[TrueScotsman]

In the recent "Pedal Feedback" thread Socket explained that even 100% anti-squat can't take into account the vertical accelerations of your body mass. Therefore you still get "bob". In his estimation this peak vertical acceleration takes place around 30* out of sync with peak power output (if I understand him properly).
This lead me to suggesting (maybe crazily!) a biopace-type chain ring that is also 30* out of sync. This would increase the power output at this point but also, critically, increase the amount of anti-squat- counteracting the vertical downward acceleration. Net result- zero/less "bob". See sketch for details of how this works;

Remember, Fabian Barel did use an odd shape chainring at the Canberra Worlds because he had a knackered knee, so it could (theoretically) be done.

What do you reckon?- worth looking in to, or am I just crackers!

 

[xy9ine]

perhaps there's some merit to it; google search random snippet:

I did some testing about 10yrs ago with some singlepivot suspension bikes, and found that eggrings could eliminate bob, increase traction - really good.

We ran something on it in mbr, but I think everyone is so tied in to Shimano rings, it was never going to work. It's a shame. In terms of eliminating suspension actuation from pedaling forces it's a cool way of doing it - but people seem happier running over complex rear shocks that don't work on small bumps.

 

[Zutroy]

Part of the problem could be the Shimano Biopace rings were shaped wrong...........

 

[TrueScotsman]

perhaps there's some merit to it; google search random snippet:

I did some testing about 10yrs ago with some singlepivot suspension bikes, and found that eggrings could eliminate bob, increase traction - really good.

We ran something on it in mbr, but I think everyone is so tied in to Shimano rings, it was never going to work. It's a shame. In terms of eliminating suspension actuation from pedaling forces it's a cool way of doing it - but people seem happier running over complex rear shocks that don't work on small bumps.

xy9ine- where did you get that info from- i'd like to read more.

I thought it was revolutionary thinking on my part, but I guess there's no such thing as original thought!

 

[xy9ine]

xy9ine- where did you get that info from- i'd like to read more.

I thought it was revolutionary thinking on my part, but I guess there's no such thing as original thought!

that was a thread discussion re: egg chainrings. here's the site link; bunch of info, and a comment about reducing bob, but haven't found any detailed explanation yet. dig away:

http://www.highpath.co.uk/

 

[ChrisKring]

do you guys really have that much of a problem with pedal bob? I'm just asking because most modern bikes are pretty good in this respect. It's not like everyone is out riding GT RTS frames with Fox Vanilla shocks, or are you?

Most of the race courses I have ridden only require pedaling in relatively smooth areas. Pedalig circles rather than smashing the pedals takes care of most of the "problem" along with a properly tuned low speed compression circuit.

The problem with odd shaped chainrings is that you would preceive the difference. Another problem would be with the chainguide.

With regards to traction, are you really loosing traction under forward accelleration? If so, are you a track racer? What tires are you running?

Sorry for the sarcasm, it's just that i don't see this as a real life problem today and more of a problem that can only be seen in an equation. Braking and cornering preformance seem to be much more critical.

 

[iRider]

Those Rotor rings are made to eliminate the dead spot, but because they can be turned in whichever angle you fancy they are perfect for your experiment I reckon.

 

[davec113]

It'll work to a degree, but it will probably feel weird to have an elliptical chainring.

I agree that it's not really an issue, I don't care about efficiency that much I guess.

 

[dropmachine]

I think this is a case of people overthinking things. Its a suspension bike. ITs gonna bounce. End of story.

 

[TrueScotsman]

do you guys really have that much of a problem with pedal bob? I'm just asking because most modern bikes are pretty good in this respect. It's not like everyone is out riding GT RTS frames with Fox Vanilla shocks, or are you?

Most of the race courses I have ridden only require pedaling in relatively smooth areas. Pedalig circles rather than smashing the pedals takes care of most of the "problem" along with a properly tuned low speed compression circuit.

The problem with odd shaped chainrings is that you would preceive the difference. Another problem would be with the chainguide.

With regards to traction, are you really loosing traction under forward accelleration? If so, are you a track racer? What tires are you running?

Sorry for the sarcasm, it's just that i don't see this as a real life problem today and more of a problem that can only be seen in an equation. Braking and cornering preformance seem to be much more critical.

