[fluider]

How would that feel if chainline was changing its vector two times per 360° cycle (one change per one crankarm, isn't it) ? I'm more inclined to idea of using this to even out rider's uneven power output (as that is a problem) and have only one chainline (and not make other issues), which would need jackshaft style chaindrive with gearbox ofcourse .

 

[Transcend]

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.

Yup, and a bunch of big name XC guys who are on the podium every week are running them. I believe most of the specialized factory XC squad are running them.

 

[TrueScotsman]

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

Ha, not great- it's been a while since school. To put things into context- I trained as an Architect, we were taught to come up with unconventional ideas and then pass them to an Engineer to make them work!!!!

 

[TrueScotsman]

Right- I've been doing a little research into pedaling forces. I found a study by R.R. Neptune and M.L.Hull that measured the maximum vertical force angle at the cranks. Here's the results;

If you concentrate on the centre graph you will see that the maximum vertical force occurs at around 150* (using the "subjects" data). Combine this to previous graphs showing maximum chain force to take place around 100* (when seated) to around 135* (when standing) and you will see that Socket's estimation of the cycles being 30* out of sync isn't far wrong! Well done that man!

So, if you were to make a chainring that helped increase Anti-Squat to counteract the maximum vertical load that causes bobbing, it would look a little like this;

 

[Bikael Molton]

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.

agree. I used RR for a couple years and never had the problem he mentioned. Never rode that bike on asphalt once.

 

[no skid marks]

So am I right in saying you're opting to swap anti Squat or pedal kick back for a similar feeling(non smooth pedal stroke feel)when pedaling?

 

[DirtyMike]

Ok so check it out...... the eliptical rings, biopace, rotorrings etc..............

Ran them years ago on the road bike

Ran them years ago on the XC bike

Ran them years ago on the DH bike.

I will keep it simple, the idea is great, but they can be a super pain in the ass. First part is getting a front mech to work right... Needs a really good fine tune, but once its set, your good.


There is Merrit behind the idea of making more effeicient power, Problem though..... its only really effective when in the saddle. I ran these on my old madone.... I loved how much faster I could keep my pace, and my cadence.... BUT I never left the saddle, in fact because out of the saddle felt so weird, it started making me stay in the saddle during a climb... found myself downshifting instead of powering up the climb.

So I tried them on a hardtail xc bike.... felt exactly the same as the road bike, same effect, found myself wanting to stay in the saddle and never come up for a burst of power.....

Full sqush xc bike... this is were it started to get weird... again out of the saddle just felt weird when peddaling, in the saddle however felt ok, but was just off with the bit of bounce that it did have, I cannot say I gained any benefit of having them on the fullsquish of those days< no anti squat, no anti bob shock, not the greatest geometry>...

Now for the DH bike........... Made three runs at big bear, went home, and took the eliptical off. Pissed me off to hi hell, I was thinking the same thing as this thread was... using to eliminate pedal induced bob.... Welll...... it ddidnt reduce anything, felt almost as if it actually made it worse, gave an interesting pedal feedback when exploding out of corners, so much that it was hard to pedal out of corners.

My biggest problem I can say with the eliptical rings for DH........ the ever changing body position needed. When the numbers are exact it works great... but with DH they are never exact, angles of attack on an obstacle, steepness, speed, body position, bicycle geometry.... all this has to be taken into account.


Not trying to slam, not trying to say this is all ever exacting info....... this is just what I noticed when runing these types of rings on different bikes.......

also not tryuing to say the OP is crazy with the idea...... its an idea that will work great on paper... personally I am curious how it would feel to run one that was tuned for my bike and myself..... If you have an idea, chase it... tahts how we have seen some of our biggest innovations in the bicycle industry....... sometimes with ideas such as this, you just gotta nutt up and go for it.

 

[DirtyMike]

Yup, and a bunch of big name XC guys who are on the podium every week are running them. I believe most of the specialized factory XC squad are running them.

This is true..... also on bikes with suspensions and shocks that were first designed to eliminate pedal bob to begin with...... I can absolutly see it working well on what the top pros are riding with Speccy

 

[TrueScotsman]

So am I right in saying you're opting to swap anti Squat or pedal kick back for a similar feeling(non smooth pedal stroke feel)when pedaling?

No, quite the opposite;
I am opting to increase anti-Squat (momentarily) for a smoother pedal stroke feel when pedaling.

(At least that is what I am asking if it is theoretically possible?)

