You don't need to be an 'investor' to invest in Singletrack: 6 days left: 95% of target - Find out more
it might help to think of it like this: the sum of the force on the saddle plus that on the pedals and the bars must always equal your body weight. this can't change. you can move it around between the 5 points, but it's not going to change, bar lunging about. it's this downward force that is equal to that upward force from the tyres, passed though the shocks of the suspension, sagged.
sometimes you pull up hardon the bars to get more pedal downforce, and this causes extension of the forks. then you can get a bit of "rocking horse" as you pull up, accelerate forwards, squat the rear suspension a bit, etc etc.
ok I have drawn this out
i have assumed the swinging arm is horizontal
and the other assumption is the chain pulls directly to the rear axle the chain meets the rear axle at 10.5 deg
so we have a force of 200kg ( from MTG ) at 10.5deg split into a horizontal and virtical component
this give 196.54 kg of compressive force in the SA
and 29.04 kg of virtical upward force which is 29.04 % of the riders body weight,, this is not in my opinion a small insignificant number,,
you can push it further than that can't you ade? assuming a 400lb in spring and a 3:1 leverage ratio?
If the riders weight gives 1/3 sag and this force is 1/3 of the riders weight then travel thats 1/9 of the travel (assuming its a constant spring rate). So just over 11 mm on a 100mm travel bike or 22mm on 200mm travel bike
As I said I can see it move leaning against the shed....
This is easily the worst thread I've ever read.
I think I'm going to cry.
Thread hijack: sorry: bullheart YGM
This is easily the best thread I've ever read.
I've trawled through the whole lot and there is a lot of guff but also a few people who I believe have some idea of what is going on. I love a good debate about suspension as it makes me question what I’ve tried to work out myself.
So, I agree with some points and disagree with others. There are too many points to quote and comment on so I'll just write out my thoughts here as if I hadn’t have read the previous posts.
Stating the obvious:
No one suspension design is the best, despite what manufacturers will tell you. Having said that, there are some good ones and some complete dogs.
Suspension design is a compromise and it is up to the designer to decide where those compromises should be based on what the bike should be used for and the target customer.
Some marketing has a scientific basis, most is complete and utter guff.
People misuse and misunderstand the terminology leading to all sorts of confusion. My biggest gripe is use of the term “brake jack”.
Firstly, antisquat. This is the ability of the suspension system to resist squat (compression) of the suspension resulting from mass transfer due to acceleration. When I say "suspension system" I mean the linkages (arrangement of pivots) and effect of the force in the chain line. It is possible for the linkage to have inherent antisquat properties on its own without the need to utilise the chain force (more on that later). The definition of antisquat properties are as follows (this has probably been covered, but hopefully this will make it black and white for anyone who still isn’t sure):
[b]100% antisquat =[/b] the arrangement of the suspension system exactly counteracts the compression of the suspension resulting from mass transfer due to acceleration. The suspension is therefore balanced and neither compresses or extends.
[b]>100% antisquat =[/b] the suspension system has more antisquat ability than it needs to fully counteract the mass transfer effects so the suspension will extend.
[b]<100% antisquat =[/b] the suspension system does not have enough antisquat ability to fully counteract the mass transfer effects so the suspension will still compress. The more antisquat ability, the less compression there will be under acceleration.
[b]0% antisquat =[/b] the suspension system has no antisquat capability at all and so the suspension will see the full effect of the mass transfer and compress.
[b]negative antisquat =[/b] otherwise known as "pro-squat". If the antisquat value is negative, the chain line force will act to forcibly compress the suspension and so act in addition to the compression resulting from mass transfer. This is not a desirable behaviour for a suspension bike for obvious reasons. Despite this, there are still some designs that have pro-squat (i.e. concentric BB/pivot bikes).
It should be noted that these anti/pro squat percentage values vary with suspension travel on all bikes so it is very difficult (although I believe not impossible) to design a bike to have 100% antisquat throughout the whole travel range.
So with that in mind, I shall now move on to this ongoing issue with chain line angle (I will be using the single pivot example for simplicity). I’m fairly confident but not 100% sure on this but here I go anyway .
