[hmcleay]

Hey everyone,

There's too much lizard talk and not enough armchair engineering going on in this forum, so I've decided to steer things back on track.
Here's something I'll be working on over the next few months. Throughout the series I'll be discussing all things kinematics, mostly from the designer's perspective, and I'll try and cover most of the different suspension configurations available on the market.
If you're already familiar with Linkage, then skip to 4:20 to see a handy tool that I've developed.

Cheers,
Hugh.

 

[Happymtb.fr]

Will you make this tool available for mere mortals, aka armchair engineers of the interwebz ?

 

[wiscodh]

[QUOTE="hmcleay, post: 4333279, member: 28424] then skip to 4:20

Cheers,
Hugh.[/QUOTE]

 

[Udi]

This is actually super cool, because previously a degree of creativity or guess/check was required to know where to position the pivots to achieve a certain set of curves simultaneously. Being able to set a practical/manufacturable boundary and then have the software estimate the best locations within the boundary could cut kinematic development time solidly. Nice work Hugh.

That said, Santa Cruz only learned what a leverage curve does 2 years ago, we might be waiting a while for any non-armchair benefit to come from it.

 

[norbar]

This is awesome Hugh.

By the way. Any new projects in the pipes? Or any new ITrack bikes?

 

[hmcleay]

Will you make this tool available for mere mortals, aka armchair engineers of the interwebz ?

I don't plan on making it publicly available, however people can gain access to it through my consulting services.
The main reason for this, is that many clients would have their own additional design constraints (more than just limiting pivot locations to +/- Xmm) that they need to adhere to.
For example, there are quite a few patents that relate to kinematic features, and so company X might want to limit the output so that it only produces configurations that avoid infringing a certain patent(s), or they might want to limit it to produce results that are within the claim set of their own patents. If that's the case, then I can write some additional code to ensure that the results meet the clients requirements. Basically, I can tweak the software to help designers overcome specific challenges, on a case-by-base basis.

Happy to help out any monkeys with personal projects though. PM me if you've got a specific problem you're trying to solve.

 

[hmcleay]

[QUOTE="hmcleay, post: 4333279, member: 28424] then skip to 4:20

Cheers,
Hugh.

[/QUOTE]

I wish I was witty enough for that to have been intentional!

 

[hmcleay]

This is actually super cool, because previously a degree of creativity or guess/check was required to know where to position the pivots to achieve a certain set of curves simultaneously. Being able to set a practical/manufacturable boundary and then have the software estimate the best locations within the boundary could cut kinematic development time solidly. Nice work Hugh.

That said, Santa Cruz only learned what a leverage curve does 2 years ago, we might be waiting a while for any non-armchair benefit to come from it.

Thanks Udi.

Yep, it certainly cuts down design time. For those people that have designed a bike, and then proceeded to build it, you'll know that there's usually a point in the process where you need to go back to the drawing board and adjust pivot locations to resolve a structural/packaging challenge. With this tool, it makes it really easy to relocate a troublesome pivot, and then work out what needs to be done to the other pivots to get back as close as possible to the original target characteristics.

 

[shirk007]

I found part of the fun in designing the bike has been fiddling in Linkage to move around the pivots to learn when it does. Spent many many hours fiddling with the pivots to get the desired outcome for LR and AS.

Look forward to see what you have next.

 

[SkaredShtles]

<snip>

That said, Santa Cruz only learned what a leverage curve does 2 years ago,

I lol'd.

 

[norbar]

I lol'd.

It's funny because it's true

 

[Kurt_80]

"That said, Santa Cruz only learned what a leverage curve does 2 years ago"

Is this about the S-shaped wheel path? Or is there some other tom-foolery I'm not aware of?

 

[William42]

Their V10 was the only bike that had a sensible leverage rate for a long time. All sorts of wonky shenangins had to go on with shock tuning to make shocks that didn't blow through the travel on them. Recently, they started bringing their V10 leverage curves to their trail bikes and they've improved quite a bit.

 

[norbar]

"That said, Santa Cruz only learned what a leverage curve does 2 years ago"

Is this about the S-shaped wheel path? Or is there some other tom-foolery I'm not aware of?

First part of travel on many SC bikes leverage ratio went in the opposite direction to what was commonly known as good. Instead of starting at a high leverage ratio and going down to overcome stiction they started at a lower and went slightly higher because reasons.

 

[Kurt_80]

Ah, so the famed Nomad leverage ratios (before the current model).

 

[Happymtb.fr]

Today's flatish leverage curves is what we get when bikes are targeted towards people riding flow trails and wanting to bottom out on a 2 feet drop...

