[Udi]

Good point - I should have added the word numerical in front of ratio, but given this is the internet I trust you wholeheartedly.

If only Specialized put this much effort into firing their suspension guy and hiring someone good. They have some of the most advanced manufacturing processes in the industry, very good build quality and structural engineering (not hard with companies like Intense and Yeti out there in fairness), but continually produce the same mediocre suspension. They may as well license the magic link.

 

[Dogboy]

If only Specialized put this much effort into firing their suspension guy and hiring someone good. They have some of the most advanced manufacturing processes in the industry, very good build quality and structural engineering (not hard with companies like Intense and Yeti out there in fairness), but continually produce the same mediocre suspension. They may as well license the magic link.

About 6 years (think Iron Horse implosion) ago I thought maybe Specialized was going to license the DW Link. Can you imagine how good the bikes would have been? Or how good they would be now a couple of generations later?

 

[demo 9]

About 6 years (think Iron Horse implosion) ago I thought maybe Specialized was going to license the DW Link. Can you imagine how good the bikes would have been? Or how good they would be now a couple of generations later?

bradflynn licensed it

 

[Samoto]

Good point - I should have added the word numerical in front of ratio, but given this is the internet I trust you wholeheartedly.

If only Specialized put this much effort into firing their suspension guy and hiring someone good. They have some of the most advanced manufacturing processes in the industry, very good build quality and structural engineering (not hard with companies like Intense and Yeti out there in fairness), but continually produce the same mediocre suspension. They may as well license the magic link.

exactly my thought. Their specialty is road racing market which is much larger in revenues.

 

[mtdh]

Good point - I should have added the word numerical in front of ratio, but given this is the internet I trust you wholeheartedly.

If only Specialized put this much effort into firing their suspension guy and hiring someone good. They have some of the most advanced manufacturing processes in the industry, very good build quality and structural engineering (not hard with companies like Intense and Yeti out there in fairness), but continually produce the same mediocre suspension. They may as well license the magic link.

Is there something you can point out that is wrong with the design. I looked a few other design and they and they have caveats as one would expect.

 

[Sandwich]

negative anti squat

 

[jackalope]

^^

Double negative. Thus, it has positive squat.

 

[jonKranked]

quantum state squat.

 

[mtdh]

negative anti squat

Anti squat requirement is subjective and seems to be lot marketing BS (brain washing). There s a lot of marketing on the benefits of anti squat but it certainly does have equal amount of negatives that are not discussed. Question is what type of suspension one want in the end. For example, I would want something that is very responsive an active all the time and sacrifice any other requirements. The FSR would be better choice in my case and LW link would not but that is another debate.

UDI has perspective on his comments so curious on what his view is.

 

[jackalope]

 

[Sandwich]

Anti squat requirement is subjective and seems to be lot marketing BS (brain washing). There s a lot of marketing on the benefits of anti squat but it certainly does have equal amount of negatives that are not discussed. Question is what type of suspension one want in the end. For example, I would want something that is very responsive an active all the time and sacrifice any other requirements. The FSR would be better choice in my case and LW link would not but that is another debate.

no

too much anti squat is debatably bad, but quantifying "too much" is pretty hard. You can get away with quite a bit, if you look at the voltage and grumbler. The demo has less than zero as it gets deep in travel. I can't figure a good reason to do that. I could see less, like 50%, or more, like 150%, and rationales for each, but I can't see -50% as a good thing.

This excludes any focus on leverage rate, of course.

 

[William42]

basically the moral of the demo is "we can get away with really sh1tty suspension kinematics as long as we have somebody make us a really really good shock for it and have good geometry"

 

[rbx]

Has any company ever provided "efficiency" real world number differences between 0,50 and 100% AS suspension systems?

I mean is it like a 5 watt power difference(loss) 0 vs 100% AS?

 

[Udi]

Is there something you can point out that is wrong with the design. I looked a few other design and they and they have caveats as one would expect.

