VeloClub is CyclingTips’ membership program which brings us closer to our members, and connects likeminded cycling enthusiasts.
by James Huang
April 17, 2018
Photography by James Huang
I’ve long thought that carbon-fiber road frames have been on an asymptotic development path; even mid-range models these days are exceedingly good for the most part, and whatever gains engineers do make at the high end seem to require disproportionate amounts of work to get there. But nevertheless, manufacturers continue to insist that new bikes continue to get “better” with each passing iteration.
One of the areas that I regularly see touted as undergoing steady improvement is frame “compliance” — otherwise known as how much the chassis flexes beneath you vertically in an effort to enhance comfort. But the downward trend in tire inflation, and the upward trend in tire size, got me thinking lately: Does frame compliance still matter at all? Can’t we all run super-stiff frames and adjust the tires to get the ride quality we want?
Intuition might suggest that to be the case, but the complete picture is a bit more complicated.
Road bikes consist of a multitude of different components that each contribute in some way to ride quality — the frame, the fork, the wheels, the seatpost, the tires, etc. While some parts are obviously “softer” than others, all of them will yield to some degree given enough applied force. But when it comes to analyzing the individual contributions of each of those parts, the “stack of springs” analogy is often used. Each component is represented as an individual spring, all of which are laid end to end. In that configuration, it’s really only the softest spring in the series that matters in practical application — or at least so goes the widely accepted view of the matter.
Silca owner Josh Poertner used the same analogy in a blog post back in 2016. He analyzed much of the relevant data that was available at the time, not only from his own company, but also drawing on his experience as former technical director at Zipp, and the wealth of objective lab measurements conducted by popular German publication Tour Magazin.
According to Poertner, while there are a multitude of factors affecting ride quality on a road bike, there are really only three main ones: the tires, the frame/fork/seatpost, and the wheels. Of those, the tires comprise the softest spring in the series by far; the frame/fork/seatpost is roughly 2-5 times stiffer, and the wheels 5-10 times stiffer still.
Not surprisingly then, Poertner’s experiments concluded that tire pressure had the biggest influence on ride quality, with even modest differences (roughly 15psi) yielding as dramatic an effect on comfort than the total difference between the softest and firmest frames he included in the analysis.
“Clearly the differences between the stiffest third of bikes tested and the most comfortable third of bikes tested is very real,” he stated in his post, “but it just isn’t very large in magnitude when you consider that you can go up a tire size and down 1bar [14.5psi] of pressure and be roughly equivalent or even better off.”
Can just changing your tire pressure completely change the feel of your entire bike? Actually, yes. Photo: Matt Wikstrom.
Poertner’s argument certainly bolsters the commonly-accepted notion that ride quality is almost completely determined by the tires. One could also draw the conclusion, then, that all of the attention surrounding “frame compliance” is just a bunch of marketing hype designed to sell people bikes they don’t really need.
However, while Poertner — and others — rightfully insist that tires do make the biggest difference in ride quality, it’s not the only thing that contributes significantly to the overall feel.
It’s worth pointing out that the range of frame/fork/seatpost stiffness included in Poertner’s analysis — 150N/mm at the low end, to 250N/mm at the high end — is meant to represent averages for the Aero Race, Race, and Comfort categories as designated and measured by Tour Magazin, not the total range of bikes on offer. Poertner is certainly no stranger to rigorous data analysis, but in effort to make the information a little more palatable to the majority of his audience, some simplification was warranted.
Even straight tubes will flex under the right loads, but pre-bending them, as GT did on the back end of this Grade gravel bike, promotes the movement even further.
Fat chainstays and skinny seatstays promote the idea of a smooth and buttery ride, but looks can be deceiving.
Another increasingly common trick is lowering the seatstays, which helps the entire seat tube flex when you hit a bump.
D-shaped tube profiles are supposedly very good in terms of aerodynamics, but it turns out that they’re very good for bump compliance, too.
Seatposts are still a major source of ride comfort on road bikes. Having more seatpost extending out of the frame makes for a longer, and softer, lever arm, as does reducing the diameter. In other words, the short extension and oversized 31.6mm diameter shown on this Kinesis Aithein isn’t doing much for the bike’s comfort level.
“Our 150/200/250N/mm range was done to build a matrix out for people using the most common stiffness ranges we had seen,” Poertner explained. “I wanted to give a sort of general guide and then also break out the smaller ranges to show effects within a grouping. The vast majority are in the 150-300N/mm range.”