 

[Westy]

This is actually super cool, because previously a degree of creativity or guess/check was required to know where to position the pivots to achieve a certain set of curves simultaneously. Being able to set a practical/manufacturable boundary and then have the software estimate the best locations within the boundary could cut kinematic development time solidly. Nice work Hugh.

That said, Santa Cruz only learned what a leverage curve does 2 years ago, we might be waiting a while for any non-armchair benefit to come from it.

This is actually super cool, because previously a degree of creativity or guess/check was required to know where to position the pivots to achieve a certain set of curves simultaneously. Being able to set a practical/manufacturable boundary and then have the software estimate the best locations within the boundary could cut kinematic development time solidly. Nice work Hugh.

That said, Santa Cruz only learned what a leverage curve does 2 years ago, we might be waiting a while for any non-armchair benefit to come from it.

I took a kinematics design course as an elective in grad school. It was pretty simple to take an optimization function from Matlab or Mathematica, plug in 4 bar kinematics, pivot constraints, desired axle path and get an optimized design in a few tenths of a second. The only real trick was having constraints that would produce a solution.

Professor called it the patent cracker.


One could get creative and iterate possible pivot constraints or weight points on the path if a solution wasn't readily available.

If anyone is interested I can see if I can dig up my notes.

 

[Flo33]

If anyone is interested I can see if I can dig up my notes.

That would be awesome. I have little knowledge in Mathematica, but that's when a brother with a PhD in Mathematics who is working as a simulation specialist at a certain star sporting car manufacturer with said software comes in handy.

 

[Udi]

If anyone is interested I can see if I can dig up my notes.

Yeah please do, I'm doing some (very basic) work in Mathematica at the moment too.

 

[Tantrum Cycles]

The only real trick was having constraints that would produce a solution.

BWAHAHAHAHAHAHAHAHAHAHA.

too funny. Real life.

 

[Westy]

Not sure if this will work

https://www.dropbox.com/s/luiqms8seot38z0/youssef_1973_optimization_func_path.pdf?dl=0

https://www.dropbox.com/s/faf8les59is4pic/youssef_1975_opt_kin_syn.pdf?dl=0

Found my old notes but my Mathmatica licence expired, will see if I can find away to open them.

I understood this stuff at some point, can't say that point is now.

 

[tacubaya]

Not sure if this will work

https://www.dropbox.com/s/luiqms8seot38z0/youssef_1973_optimization_func_path.pdf?dl=0

https://www.dropbox.com/s/faf8les59is4pic/youssef_1975_opt_kin_syn.pdf?dl=0

Found my old notes but my Mathmatica licence expired, will see if I can find away to open them.

I understood this stuff at some point, can't say that point is now.

Ahhhhh that brings back memories.

 

[tacubaya]

Hey everyone,

There's too much lizard talk and not enough armchair engineering going on in this forum, so I've decided to steer things back on track.
Here's something I'll be working on over the next few months. Throughout the series I'll be discussing all things kinematics, mostly from the designer's perspective, and I'll try and cover most of the different suspension configurations available on the market.
If you're already familiar with Linkage, then skip to 4:20 to see a handy tool that I've developed.

Cheers,
Hugh.

Amazing work Hugh!

 

[norbar]

Today's flatish leverage curves is what we get when bikes are targeted towards people riding flow trails and wanting to bottom out on a 2 feet drop...

Flattish were popular before and that's not the problem that was old SC. Seriously look at SC bikes pre 2010 and it's like Halloween for anyone understanding basic physics.

Also I think the best curves for flow trails were intense 951 wall of unbottomability that resulted in you never getting full travel.

 

[norbar]

Ahhhhh that brings back memories.

Electronics courses only thought me matlab not mathematica. No memories of sweaty dudes in flannel shirts here

 

[Happymtb.fr]

Flattish were popular before and that's not the problem that was old SC. Seriously look at SC bikes pre 2010 and it's like Halloween for anyone understanding basic physics.

I have looked at these curves and I am mostly amazed that SC managed to keep its fan base after doing such horrible things...

I was referring to more recent bikes like Mondraker Foxy 29", Transition Sentinel, SC Hightower, Orbea Rallon, Ibis Ripmo, etc... which are all marketed as enduro capable bikes.

Also I think the best curves for flow trails were intense 951 wall of unbottomability that resulted in you never getting full travel.

If unbottomability is your main target, just go for it!

 

[norbar]

I have looked at these curves and I am mostly amazed that SC managed to keep its fan base after doing such horrible things...

I was referring to more recent bikes like Mondraker Foxy 29", Transition Sentinel, SC Hightower, Orbea Rallon, Ibis Ripmo, etc... which are all marketed as enduro capable bikes.


If unbottomability is your main target, just go for it!

Yeah in the good old days of "mini dh" and "park bike" ultra unreasonable progressive curves were hot shit.

