[Steve M]

So after reading this and a few other threads, reading a moto suspension book or two, and a couple of actual prototypes under my belt I think I am starting to get a good understanding of the the pedal/brake induced forces and the design to enhance/counter those forces.

I would like to hear peoples thoughts on the brake induced forces. I understand that "brake jack" (where the suspension extends or firms up under braking) is bad. I personally like to have a bit of "brake squat" to balance out the compression of the front fork under braking. I would imagine that to much of this would be bad for traction and in the event of larger hits under braking, would probably allow the suspension to bottom easier. I personally don't use a ton of rear brake, so I don't think I would notice the disadvantages of this as much, but everyones riding styles are different.

What are your thoughts? Others talking about it helps me (and others I would imagine) to understand the true advantages/disadvantages in "real-world" bike riding, as opposed to all of the marketing hype in the bike industry.

Thanks a ton,

-Hollis

As with pedalling acceleration, obviously when you brake your weight shifts forwards. In other words, whenever you touch the brakes, the normal force on the front tyre increases, and on the rear tyre decreases (no surprises here). In a steady-state geometric sense (ie constant braking on a smooth surface without wheel lockup), having 100% anti-rise on the rear end of the bike would completely negate the effect of the rear brake (though there would still be some extension of the rear suspension from the effect of the front brake shifting more weight forwards). While this sounds good in a sense, it's only basically a static analysis.

Considering the suspension system dynamically - when you first hit the brakes or when, for whatever reason, the braking force rapidly increases (such as the wheel regaining contact with the ground after skipping off a bump), the initial reaction of a bike with any degree of anti-rise is that the rear suspension tries to compress itself (in other words, prevent itself from extending due to weight transfer). This means that, due to the inertia of the bike/rider mass (which can only accelerate towards the ground due to gravity, since the tyre can't actually "pull" the bike/rider towards the ground since it just rests on top of whatever surface it's on), the rear wheel will try to move itself upwards. If that upwards, compressive force on the suspension is applied too quickly and too hard (too much anti-rise), the wheel could actually hop off the ground. This is a pretty extreme case on flat smooth ground when the rider just grabs the brakes, because the rider can't actually apply the braking force all that suddenly. However, with stutter bumps, the fact that the wheel is constantly breaking and regaining contact with the ground (and as a result, also breaking/regaining traction) very rapidly means that you can get the sharp compressive forces required to cause wheel hop, decrease traction, and cause excessive harshness to be felt by the rider.

There's more detail in this post http://www.ridemonkey.com/forums/showpost.php?p=3127717&postcount=20 if you're interested.

 

[ChrisKring]

Socket

Good explaination.

Everyone needs to remember that with a bicycle, the rider has a significant part of the overall mass. Much more than most other vehicles. Therefore, the rider can, and will, effect the dynamics of the bike. Most riders will instictively shift there mass during braking in an effort to distribute weight to the rear wheel. Still as you point out, the tire can not "pull the rider" down. However, the tire does not induce a force pushing the rider either. Therefore, the skilled rider will squat the rear suspension under braking by shifting the COG. In addition, the skilled rider will not "stab" the brakes causing the wheel to loose traction to begin with (normal force wheel to ground X friction < braking force). That said, you are correct that you could have too much brake induced squat (ie. mantra) which would make braking without intermentant skidding more difficult.



Anyway, you can't fight physics and win. Pro level riders ride in a method that physics proves is the best way. Wether they intentionally do it or not, they all do it. Regardless of the suspension design, modulate the brake to avoid skidding, don't brake in braking bumps and brake in a straight line. The job of the suspension designer is to give them a design that does not actively work against the rider while providing stability for the paying consumer.

At the end of the day, I prefer a moderate brake induced squat since it rakes the fork out, thus more trail. Sorry if there are random thoughts above. I had a long day at work. It is way more fun to read and discuss bike physics than the physics of interior car parts. Thanks to all of the posters for improving my day.

 

[JCL]

A great couple of posts Socket thanks.

 

[dw]

(and there are certainly other means that may have never been built that could actually replicate the DW-link pedalling characteristics without infringing on the patents),

Plenty of patents pending still, 26 to be exact, and that number is growing. At the heart of the pending patents is claim language protecting the shape of the dw-link anti squat curve. I've researched literally every full suspension design that I have ever been able to find, back to the 1800s. I've never been able to find a linkage arrangement that achieves what has been done with dw-link. I'll be sure to note them on the dw-link site when they publish.

 

[dw]

I don't agree with that 100% - "mashing" on the pedals, as well as generating oscillatory power output, also tends to create vertical motion of the rider/bike's CoM. Unless the maxima/minima of the vertical accelerations involved were in phase with the chain tension maxima/minima (which I think they'd be somewhere near 90deg out of phase since max chain tension would be at about 9/3 o'clock pedal position, whereas max vertical acceleration of the rider's mass would be at closer to 6/12 o'clock) then the vertical motion alone of the rider would move the bike up and down in response, in addition to varying your anti-squat amounts. I don't think an ideal geometric solution exists for this - biopace sounds good in theory but screws up other things, like constant chainline/anti-squat anyway. Aerodynamic load can certainly be countered, being a force that changes at typically low frequencies.

