detensioning becomes acute (80 psi)?
Very good questions and too much to address right now! I will dedicate the next post to these issues. Timing is right! Thanks.