A more complete picture is provided by Damon Rinard, engineering manager for Cycling Sports Group, the parent company for Cannondale, GT, Mongoose, Schwinn, and several other popular brands. Readers may choose to take his words with a grain of salt given his position in the industry, but numbers are numbers, and his are compelling.
Metal bikes are often lauded for their superior ride quality, but that’s often because their skinny tubes flex in all directions, not just the ones you want. Even then, modern carbon fiber frames still have them beat, since the material can be more dramatically tuned. Still, traditional metal frames like this Otso Warakin can ride like a couch if you put the right tires on it — in this case, 650×47 beasts like the WTB Horizon 47.
While some degree of “compliance” is generally regarded as a good thing, not everyone has the same preferences. Specialized made little effort to soften the ride of its Allez Sprint frames, for example, but it’s unlikely that the racers who gravitate toward it will care much.
The Ritchey Logic steel road frameset looks like it should be very comfortable, but in reality, it isn’t as smooth on rough ground as modern carbon-fiber frames that have been specifically tuned for the purpose. That doesn’t mean that bikes like this are bad, however; it’s more a matter of personal preference.
“Frame compliance used to not matter, in my old opinion, but in my new opinion, it does matter, and here’s why I say that,” he said. “Back in the late 1980s/early 1990s, Jobst Brandt published Bob Bundy’s FAQ on frame stiffness. They measured a bunch of frames and wheels and tires, and the frames were many, many times more rigid than the tires — so much so that even the differences between frames would shrink in comparison to the tires, so we focused on tires, not frames. But modern frames are different. We’ve tried over the years in the bike industry to engineer comfort into frames, and slowly, over decades, it’s worked.”
Like Poertner, Rinard also compiled frame stiffness data from Tour Magazin, including the full range of frame/fork/seatpost results from the past two years — nearly 100 in total — but given that his analysis is more recent, the bikes included enjoy the benefit of additional development work. Either way, when you look at the complete set of data, rather than representative subsets, the span of frame stiffnesses is substantially broader. The stiffest frame recorded is a literal bone shaker at a whopping 423N/mm. However, the softest-riding frame posts a comparatively marshmallow-like 69N/mm.
Going along with that is Rinard’s own tire stiffness data (measured on a flat surface), which he says constitutes a range of about 100N/mm (25c tire inflated to 87psi), up to about 150N/mm (23c tire inflated to 116psi).
Data compiled by Cycling Sports Group engineering director Damon Rinard blows apart the idea that frame compliance is inconsequential as compared to tire stiffness. Photo: Damon Rinard.
Based on Rinard’s analysis, there’s more overlap between tire stiffness and frame stiffness than one might otherwise assume. And according to Rinard, while the total spread of frame stiffnesses is very broad, most of them fall on the lower end of the scale.
“The range of tire stiffnesses overlaps the bulk of the range of frame stiffnesses, so they’re in the same ballpark, when they didn’t used to be. [The frame/fork/seatpost] is one of the softer springs in the series, [and] the tires are now about the same order of magnitude. Neither one dominates anymore.”
Riders often like to say that certain wheels ride stiffer or softer than others (and I’m sure I’m guilty of saying the same thing in the past). But the reality is that wheels, regardless of how they’re made, contribute very, very little to the overall ride quality of a bike.
The stack-of-springs model may be a popular one, but it’s important to keep in mind that it’s just that, a model meant to simplify a more complicated real-world scenario. While the three major components that contribute to ride quality — the tires, the frame/fork/seatpost, and the wheels — can all be viewed as springs with vastly different stiffnesses, we’re still talking about fairly complex structures here.
One issue is that the spring rates of those components aren’t always linear. In other words, it might take a certain amount of force to compress one of those components by 1mm, but it might take more than double that effort to compress it another millimeter. This is particularly true for pneumatic tires, which, like every air spring, exhibit a distinctly progressive spring curve.
“The springs-in-series theory is roughly a good model, but the spring rates of the tires and frame system aren’t as disparate as commonly assumed,” said Specialized integrated technologies director Chris Yu. “A tire doesn’t behave like a linear spring, so its benefits start to diminish as it compresses. Just beyond the first few millimeters, the frame-plus-seatpost stiffness is between 50-200% greater than the tire, and they become close to equal as the tire compresses more.”