See, I used a huge amount of data aquisition and video back in the early 2000s to characterize this issue. (The final effort was the Sunday prototype from 2003 courtesy of Gordo Spomerino and Littermag.com)

Anyways, what I found was that acceleration and mass transfer forces put aside, the rider's hips and the movement of his or her hips in relation to the power stroke was the biggest contributor to suspension compression. Any rider with a dw-link bike can see this for themselves. Get coasting, stand up, and backpedal. Notice the suspension compressing wildly. Then do the same accelerating forward. Notice the suspension not responding to your input the same way as when you backpedal.

In theory, the ideal value of anti-squat is about 100%.

I haven't written down all my tricks on Ridemonkey yet! (Although the bikes exist to measure and decipher)

What's really clever about Dave's design - the part that IMO is absolutely ingenious and he may well be pissed about me posting in public since it's a concept that's pretty hard to patent - is the way it decreases after the usable sag range. What this means, potentially/theoretically at least, is actually BETTER bump absorption if you're pedaling than if you're coasting, as well as a combination of efficient, firm pedalling and low pedal feedback.

That's actually EXACTLY what I've patented. That's the heart of the invention, and the first independent claims in the patent applications aim to cover this specifically. The patent process is ridiculous. They broke down all of the claims into 4 groups, so I had to file 4 patents to get all the claims. The office takes forever it seems. We've been at this patent for 5 years!! Like I said in my post before, there are a lot more patents in the dw-link family in process. I wish it all could have been in one, there would be significantly fewer dw-link lookalikes out there, but what can you do. Have to play the game.

I'm not so sure I agree with some of Dave's damper curve setups when it comes to what's ideal, but there is no question about the pedalling efficiency.

I'd love to see what you think about some of the newer bikes, and I'm sure you would appreciate some of the things that have gone into bikes like the Sunday. I don't think that many bicycle suspension people other than myself were looking at combined wheel rates of springs and position sensitive dampers and designing specific leverage rates to achieve an end result in 2002. Too bad the progressive damper went away. Lots of potential in that design. I was especially bummed seeing as I designed the Sunday around it, no other damper after it had the end travel high speed compression increase that the leverage ratio was designed around. Hence the dual progressive curve on the new DHR and Revolt. I'm rambling now, need to get to work..

It's a pleasure reading your posts man, you've become quite knowledgeable over the years, I'd like to think that I helped you along the way.

 

[Steve M]

Plenty of patents pending still, 26 to be exact, and that number is growing. At the heart of the pending patents is claim language protecting the shape of the dw-link anti squat curve. I've researched literally every full suspension design that I have ever been able to find, back to the 1800s. I've never been able to find a linkage arrangement that achieves what has been done with dw-link. I'll be sure to note them on the dw-link site when they publish.

As I said, I believe there are ways to achieve the same ends with a different suspension layout - whether or not that has been done in the past is another matter entirely. I'd be inclined to speculate, without having done the same research that you have, that it hasn't ever been done before, simply because motorcycle suspension has never focused on those same issues (due largely IMO to not having the bobbing issue that bikes do) and because MTB suspension is well-documented and has only been around for <20 years.

See, I used a huge amount of data aquisition and video back in the early 2000s to characterize this issue. (The final effort was the Sunday prototype from 2003 courtesy of Gordo Spomerino and Littermag.com)

Anyways, what I found was that acceleration and mass transfer forces put aside, the rider's hips and the movement of his or her hips in relation to the power stroke was the biggest contributor to suspension compression. Any rider with a dw-link bike can see this for themselves. Get coasting, stand up, and backpedal. Notice the suspension compressing wildly. Then do the same accelerating forward. Notice the suspension not responding to your input the same way as when you backpedal.

I agree that the rider's mass moving is a huge component of the bobbing phenomenon, but I don't think backpedalling is a fair representation of the way the rider's mass moves when squat forces are discounted. My reasoning is that when you backpedal, you can't actually exert enough force against the pedals to move your body the same distance, with the same acceleration, or in the same phase (albeit reversed) as when you pedal forwards and have the resistance of the pedals. I'm fairly sure you could replicate/quantify it by somehow providing a resistance torque to the cranks that wasn't attached to the rest of the bike though, and I'm also sure you're aware of this.

The data acquisition stuff is freaking cool, I spent most of last year messing around with DAQ gear as part of my thesis. My thesis, though not related to pedal efficiency, was basically aiming to develop a means of simulating the relationship between the bike and rider in terms of suspension response. Budget limitations meant it was confined to low speeds but the results were pretty interesting. If I had the money I'd love to actually get the DAQ gear on while I was riding (I had to build a test stand to excite the suspension) and begin to quantify that kind of stuff in a real-world environment. Don't suppose you're looking for a recently qualified engineer to handle such duties?