No, I don't have a great problem with "bobbing", but my DH bike hardly pedals like a hardtail. I like looking at the theory behind bike design and if you ain't aiming for 100% efficiency you ain't exploring all avenues, no matter how crazy ideas may sound.

Pedalling "circles" would indeed help but DH is a sport where a lot of times a quick mash on the pedals is all you can fit in. As for "properly tuned LSC"- this is a fable- if you are adding damping to counteract pedal bob, your suspension is not working at it's optimum at what it should be- soaking up bumps.

Read the "egg-rings" link that xy9ine posted- due to the nature of pedalling power curves and "dead-spots" an elliptical chainring may actually feel smoother (and give a smoother power output)

I agree that cornering and braking performance are very critical, BUT you have to remember that all these small advantages add up. And they are not to the detriment of braking or cornering- you can still concentrate on these areas independently. As for the sarcasm- don't worry, I can rise above it!

Surely, this forum board is the ideal place to bring up such discussions- those who wish to contribute can, and those that don't, don't. They can go on and discuss whatever takes their fancy, be it Sam Hill's pyjamas or the advantages of teflon-coated carbon rims.

 

[TWeerts]

Those Rotor rings are made to eliminate the dead spot, but because they can be turned in whichever angle you fancy they are perfect for your experiment I reckon.

this

http://www.rotorbikeusa.com/

 

[TrueScotsman]

Braking and cornering preformance seem to be much more critical.

Just thought of another advantage- if your bike doesn't "bob" as much under pedaling you can have a lower BB height. This would lead to much better cornering ability! Oh, and your COG would be lower- better for heavy braking!

Ha!

 

[StabprimoMonkee]

Eliptic rings are the most retarded idea ever. It is a classic example of over-engineering. I think that Barel himself is guilty of falling to the placebo (he is an mechanical engineer himself).

They might work in theory - but in practice, they are just goddamn awful for a million reasons.

Anyone riding one of those and hyping them up with a straight face looses half his credibility, in my eyes in an instant.

 

[marshalolson]

it's just that i don't see this as a real life problem today

The internet sounds like a hard place to ride

just saying.

 

[xy9ine]

^i'm curious as to your reasons (stabprimo). at a glance, the stated function kinda / maybe / possibly makes sense for applications such as dh when you're standing & mashing on flats rather than spinning spinning circles in the saddle. alas, i'm in no means an enginerd. i seek illumination and enlightenment!

 

[StabprimoMonkee]

@xy9ine: Whatever you gain on paper - you lose in awkward feeling ride, chainguide problems, etc.

Like ChrisKing said - thats what low speed compression is for. You will gain far more, by inflating your tires a bit, going tubeless, using lighter rims, etc...

As long as 95% of riders I see have their suspension tunes completely wrong - elliptic chainrings are snake oil. /thread

 

[TrueScotsman]

@xy9ine: Whatever you gain on paper - you lose in awkward feeling ride, chainguide problems, etc.

Like ChrisKing said - thats what low speed compression is for. You will gain far more, by inflating your tires a bit, going tubeless, using lighter rims, etc...

As long as 95% of riders I see have their suspension tunes completely wrong - elliptic chainrings are snake oil. /thread

Awkward feeling ride- as I said before this may not be true. They MAY reduce dead-spot feeling.

Chainguide problems- maybe, who knows?

LSC- Low speed compression damping is a compromise between anti-bob and low speed bump sensitivity.

Lastly i have to say I am not saying that eliptical/odd shaped chainrings ARE the answer- all I'm doing is putting the question out there. I am willing to be swayed by sensible argument either way.

 

[MarkDH]

I'm on the fence with this one, sell it to me. So if I run a weird chainring I get less pedal bob, so I can run less LSC, so my suspension can work better? The thing is, I don't run LSC to counteract pedal bob, I run it to keep the bike stable and stop it wallowing or diving. So if I have to run the LSC anyway, that also helps with the pedal bob (which tbh I never notice), how does the elliptical chainring help?

 

[TrueScotsman]

@MarkDH-Thing is, I don't think anyone would be be able to sell it to you yet. No-one is saying this IS the answer, I just started the thread to discuss pros and cons of my idea.

For instance, you make a good point- for you, LSC is not used to control bob, and you don't seem to notice any bob- maybe you should just stick to round rings.

