Posts
1671
Joined
10/26/2015
Location
CA
US
Edited Date/Time
6/27/2025 1:41pm
All things flex stay; pros, cons, considerations, variants, whatever.
Initial discussion is here: https://www.vitalmtb.com/forums/The-Hub,2/2020-MTB-Tech-rumors-and-inno…
|
kperras had some good points in the tech rumors thread.
"- The spring force of the flex stay is minimal, relative to the spring force of the shock. As in it barely registers in a meaningful way to the ride quality of the bike. This was measured with instrumentation. The shock tune is all that matters here. "
The whole argument boils down to what is a meaningful impact on ride quality. what can instrumentation measure vs. rider perception? how much of that impact is lost in the noise of the measurement? does "noise" in a digital measurement really equate to a lack of perception on the riders' parts?
The second thing I am unconvinced about is "the spring force of the flex stay... relative to the spring force of the shock." no argument here that the forces aren't comparable. what myself and others are trying to communicate is that the problem isnt the contribution to the overall spring force, it is that it is still an undamped spring in the system. you can make the argument that it is coupled to the damper in the shock at times, but it is also decoupled from the damper at other times. again, I can see how this is only a few mm at times, but in my mind this is adding another dimension into a system that kperras and others have admitted is already chattery due to other factors.
where am i thinking about this wrong? also to reiterate: i had the same issue on stumpy alloy vs carbon, so its not just the EE that was problematic for me. the stumpy exhibited the same chattery feeling as the EE.
The talk of a flex stay rear triangle as undamped is not really true. It is only undamped in the same way that your coil spring on a shock is "undamped". Both are constrained springs; neither can move on their own without being impacted by the damper of shock.
This is completely different to something like the spring effect from a tire which is fully isolated from the shock damper (the only damping is from the tire casing itself).
As an extension of this you could fairly easily design a bike that uses a bare damper with no coil spring at all and rely exclusively on the rear end flex as the spring (this is used in other vehicle systems with good success). Hysteresis effects aside its viable to design it in such a way that the feel to the rider is identical.
I can't speak to how Ken and crew actually measured any effects of the flex stay spring but as someone who designs instrumentation the systems are more than capable of measuring things that well below the perception range. The much harder part is determining what (if anything in some cases) the rider is feeling; and then what part of the systems is actually contributing to the feeling. It's easy to say "oh I feel this on that bike it must be the flex stay (or any other thing you could point at)" but that doesn't mean it's actually the root cause.
Off road suspensions is fairly complicated on its own. When you add very adaptable humans to the system it gets even harder to model individual effects. It's particularly hard on bicycles because the human component is both very dynamic and a way higher proportion of the sprung mass.
You need to get past that it's somehow undamped. The stays can't move without the shock moving. Full stop. If the stay moves the shock moves so there is damping.
I have build a steel fs with 160mm travel that replaces not just one but two pivots with flex in the seatstays. It sit's neutral at 40% travel and takes just light finger pressure to get it to move through the travel. The additional spring force is NOTHING compared to the shock spring.
I have some video of the rear end flexing from last year when I first built it that I will try and post up.
https://photos.app.goo.gl/vnJSHQKwPbHBKTY86
So at the end, when you compress the springs and let them rebound, you think the damper in the shock would control that?
The flex in the RT cannot happen unless the RT moves through it's travel, so at no point is the flex undamped. Think of the spring force of the flex stay RT as an extension of the air spring in the shock, and thus connected to the bike.
Next time I pull the frame apart I'll pull the spring off the shock and run that same flex.
Stop and think about it. The frame runs a 450lb spring....I am flexing that rear end with one finger. I certainly can't compress a 450lb spring any measurable amount with one finger and the shock damper handles that rebound. In the flex video I am also flexing it well past bottom out, I run a 55mm travel shock.