Ade is almost correct when he describes the pivot above and below the chain line extending and compressing the suspension respectively. The exceptions occur when the chain line is parallel to the line between the rear axle and the main pivot (not the line between the rear axle and the BB as TJ said earlier, I assume this was a typo TJ?) and also when the chain line lies between the point at which it is parallel and the point at which it passes through the main pivot. To better describe that second scenario, it’s when the chain line and swingarm line converge towards the FRONT of the bike.
To summarise:
[b]main pivot above chain line = [/b] chain force will always act to extend the suspension.
[b]main pivot on chain line = [/b] chain force will not extend or compress the suspension (ignoring the fact that as the suspension moves through its travel the chain plays off the top of the front chain ring and so moves in relation to the fixed pivot).
[b]chain line above the main pivot but below the point where it would be parallel to the swingarm line =[/b] chain force will act to extend the suspension.
[b]chain line above the main pivot and parallel to the swingarm line = [/b] chain force will not extend or compress the suspension (as long as the lines are parallel, this could only be an instantaneous point in the travel).
[b]chain line above the main pivot and above the point at which the chain line and swingarm lines are parallel = [/b]chain force will act to compress the suspension.
Now, there has been some talk of examples to describe this effect but you have to be a little bit careful as the fact that the chain is pulling on a spinning wheel makes it all difference. A moment of the force through the chain about the distance from the rear axle to where the chain leaves the rear sprocket CANNOT be transmitted into the swingarm, assuming you are ignoring the friction in the bearings. It is therefore not valid to compare this situation to pulling on the end of a rigid “L” shaped beam pivoted at one end with a piece of string.
I’m on a roll so I may as well continue with the example of a concentric BB/main pivot on a single pivot bike. Consider this simple set up with the swingarm horizontal (bike is at sag) and a typical gearing of 36 at the front and 22 at the rear. The chain line will obviously be angled down towards the back of the bike making the intersection of the chain line and the swingarm line some way behind the rear wheel. The line now drawn through this intersection and the rear tire contact patch will indicate pro-squat as defined earlier. (If people don’t understand what I’m talking about here, I’ll have to draw a diagram but those familiar with the antisquat calcs/diagrams should hopefully be on the same page of this long, fulfilling and interesting book). This goes hand in hand with my statement made above: [b]chain line above the main pivot and above the point at which the chain line and swingarm lines are parallel = [/b]chain force will act to compress the suspension.
Now consider the chain ring sizes are the other way around (lets be honest, unlikely unless you’re trying to do trials on a slopestyle bike). The chain line is now angled down towards the front of the bike making the intersection of the chain line and the swingarm line some way in front of the BB. Draw the line through this intersection point and the rear tire contact patch and it’ll indicate some percentage of antisquat. This goes nicely hand in hand with my statement above: [b]chain line above the main pivot but below the point where it would be parallel to the swingarm line =[/b] chain force will act to extend the suspension.
Finally lets consider the situation where front and rear chain rings are the same size resulting in the chain line being parallel to the swingarm. In this case the chain line and swingarm line never intersect. The line through the rear tire contact patch must therefore be drawn parallel to the chain line and swingarm line giving exactly 0% antisquat. This goes hand in hand with my statement above: [b]chain line above the main pivot and parallel to the swingarm line = [/b] chain force will not extend or compress the suspension (as long as the lines are parallel, this could only be an instantaneous point in the travel).
So in that last example, neither the arrangement of the pivot or the angle of the chain line give any antisquat ability. Consider the same system but with a higher pivot so the swingarm now slants downwards towards the rear wheel. The line through the rear tire contact patch will still run parallel to the chain line and swingarm line but it will indicate more than 0% antisquat. I believe (and I may be wrong) that this antisquat is due ONLY to the pivot position and chain force has no affect.
Now I suppose I better briefly touch on "rise" and that dreaded “brake jack” term. When you slow down using the front brake only you will obviously get fork dive but you will also get “rise” (extension) in the rear suspension resulting from that mass transfer due to acceleration. When you apply only the rear brake that mass transfer will still act to extend the suspension but, depending on the design, the braking force transferred into the swingarm through the caliper will either counteract that extension (antirise) or add to it (horrible “jacking” of the rear end, this is where the term “brake jack” is sometimes used). Antirise is calculated in the same way as antisquat but you just don’t need to take into account the chain line.