Despite efforts to improve frame comfort in and of itself, the biggest influence on ride quality is still the tires – the size, the inflation pressure, how they’re made, etc.
Furthermore, how a tire reacts to bumps in the road depends not only on the size of the bump, but also the shape.
Poertner is one of many engineers who have quantified that relationship. According to his testing, it takes 184N of force to compress a 25c tire inflated to 101psi by 1mm on flat ground, but just 102N to compress the tire the same amount against a bump with an 8cm radius. When held against an even-smaller bump with an 8mm radius, the force required to compress the tire that same 1mm falls to just 44N.
How much the tire compresses over a bump depends greatly on the tire width, the inflation pressure, and the bump shape. Photo: Josh Poertner.
“The frame stiffness is what it is; it doesn’t matter if it’s washboard, cobbles, a road seam, or the rider bouncing his weight. It’s all the same,” explained Poertner. “The tire, though, has this decreasing spring rate with bump size, so while the tire and frame are pretty matched for stiffness on the flat surface, the tire spring rate is roughly 50% less on the 8cm cobble and 80% less on the 8mm bump shape. This means that over smaller bumps, the system stiffness becomes even more disproportionately controlled by the tire, while on smooth surfaces the tire stiffness is much more inline with that of the frame.
“My theory is that this is why we generally don’t notice low tire pressures or are late to notice them on a ride,” he continued. “On good-quality flat surfaces, there just isn’t that much difference in system stiffness between pressures. The difference between 6 and 8 bar for the [28c] tire is only 167N/mm compared to 235N/mm. On a [Cannondale] Synapse with a vertical stiffness of 158N/mm, the frame is driving the stiffness on a smooth flat road. But then when hitting a road seam with 8mm radius (which is actually still bigger than most things you hit), the tire stiffness is now 37 or 50N/mm, and the system is heavily driven by the tire.”
The engineers at Trek put it in even simpler terms: “To envision this effect, think about how much easier it is to poke your finger into a balloon compared to pressing your open hand against that same balloon.”
One major factor affecting ride quality that often isn’t discussed is frame size. Industry-wide, 56cm frames are considered “average,” and are generally the ones used for published manufacturer tests. However, larger frames are almost always more flexible than smaller ones, whose correspondingly smaller triangles are inherently more resistant to flexing. That trend is slowly changing as companies with more engineering resources devote more attention to the challenge, but geometry is still geometry.
“Frames are typically not extremely proportionally responsive to different frame sizes,” said Rinard. “The [carbon fiber] lay-up does change from size to size, but vertical compliance hasn’t been the goal for those lay-up changes; the strength has been the goal of those lay-up changes. More recently, companies are trying to tune not only the strength, but also the stiffness of those different sizes. But just the fact of a frame size being smaller unfortunately makes it more rigid vertically; on bigger frames, it’s the opposite.”
Specialized claims that the new S-Works Tarmac SL6 rides more smoothly than the previous Tarmac SL5, but on the 52cm size I rode, I couldn’t tell a difference. Both ride very firm to me, but that’s not entirely surprising given that smaller frames are inherently more resistant to flex than bigger ones, plus the fact that I weigh a modest 70kg.
Rider weight plays a role as well, especially given that taller riders aren’t necessary heavier ones, and vice versa. In other words, a tall rider who is particularly light will experience a very different ride quality on the same bike as compared to a rider who is the same height, but 20kg heavier.
Take, for example, former top professional road racer Emma Pooley, who weighs approximately 48kg (106lb), and stands at a modest 1.57m (5ft 2in) in height.
“When we were sponsoring Cervelo Test Team back in the day, I asked her about frame stiffness,” Rinard recalled. “She said, ‘I’ve never ridden a bicycle that felt too flexible in any direction.'”
Frame designers have sought the holy grail of “lateral stiffness and vertical compliance” for decades, and it’s truly admirable how far the industry has come toward achieving that magical combination. But as good as modern frames are in that respect, and as much design flexibility as carbon fiber allows, there are still limits to what can be done. Now more than ever, bike companies are turning more toward mechanical solutions to increasing rider comfort, instead of just trying to figure out how to inject more selective flex into an otherwise rigid structure.
Trek’s first-generation Domane endurance was truly revolutionary when it launched in 2012 specifically for that reason. The frame’s key feature was the IsoSpeed “decoupler,” an axle-and-bearing pivot at the seat cluster that allowed the seat tube to bend under load much more than traditional joints, without affecting the flex characteristics of the rest of the structure.