Granted this forum is mostly geared towards all-mountain and Enduro side of things. Riders who are primarily Enduro riders and like to smash into things tend to dislike these flex stay or XC style bikes. In XC, we don't smash into things like a big Enduro bike. It requires finesse, lots of active riding by unweighting/hopping over obstacles instead of plowing through them. You also have to compromise some of that downhill smashing ability for uphill and out of saddle performance which is where these races are won. Being an XC guy at core who now does equal part Enduro riding, I love them. They can be fast too but you just need be in that mindset.
I had the Epic Evo then got Transition Smuggler C and finally settled on the Stumpjumper Flexstay. For context I am 165lb that exclusively rode and raced XC for most of my riding years then moved to the mountains. Epic Evo geometry felt a bit lackluster for our local trails so I decided to build a light weight trail bike i.e. Smuggler. The horst link design excelled on the descents but left me wanting that "zing" and sharpness of my previous XC bikes. So I sold the Smuggler and built a Stumpy Flex which got me what I wanted. Better geo for our terrain, direct feeling acceleration and good downhill performance if you are willing to ride it like a XC bike. Now this bike is unique as it delivers 130mm travel which caused Specialized some flex stay spring back issues hence the very specific RX tune. See "Suspension Design" section here: https://nsmb.com/articles/2021-specialized-stumpjumper/ I just raced a local virtual enduro on blue trails using my Stumpy and placed near the top against folks with 160+mm bikes.
I think a lot of people get caught up in viewing the shock and spring as one unit. You can look at them as separate entities and the impact of a flex stay makes more sense. The total spring forcing the bike back to top of travel is the combination of the flex stay and the spring on the shock. The damper controls rebound of the entire rear end, so as long as the rebound tune is good for the combined spring rate you’re good.
the "subtle negative spring" is what has always had me curious about the chatter. i remember reading your review (its really good btw) and it really clued me in to a possible place where the negative quality was coming from, leading to my purchase of an evo (i would have gone back to stumpy alloy but i felt i needed more travel.) my feeling is that the subtle negative spring element, as well as the rebound mentioned in the article, is where this chatter comes from. i still cannot see how the damper can control the movement of that spring if the rear triangle is unloaded. please someone explain.
I wonder if you understand how the flex stay works? Maybe you could describe (with a picture?) where you think the flex is occuring.
The seat stay is effectively a cantilever beam, the peak strain will be at the end without a pivot i.e. at the axle. Because of this the flex stay really does behave like a revolute joint with a torsional spring in parallel.
I think you assume the flex stay spring is somehow in series to the the damper, this is incorrect.
N.b. The manufacturer manufacturer will likely try and engineer a flex region into the stay as well to improve robustness.
A lot of the harshness etc could be simply going from a Horst link set up to what is now effectively a linkage driven single pivot. Where you therefore are limited on the ability to decouple braking effects from the suspension.
O/T-ish: Speaking of leaf springs, how about this one? This will never be a Spot on my radar screen, but I am curious about how it feels / what sort of shock sorcery is needed to make it feel right.
I've been an engineer on a few flex stay bikes and have seen the "negative spring" narrative evolve. Let's say the distance between the pivots on a seat/chainstay part changes 4mm. It's smart then to design the part at the 2mm position to equalize (and thus reduce) the strain at each extreme.
From there it's a short walk to a marketing story where "the rear triangle is in it's neutral position at sag with negative spring effect in extension and positive in compression. It's not untrue, but in reality it's inconsequential compared to the force from the spring in shock absorber because the displacement across the rear pivots is tiny.
The rear triangle stiffness needs to be adequate to keep the wheel in line with the frame and prevent the linkage from binding. The left and right sides flexing independently would be particularly troubling, which I think is why the design has become popular since the introduction of rear through axles. I've heard people hint that the Busby flex-pivot patent expiring was a factor too, but I'm not sure it applies if it's the entire seat/chainstay boomerang flexing.
Every flex design that's ever made it to market is just replacing a bearing joint that has a very very low angle of rotation with a flexure. The pivot point will have an angle of rotation in the 1-5 degrees range. The rest of the kinematics play a much bigger role than the bearing that has been replaced ie Horst / Faux bar / VPP will determine the ride characteristics and not the flex.