There is also another braking effect that some people describe as “brake jack” and that is the apparent stiffening of the suspension when applying the rear brake. This is the brake pads resisting the natural movement of the disk through the calliper as the suspension compresses. Single pivots are often worse for this whereas FSR and Trek’s ABP/DW’s split pivot are designed to reduce it as much as possible (with the added advantage of being able to fine tune the antirise properties of the suspension). It has been argued however that in reality this effect is minimal as it is only a problem if the tire is firmly and constantly in contact with the ground which is often not the case when traversing rough terrain at speed.
Anyway, although that’s a small 1700 word essay I just wrote, I hope people find it interesting and I would be keen to know whether people are in agreement or not. One last point, as Brant said, other things like geometry and change in leverage ratios/damper matching are often regarded as being more important/significant than all that stuff above. I think it’s useful to consider the antisquat stuff though, even if the design is not based around theory like the DW link. Remember that all these antisquat/antirise principles are based around a few highly variable assumptions (namely COG height and chain line angle) which the marketing chaps don’t like to tell you about. Geometry, leverage ratios etc. don’t need any assumptions, they are what they are.
I make no apologies for being a geek.
seriously, i just realised that i could have gone and learnt some new skills in the time it took me to read that - which would be more effective than any pivot based BS.
i have a vpp and have had SP and will be going single pivot next time for 5 reasons
1. maintenance
2. i would say vpp is better at different things than SP, but sp has more consistent feel, and mid stroke support.
3. maintenance (run vpp for 5 years and you will know)
4. I am not convinced the axel path makes anywhere near as much difference as a good well set up shock - ie nomad with rp23 probably not as good as alpine 160 ccdb
5. maintenance - seriously £100 every 12months on bearing replacement
agreed except it's fun to try and understand this stuff as long as you keep it in perspective.
photo tim that was a long read on a mobile. I'll try and follow it with my own diagrams later. I'm glad you agree about the difference that the rear axle makes to the forces on the swing arm.
ampthill, glad to see someone is still almost interested, I was beginning to think I had killed off the thread with my excessively long post, but perhaps that's a good thing 🙂 Seems like munkyboy enjoyed it at least!
Interested to hear what the other contributors opinions are too...
Cracking post Tim. You a mate of Joes?
Thanks Brant, does it make sense? Yes I know Joe, YGM...
I'm gonna need diagrams.
chain line above the main pivot but below the point where it would be parallel to the swingarm line = chain force will act to extend the suspension.
I really can't see how a chain under tension above the pivot point can pull the swinging arm downwards.
What am I missing ?
munkyboy - Member
seriously, i just realised that i could have gone and learnt some new skills in the time it took me to read that - which would be more effective than any pivot based BS.i have a vpp and have had SP and will be going single pivot next time for 5 reasons
1. maintenance
2. i would say vpp is better at different things than SP, but sp has more consistent feel, and mid stroke support.
3. maintenance (run vpp for 5 years and you will know)
4. I am not convinced the axel path makes anywhere near as much difference as a good well set up shock - ie nomad with rp23 probably not as good as alpine 160 ccdb
5. maintenance - seriously £100 every 12months on bearing replacement
Maintenance was, I believe a big issue on the older SC VPP bikes. I've had my Blur LT2 for 2.5 years now, and have done no more than:
1) use the (supplied) grease gun on the nipples of the lower bearings every couple of months (about 2 minutes each time)
2) Three times tweaked the top bearings with an allen key (also about 2 minutes each time)
The Nomad now has the same bearing set up so would expect the same lack of need for maintenance. Therefore cost/time to maintain should no longer be a reason for not choosing VPP
I've 2 Intense VPPs and a Turner Flux 4BL amongst my stable of bikes - the VPPs are 140 & 170 mm travel and work best in rocky/ Dartmoor/ Welsh trails. The 4 bar is a 100mm Polaris/ marathon type bike, and is still running on the original set of bushes over 5 years after I bought it. The VPPs are brilliant at climbing - helpful for this fat bas***d when dragging so much travel up the hills. My VPP bearings last reasonably well - 18m typically for a bottom set, twice that for the top ones. They used to last less long, but since I learned how to fit them, and order Enduro Max rather than the wheel bearings my LBS used to fit, all has been well. So imho it's all about horses for courses - if you want lots of travel, VPP is a great way to take it, the cost being some reliability, but at £15 per bike pa the cost of bearings is not exactly exorbitant – I spend far more on brake pads and lube!