EDIT- I know it looks even crazier, but could this sketch be the ultimate! (Please no flaming- I know it looks bonkers!!!!) :

Man, I'm gonna look crazier than that Lahar guy!!!!

 

[TrueScotsman]

@DirtyMike-
Cheers for adding your real-world findings- I too have tried biopace rings but that was years ago only on a hardtail.
BTW- what type of rings did you use on the DH bike (biopace, rotor, etc.) as I believe that their elliptical-ness and orientation differ greatly?

The fact that it was so terrible on the DH bike kinda reassures me- could it be that they were 90* out of whack? If it made that much difference in the wrong way, it shows that it there is potential if done in the right way. A bigger problem could have been if you noticed NO difference! If that makes any sense?

Man, I am going to get the name of a serious fruitcake on here! Believe me, I do know how crazy all this sounds

 

[no skid marks]

No, quite the opposite;
I am opting to increase anti-Squat (momentarily) for a smoother pedal stroke feel when pedaling. Sorry, I wasn't really thinking(again) in saying "anti squat". I should have said pedal kickback when not pedaling, for pedal kickback when pedaling
I think for most bikes your idea has merit, especially if used moderately, so the gain is minimal, and same with any flaws.
Depending on crank location, pedal kickback will be increased or decreased from the extra leverage of the high spots won't it? Probably a negligible amount.
Man, I'm gonna look crazier than that Lahar guy!!!!

Hmmm, seems to be lots of bikes made from carbon now days, super slack head angles, press fit BBs, and the brilliant Zerode. I wouldn't call Lahars non conformed ideas in regards to frame design crazy in any way. The Lahars design is still better than 99% of current bikes, doesn't really need any bio pace concept. It's only flaws(design wise)IMO were chainstay length and BB height, and obvious personal dislikes like any bike suffers from like geo, top tube length, floating or non floating brake, etc.

 

[TrueScotsman]

Hmmm, seems to be lots of bikes made from carbon now days, super slack head angles, press fit BBs, and the brilliant Zerode. I wouldn't call Lahars non conformed ideas in regards to frame design crazy in any way. The Lahars design is still better than 99% of current bikes, doesn't really need any bio pace concept. It's only flaws(design wise)IMO were chainstay length and BB height, and obvious personal dislikes like any bike suffers from like geo, top tube length, floating or non floating brake, etc.

Aye, sorry- was't meaning to bash the Lahar bike- it was ahead of it's time in so many respects- revolutionary you may say. It was just the guy came on to RM and got the reputation (rightly or wrongly) for being nuttier than squirrel poo!
This was my point- no matter how good certain ideas you have are, you can still get a name for being cuckoo!

Edit- As regards pedal feedback, I think the above chainring would actually slightly decrease pedal feedback at the coasting positions (3 and 9 o'clock) due to the larger effective radius.

 

[Steve M]

Quick question... how are you gonna get your chainguide to keep the chain on?

 

[TrueScotsman]

Good question!- Bear in mind this is currently just a theoretical chainring so doesn't have to deal with real life problems like chains coming off!

You could use a sandwich-type chain device (think old-school MRP) with 2 round plates and rollers, or a wonky chainring integrated with a round bash guard. Not sure if these would work but there's only one way to find out.......

 

[w00dy]

Good question!- Bear in mind this is currently just a theoretical chainring so doesn't have to deal with real life problems like chains coming off!

You could use a sandwich-type chain device (think old-school MRP) with 2 round plates and rollers, or a wonky chainring integrated with a round bash guard. Not sure if these would work but there's only one way to find out.......

That was my assumption. Seems like it worked for Barel.

 

[fluider]

Quick question... how are you gonna get your chainguide to keep the chain on?

Another point for jack-shaft style gearbox bike ala Lahar

 

[Steve M]

Another point for jack-shaft style gearbox bike ala Lahar

Won't work unless the elliptical gear is on the secondary drive output, in which case you still need some kind of chainguide anyway...

 

[TrueScotsman]

Also, if it was on the secondary drive output the periods of higher anti-squat wouldn't sync with the maximum vertical force.

I think that this idea is limited to standard-drive suspension designs unfortunately.

 

[fluider]

You, Socket and TrueScotsman, would use elliptical chainring to gain variable antisquat curve via variable chainline.
While I would use it just to even out rider's power output coming in to gearbox.