Would you mind making or showing me a diagram?
The Busby patent for dorks like me that were curious about it.
https://patents.google.com/patent/US6092823A/en
We talking flex stays now? Imo they're like the cooler, easier to maintain older brother of the split pivot. Best if done with steel, I don't necessarily trust the idea of carbon flex stays lol
Carbon can some impressive stuff, not too worried about a few mm of flex
Ha, yes that link would've been helpful - I forget not everybody is burdened with a mental catalog of all the suspensions!
Yeti made these flex-pivots in the early 2000's.
I saw a number of these flex pivots come unglued from the aluminum tubes. IIRC GT sued Yeti so they had to make the whole rear triangle flexible (I don't think they were the first, but maybe).
Can you use this diagram to explain your point? Earleb, I’m curious for you to explain this as well.
https://bikerumor.com/wp-content/uploads/2024/03/Yet-ASR-Frame-Flex-Sta…
These explanations are helpful; is it roughly correct to say that the swingarm is ALWAYS under either tension or compression (except at the vertex of the curve where it switches from one to the other) so that the shock damper is always engaged?
Regardless of whether any given frame member is in tension or compression, for the flex stay to move as a joint in the suspension the shock must also move. Thinking about it as a torsion spring, if you say at full compression the spring is wound up its maximum amount, for it to unwind the shock must extend. The extension of the shock is controlled by rebound damping so the unwinding of the torsion spring is also controlled by that rebound damping
The seatstay is flexing forward of the riser at the dropout. This is where a pivot bearing would be if it wasn't a flex design. The one thing to point out is that the seatstay doesn't likely have this much visual deformation when it flexes. The CAD simulation will have a deformation scale, this exaggerates the visual of how much something is flexing. Something can look like it's flexed like a pasta noodle in CAD but it's only moving a few mm.
I feel like I’m getting closer to understanding the way you see it:
You’re saying that if we put a completely rigid shock in the bike that the flex stay wouldn’t flex?
Correct. That’s making the assumption the flexure is designed well of course.
Ok so you’re saying that depending on layup and other factors it may not work the way y’all are talking? That it could flex separately from the linkage? Also, how is it designed to avoid the issue I’m talking about?
@dolface think about it like a wishbone.
If you pull the ends apart it's under tension. If you push them together it's under compression. If you neither push nor pull it's neither under tension or compression.
It's two wishbones though, so it's important that they flex together.
Also one side had the brake attached so it's either important that it braking doesn't upset things - or that the importance of weight/cost/other outranks braking effects.
The displacement across the wishbone (rear triangle) is normally tiny, especially on short travel bikes, so the force it imparts to the system is inconsequential.
@CascadeComponents I agree your torsion spring analogy is the way these should operate.
I think perhaps whoever brought this up was imagining a scenario where if the shock had bottomed out, the flex of the rear triangle could still provide movement (like an flexural incarnation of the twin shock Cannodale Gemini prototypes). I don't remember the name of it but I think in the last decade there was a bike with a suspension which let the rear axle move within an spatial envelope rather than along a path. Maybe they're thinking of that.
@earleb in that Yeti image the seat stay is shown as the focus of flex. Some spread the flexing across the whole part. Some focus the flex on the chainstays at faux-horst links, like this Cannodale
at 9" travel, this is the most extreme implementation of a flex stay i've seen. not that there were enough examples to generate meaningful durability data, but by all accounts i'm aware of, these were bomber. this one saw 10 years of regular use, and is still intact 18yrs later (only dusted off for the occasional nostalgic retro ride these days). super stiff laterally, silly supple vertically. as a proof of concept, it certainly succeeded.
That's great and now I get it; thank you!
I brought it up and your description of how I was thinking about it is correct; really appreciate all the thoughtful responses and discussion here.
I’m still having trouble understanding how that wishbone wouldn’t chatter as you are descending. If the flex stay is positioned in between two ends it can oscillate to, couldn’t vibrations rattle that thing around?
I’m also still not seeing how a flex stay wouldn’t flex on a rigid bike.