well written Tim thats what this forum needs a well written reasoned description
This goes hand in hand with my statement above: chain line above the main pivot and parallel to the swingarm line = chain force will not extend or compress the suspension (as long as the lines are parallel, this could only be an instantaneous point in the travel).
this bit was very interesting
i feel it is important to look at the dynamics of the system on or around ride height ( rider on and suspension sagged) as this is the postion where most of the pedaling happens,, and the suspension spends most of it's time at this postion,,
on a side note it actualy speend 99 % of it's time topped out without a rider on it sitting in the garage or shop,, virtualy all the pictures you see of a bike have no riders on them and there is some mileage in making a bike look good in this postion,,
I really can't see how a chain under tension above the pivot point can pull the swinging arm downwards.
What am I missing ?
Graham, it is basically down to black magic and witch craft. If the chain was attached to a rigidly attached extension off the rear axle end of the swing arm (equal in length and position to the distance from the rear axle to the point where the chain plays off the rear sprocket), a moment would be induced at the rear end of the swing arm and reacted at the main pivot causing the swing arm to rise/suspension compress. In reality, the force pulls on the top of the sprocket, no moment can be induced into the swing arm but instead its reacted at the rear tyre contact patch.
In your mind, extend the chain line and swing arm line forwards to infinity so they cross...does that make it clearer to visualise?
this bit was very interestingi feel it is important to look at the dynamics of the system on or around ride height ( rider on and suspension sagged) as this is the postion where most of the pedaling happens,, and the suspension spends most of it's time at this postion,,
on a side note it actualy speend 99 % of it's time topped out without a rider on it sitting in the garage or shop,, virtualy all the pictures you see of a bike have no riders on them and there is some mileage in making a bike look good in this postion,,
Ade, I've been racking my brains about the parallel thing and I'm now very almost certain that what I have written is correct as it works out when you consider the concentric BB/pivot antisquat example.
What you say about designing for the the sag position is spot on. Lots of bikes are designed to have 100% antisquat at roughly 30% sag (so about 130-140% antisquat at zero travel). You can also obviously tweak change in leverage ratios and antirise characteristics to coincide with the sag point.
Phototim, I understand what you're saying, I just don't believe it.
I know I'm right even though everyone else is telling me I'm wrong.
I'm going to have to either admit I'm wrong or spend the rest of my life all bitter and twisted about it.
Somehow, I can vaguely remember hearing that a pulley is just a lever that never changes its angle, although I can't find the explanation anywhere on the web.
If that's the case, then a sprocket does act exactly like a lever extending from the rear wheel centre.
Hi Graham, as I said in my original post, I'm not 100% sure on it and welcome any reasoned argument to the contrary! Its where my understanding gets a bit ropey. I'll try and figure it out once and for all.
Just a quick note on that linkage program, don't rely on the supplied examples to be accurate. They are nearly all taken from side on photographs and so pivot positions will all be slightly out. It doesn't take much to give you completely inaccurate answers. In my opinion, the program is only really worth using if you are working from a set of drawings or a CAD model. I ended up writing an Excel spreadsheet that did the same thing but it was only specific to one type of linkage driven SP and I would have to completely re-write all the trig equations to adapt it to other designs. 😯
i have found linkage very good,especaily as my maths isnt strong
but as you say you have to be carefull with drawings,and who has drawn them
i do transfer from cad for my stuff as it gives x y cordinates for the main points on each drawing it certainly lets you move points and see changes very quickly
but with anything like this garbage in garbage out
ok have modified a bike on linkage so main pivot and bb are in the same place front and rear gears are both 36
and the anti squat graph is a dead straight line throughout the travel on 0%
Yarp, garbage in, garbage out!
ok have modified a bike on linkage so main pivot and bb are in the same place front and rear gears are both 36and the anti squat graph is a dead straight line throughout the travel on 0%
So the swingarm is horizontal yes? Now try it with a smaller front chain ring and it should show some percentage of antisquat. I believe that this indicates that only the chainline is causing this antisquat as you have already shown in the example above that the pivot position is not giving any antisquat ability...so with the chainline below the parallel position but above the pivot it MUST act to extend the suspension, surely?