In short, you've got the wrong idea about how squat (and AS) works. To explain very briefly, acceleration in any vehicle causes rearward mass transfer, which causes a compression of the rear suspension (and some extension of the front). Unlike a motorised vehicle, the torque input of a bicycle is sinusoidal, thus there's a large wastage of energy when propelling these vehicles forward unless the squat forces are balanced, or at least counteracted to a high degree.

Without getting into too much detail, a carefully designed rear suspension system will extend the rear suspension during acceleration by an approximately correct amount to counteract the mass transfer related compression that would occur purely as a result of the torque input. This opposing squat force (anti-squat) is applied as a carefully engineered curve on something like a dw-link bike to target regions of travel where mass transfer is high, and fall off in areas where it is not necessary to prevent unwanted effects such as pedal feedback or excessive extension of the suspension (which would still cause a loss of acceleration energy input).

This may be the part you are missing - it means that you can build a bike that will accelerate well without sacrificing bump absorption provided you know what you are doing. A shocking number of MTB kinematic engineers don't, and unfortunately it's one of those things where unless you understand every single factor involved (and their interactions), you don't really know anything at all.

To comment on FSR in particular, bump absorption on most bikes is primarily a function of leverage curve and axle path, and acceleration performance (on conventional chain-pathed bikes) is a function primarily of axle path and gearing. The previous demo didn't employ either variable particularly well to achieve these goals, and this one unfortunately only looks slightly better from my quick look at pivot locations.

That's not to say the bike won't work reasonably well, like I said the company does other things very well eg. structural and geometrical factors, but when it comes to suspension kinematics in particular, greater optimisation is possible without actually impeding on other performance variables.

There s a lot of marketing on the benefits of anti squat but it certainly does have equal amount of negatives that are not discussed. Question is what type of suspension one want in the end. For example, I would want something that is very responsive an active all the time and sacrifice any other requirements.

In layman's terms, the bottom line is that generally speaking, making one thing better will make another thing worse (as you correctly assume) once that one thing reaches a certain threshold (which many do not realise). However with good engineering and optimisation you can maximise single performance factors to that threshold without bringing down other factors - a majority of designers fail to do this.

Hope that's remotely helpful in clearing some things up for you.

 

[kidwoo]

In short, you've got the wrong idea about how squat (and AS) works. To explain very briefly, acceleration in any vehicle causes rearward mass transfer, which causes a compression of the rear suspension (and some extension of the front). Unlike a motorised vehicle, the torque input of a bicycle is sinusoidal, thus there's a large wastage of energy when propelling these vehicles forward unless the squat forces are balanced, or at least counteracted to a high degree.

Without getting into too much detail, a carefully designed rear suspension system will extend the rear suspension during acceleration by an approximately correct amount to counteract the mass transfer related compression that would occur purely as a result of the torque input.

This has bugged me since the daveW first started talking about this. Bikes aren't motorcycles. People almost always lean forward when pedaling hard, which already transfers mass away from the rear end. It's not like the bike is accelerating out from under you like on a motorcycle. I fully understand the theory but the applicability to bicycles just always seemed a little off for that reason.

I don't know about you guys but I've never been able to pedal so hard my bike shoots forward out from under me.

 

[EVIL JN]

This has bugged me since the daveW first started talking about this. Bikes aren't motorcycles. People almost always lean forward when pedaling hard, which already transfers mass away from the rear end. It's not like the bike is accelerating out from under you like on a motorcycle. I fully understand the theory but the applicability to bicycles just always seemed a little off for that reason.

I don't know about you guys but I've never been able to pedal so hard my bike shoots forward out from under me.

I would say it depends on the situation, flat out in a high gear, sure I have most weight on the bars. A snap out of a corner I have more weight on the rear, driving the power down and pulling on the bars. So I guess acceleration on bikes are very situational since it is a human that is the power and we dont tend to do things in a single way/pattern.

