## Myth 9: Fork Blades Don’t Flex

When we first started talking about shock absorption and fork blades, it was commonly believed that fork blades didn’t flex significantly. Experts told us: “All the flex in a fork is in the steerer tube, where the lever arm is longest.” And yet, when we rode bikes with flexible fork blades, they clearly took the edge of bumps. Was this another myth in need of debunking?

We designed a simple test to measure the flex of fork blades. By combining a small bag-support rack with the hoop of a low-rider, we could easily visualize (and measure) the flex of the lower fork blades: The two racks will move against each other only if the fork flexes between their attachments – in the lower 2/3 of the legs. (If the flex occurs higher on the fork, both racks will move in unison.)

The video above shows the same test on my ‘Mule’ with its flexible Kaisei ‘TOEI Special’ blades. As the camera zooms in, you can see how much the fork blades actually flex. That is what takes the edge off bumps that are too large for the tires to absorb.

And since more comfort equals more speed, flexible fork blades make your bike faster, too – even on smooth roads. We quantified that when we ran a stiff fork, a suspension fork and a fork with flexible blades both on rumble strips and on the brand-new, ultra-smooth pavement next to them (above).

It didn’t come as a great surprise that compared to our stiff fork (Trek unpadded), the suspension fork (RockShox) and the flexible steel fork (Singer) each saved about 50 Watts on the rough surface of the rumble strips (left graph).

What we didn’t expect was that both flexible forks saved somewhere between 20 and 35 Watts on the smooth pavement (right graph). That is a very significant difference in power output – in fact, it was large enough to pass the all-important test of statistical significance.

In the graphs above, you notice a fourth bar: Trek (bar padding). We also tested thick foam on the handlebars to simulate the maximum padding you could put on your bars, but we found that the effect was much smaller than that of the forks’ suspension. It was not statistically significant, and neither were the small differences between the RockShox and Singer forks.

So it’s clear that the Kaisei ‘TOEI Special’ fork blades make a bike more comfortable and faster. That brings us to the next question: Do flexible fork blades negatively affect the handling of the bike? Read the next post in this series to find the answer!

Further information:

• Bicycle Quarterly 23 with our original experiment measuring the flex of fork blades
• Bicycle Quarterly 29 with our rumble strip tests of suspension losses that included forks, tires and other equipment
• Compass framebuilding parts, including Kaisei ‘TOEI Special’ fork blades
• Other posts in this series:
Myth 1: Wider tires are slower
Myth 2: Titanium is lighter than steel
Myth 3: Fenders slow you down
Myth 4: Stiffer frames are faster
Myth 5: An upright position is always more comfortable
Myth 6: Tread patterns don’t matter on the road
– Myth 7: Tubeless tires roll faster
– Myth 8: Modern components are lighter
Myth 10: Stiffer forks steer better
Myth 11: Rear tires should run at (significantly) higher pressures
Myth 12: Disc brakes work better than rim brakes
Myth 13: Leaning without Countersteering
Myth 14: More lumens make a better light
Myth 15: Marginal gains
Myth 16: Higher tire pressure is faster
• The Alex Singer fork in the original test used old Reynolds 531 ‘Super Resilient’ fork blades that are no longer made. The Kaisei ‘TOEI Special’ blades have the same diameter and profile with slightly thinner walls, making them yet a bit more flexible.
• The RockShox fork was a brand-new Ruby. With age, the performance of the suspension fork will probably deteriorate.

## About Jan Heine

Spirited rides that zig-zag across mountain ranges. Bicycle Quarterly magazine and its sister company, Rene Herse Cycles, that turns our research into the high-performance components we need for our adventures.
This entry was posted in Myths in Cycling, Testing and Tech. Bookmark the permalink.

### 60 Responses to Myth 9: Fork Blades Don’t Flex

1. Billy Wattson says:

This is quite intriguing.

One thing I’ve been wondering about is flexible fork blades vs disc brakes. As a long time follower I’m well aware of your take on even the fork leg stiffness that even canti brakes need to perform, compared to higher mounted centre pull, and disc brakes obviously need even stiffer legs.