Maybe I'm just over-estimating it, but I think a variable AS curve would bring a kind of strange traction behaviour or feel. I'd prefer mechanical/construction simplicity of elliptical chainring on primary drive and 'one' behaviour of AS curve with smooth engine output.

 

[TrueScotsman]

You, Socket and TrueScotsman, would use elliptical chainring to gain variable antisquat curve via variable chainline.
While I would use it just to even out rider's power output coming in to gearbox.

Maybe I'm just over-estimating it, but I think a variable AS curve would bring a kind of strange traction behaviour or feel. I'd prefer mechanical/construction simplicity of elliptical chainring on primary drive and 'one' behaviour of AS curve with smooth engine output.

Thanks fluider, this is the kind of feedback I was looking for. I don't know how the variable AS curve would affect traction. As for feel, the idea is to level out the bobbing, therefore it would actually "feel" more even than a static AS curve.

 

[fluider]

Yes, to even out bobbing which has been a topic of this thread.
So then I question why to try to even out vertical acceleration using a horizontal acceleration? The resulting variable chainline shall be carefully studied because along with desired "anti-bobbing" effect you can get too high AS, as well.
I must admit that I am fueled by desire of getting rid of derailuers, so if anti-bobbing requires saying them Pa Pa, I'm not going to cry . My estimation is that with variable chainline (and AS as well) you won't get 100% anti-bobbing without effecting actual anti-squat. But that's just my estimate! Anyway, I like the images you posted here, will study them little bit ...

 

[TrueScotsman]

Yes, to even out bobbing which has been a topic of this thread.
So then I question why to try to even out vertical acceleration using a horizontal acceleration? The resulting variable chainline shall be carefully studied because along with desired "anti-bobbing" effect you can get too high AS, as well.
I must admit that I am fueled by desire of getting rid of derailuers, so if anti-bobbing requires saying them Pa Pa, I'm not going to cry . My estimation is that with variable chainline (and AS as well) you won't get 100% anti-bobbing without effecting actual anti-squat. But that's just my estimate! Anyway, I like the images you posted here, will study them little bit ...

Anti-Squat isn't a horizontal acceleration- it is a reaction to a horizontal acceleration. I think that this same reaction could resist the vertical forces.

I am proposing that the "too high AS" isn't too high as it is the exact cancelling force for the vertical accelerations. Therefore the behaviour of the bike will be the same as the rest of the pedal stroke. i.e. you only get a higher AS when you need it. (when it would normally "bob")

I must also admit that I too am fueled by a desire to get rid of derailleurs, but I had this idea for standard drivetrains and thought it merited investigation.

Anyway, I took a short spin on the DH bike tonight on flat tarmac with the specific task of trying to "feel" what it was like to pedal/sprint. Although you may not be aware of it DH bikes (even with a high degree of AS and LSC damping) do indeed bob a fair bit. Most of us are used to it and reckon that is just "how it is" and get on with it. Good for them- unfortunately my brain is overactive and my legs are lazy meaning that i look for a mechanical way out!
Also I noted the maximum "bob" seemed to occur like the graph said (just before the lowest pedal point at around 150*.

I would welcome others to give me their feedback on doing the same thing on different bikes- sprint hard and look down at the shock (don't crash!)- does it "bob" and if so what angle is your pedal at?
Cheers.

 

[fluider]

I've put three components of your last lab-data together, to have both right and left crankarm in one plot. Now I'm gonna sound as a mistake-seeker or doubter but ... I think you should look at both horizontal and vertical force plots simultaneously and have in mind actual angle value, because sinus of an angle can make even the biggest force negligible. sin(150°) = 0.5.
So, sin(150°).600N = 300N < sin(90°).400N = 400N. And better is to look at torque plot in 150° angle, it's not far from 0 N.m.

I think your feeling that maximum bob occured in 150° was caused by some 'inertia' of human perception. I'm trying to say that it was the maximum torque range <70°,130°> when your suspension started to bob and what you was perceiving was a bobbing already in action. And that is what IMO those graphs show, an asynchronity of human power output and resulting torque (and torque is what transmitts work) so by synchronising these two a more effective power transimision could be achieved.

In this image, the red line shall describe sum of torque values from both right and left crankarm. Green line depicts a torque curve that anti-bobby chainring could achieve for me .

Huh, does it make sense ?

 

[TrueScotsman]

I think I see what you're saying. With an elliptical chainring you could modulate the crank torque to get a relatively even output. Therefore reducing the bobbing due to constant acceleration/deceleration that a bike usually undergoes. Yes?