 

[mtdh]

In short, you've got the wrong idea about how squat (and AS) works. To explain very briefly, acceleration in any vehicle causes rearward mass transfer, which causes a compression of the rear suspension (and some extension of the front). Unlike a motorised vehicle, the torque input of a bicycle is sinusoidal, thus there's a large wastage of energy when propelling these vehicles forward unless the squat forces are balanced, or at least counteracted to a high degree.

Without getting into too much detail, a carefully designed rear suspension system will extend the rear suspension during acceleration by an approximately correct amount to counteract the mass transfer related compression that would occur purely as a result of the torque input. This opposing squat force (anti-squat) is applied as a carefully engineered curve on something like a dw-link bike to target regions of travel where mass transfer is high, and fall off in areas where it is not necessary to prevent unwanted effects such as pedal feedback or excessive extension of the suspension (which would still cause a loss of acceleration energy input).

This may be the part you are missing - it means that you can build a bike that will accelerate well without sacrificing bump absorption provided you know what you are doing. A shocking number of MTB kinematic engineers don't, and unfortunately it's one of those things where unless you understand every single factor involved (and their interactions), you don't really know anything at all.

To comment on FSR in particular, bump absorption on most bikes is primarily a function of leverage curve and axle path, and acceleration performance (on conventional chain-pathed bikes) is a function primarily of axle path and gearing. The previous demo didn't employ either variable particularly well to achieve these goals, and this one unfortunately only looks slightly better from my quick look at pivot locations.

That's not to say the bike won't work reasonably well, like I said the company does other things very well eg. structural and geometrical factors, but when it comes to suspension kinematics in particular, greater optimisation is possible without actually impeding on other performance variables.



In layman's terms, the bottom line is that generally speaking, making one thing better will make another thing worse (as you correctly assume) once that one thing reaches a certain threshold (which many do not realise). However with good engineering and optimisation you can maximise single performance factors to that threshold without bringing down other factors - a majority of designers fail to do this.

Hope that's remotely helpful in clearing some things up for you.

I appreciate the response and your comments. FWIW, I do have a full understanding of anti squat but the application of a moving mass (the riders) makes this unique. My back ground is in mechanical engineering and have spent a fair amount of time working with automotive suspension analysis and design which really doesn’t not translate to MTB at all expect some real basics calculations. I would tend to agree with engineering abilities on certain companies. I have the knowledge and resources to analysis a MTB design with multiple iterations based on defined targets. The problem is what looks good on paper may not work well in the field for many reason. I don’t have the back ground knowledge on what works well and why which is not going to be explained with calculations. I can only assume the people doing the dynamic analysis are not using the best software tools to do the job as they can run thousands of design iteration and get multiple solutions. The other side is that engineers do not know what a good or bad design is.

I have not seen any reliable data of the old demo to make any judgment as my cured measurements are not good enough. I am not much in the guessing game and making assumptions on what the new demo will be but I not seeing the same thing you are. I do have CCM measurement of Enduro Evo Expert. I posted some data in this form. This setup is something I am very happy with and it is progressive from start to end (20% rate increase) with a slight rearward axel path for the majority of the travel. They seemed to have done this design well.

Again thanks for your response…

 

[gemini2k]

Looks like this bike is still a career killer. Has Gwin won a race on a demo yet?

 

[Kanye West]

Remember.....INNOVATE OR DIE.

 

[sethimus]

Looks like this bike is still a career killer. Has Gwin won a race on a demo yet?

short term memory loss?

 

[gemini2k]

short term memory loss?

Please enlighten me? He won PMB on an endruo-whatever it's called. I don't think he won a single race last year.

#demo=careerkiller

 

[Samoto]

demo is supposed to be demo (not to be production capable, just a demo-ex) Lol

 

[kidwoo]

A snap out of a corner I have more weight on the rear, driving the power down and pulling on the bars.

For sure. I'd still say the other scenario is more common though throughout a run though.

I'd just feel better (on the inside, where it counts) if those scenarios were discussed a little more in a bike specific context when the antiskwats conversation starts. Assuming weight transfer from all varieties of pedaling just doesn't seem accurate......especially in a dh bike context. I get wanting a good bit of antisquat just so you're not sending all that power into your dampers, but weight transfer isn't the reason for it IMO. We're not exactly passive sacks of taters when we pedal.