As for SC fans. SC is the apple of bikes. For a long time I thought NS bikes were it since they had good graphic design but SC has such an amazing branding they can sell any shit and like apple their bikes have some positives - usually good customer service, they are fairly durable, they pedal quite well. So people loved them. I remember 12+ years ago everyone wanting a VP free despite that bike being bad as hell.

 

[StiHacka]

Say what you want, my Lowtower is a fantastic fun bike, curves or not.

 

[norbar]

Say what you want, my Lowtower is a fantastic fun bike, curves or not.

It's probably one of the new ones without wonky curves though I'm too lazy to check (and at work).

Not that SC owners were not saying that for every SC bike they had. Until they got a non sc bike that is

 

[toodles]

It's probably one of the new ones without wonky curves though I'm too lazy to check (and at work).

Nah they haven't released a 29er with the new LR curve yet.

Bear in mind, people are allowed to have different preferences on leverage curves. I don't agree that there's a perfect leverage curve - some riders prefer more feedback in different parts of the stroke and are happy to trade-off small bump compliance or traction. Others simply don't care that much about what the shock feels like as long as they aren't breaking their ankles off hucks.

Remember when the Orange 222 and 223 were winning WCs?

 

[hmcleay]

Episode 2 - Single Pivots:

 

[norbar]

Nah they haven't released a 29er with the new LR curve yet.

Bear in mind, people are allowed to have different preferences on leverage curves. I don't agree that there's a perfect leverage curve - some riders prefer more feedback in different parts of the stroke and are happy to trade-off small bump compliance or traction. Others simply don't care that much about what the shock feels like as long as they aren't breaking their ankles off hucks.

Remember when the Orange 222 and 223 were winning WCs?

Yeah I understand people have preferences but the curves on old SC bikes had no benefit. I am talking about leverage curves and especially about their tendency to start with a lowish leverate ratio and then going up. Also when I say no benefit I don't mean "measured on race track". I mean they didn't even make the bike feel better. They made the bike feel less supple. It literally made no sense.

Also no mid stroke support has nothing to do with what happened early in SC leverage curves. If it did more companies would adopt this idea and SC would not be abandoning it.

Also if you don't care what your shock feels like then ok but that doesn't mean the suspension should not get criticized.

Finally - bikes winning races speaks little about those bikes and more about the rider. CG rode the rampage on a C-dale Gemini and that bike was as rigid as a wet noodle. Not to mention old orange bikes had good geo for the times.

Also I'm not trying to shit on SC as a whole company. Their bikes are durable. Their bikes pedalled well early on where everything else was a pogo stick. Their new susp design is good. It's just their old suspension designs were really really wonky and it was really surprising they pushed ideas developed in the days of guesswork design for so long.

 

[toodles]

Yeah I understand people have preferences but the curves on old SC bikes had no benefit. I am talking about leverage curves and especially about their tendency to start with a lowish leverate ratio and then going up. Also when I say no benefit I don't mean "measured on race track". I mean they didn't even make the bike feel better. They made the bike feel less supple. It literally made no sense.

I'm not disagreeing that I prefer the commonly accepted progressive or dual progressive curves, just stating that there are riders that for some reason actually prefer the weird digressive designs. Granted a coil shock tends to cover up a lot of poor designs as well, but I can see why some riders would prefer the high levels of initial feedback. It is possible to make a bike too supple on the initial stroke and rob the rider of feedback from the rear wheel and some riders tolerate that and others don't.

 

[djjohnr]

Regardless of what they've done in the past, the latest Bronson is the most confidence inspiring bike I've ever ridden in it's class by a long shot. The difference in lean angles I'm achieving with it is something I wouldn't have believed possible previously.

 

[norbar]

I'm not disagreeing that I prefer the commonly accepted progressive or dual progressive curves, just stating that there are riders that for some reason actually prefer the weird digressive designs. Granted a coil shock tends to cover up a lot of poor designs as well, but I can see why some riders would prefer the high levels of initial feedback. It is possible to make a bike too supple on the initial stroke and rob the rider of feedback from the rear wheel and some riders tolerate that and others don't.

You can see why some riders prefer a suspension that has high initial stiction? Please tell me. Outside of liking strange things like no damping fox dhx 5.0 set to fully open I see no reason. I am being honest here.

Also what is this feedback from the rear wheel you are talking about? Not trying to be mean. I seriously have no clue what do you mean here. As for too supple - even if it's true old SC bikes were far from too supple. They were on the opposite end of the spectrum so you worrying about too supple is not a problem with old SC designs. We can discuss merits of different gradients of leverage curves if you think one design may be to steep but in SC designs the curve basically goes upside down.