SA is horiznotal ok change front ring to 34 so 36 rear 34 front
gives anti squat figures start 1t +8 and got to -8 at full bump
good job Tim
Graham, all is not lost for you, I think there is something wrong with my analysis. Certainly the concentric example I gave checks out but there is something wrong somewhere as I have come up with another example that contradicts some of the stuff I've said above 😕
This stuff is a nightmare, you think you've got it right and then you go and read a thread like this 😯
If I have some time over the next couple of days, I'll scratch my head and see if I can work it out. Right now though I have to finish my woodlouse new year costume...
Shirley a designer must have simulated all of this?
Can't be that tough for a pro?
Shirley a designer must have simulated all of this?Can't be that tough for a pro?
You'd be surprised how little understanding there is! As you said before though Al, important to keep it in perspective...is it really as important as leverage rates, damper choice, geometry etc? I personally don't think so but I also believe it is important to be mindful of it.
Tony Foale understands it (well, he's written a book on it at least) and apparently is in agreement with Dave Weagle about antisquat theory. I'm re-reading an extract from his book now to see where I've gone wrong.
"£100 every 12months on bearing replacement"
You've been ripped off. My maestro ones lasted 2 years and even then it was only the main bearings, though the linkage bearings were getting gritty so I had them all changed by the shop. It cost a chunk less than £100, though I forget exactly how much.
been looking for my Tony Foale book i seem to remeber lending it to someone who never returned it,, have been reading john robinsons book on chassis tuning, he only briefly mentioned pro squat as used on honda rc30
when i ran the ohlins race service in 85-90 most people then disconnected that linkage used on the rear brake and bolted it to the swinging arm,, mainly those who balenced the bike midcorner on the rear brake,
Mmmmm, RC30!!
A couple of things I've been thinking about which may or may not be relevant. Could someone check my maths to see if I'm somewhere near right.
It was claimed earlier that a cyclist and a motorcycle engine produce about the same torque.
Using my example of a 200kg load on an 85mm radius chainring, I make that 17Nm of torque for the cyclist.
50Nm would be typical for a motorbike, so that's three times the cyclist.
A cyclist only produces maximum torque when the cranks are horizontal, and near enough zero torque when they are vertical, so the average will be about half the maximum..
A multi cylinder motorcycle engine with a flywheel will produce fairly constant torque.
So now the motorbike is producing six times the torque of the cyclist.
A motorcycle has a primary drive as well as a gearbox so that the final drive (chain and sprockets) rotates at between 1/2 and 1/10 of the engine speed, or has between 2 and 10 times the torque.
So, all together the motorcycle is putting out between 12 and 60 times the torque of the cyclist at the gearbox output sprocket/chainring.
So, again, we come to the bicycle suspension being designed for 1/3 of the weight of the motorcycle while dealing with 1/12 to 1/60 of the tension on the chain.
I still can't see how chain tension has a significant effect on bicycle suspension.
The other thing is that no one has mentioned torque reaction at the rear wheel yet.
As the torque at the rear sprocket tries to rotate the wheel forward, it is also trying to rotate the swinging arm backwards. This is why bikes wheely under power.
It's hard enough getting a bicycle to wheely solely under pedal power without deliberate weight transfer and pulling up on the bars.
It would be impossible for a cyclist to wheely something as heavy as a motorbike using pedal power, but it's easy for a motorbike engine.
You never saw me trying to wheelie my Gixxer6.. easy my arse!
Graham, you've got your maths/mechanics a bit iffy.
i'm not sure where you get 200kg from but i weigh 75kg, so let's round it up to 100kg.
100kg = 981n = 1000N (give or take)
i use 175mm cranks = 0.175m
torque = 1000 x 0.175 = 175Nm
(the diameter/radius comes in if you want to calculate chain tension)
(tension = torque / chainring radius (= 0.065 metres for a 32t))
chain tension = 2700N
i've never done the maths before, that's loads! - that's roughly the 'weight' of 3 people.
so you design a suspension system to cope with the weight of 1 person, and then pull it up/down with a chain pulling with the equivalent of 3 people.
when i've looked at motorbikes i've seen that some clever engineer has done a lovely job of getting the tiny output sprocket almost concentric with the swingarm pivot.
because a concentric pivot won't cause any chain 'growth'/ swingarm compression/extension under chain tension.
i'm an engineer, just trust me...