 

[stiksandstones]

Looks like this bike is still a career killer. Has Gwin won a race on a demo yet?

Not the race you are looking for, but he did win U.S. Champs on the demo.
You did ask.

 

[Kanye West]

This has bugged me since the daveW first started talking about this. Bikes aren't motorcycles. People almost always lean forward when pedaling hard, which already transfers mass away from the rear end. It's not like the bike is accelerating out from under you like on a motorcycle. I fully understand the theory but the applicability to bicycles just always seemed a little off for that reason.

I don't know about you guys but I've never been able to pedal so hard my bike shoots forward out from under me.

Likewise.

The only situation where I've felt that this really applies is small short accelerations with seated pedaling. So....XC.

Er, wait. "Enduro". Yes, that's what I meant to say.

I'm convinced it's about as useful as an asshole on your elbow while riding a DH bike, especially if it comes at the expense of any other aspect of performance.

 

[gemini2k]

Not the race you are looking for, but he did win U.S. Champs on the demo.
You did ask.

Yeah, that I knew. I meant real races. Not races with one pro (or 2 if Neko was there) beating up on a bunch of weekend warriors.

 

[CBJ]

But isn't the point you can incorporate it without compromising the other suspension attributes like UDI describe thus helping create a faster overall bike. If races are won with seconds every little bit counts.

I remember as if DW did write the DW link was even more beneficial in an XC setting but not sure. I don't think he post here anymore unfortunately.

 

[jackalope]

Well, Brosnan did win Ft. Bill on the previous Demo, and seems to be getting along well with the new (proprietary) shock changer model.

 

[Udi]

I'd like to preface this post by pointing out that my current riding stable includes zero bikes that have suspension designed by DW, zero bikes featuring FSR, and zero bikes using the I-TRACK suspension system (I'm not sure you can even buy one). I've owned many of the former two but have no allegiance to any, I just find this stuff interesting and am open to correction on any of the points made.

I genuinely think the biggest mistakes people (myself included) make on this topic are a) assuming making something better will always make something else worse, and b) writing off claims as marketing without actually making an honest and objective attempt at calculating the plausibility of the claim themselves.

Finally - I realise that acceleration performance is not the most important factor on a bike. However this is one of the most misunderstood areas so I'll focus on that.

This has bugged me since the daveW first started talking about this. Bikes aren't motorcycles. People almost always lean forward when pedaling hard, which already transfers mass away from the rear end. It's not like the bike is accelerating out from under you like on a motorcycle.

You've got the wrong idea here, but I think it's mainly because the key part is not very obvious intuitively, or at least I struggled with it personally until I drew an FBD.

I was vague with my definitions before but the transfer can be broken down into load transfer - a change in vertical force experienced by tire contact points; and mass transfer - a physical variation in the CoM with respect to the contact patches. The load transfer is hard to picture intuitively (because it's generated in one place but acts somewhere else), and the mass transfer itself while easy to picture, has a smaller effect on the total squat force than the load transfer.

The squat force we are talking about here is generated by any wheeled body that is accelerated (even without suspension), and it is a function of three key things - the traction force generated at the tire, the wheelbase, and the vertical distance from the ground to the vehicle's CoM. A moment is generated by the CoM (with the vertical distance being the lever arm), then this rotation generates an increase in the normal force at the rear tire (and a corresponding decrease at the front tire) - which in the case of a vehicle with suspension will cause the rear to squat.

You talk about leaning forward, but this has only a small effect on the vertical location of the CoM, and this is what matters when calculating the largest component of the squat force. The only large effect of leaning forwards is on static mass distribution, so the net effect is that your "bob" will occur in a different region of travel, the peak amplitude of oscillation will change very little!