Rather than answer the question now, let's discuss the +/- of the idea and then come to a conclusion.

For example, do any of the enginerds (xy9ine term!) have any data that supports/disproves the vertical acceleration of body mass being around 30* out of sync with the peak power output from the cranks?

 

[Commencal-guy]

Does chainring size effect the amount of pedal-feedback as well?

Lets say a single pivot bike with the main pivot location above the BB. Would there be less pedal-feedback from a 36T chainring to a 38T? same with cassette size. 11-21T verses a 11-25T

 

[TrueScotsman]

Does chainring size effect the amount of pedal-feedback as well?

Lets say a single pivot bike with the main pivot location above the BB. Would there be less pedal-feedback from a 36T chainring to a 38T? same with cassette size. 11-21T verses a 11-25T

Check out the pedal feedback thread;http://www.ridemonkey.com/forums/showthread.php?t=244131

We were discussing the different ways that pedal feedback is calculated. Using the Linkage program definition (maybe wrong?) pedal-feedback would reduce with a bigger chainring (same chain "pull" divided by a larger ring = less angle of rotation)

 

[davec113]

Does chainring size effect the amount of pedal-feedback as well?

Lets say a single pivot bike with the main pivot location above the BB. Would there be less pedal-feedback from a 36T chainring to a 38T? same with cassette size. 11-21T verses a 11-25T

More pedal feedback w/36 vs 38 for most current low single pivots like Session 88.

TrueScottsman, maybe you can calculate the difference in anti squat an elliptical chainring would produce, and if you think it'll actually be noticeable in real life then try it out and see...

 

[StabprimoMonkee]

@TrueScotsman: Ok /flaming: I have ridden bikes with elliptical rings and they felt retarded. As far as eliminating pedaling deadspot goes I'm sceptical too.

I'm with MarkDH on this one - I run Massive LSC on my bike - because of stability and general high energy g-out scenarios that are occurring while riding DH. A properly set DH bike for riding fast is has so much LSC that pedalling bob, down the hill mind you, is completely non existent.

Also proper pedaling technique with high cadence helps a lot.

my 0.02$

Edit: Also don't forget that elliptical chainring might amplify the problem in some scenarios (e.g.: berms with brake bumps) if ridden "out of phase".

 

[Commencal-guy]

So in theory wouldn't a longer chain help cope with the anti-squat feel since it has more links to tension before the derailleur pulley is acted upon, which effects the suspension rate? verses a shorter chain?


Sorry if this has been discussed before im just a terrible follower.

 

[TrueScotsman]

TrueScottsman, maybe you can calculate the difference in anti squat an elliptical chainring would produce, and if you think it'll actually be noticeable in real life then try it out and see...

yup, i'd like to try it out but I would need to know the magnitude of the vertical body mass acceleration and the angle of crank rotation that this happens at in order to work out the required increase in anti-squat. Unfortunately, this is outwith my capabilities!!!!

Anyone able to help?

I am also thinking that a true ellipse is not the ideal shape, but rather a double-flatspotted circle if that makes sense- more like the osymetric rings.

 

[TrueScotsman]

@TrueScotsman: Ok /flaming: I have ridden bikes with elliptical rings and they felt retarded. As far as eliminating pedaling deadspot goes I'm sceptical too.

I'm with MarkDH on this one - I run Massive LSC on my bike - because of stability and general high energy g-out scenarios that are occurring while riding DH. A properly set DH bike for riding fast is has so much LSC that pedalling bob, down the hill mind you, is completely non existent.

Also proper pedaling technique with high cadence helps a lot.

my 0.02$

Edit: Also don't forget that elliptical chainring might amplify the problem in some scenarios (e.g.: berms with brake bumps) if ridden "out of phase".

Cheers, that post seems more considered!

You're right about a proper pedaling technique and a higher cadence.

However, I don't think that they could be ridden "out of phase" as the higher Anti-Squat values would always happen at the times of vertical acceleration of body mass (i.e. angle of crank rotation of heaviest "mashing" on pedals)

@Commencal-guy; try posting in the Pedal feedback thread- you may get more relavent answers there.

 

[MinorThreat]

I just wonder how Bob is holding up under the pressure of all this anti-Bob sentiment.