But for many riders disc brakes add “value”, be it in wet/dirty conditions, lesser force required at the lever etc.

Do you think the use of a SRAM “predictive steering” hub that clamps the fork legs could help, so flexible fork legs can be used with disc brakes?

• We once tested a bike with a Wound Up carbon fork and hydraulic SRAM discs. The fork blade flexed so much in reaction to the brake forces made the bike turn when braking from high speeds. I learned to compensate for this by turning the bars when I braked, but it’s not ideal. You can read the review in the ‘sample issue’ at http://www.bikequarterly.com

Most forks for disc brakes are very stiff to avoid this issue, but you lose the benefits of shock absorption.

2. Klaus says:

It doesn’t surprise me that fork flex makes your bike faster….it astonishes me that a handful of people around Paris and Lyon already knew this many decades ago.

• Not just the French – the Reynolds ‘Super Resilient’ fork blades used by Alex Singer and others were made in Britain, after all.

3. Scott Williams says:

The RS Ruby is an old, coil spring fork. Would an air fork be more beneficial vs. the Singer? With bike weight distribution biased to the rear, would rear suspension produce significant, or even greater, additional energy savings?
You demonstrate that the Singer front fork, with its simplicity, cost and weight savings, produces substantial energy savings, on a par with a mechanical suspension fork. Can this be replicated at the back of the bike?

• Good question! While there is more weight on the back, the rider’s pedaling legs already provide suspension. Also, once you get meaningful suspension travel with a mechanical suspension, you get into issues of suspension bob. I suspect that is why mountain bikes lock out their suspension when pedaling hard on relatively smooth terrain.

• morlamweb says:

I get plenty of suspension bob when riding my wife’s bike. It has fork suspension, springs under the saddle, and a small spring in the seatpost. I get the strong feeling of pedalling a marshmallow when I ride it. It’s especially strong when riding out of the saddle.

• The key for suspension on a road bike is to have enough to enhance the speed and power output, but not so much that it detracts. We once tested a Trek with an elastomer in the rear triangle. It came with three elastomers, plus we made a Delrin block to set up the bike without suspension. Both the Delrin block and the hardest elastomer had no effect. The middle one was great, actually enhancing the performance of the bike. The softest elastomer was more comfortable, but the suspension bob made it difficult to put down power.

Fortunately, the fork of a bike is less affected by the rider’s power, and these fork blades have decades of experience on their side. You’ll notice a little flex when riding out of the saddle with high power and low cadence – for example when you try to get over the top of a hill without shifting, but for most riding, it’s not an issue. I wouldn’t hesitate to race on them – as did many pros in decades past – but I wouldn’t recommend them for a track sprinter.

4. marmotte27 says:

I wonder how the myth came into being. Just looking at one’s fork (which doesn’t even have to be especially flexible) while riding one can see that it moves quite a bit. Those movements cannot happen at the steerer tube, and if they did I’d be very worried.

• Actually, the visible back-and-forth movement of the fork could be in the steerer tube. There were a number of suspension forks for mountain bikes that used a pivot at the fork crown. So it was important to show that – with well-designed fork blades – the blades actually did move in their lower sections.

5. Dan says:

Very interesting test. What are your thoughts regarding the Kassei fork’s ability to withstand disc brake forces? Can a comfy flexible fork hold up to disc brakes?

• I don’t think the Kaisei forks would break, but they might flex so much that you get auto-steer every time you brake hard – see the comment above about the Wound Up carbon forks we had on a BQ test bike once.

6. Andy Stow says:

On my steel touring bike with disk brakes, I can see the fork twisting when I brake, as my fender moves relative to my front tire.

7. Derek says:

It looks like the best set up would be flexible forks AND padded bars. Why didn’t you test that?

• If we were to do the tests again, we’d probably test flexible forks with padded bars.