It's also worth noting that the CoM on a bicycle is very high compared to other vehicles, so when you and mtdh suggest that conventional vehicle dynamics are less applicable to bicycles, you aren't considering the fact that if anything - the effect in question here is amplified rather than reduced! There are variables related to the sinusoidal nature of the acceleration input and moving CoM (in relation to the wheels) that do complicate matters, but these are secondary considerations that have a smaller impact.

I am not much in the guessing game and making assumptions on what the new demo will be but I not seeing the same thing you are. I do have CCM measurement of Enduro Evo Expert. I posted some data in this form. This setup is something I am very happy with and it is progressive from start to end (20% rate increase) with a slight rearward axle path for the majority of the travel. They seemed to have done this design well.

Again thanks for your response…

You're more than welcome, I imagine everything above is incredibly elementary for you but I suspect you just haven't considered it in detail previously. My area of interest is primarily DH bikes, so the data I'm most familiar with is in that genre. If you want to PM me your email address I'd be more than happy to forward some some comparative graphs I've sent around before.

Please don't take this as being condescending towards you or your bike (in fact I know very little about it) but when you say "slight rearward axle path" - this is somewhat subjective, is this the optimal amount for the best bump absorption? How is the 20% rate increase distributed over the leverage curve, and is this amount and its placement optimal? Which bikes have you compared these values to and why is one better than the other?

It seems a little unfair to dismiss something as marketing BS if you haven't considered these points in great detail. You may very well be correct but without analysing and calculating things yourself (or doing a lot of objective back to back testing with different designs), you don't really know right?

As I stated earlier, the forces generated to oppose load transfer based squat on a bicycle (with conventional drivetrain) are primarily due to axle path and gearing. This means that the axle path needs to change proportionately to the squat forces generated. If the shape of this curve does not precisely correlate to to the squat forces, you are going to generate oscillations and lose energy. Obviously every bike will have losses, but some have less than others, and as I said earlier - many bikes can be made much better before other performance factors are impeded upon.

A useful link which you should definitely read:
http://www.i-tracksuspension.com/suspensiontheory.html

Most of this summarises Dave Weagle's work (in DW's patents which are referenced within that link), but Hugh from I-track has developed upon this work by allowing the antisquat curve to be defined (precisely) independently of the axle path - which is incredibly clever and in my opinion represents one of the first substantial forward steps in MTB suspension kinematics since Weagle's original developments. He has no connection to DW and (from my understanding) is an independent mechanical engineer who has been doing his own R&D.

As a sidenote, before someone like noskidmarks on here decides to extrapolate on the above as endorsement for huge variations in axle path (and thus chainstay length), it's not. The brilliance in Hugh's work (in my opinion) is not necessarily in his prototypes, but in the decoupling of certain performance factors so that they are not limited by the point where they would negatively affect each other. This still leaves the problem of finding which amount of each factor is *actually* the optimal amount which I suspect is a work in progress.

The only situation where I've felt that this really applies is small short accelerations with seated pedaling.

Come on, you of all people should understand physics does not pick and choose when it applies. The concept of load transfer applies anytime you accelerate a body whose center of mass is above or below the tractive force's point of action.

I've personally found the differences very obvious when standing up and cranking hard as you would on a DH bike, but it's not isolated to either of the scenarios we described. If you do a truly objective comparison it's pretty easy to feel significant differences in acceleration performance between a bike that optimises for it and one that doesn't.

I'm convinced it's about as useful as an asshole on your elbow while riding a DH bike, especially if it comes at the expense of any other aspect of performance.

See my point at end of #214, it doesn't need to come at the expense of other aspects, and it can (and is) managed much more poorly than it could be on a lot of bikes - i.e. nowhere near the level of impeding on other performance aspects. As for your elbow, see quote below. Hill won by 0.5s today.

If races are won with seconds every little bit counts.

Absolutely nailed it. I'm consistently amazed by how few people appreciate this fact.

I fully understand the theory but the applicability to bicycles just always seemed a little off for that reason.

You've proved the opposite, but cue the typical kidwoo response that avoids admission of this in 3...2...1...