The IsoSpeed suspension mechanisms built into the front and rear ends of the Trek Domane SLR are tacit admissions that frame designers are near the end of their potential when it comes to engineering specific types of flex into carbon fiber frames.
Trek’s innovative IsoSpeed “decoupler” was derided as a gimmick by casual onlookers when it was first introduced in 2012. But in reality, it works remarkably well.
Trek has recently expanded the IsoSpeed concept to the front end, too. Carbon fiber offers a wide range of tuning in terms of selective flex, but there are still limits. Once those limits are reached, frame designers have to look elsewhere for additional gains in rider comfort.
According to Trek’s official figures, the range of adjustment on the latest Domane equates to vertical stiffnesses between 99 and 144N/mm, which fall well within the range of tire stiffnesses measured by Poertner and others. That movement can readily be both seen and felt, and equates to a dramatic difference on the road.
Specialized has taken a somewhat more conventional approach with the FutureShock mechanism integrated into the most recent Roubaix and Diverge models. Situated up front, inside the fork, FutureShock comprises a telescoping steerer tube section and a set of steel coil springs, providing 20mm of total axial movement. Given the location of the FutureShock mechanism, test numbers aren’t directly comparable to the test results from Tour Magazin’s frame tests, which measure the displacement at the saddle rails to quantify movement at the rear end of the bike, not up front.
The latest Specialized Roubaix is an intriguing exercise in trying to build comfort into a frame.
Mock the FutureShock all you want, but it works as advertised, offering far more front-end comfort than you could ever hope to get from tuned flex alone.
The unusual CG-R seatpost has been in the Specialized stable for several seasons now. The kinked head is supposed to flex when the rider encounters a bump in the road, but its effectiveness is highly dependent on rider weight and saddle position. Arguably more effective is the lowered seatpost collar and open-top seat tube, which leaves more of the seatpost shaft exposed to flex on bumps.
But that said, the three possible spring rates for the system range from just over 2N/mm up to 5N/mm, all of which are much softer than even high-volume road tires and unusually low pressures, and certainly vastly softer than what you typically get out of a conventional road frame.
Keen-eyed readers will also note the presence of Cannondale Lefty suspension fork data in Rinard’s stiffness chart above. Cannondale uses the 30mm-travel Lefty Oliver fork on its Slate gravel bike, and while that machine is undoubtedly polarizing for several reasons, few argue the effectiveness of the fork on rough terrain. According to Rinard, the spring stiffness for the Lefty Oliver ranges from just 9N/mm at the low end, up to 13N/mm at the high end, on average. Again, that’s much softer than what we see either from either tires or frame compliance, and with the added benefits of damped movement so as to minimize bounciness, as well as the ability to lock out the movement on smooth terrain.
Love it or hate it, Cannondale deserves credit for having the courage to include a proper suspension fork on the Slate gravel bike. Photo: Matt Wikstrom.
The Lefty Oliver suspension fork on the Cannondale Slate is said to provide up to 30mm of travel. Photo: Matt Wikstrom.
According to Rinard, the Lefty’s spring stiffness is lower than that of the tires, meaning it provides more effective ride comfort benefits than what you could typically expect to see from adjusting tire inflation alone. Photo: Matt Wikstrom.
Frame designers will increasingly turn to dedicated suspension devices as riders continue to head off-pavement.
As far as gravel machines go, the Scott Addict Gravel 10 Disc leans more toward the sportier end of the spectrum with its highly responsive carbon fiber chassis and stiff ride, relying more on the tires to provide a modicum of comfort.
Can’t you just get the ride quality you want by carefully choosing the right tires and pressure? Yes and no.
Will we see more development on the suspension front moving forward? If the current trend of riders moving further away from paved surfaces continues, it seems likely. Tuned frame flex can only do so much, so engineers will have no choice but to resort to other means to improve ride quality further.
I should point out at this stage that while the question of ride quality has been stirring about in my head for some time, what really got me thinking about it in earnest was a comment from reader Tom Anhalt on the recent review we published on Specialized’s latest S-Works Tarmac SL6 Disc. During that discussion, he questioned my practice of systematically over-inflating tires when testing road bikes. I do this in order to minimize the effects of the tires on ride quality, and instead cast a sharper focus on the supposed compliance benefits often touted by bike manufacturers.