The testing requires constant conditions, so the quicker it’s done, the better the results. If it gets warmer, we have to correct for that, and that introduces potential error. If even a slight wind comes up, our testing is over, and we have to come back another day. That is why we prepare the bikes beforehand, then run the tests, and only look at the results once the testing is complete. So we can’t change what we test based on the results, unless we do another round of testing, as we did with our tire tests.

However, it’s important to note that the padded bars were not statistically significant. That means the lower power required with them could just be ‘noise’ in the data. The forks were statistically significant, so we can be confident that they required less power.

For this test, the padded bars used pipe insulation – much thicker than most handlebar tape. Our goal was basic research in how bikes work, not a practical setup, and so we exaggerated the bar tape in the hope of getting more reliable results. I suspect that the bar tape commonly used on bikes has little effect on suspension losses and speed, because the unsprung mass of the front of the bike is too great. What bar tape can do is reduce pressure points as you grip the bars.

8. Dr J says:

I don’t even know why that would be surprising. The fork you have – with so much rake and made out of lightweight tubing will flex for sure. On the other hand, carbon forks with straight, wide blades and aluminum steerers will flex much less in the blades but quite a lot in the steerer (I know from my own experience).

But on another note – why so many of those myths focus on speed? On what makes us faster? Most cyclists never race and don’t care much about speed anyway.

• Most cyclists never race and don’t care much about speed anyway.

You are right: Most cyclists don’t race. And yet, fundamentally, the appeal of cycling lies in its speed. Even a commuter would just walk if the bike wasn’t faster.

Beyond that, the effortless glide of a great performance bike enhances the enjoyment of cycling, whether you race or not. Especially when that speed goes hand in hand with more comfort, too, as it does in this case.

• Mitch says:

“why so many of those myths focus on speed? On what makes us faster? Most cyclists never race and don’t care much about speed anyway.”
If you don’t care about speed you likely still care about ease of cycling. The same things that makes you faster make cycling easier for any given speed. If neither ease nor speed were important to cyclists, pneumatic tires would never have caught on.

• morlamweb says:

Not being interested in racing doesn’t mean that one isn’t interested in speed. I’m basically an Unracer and don’t have a competitive bone in my body; and yet, just watch me pedal when I’m late for work, or an appointment…

Thats not what I see. The industry default setting for a road bike is a carbon fiber racing bike. Scads of people buy them that have no intention of racing, even though the price per mile of a carbon bike is crazy expensive. So people must care about having a fast bike.

• Nik says:

Even if you don’t want to go faster, it’s probably still appealing to go at the same speed with less effort.

• Sukho in PDX says:

I don’t race, but I enjoy riding fast; both for weekend rides and daily commuting. I love the sensation of going fast; in fact I’ve tried riding slow and basically can’t do it. My enjoyment of riding a bike goes down considerably. Maybe I’m weird in that respect. So I’m always wanting a bike that performs well for me, and a lot of that means the bike makes it easier for me to go fast.

And I agree why the topic (that forks do flex) is not surprising. My bike has a Jeff Lyon fork with 63mm of rake (Kaisei blades). And I can look down and see it “dancing” as it absorbs all kinds of bumps/road chatter. It’s VERY obvious. With 42mm tires and this fork, my bike is super comfortable and fast.

9. Stefa says:

There is way more to the topic then “static” flex. It´s a dynamic process when you ride, and each fork with different stiffness will absorb different frequencies better (think of it as base isolation, in the simplest case as a two mass oscillator). There is always some sort of transfer spectra, means that for example a fork that is too flexible would not absorb a lot as well as a fork which is too stiff.

It would be interesting to do measurements with accelerometers on position of the hub and the crown to seen what´s coming in and coming out while riding on different surfaces, calculate transfer functions etc.

This could lead for excample to forks with adjusted stiffness for certain undergrounds and rider weights. Think there is room for a lot of improvement, just take a look at structural dynamics and don´t concentrate too much on the static case.

On the other side: One of the reason why I love bikes is because they are so efficient and simple machines. Maybe it would reduce a lot of fun when looking at everything just under the viewpoint of optimization.