 

[Sandwich]

Hey Udi-
Do you think there's more value in representing anti-squat as a percentage or a numerical value based on units? I've often wondered that as I question what 100% is...what if a rider is heavier than another, or has more power than another? Certainly the way the suspension behaves, as a percentage, would be affected, correct?

 

[Lelandjt]

the asymmetrical frame thing just isn't jiving with me. is their reasoning that it makes it moar easy to access the shock for maintenance and shiz? Cause for the life of me, I can't think of anything you would need to access on a shock while its mounted, other than the knobs and mounting hardware. It looks ****ing ugly. The side with the bracing is too beefed up to compensate for the lack of support on the other side, and the side without it just looks ****ing stupid. No need to reinvent the ****ing wheel, put support on both sides and slim down the structure, would have looked much better IMO. Owell, maybe the spring will castrate a few Specialized customers and keep them from producing offspring.

A single bigger structure is lighter than two slimmer structures. Specialized went for performance but you want them to go for looks

 

[Nick]

Specialized went for performance but you want them to go for looks

dude. it's ridemonkey.

 

[kidwoo]

You've proved the opposite, but cue the typical kidwoo response that avoids admission of this in 3...2...1...

Nah. Like you said, it sounds like I was misunderstanding (or it was poorly explained previously ) what weagle always referred to as mass transfer under acceleration. I mean he WAS always bringing up motos.



edit: So yeah! You're wrong!

 

[Jm_]

Isn't every single pedalstroke, whether it's uphill or downhill, an "acceleration"? I know it's been measured before, looking at the actual speed varying slightly with each stroke, even though it's a very small number, still an acceleration. We can't adjust our body output to do anything else.

 

[Kanye West]

My point was that when standing up and pedaling and yanking on the bars, there is so much other weight transfer going on, whatever benefit might come from a slightly better anti-squat character would be entirely masked by all that. When the mass of the rider is more static, then it becomes a more prominent factor

What makes the MOST difference in stand-up pedaling on a DH bike is a proper fit between the cranks and the bars, and being able to use your back and your core to push down WAY more power with your legs. When your knees are almost smacking the bars because they're too close/too low, that's going to sap power transfer just from poor back ergonomics far far more than a little bit worse anti-squat character ever will.

Not to say the theory isn't valid, just pointing out that in practice, on a DH course, I think you'd have to REALLY stretch to justify a claim that it's making a marked difference over the other combined factors.

 

[mattmatt86]

You guys must me a delight to go on rides with, "I would have totally KOM'd that Sh!t if I only had more Anti-squats!"

 

[csermonet]

A single bigger structure is lighter than two slimmer structures. Specialized went for performance but you want them to go for looks

How is having an asymmetrical support structure more performance oriented than having one that is symmetrical? A triangle is the strongest geometric shape in the world, not a "C". It just so happens that having a symmetrical structure would be more aesthetically pleasing than that stupid **** they have going on. Who gives a **** if it's lighter, I don't want my bike to have asymmetrical stiffness characteristics. That's like doing what F1 went through in the 60's and 70's, doing **** to make it lighter without giving any regard to fatigue and stiffness, thus failure after failure.

There's a reason **** like this doesn't catch on

 

[Jm_]

There's a reason **** like this doesn't catch on

But if Cannondale made airplanes...

 

[jackalope]

 

[Samoto]

How is having an asymmetrical support structure more performance oriented than having one that is symmetrical? A triangle is the strongest geometric shape in the world, not a "C". It just so happens that having a symmetrical structure would be more aesthetically pleasing than that stupid **** they have going on. Who gives a **** if it's lighter, I don't want my bike to have asymmetrical stiffness characteristics. That's like doing what F1 went through in the 60's and 70's, doing **** to make it lighter without giving any regard to fatigue and stiffness, thus failure after failure.

There's a reason **** like this doesn't catch on

A LOT PPL GIVE **** FOR WEIGHT!!!!

otherwise we wont have carbon fiber nowadays.

Also it is not important with that upper part of seatstay, they could remove it as well and they'd attach a funny thing on headtube to support rider's ass aka Moto style!