I haven't written down all my tricks on Ridemonkey yet! (Although the bikes exist to measure and decipher)

haha yeah I don't doubt that for a second... I've read a couple of the patents and have noticed the lack of specifically quantified data in them. If you have application #s for all the patents that are going up, I'd be interested to read them?

That's actually EXACTLY what I've patented. That's the heart of the invention, and the first independent claims in the patent applications aim to cover this specifically. The patent process is ridiculous. They broke down all of the claims into 4 groups, so I had to file 4 patents to get all the claims. The office takes forever it seems. We've been at this patent for 5 years!! Like I said in my post before, there are a lot more patents in the dw-link family in process. I wish it all could have been in one, there would be significantly fewer dw-link lookalikes out there, but what can you do. Have to play the game.

I've seen some of the example anti-squat curves in one of the patents that would generate that characteristic, as well as the linkage geometry specifications (IC starting outside the lower link and moving onto it at a later point in the travel, with no specification as to whether the IC must be on the lower link at bottom out - possibly one point of contention with the Maestro linkages?) that provide the reversal in the rate of axle path radius of curvature increase (unless I've wrongly identified this characteristic). However, in the patents I read (two or three of them), I don't recall reading the same explanation of how that decreasing rate of anti-squat assists the lack of chain extension, that seemed to be almost implied rather than explicated. Again, correct me if I'm wrong, it's been a while since I last looked at any of the patents.

I'd love to see what you think about some of the newer bikes, and I'm sure you would appreciate some of the things that have gone into bikes like the Sunday. I don't think that many bicycle suspension people other than myself were looking at combined wheel rates of springs and position sensitive dampers and designing specific leverage rates to achieve an end result in 2002. Too bad the progressive damper went away. Lots of potential in that design. I was especially bummed seeing as I designed the Sunday around it, no other damper after it had the end travel high speed compression increase that the leverage ratio was designed around. Hence the dual progressive curve on the new DHR and Revolt. I'm rambling now, need to get to work..

I am definitely a fan of MOST of the things about the Sunday - however I'm yet to ride one where I thought the small bump sensitivity was anything special. The ones with the 5ths tended to be quite sticky, and the two of those that I rode after being revalved with the correct tune still didn't feel very nice. My own dyno testing of 5ths revealed that right off the bat they are VERY heavily damped shocks (literally double, roughly, that of the DHX I tested, in minimum and maximum compression and rebound) and I think that even when revalved to have much lighter damping, they still just weren't a great mix in the real world. That is why I wasn't a huge fan of the leverage rate curve - the average ratio seemed too low for the shocks available at the time (though no doubt with some minor shock mods you could improve that quite a bit) and they just didn't seem progressive enough to be able to set up soft/with low compression, without bottoming out all the time. Again, better setup might have helped, but I didn't especially like the feel of the ones I rode (now totalling about 6 or so different Sundays). I think the theory behind the idea was sound, but the stickiness of the 5ths and the lack of ramp up/compression of the DHX meant that there was still a fair bit of room for improvement for the whole package.

I agree that nobody else really seems to consider wheel rate, in fact it is pretty well completely neglected in all marketing literature which would tend to indicate that it's not even considered by most people. I do have a question though - in your opinion, does the wheel rate of the bike have any SIGNIFICANT bearing on pedalling efficiency? Because I recall the "simultaneously engineered anti-squat and leverage rate curves" marketed on the dw-link website, which I initially took to mean that the pedalling characteristics were somehow affected by the leverage rate of the bike, but I am now more inclined to think that you meant the two were simultaneously achieved with the one 4-bar linkage? Also, have you done any testing of the way the rider affects the suspension response? As I said, I've done some basic testing at low-speed, I'm interested to see whether your wheel-rate optimisation was based solely on data acquisition/trial and error or based on simulations that incorporated the deformation of the rider. It seems to me that analytical optimisation of the mechanical components of the suspension (ie the spring/damper and their resultant wheel rate through the linkage) would require a model of how the rider himself actually reacts to inputs at the wheels. This is pretty much what I was trying to create in my thesis.

I'd be pretty keen to try the DHR or the Revolt, if you want to ship one my way for official Farkin.net testing/review let me know

It's a pleasure reading your posts man, you've become quite knowledgeable over the years, I'd like to think that I helped you along the way.

Thanks, and no doubt. Thanks for taking the time to explain this stuff to us plebs.

 

[amnonis]

I went through all the thread and I think I understood most of it. Does anti squat is exactly what the new Kona CoilAir achieves?
watch http://www.konaworld.tv/ -> training technology -> magic link theory
I rode the bike for few weeks and I can tell they have endless travel feel like a 7.5" bike and they pedal like a 5.5" bike. My feeling that they mange to achieve that by anti squat force from the pedals. all the rest IMO is marketing since every bike change its geometry while compressing.
I don't really understand why they had to add the coil. can't they achieve the same anti squat force only with a floater?
Can you help me to figure it out?