10. John C. Wilson says:

At what speed was this test? How was the experiment controlled for constant speed?

Other queries that I can only suggest and can’t test myself would be how much of the fork travel comes from the light gauge tube and how much comes from the rake? In the 1950s (and maybe other times and places, I don’t know) some English bikes were made with undersized fork blades and pencil stays. Then there were the Russ forks with a sharp low bend that looked as if the fork would hinge right there. Some current production customs are built with extremely light seatstays. Any thoughts? You’ve started a fascinating line of inquiry.

• The testing was done at 31 km/h. Constant speed was easy, since the same bike was used with different forks. Just pedal at the same cadence. There was a computer on the bike that helped, but on the rumble strips, it turned into a blur that was impossible to read. Still, the speed was recorded and correlated to the power in the analysis.

To figure out whether the diameter or wall thickness is more important, you could just calculate the stiffness, rather than measure it. You’ll find that slim diameter trumps wall thickness. However, in all this, we mustn’t forget the durability of the fork. Fortunately, we can build on decades of experience, and we know that the fork blades we are using have a close-to-zero failure rate. And in the rare cases where they fail, it’s usually at the fork crown due to poor construction.

11. Please add the Lauf fork to your next testing session.

• Greg says:

And the Fox AX gravel fork . . . And the Cannondale Slate Lefty. A BQ suspension shootout! Only half joking here. Jan has said he’d be curious to try one (or all?) of these.

12. What is the common weight of this kind of fork ? The modern equivalent and disc brake compatible will be the Lauf Grit SL fork (850g claimed) but less aero with a sort of ugliness, or at least, less simple.

• The weight of a fork always depends on the length and diameter of the steerer. We once weighed a bunch of forks – I’ll try to look it up. From what I recall, a good steel fork weighs in the upper 600s (in grams) for a tall frame (60 cm). So you’ll probably save some weight by going with steel.

13. Vicnent says:

“fundamentally, the appeal of cycling lies in its speed”.
With all due respect Jan, I think that the way you present it is flawed.
What DrJ meant (I think) is: Since most people don’t race, the trade-off between speed and other elements of decision makes speed not always win, as is the case in racing (it is the “raison d’être” of competition where a 1/100th of a second might mean more than a lot).
Therefore the choice is not between “speed” and “slowness” (comparable to walking), but between maximum speed and “epsilon” less speed (epsilon varying greatly of course depending on factor, and to a value accepted differently by different folks).

• I see your point, but I am not sure how the added speed of flexible fork blades requires trade-offs that a non-race might not want to make. If we are talking about lowering your stem beyond what is comfortable to be more aero, then I’d agree with you. But that has never been the topic of this series.

• Vicnent says:

You’re right. I (and you also) commented a side comment about racing and speed, not strictly related about the post’s topic (fork flex brings both speed and confort).
The trade-off of a flexible fork would be with other advantages of disk brakes ( I read your answer that your test shows mechanical disk brakes perform less good as good rime brakes), unrelated with speed.

14. Stats Nerd says:

Fascinating post as usual Jan. One friendly criticism: if you’re going to talk about statistical significance, you should include some sort of numeric information about your significance level. A p value or alpha value, ideally with a confidence interval, would be great. Otherwise you’re really not saying anything and it gives the impression of “hand waving.”

• As per standard practice, we are looking at a 95% confidence interval. For more details, I’d have to ask Mark Vande Kamp, who did that statistics. (He has a Ph.D. with a minor in statistics.) The original report in BQ 29 included a one-page article where he describes the methods he used for the data analysis.

• Stats Nerd says:

good enough for me, thanks!

Its a bit of a trade off between the awesome braking power of disc brakes and the lower weight, better reliability, and improved handling of a properly designed steel fork with rim brakes. Personally I go for the latter. Discs are absolutely the way to go with a suspension fork but I’m surprised that so many prefer discs on a rigid fork.

• The stopping power of disc brakes varies widely. The hydraulic models usually have plenty of power, but all the mechanical ones we’ve tested had less braking power than a good rim brake.