The Retrobike forum had a interesting discussion about eliptical/oval/non-round chainrings here;http://www.retrobike.co.uk/forum/viewtopic.php?t=47932 Worth a look.

I will have to back to you about the rest after I investigate it further!

 

[Vrock]

This is the second thing you need to do your math... COM position while pedaling (Seated). 40.484 minimum (120º) and 40.672 maximum (210º). The weight of the whole group is 85kg.

 

[TrueScotsman]

fluider- okay, I have looked at the graphs again. I don't see how the horizontal force graph is relevant for my purposes as I am talking about lessening the vertical forces (the second graph) with increased A-S at that point (requiring the third graph).
So I took a leaf out of your book and plotted some figures; firstly the vertical force graph;

As you can see the green line is roughly a combined sum of both the red force lines (each pedal). Interestingly, there is generally still weight forcing down on the upward stroke. You can see there is a definite peak (trough?) at the maximum downward force (around 700N) at around 150*. It is this that I am trying to cancel out.

Next the torque output graph;

On this graph I have plotted the torque levels at 150* (lime green lines). From the graph (by no means 100% accurate) it looks like the torque is around 33Nm, plus the negative torque on the upward stroke (confirming the previous note of weight on the upward pedal) of around 5Nm. Thus the resultant torque is 28Nm.

I therefore have to calculate how much the anti-squat would have to increase by (assuming that the theoretical bike has 100% anti-squat at sag) to make the pedal torque of 28Nm cancel out the downward force of 700N. If it is at all possible!

Also, I would have to decipher what mean forces and averages were used to calculate the bike's A-S figures in the first place!

Vrock- I am not sure what you mean by your figures, but this could just be my fried brain not seeing clearly!

 

[TrueScotsman]

Okay, looking at the figures again (specifically) the Maximum Vertical Force one and overlaying the figures it becomes clear that to have a chainring that counters bobbing you would have to take account of the entire graph (high and low spots). If you look at the graph you will see that these high and low points do not occur at 90* from each other. The minimum vertical forces occurs at around 25* and 205* and the maximum vertical downward forces occur at around 150* and 330* (these are crank angles for one leg rotation but takes account of both legs). See graph;

Because the max and min values are not 90* out of sync a true elliptical chainring would not be ideal. It would require a chainring like this; (bear in mind that this is not trying to even out pedalling forces but ONLY concerns varying A-S levels to cancel out vertical bobbing) ;

Please note- the size of the rises and dips on the chainring are not to any scale or the product of specific measurements, but do show the general shape required.

A true "anti-bobbing chainring" would have to take account of both vertical force AND uneven horizontal acceleration force.

 

[Vrock]

If you have 100% Anti-squat, it doesn't matter that the horizontal aceleration is uneven. Yo have to calculate first the vertical aceleration due to the movement of the legs and then, try to compensate it with the shape of the chainring...

 

[fluider]

Vrock, what software is it on your last screenshot? Jeronimo? I can't read it.

 

[TrueScotsman]

If you have 100% Anti-squat, it doesn't matter that the horizontal aceleration is uneven. Yo have to calculate first the vertical aceleration due to the movement of the legs and then, try to compensate it with the shape of the chainring...

Yes, but if you modified a chainring to vary the Anti-Squat it may not be 100% at the point you need it to deal with horizontal accelerations. Therefore the best chainring would be a compromise between getting 100% A-S when needed (for horizontal accelerations) and reducing/increasing the A-S to compensate for the vertical forces.
Or....... you modify the chainring to also deal with the uneven power output, therefore lessening the horizontal accelerations.

(I think!)

 

[Vrock]

Yes but you have to calculate first the vertical acereration and then you can play with the chainrings... you are going to need 100% all the time except at Xº where you are going to need X% A-S. You don't need to worry about uneven horizontal aceleration because the AS is going to be uneven too, if the squat force is 50N the AS is going to be 50, if it is 100N the AS is going to be 100N too... It's 100% all the time.

Fluider, the program is Working Model 2D.

 

[TrueScotsman]

Yes but you have to calculate first the vertical acereration and then you can play with the chainrings... you are going to need 100% all the time except at Xº where you are going to need X% A-S. You don't need to worry about uneven horizontal aceleration because the AS is going to be uneven too, if the squat force is 50N the AS is going to be 50, if it is 100N the AS is going to be 100N too... It's 100% all the time.

Fluider, the program is Working Model 2D.