 

[Zutroy]

@TrueScotsman: Ok /flaming: I have ridden bikes with elliptical rings and they felt retarded. As far as eliminating pedaling deadspot goes I'm sceptical too.

Who's rings did you ride? They all are a bit different from each other. I've had some time on the Rotor and Ossometrics ones on a road bike and really didn't feel much of a difference in them.

 

[wood booger]

It is hard enough getting a front derailleur to work w/ rotor rings (which are pretty mildly elliptical) on a road bike, I can't imagine a chain guide functioning correctly on a DH bike.

 

[w00dy]

Rather than answer the question now, let's discuss the +/- of the idea and then come to a conclusion.

For example, do any of the enginerds (xy9ine term!) have any data that supports/disproves the vertical acceleration of body mass being around 30* out of sync with the peak power output from the cranks?

I'll discuss this, but there is no chance a consensus will be reached. This IS the internet.

Not sure if the 30 deg. point is true, but it makes sense that a person will give an extra jounce at the end of the stroke to get their weight back up high for the next manly, knee-sploding pedal mash.

My first problem with the arguments against is the "proper pedaling technique" crap. Legs don't move in perfect circles, they're not designed to. They produce the most power pushing away from the body in a straight line. If you're taking round chainrings as a given, then technique will help greatly. This idea doesn't follow the same rules, so drop it.

Benefits:
1. Elliptical chainrings can eek more power out of a person, at least as far as Rotor and Shimano have found. Both seem to know what they're talking about, so I'll take them at their word.

2. Could make bike lighter, more reliable. It modifies an existing part without adding complexity or extra mass. It can even take the place of some extra linkages or shock components.

3. no reason you can't still use a round guide. Just so long as the guide plates are large enough to handle the taller points.


Caveats:
1. Rings may have to be tuned to each bike. The pivot location would affect this a great deal.

2. Ugly. I've always thought this is what killed biopace. Could be covered up by chainguide.

3. Will feel funny to established riders. This DID kill rapid rise.

 

[marshalolson]

I
3. Will feel funny to established riders. This DID kill rapid rise.

rapid rise died because it didn't work as well. the derailleur spring tension was trying to pull the chain down the cassette whenever the chain slackened, and made for constant skipping while peddling on anything but perfect asphalt.

 

[TrueScotsman]

Cheers w00dy- some good points there! Especially about pedaling force not being even- this is particularly the case with flat pedals (no upstroke). Also, you are right- rings may have to be tuned/tailored to particular bike designs (with varying amounts of anti-squat)

After a few minutes searching on the internet I have found a few graphs that seem to indicate that maximum power is reached just after 90* (around 100*).

If we were to assume that the maximum vertical body mass acceleration IS at 30* less rotation I reckon that an anti-bob ring should have a profile like this;

Remember- this is not a chainring to even out pedalling force (although this could perhaps be integrated) this is just to try and counteract vertical "bobbing" force with increased anti-squat.

 

[no skid marks]

Don't you want it to rob max power, so the flat's where the power is?

 

[w00dy]

rapid rise died because it didn't work as well. the derailleur spring tension was trying to pull the chain down the cassette whenever the chain slackened, and made for constant skipping while peddling on anything but perfect asphalt.

I've used nothing but rapid rise since the late 90s. Works like a charm for me.
Replace your derailleur cable housing once in a while ya cheap bastard.

 

[TrueScotsman]

Don't you want it to rob max power, so the flat's where the power is?

No- that's my point- The "flat spots" are not there to even out power output but to increase the Anti-Squat response at the point of maximum vertical acceleration (bobbing due to mashing the pedals).

100% Anti-Squat would take care of the rearward mass transfer due to horizontal acceleration (most prevalent where the max "power spike" is).
But IF the max vertical acceleration of body mass does indeed take place 30* out of sync, we could do with >100% Anti-Squat (suspension extension force) to counteract this.

But, as I have said, I am willing to be proved wrong on this!

 

[I.van]

No- that's my point- The "flat spots" are not there to even out power output but to increase the Anti-Squat response at the point of maximum vertical acceleration (bobbing due to mashing the pedals).

100% Anti-Squat would take care of the rearward mass transfer due to horizontal acceleration (most prevalent where the max "power spike" is).
But IF the max vertical acceleration of body mass does indeed take place 30* out of sync, we could do with >100% Anti-Squat (suspension extension force) to counteract this.