16. Axel Reichert says:

Another nice part of a nice series! Combine it with

https://www.cervelo.com/en/Engineering-Field-Notes/Engineering-Fundamentals/Ride-Quality

to get the following bottom line for a suspension with near-zero maintenance:

With wide, supple tires, a flexible fork, and a good leather saddle you have accounted for roughly 60 percent of both front and rear compliance.

17. Alex Thompson says:

This reminds me of comments made regarding the Vitus 979 frame 30 years ago. Too flexible said some, but Sean Kelly used that comfort to win Paris Roubaix twice. I am 6’4″, 250lbs, and it is one of my favorite rides. I just wish it took tires wider than 28mm, and fenders,

18. What about the Trek Domane & Specialized Diverge? The Domane uses a pivot at the top of the head tube to exaggerate the flex in the steerer tube. It’s a light & simple mechanism, and in my experience over two years has been reliable. The Diverge has a spring at the top of the head tube. Does it matter where the suspension movement comes from? And if the Domane & Diverge mechanisms work don’t they overcome some of the problems with disc brakes? I find the Domane quite smooth, though it’d be better if I could fit larger than 32 mm tyres, and there’s a part of me waiting for either the pivot or the steerer tube to snap. I try not to think about that too much—I guess it’s like worrying that a steel fork might fail. (All things considered though, esp. carbon’s vulnerability to knocks &c which I’ve experienced first hand, I’d rather have one of my old steel or titanium bikes.) BTW, when I checked the details of the Diverge, I was interested to see Specialized using a Rebour-like line drawing to advertise the Diverge.

• Tim Quinlan says:

It does matter where the suspension action is, it determines the amount of unsprung mass (https://en.wikipedia.org/wiki/Unsprung_mass). If you take a rockshock for example, the unsprung mass is the front wheel + the sliders of the shock. If you consider a suspension stem, then the unsprung weight is the front wheel + the entire fork + a considerable amount of of the rest of the bicycle.

It would be interesting to do a comparison between a rockshock, then on the same bicycle a stiff fork combined with a suspension stem.

• Absolute. That is why tires are the only thing that works for ‘road buzz’ – the high-frequency vibrations of the pavement irregularities. The forks augments the tire suspension on large, infrequent bumps like ‘frost heaves’ and the expansion joints in concrete road section, as well as larger rocks on gravel roads.

19. Mat Grewe says:

Are there any particular pannier front rack designs that don’t inhibit fork flex?

• That is an issue. On the Mule with its flexible fork blades, the bolts for the low-rider tend to come loose, as the fork wants to flex. If I were to build a camping bike, I’d use stiffer fork blades. With the extra weight in the panniers, the ride is smoother anyhow – the tire has to work harder!

René Herse did design some racks in the late 1940s that attached to the fork crown and to a braze-on about 3/4 of the way down the fork, leaving the very bottom to flex independently of the rack. It’s something worth thinking about, even though Herse abandoned that design later.

• Murray Watson says:

I believe the Crust Bikes front rack is designed with this in mind.

20. Mat Grewe says:

It appears that Mitch Pryor did that with his infamous “Rambonneur.” Do you have any idea how much weight that could durably hold (over all kinds of terrain of course)?

• If I understand your question correctly – the Kaisei fork blades can handle a heavy rider. I wouldn’t recommend them for tandems, though. As to Mitch Pryor’s bikes, you’d have to ask him.

21. John Duval says:

It seems like some steel forks have a tendency for brake rub, and others do not. I wonder what construction details might maintain wheel alignment but allow greater suspension effect.

22. Franklin Miller says:

Crust Bikes make a front fack that is supposed to attach in a way that leaves the fork free to flex.