Cheers Vrock, you are making me think today! I see what your saying about the chainring- the maximum radius of the chainring must give around 100% Anti-Squat and the other (lower effective radius) parts must give X% above 100% Anti-Squat.

Yes, you're right I will have to work out the vertical accelerations.

However, one part I don't agree with is this- I think you are mixing up vertical and horizontal forces. Remember the horizontal and vertical forces are not in sync with each other- they are around 15*to 50* out of sync (depending on whether you are seated or standing). Therefore the point of highest acceleration may not have the highest A-S level. This is why the chainring would have to be a slight compromise between both vertical and horizontal forces.It would still be a lot better than a round ring!
I will do another doodle to explain myself better!

EDIT- Here's the doodle;

Note this shows the orientation only of what I call the "power ellipses", i.e. the orientation of an ellipse chainring to even out horizontal accelerations. These are laid over the Anti-bob chainring to show how "out-of-sync" they are- around 15* for standing pedalling and around 50* for seated pedalliing.

2ND EDIT- Sorry I have the orientations wrong! D'Oh. I will re-draw and re-post

 

[Vrock]

Forget about the standing position for a moment, it's too complicated. The seated position is much easier to calculate, and that's how most of the people pedal.

 

[xy9ine]

based on the previous pages of discussion, i believe we're ready to go into production with these. we've got fancy graphs and lots of jargon, so the marketing end is pretty much taken care of. ts's drawings look pretty wild, so you've got that essential visual differentiation to justify greater margins nailed. the "ABC" acronym is perfect as well. lets do this! riches will inevitably ensue!

 

[TrueScotsman]

Ha ha ha xy9ine!

My pictures may look pretty, but as I said the sizes aren't based on any real data. But then, when has that actually stopped anything from happening in the bike world!

ABC.......hmmm, has a nice "ring" to it!

 

[TrueScotsman]

BTW, here's a wee graph showing what I hope an anti-bob chainring would accomplish.

The blue line is a copy of the combined vertical force (in Newtons) seen in post#67 and the green line represents the theoretical variable anti-squat percentage an Anti-Bob Chainring would give. 100% Anti-Squat line is assumed from the red circle on the ABC drawing (representing a round chainring).
Notice that the resultant line (amber) still dips and rises a small amount. The size of these dips and rises are not yet figured out. My point to Vrock was that the point of maximun horizontal acceleration would still occur slightly in these dips/rises. How significant this would be is not yet clear.

I am in the process of gaining data from various sources to discern how viable in real life an ABC chainring would be.

EDIT- the red line should actually be 100+X% so that the graph never dips below 100%. oops!

 

[Steve M]

This is actually getting rad! You've got my email if you want to discuss any ideas...

 

[davec113]

BTW, here's a wee graph showing what I hope an anti-bob chainring would accomplish.

The blue line is a copy of the combined vertical force (in Newtons) seen in post#67 and the green line represents the theoretical variable anti-squat percentage an Anti-Bob Chainring would give. 100% Anti-Squat line is assumed from the red circle on the ABC drawing (representing a round chainring).
Notice that the resultant line (amber) still dips and rises a small amount. The size of these dips and rises are not yet figured out. My point to Vrock was that the point of maximun horizontal acceleration would still occur slightly in these dips/rises. How significant this would be is not yet clear.

I am in the process of gaining data from various sources to discern how viable in real life an ABC chainring would be.

EDIT- the red line should actually be 100+X% so that the graph never dips below 100%. oops!

So what are min/max values of the actual chainring? For example would it vary between a 34-38t ring at it's min/max diameters?

Also, it seems like the chainring diameter would need to get smaller to produce more anti-squat, wouldn't that be the opposite of what you'd want for maximum pedaling efficiency?... as you're standing and pedaling, the bike gets easier to pedal during the parts of the pedal stroke where you can produce the most power and then harder to pedal during the times when you are producing the least power?

 

[TrueScotsman]

Oops- sorry you are right! Well spotted davec113.
I showed the "power ellipses" diagram rotated wrong. Easy mistake to make, hell Shimano did it with their Biopace rings for years!!!! They should be turned around so the orientation is more like the Rotor/O'symetric/Egg rings.
I will amend the drawings and repost.

(Damm- I knew things were going too well!)

 

[Vrock]

This is the idea... Find out how much bob do you have (0.1''), and when does it happens (70º-160º). Then find out how much AS do you need and design the ring...