But, as I have said, I am willing to be proved wrong on this!

Why don't you just engineer more than 100% anti-squat into the suspension design, rather than ****ing with the rings? try 130%.

 

[TrueScotsman]

Why don't you just engineer more than 100% anti-squat into the suspension design, rather than ****ing with the rings? try 130%.

Some suspension designers do this, but it will still suffer from some bobbing IF (as I understand) the maximum vertical and horizontal accelerations ARE out of sync. i.e. the suspension would extend a small bit more when under the horizontal acceleration "power-spike".

What I am looking at (in theory only, mind you) is whether it is possible to change the Anti-Squat response to be in sync with these differing horizontal and vertical accelerations.

 

[Steve M]

Cheers w00dy- some good points there! Especially about pedaling force not being even- this is particularly the case with flat pedals (no upstroke). Also, you are right- rings may have to be tuned/tailored to particular bike designs (with varying amounts of anti-squat)

After a few minutes searching on the internet I have found a few graphs that seem to indicate that maximum power is reached just after 90* (around 100*).

If we were to assume that the maximum vertical body mass acceleration IS at 30* less rotation...

Couple of things here:
1. I basically made up the value of 30 degrees, it's my estimation of what the phase shift is, not based on any measurements whatsoever. It could be 10, it could be 80, but it's almost certainly not going to be 0 or 90.
2. Something is up with that graph - what "force" exactly is being measured? I'm guessing it's pedal load because otherwise the period should be 180 degrees rather than 360 - unless they're only pedalling with one pedal or something whack. If the force value is chain tension (and I suspect it is because the force at zero matches up with the force at 360 but not at 0), then the values on the X axis are incorrect, and should be half of what they say - assuming again that we're measuring actual crank ANGLE rather than using degrees as a means to divide up a "cycle", which is half a revolution of the cranks.

If it is chain tension, however (and assuming it's correct), you could roughly derive (probably almost 100% accurate at 90 degrees but almost 0% accurate at 0 and 180 degrees) pretty much all the info you need from that curve alone. Calculate chain tension force as a factor of horizontal moment arm (ie sine of the crank angle, assuming 0 and 180 are vertical), work out how much vertical force you need at any given crank angle to achieve the chain force you're seeing in that graph, then plot it across the same axes and you'll be able to see your phase shift.

 

[TrueScotsman]

Couple of things here:
1. I basically made up the value of 30 degrees, it's my estimation of what the phase shift is, not based on any measurements whatsoever. It could be 10, it could be 80, but it's almost certainly not going to be 0 or 90.
2. Something is up with that graph - what "force" exactly is being measured? I'm guessing it's pedal load because otherwise the period should be 180 degrees rather than 360 - unless they're only pedalling with one pedal or something whack. If the force value is chain tension (and I suspect it is because the force at zero matches up with the force at 360 but not at 0), then the values on the X axis are incorrect, and should be half of what they say - assuming again that we're measuring actual crank ANGLE rather than using degrees as a means to divide up a "cycle", which is half a revolution of the cranks.

If it is chain tension, however (and assuming it's correct), you could roughly derive (probably almost 100% accurate at 90 degrees but almost 0% accurate at 0 and 180 degrees) pretty much all the info you need from that curve alone. Calculate chain tension force as a factor of horizontal moment arm (ie sine of the crank angle, assuming 0 and 180 are vertical), work out how much vertical force you need at any given crank angle to achieve the chain force you're seeing in that graph, then plot it across the same axes and you'll be able to see your phase shift.

Thanks for the input Socket- educational as always.

I have done some more reading/scanning (while i'm meant to be working!) and come across more data. These pedaling efficiency tests are maybe easier to read as they take account of both legs and are arranged circularly (although note rotation direction is anti-clockwise) :

Interestingly, however, is what happens when you stand to pedal (The case in the majority of DH riding/racing). The peak power output is rotated around by quite a degree. See graph;

This would mean that the vertical and horizontal accelerations are even further out of sync. Maybe this would mean different DH (standing) and XC/AM (seated) rings!

 

[Steve M]

Yep those graphs are far more useful... so how's your calculus?

edit: try get hold of the raw data behind those graphs, it's far more useful than the actual graph itself unless you're willing to spend the time manually regenerating some approximation of the data.