(yes, i can do the maths)
I should have looked that up first shouldn't I ?
I thought 1N = 1Kg
Its 10N = about 1Kg
So my calculations are out by a factor of 10. 😳
easily done.
power is torque x revs i think,,
so if two cyclists give the same power the one doing the least rpm must be producing the highest torque and chain tension ,,
can one of our tame boffins please help
power is torque x angular velocity
torque in Newton metres.
angular velocity in radians per second.
(derived from first principles: power = energy / second, and energy = force x distance)
I think that's a yes 🙂
I've worked out all that chain angle whatsit, just need to find a bit of time to explain it all. I'm sure you're all on the edge of your seats. 😯
Re. power and torque, what ahwiles said. Convert from rads/sec to rpm by multiplying by 60 to get rads/minute then divide by 2 pi to get revs/minute (there are 2 pi radians in one revolution).
so if two cyclists give the same power the one doing the least rpm must be producing the highest torque and chain tension ,,
chain tension depends on gear choice Shirley?
Only with multiple chainrings.
1x9 or hub gear, it depends entirely on pedalling effort.
correct.
MidlandTrailquestsGraham - Member
Only with multiple chainrings.
1x9 or hub gear, it depends entirely on pedalling effort.
PEDANTRY ALERT!
How many full suss bikes are being ridden with single chainrings or hub gears?
Oh and you forgot to tick me off re crank length, nor the reaction from the suspension!
Every full sus DH bike ever runs a single chainring?
Ok, here we go...
I realised the other day that something was wrong with my analysis as I started to wonder whether the concentric pivot example was a special case. Consider a horizontal swingarm but with the BB directly below the pivot and a front chain ring that puts the chain line right at the middle of the pivot. If my statement above about chain line running through the pivot were true (chain force will not extend or compress the suspension), this scenario would produce no anti-squat effect as I had also stated that a horizontal swingarm has no anti-squat capability. It is clear however from drawing a standard anti-squat diagram that this setup will have a certain degree of anti-squat capability.
To explain what actually happens, you have to go a bit further into the force vectors and whatnot. When you are pedalling, there is obviously a force acting along the line of the swingarm in a forward direction, into the pivot. At the same time, there is an opposing force in the chain pulling backwards. Each of these forces can be separated into their vertical and horizontal components. The horizontal component of the chain line force must obviously be less than the horizontal component of the swingarm force, otherwise the bike would not move forward. The vertical components are what are important for calculating and understanding the anti/pro-squat.
Consider the scenario where the swingarm and chain line are parallel and both angled down towards the rear wheel – not an unlikely setup. The chain line force will have a vertical component acting down pulling the main frame (sprung part of the bike) down and so producing a pro-squat effect (compressing the suspension). The vertical component of the swingarm force however will be acting upwards pulling the main frame up and so producing an anti-squat effect (extending the suspension). I have already shown that the horizontal component of the swingarm force must be greater than that of the chain line force so it follows that the vertical component of the swingarm force must also be greater. It doesn’t take a genius to therefore work out that there will be an overall anti-squat effect as the vertical swingarm force component is dominant, outweighing the pro-squat effect of the chain line force.
In the first example I mentioned above in this post, the swingarm force is horizontal so has no vertical component. It therefore has no anti/pro-squat capability, as I mentioned in my original biblically long post. The chainline force however has a vertical component acting upwards, pulling the main frame up and producing an anti-squat effect.
So in quick conclusion it IS possible to have a chainline that lies above the main pivot but below the point where it would be parallel to the swingarm line and have an anti-squat effect (i.e. extension of the suspension rather than compression). Having the chainline coincident with the main pivot does not mean the chain force won’t act to compress or extend the suspension, it just means that the chain extension (and so pedal kick-back) and variation of antisquat throughout the range of travel can be minimised.
That is basically a simplified version of what Tony Foale states in his book so I'm counting on him being right!
Diagrams would help explain this and I could probably do some if anyone is interested 😯