That comment reflects the commonly accepted sentiment I described earlier on, that tires are all that matter when it comes to ride quality. However, it seems clear at this point that the subject isn’t that simple. It’s still true that wheels contribute very little to the overall ride quality of a bike (it’s almost not even worth discussing at all), but there’s more overlap between the individual contributions of the frame/fork/seatpost and tires than what is often assumed.
Also keep in mind that all of this still only focuses on one aspect of ride quality — the amount of vertical movement generated by a given amount of force. It doesn’t take into account other factors that can influence rider comfort, such as vibration. Even Rinard, a man who has dedicated most of his professional career to finding objective answers to subjective questions, admits that there is still much work to be done here.
There are practical aspects to consider as well.
Lowering your tire pressure may be the most effective (and cheapest) way to improve the ride quality of a road bike, but it’s not an option that’s universally available. Heavier riders often have to resort to higher pressures to prevent pinch flats, for example, as do riders who regularly ride on roads that are in poor condition.
Likewise, lowering your tire pressure is easier said than done, depending on your current setup. Lower pressures are facilitated by larger tire casings, which help offset the increased risk of pinch flats those lower pressures create. Similarly, wider rims help maintain good handling characteristics at lower pressures by preventing the tire from rolling sideways through corners. However, neither of those options will mean much when your current bike doesn’t have the clearance for higher-volume rubber, or you don’t want to spend the money upgrading their older, narrower wheelsets.
Bigger tires with supple casings mounted to similarly wide rims (ideally with aero profiles) can arguably be regarded as the ideal when it comes to balancing speed and comfort on road bikes. But unfortunately, not every bike can handle the width, nor can every rider afford to upgrade.
Say your bike can only fit 23mm or 25mm-wide tires, but you still want a softer ride. What do you do then?
Performance-minded riders will want to consider, too, that while there is plenty of data available to prove the rolling-resistance advantages of lower air pressures on most real-world surfaces, those advantages are still mostly limited to high-end tires with minimal puncture protection and supple-but-vulnerable casing constructions, neither of which are all that conducive to everyday riding.
And as with almost all performance metrics, it is possible to have too much of a good thing. A softer ride certainly helps when rolling across pockmarked roads, but a ride that’s too soft can make it seem like you’ve overly isolated from the road; almost like you’re in a video game version of the real thing instead of actually riding a bike. Some riders simply want to feel more of the road, and for those that don’t, pairing a super-stiff frame with squishy tires won’t always produce the desired effect.
“With lower inflation pressures, tires offer more compliance, but also take less force to bottom out,” said Yu. “In these cases, the frame may actually need more, not less, compliance to feel balanced.”
None of this takes into account, either, the fact that cycling doesn’t happen in two dimensions. Canyon’s engineers, for example, point out that cycling often isn’t vertically static, and you’re often swinging and swaying ever so slightly. As a result, they design their frames and seatposts to incorporate a bit of specific out-of-plane flex, too.
So does frame compliance still matter? It seems the answer is “yes,” although there are still plenty of caveats to go along with that — shades of grey, if you will.
After purchasing Silca in 2013, Poertner’s livelihood now rests heavily on the importance of proper tire inflation, and how tires can influence overall performance. However, even he says it’s important to keep in mind that there’s a distinction between differences you can only measure, and ones that have a practical impact for everyday riders, as well as gains you can make in theory versus the ones that are actually available to you in practice. To him, it’s about getting everything to work together, and figuring out what works best for your own situation and equipment.
The Cannondale Synapse frame incorporates a variety of features that are proven to improve rider comfort, but they can still be squandered if you over-inflate the tires. Photo: Matt Wikstrom.
“I agree with Damon that things are better now than they were, although I’d say still tire-dominated — if your Synapse customer goes and puts 30mm tires on at 100psi, he’s just undone 90% of Damon’s work to make that bike comfortable,” Poertner said. “Also, I know that the frame guys are really working on this as I’ve talked to almost all of them. The problem for them is where to put the deflection so that it isn’t causing other problems. If you make the rear triangle too compliant, you have torsional problems. But then you still have to spread out the deflections and ensure that there aren’t interference issues under very high-G impacts, but you can’t give too much clearance everywhere as then aerodynamics and aesthetics will suffer.
“The short answer here is that while frame stiffness is far less important than tire stiffness, you obviously want to optimize both if you can. At the same time, the industry is very good at overselling the benefits of frame comfort as well as overselling the benefit of lateral stiffness.”