Overall I think you are arguing for engineering placebo. Of course some people will like a given design no matter how bad it is. People used to praise URT suspension. Some probably still do. Doesn't mean I can't call that suspension stupid. People believe in homeopathy, people believe in VP-Frees. Same thing, different dentist.

Still let's not derail it and let's talk about Hugh's videos. As fun as killing time at work is I'd want to be on point.

 

[hmcleay]

Still let's not derail it and let's talk about Hugh's videos.

It's all good.
You'll notice I'll try to refrain from passing judgment on specific shaped curves being "good" or "bad". The idea being that the content remains 100% factual, regardless of your personal preference (unless you can violate the laws of physics).
But I'm happy if it prompts some healthy discussion about what characteristics people do/don't like. So carry on.

 

[Happymtb.fr]

Santa Cruz, you said...?

NSFW

And sorry for pushing the thread off track, again

 

[Happymtb.fr]

Really cool bit of programming you did there @hmcleay !
The sound is really low in your video, by the way.

I am quite curious about wheel path. Rearward wheel path seems to be increasingly popular these days.
From memory:
- a rearward wheel path should give more time for the shock to absorb an impact
- a rearward wheel path should slow you down more on the rebound stroke since the wheel is moving against the forward direction of the bike
- some people say that it gives a strange feel while turning because of the way the ratio front center to rear center changes while the bike is pushed into a curve
- others find that a rearward wheel path gives positive flat cornering characteristics
- DW wrote that wheel path is not so important
- high pivot when combined with an idler will allow to dissociate the AS from the AR

Discuss !

 

[hmcleay]

Really cool bit of programming you did there @hmcleay !
The sound is really low in your video, by the way.

Thanks! I tried to bump up the volume afterwards, but it ended up with a lot of that 'digital' sound in the background, which was quite annoying, so I left it as it was. But I'll try and get it sorted for the next one.

I am quite curious about wheel path. Rearward wheel path seems to be increasingly popular these days.
From memory:
- a rearward wheel path should give more time for the shock to absorb an impact
- a rearward wheel path should slow you down more on the rebound stroke since the wheel is moving against the forward direction of the bike
- some people say that it gives a strange feel while turning because of the way the ratio front center to rear center changes while the bike is pushed into a curve
- others find that a rearward wheel path gives positive flat cornering characteristics
- DW wrote that wheel path is not so important
- high pivot when combined with an idler will allow to dissociate the AS from the AR

Discuss !

Great questions/comments!
A more rearward axle path will increase what I call the "roll up distance". Basically that's the horizontal distance the bike has to travel between the wheel first hitting an obstacle, and the wheel getting up onto the obstacle. The longer this distance, the slower the suspension needs to move to absorb a given bump at a given riding speed.
It's basically the same effect as going to a bigger wheel, except much more effective. For example, with the same axle path, a 29" wheel has about 2.5% more roll-up distance than a 27.5" wheel. The trouble is, to design a suspension system around a bigger wheel, and maintain the same drive ratio (smaller front chainring) and BB height and AS characteristics, then the axle path needs to arc forwards more, which diminishes the attempted gains of having a bigger wheel. 2.5% improvement becomes more like 0.7% improvement.
But more than that, a rearward axle path helps to separate rider inputs from terrain inputs. Basically, the suspension feels stiffer for vertical rider inputs like pumping/jumping/cornering, and softer for terrain inputs like 100mm high square edge hits.
I'm not sure the rebound stroke can slow you down, it's not like the wheel is pushing against anything significant when it's rebounding. If it doesn't rebound fast enough then it just lifts off the ground. Remember that we have a 'rearward' axle path on the front of the bike too. If you imagine a fork at 90deg HA, going over a sinusoidal terrain at a fixed riding speed, the compression and rebound shaft speeds would be equal in magnitude. But once you slacken the head angle, then the magnitude of the fork's compression speeds will reduce, and rebound speeds will need to increase to keep the wheel in contact with the terrain, otherwise it'll lift off the ground.
As far as 'feeling' goes, I put that in the same bucket as geometry. It's very much a pseudo-science; people are comfortable riding what they are used to.
With a conventional drivetrain, subtle changes in axle path will have significant effect on the AS curve, which is far more noticeable than the change in bump absorption characteristics due to the different axle path. So maybe DW meant that it's not worth trying to tune the axle path for improved bump absorption, because the AS curve will get out of hand before you make any worthwhile improvement in bump absorption. If you want to make a significant change to the axle path, for improved bump absorption, then you'll definitely need an idler pulley of some description to deal with the AS.
You don't necessarily need an idler to independently tune AS and AR. In the next episode on 4bars (part 1) I'll show how having a virtual pivot gives freedom to tune AS and AR independently.
Cheers,
Hugh

 

[Flo33]

Props for bringing up such a techy thread again, Hugh! And also props for the software and videos.