• I looked up the rack, and it’s an interesting idea. I worry about rack flex, though. I recall those old Blackburn lowriders, made from flexible aluminum rod, that made it almost impossible to ride the bike with more than 10 kg (22 lb) in the front low-riders. The rack flexed so much that there was a lag in the response of the front wheel when you steered: First, nothing happened due to the inertia of the load, so you turned the handlebars further, then the rack sprang back, and sent you into a huge wobble. The shape of the Crust rack doesn’t look particularly well triangulated, but they don’t list a weight, so it’s possible that they use extra-thickwall tubing to beef it up and stiffen it.

23. Jon Blum says:

Tout Terrain uses different blades on the left and right to address the twisting problem with disc brakes. Not a flexible fork, but an interesting idea.

Since you mentioned the potential impact of racks that brace the fork, are there any implications for fenders and particularly stays in terms of flexibility and clearance issues?

• If the Tout Terrain fork with the different blades did flex significantly, the bike would steer every time you went over a bump. We tested one of their bikes, and that wasn’t an issue, because their forks were very stiff. So stiff that using the different blades probably made no difference.

Fenders are quite flexible, so they don’t seem to stiffen the fork.

24. brian says:

These seem a curious bunch of experts claiming, “All the flex in a fork is in the steerer tube, where the lever arm is longest.” That’s a new one for me, and seems odd to say since the lower headset is the fulcrum, and the very stiff steerer is fixed (in terms of flex) by the head-tube.
The flex in a steel fork’s blades is easy to watch on mildly bumpy road (and fun, as long as a momentary observation doesn’t distract from what’s ahead).

• If the steerer tube bows, the fork tips will move back and forth just as they do if the blades themselves flex. Somebody may even have done an experiment using a stiff steel fork from a cheap mountain bike (with blades that don’t taper), where this probably is the case. My first reaction back then was: “If current fork blades don’t flex, shouldn’t they be redesigned so they do?” After all, stressing the steerer tube isn’t a good idea…

• Again (see above), I ask about the Trek Domane system, which uses a pivot at the upper end of the head tube to allow the steerer to bend/flex. Trek must have done some testing/resarch to arrive at this system. Is any of it available in the public domain? Does this system replicate the effects of flexible fork blades?

25. Dirk van Rossum says:

It would be interesting to also test the real impact of the Redshift Shocktop stem as well, for those of us who would be unable to retrofit a better fork to our existing bikes. Similarly with flexible seatposts or sprung seatposts like the thudbuster for the rear or how about the Rinsten spring?. Unfortately my current aero roadbike cannot even take 28 mm tires

26. Vlad Luskin says:

Then again, someone said very recently, “By comparison, fork blades flex very little.” I never know what expert to believe…

• That is why we test and even made a video to visualize this. Theoretical reasoning is a great start, but it should be confirmed by real testing.

27. Jon Blum says:

Sorry my question about fenders was vague. What I meant to ask was whether flexing of the fork could lead to the tire hitting the fender if there was not enough clearance, and whether the flex might contribute to loosening of fender attachments or fatigue failures of fenders or stays. I have no experience with these, was just curious.

• Thank you for clarifying. Yes, if the clearance is too tight, the fender will hit the tire. In fact, that is how we first figured out that fork blades do flex: For a photo shoot, I switched the tires on a classic René Herse fixed-gear bike. The tires spun nicely in the stand. Yet when I rode the bike – I always try to ride the bikes we feature – I heard something rubbing on the front. I got off the bike and spun the wheel, and nothing rubbed. When I rode, the sound was back. I removed the front wheel and saw rubber dust on the nut inside the fender that secured the headlight bolt. It was hitting the tire when my weight flexed the fork, but when the bike was in the stand, there was at least 3 mm clearance. That is when I realized that forks flex a lot more than we had thought.

As to loosening fender attachments or failures of the fenders due to fork flex, that isn’t an issue, because the fender stays can flex to compensate for the fork flex. My Rene Herse is built with the same Kaisei ‘TOEI Special’ fork blades, and its fenders have been fine for 7 years and tens of thousands of miles, including really rough roads like the Oregon Outback. I expect the fenders to last as long as the bike. Most classic bikes we’ve featured in BQ and in our book have their original fenders, even after decades of hard use.