While the theory page on this website gives a fair bit of insight into the physics of vehicle dynamics, this next section aims to show how that knowledge can be used to put theory into practice. 

This section summarises most of the common types of suspension systems available on the market today, and describes how a suspension designer would go about working with them.
It looks at the pros/cons of each system from the designer’s perspective.

Using a more complicated suspension configuration doesn't guarantee a better suspension design. A simple suspension configuration can easily be made to perform better than a poorly designed more complex configuration.

This section focusses on how using different suspension configurations affects the designer’s ability to tune the pedalling behaviour (Anti-Squat or AS), and braking behaviour (Anti-Rise or AR) only, and doesn’t go into detail about shock actuation.
With each of these configurations, there are many possible ways of implementing the shock actuation (between any two moving members). So, for simplicity, we won’t go into detail of the different shock actuation possibilities.

It’s important to remember that ultimately the pedalling and braking performance of any suspension system is defined by the AS and AR curves, regardless of the physical configuration.

Single Pivot:

• This is the most basic suspension configuration. The swingarm pivots on the front triangle in a single location (hence the name). The axle follows a perfectly circular arc during suspension travel.
• The brake caliper is mounted to the swingarm.
• The location of the pivot affects the AS and AR curves.
• AS can be made to be decreasing, constant (approximately), or increasing, and the AS curve is very linear.
• AR is typically always decreasing, and the curve is also relatively linear.
• When choosing the location of a single pivot, the designer only needs to consider what AS and AR values they want at a single positon of travel (usually sag).
• From these values, it’s possible to work backwards through the ‘graphical methods’ for quantifying AS and AR, to determine the required pivot location.
  • To achieve a particular AS value at sag, then the pivot must lie somewhere on a specific AS line (Line 1 from the AS Graphical Method).
    • Note that the pivot could be anywhere along this line, and it would produce the desired AS value at sag.
  • To achieve a particular AR value at sag, then the pivot must lie somewhere on a specific AR line (Line 2 from the AR Graphical Method).
    • Note that the pivot could be anywhere along this line, and it would produce the desired AR value at sag.
  • These two pieces of information fully constrain where the pivot must be located. Altering the AS and AR values at other suspension positions cannot be achieved without affecting the values at sag.
• If ‘reasonable’ AS and AR values are desired, the pivot location usually ends up being close to the top of the front chainring. This is pretty much coincidental, and has much more to do with other dimensions like wheelbase, CoG location, and BB location than anything to do with the pivot being at a ‘special point’ on the chainline.
• Interestingly, for a bike with ‘normal’ geometry, the slope of the AS curve can be significantly altered by making relatively small changes in the AR sag value.
  • For example, if you want 100% AS at sag, and 100% AR at sag, then the AS curve will be decreasing.
  • If AR at sag is dropped to around 85%, and the AS remains at 100% at sag, then the AS curve becomes close to a horizontal line (constant AS).
  • If AR at sag is dropped to around 80%, then the AS remains at 100% at sag, then the AS curve begins to have a positive slop (increasing AS).
• This provides a fairly effective tool for the designer to be able to tune the slope of the AS curve, by making only relatively small changes to the AR value at sag.
• While it is possible to create a stable pedalling platform (AS increasing through travel), due to the AS curve being fairly linear, it means that AS will continue to increase beyond the pedalling zone (after about 70% travel). This has quite a negative impact due to excessive pedal kickback on big suspension compressions.
• Therefore, as a compromise between having a stable pedalling platform and reducing pedal kickback, and AR values, suspension designers usually design AS to decrease throughout travel (unstable pedalling platform), which lessens the pedal kickback deeper in travel.
• In summary, any two of the following three characteristics can be tuned independently:
  • AS value at sag
  • AS curve slope
  • AR value at sag.

Axle Concentric Pivot (aka Floating Caliper, Split Pivot, ABP):

• Floating Caliper: The simplest form of this is where an additional pair of links is added to a single pivot configuration. The brake caliper mounts to a ‘brake link’ which pivots concentric to the rear axle (this means that the brake caliper always stays concentric to the disc rotor). A ‘control link’ is pivotally connected between the ‘brake link’ and the front triangle. The benefit of this design, is that the AR can be tuned by changing the dimensions of these additional links. Tuning the AR in this way does not affect the AS, therefore the design allows for independent tuning of AS and AR. This is the primary benefit of this design. A secondary benefit is that AR values can be tuned more precisely than with a single pivot arrangement. Since the axle path is still governed by a single pivot, the AS characteristics are exactly the same as what can be achieved with a single pivot design. Typically, with a ‘floating caliper’ configuration, the brake link is quite short, and the control link is quite long.
• Split Pivot and Trek ABP: For the purpose of AS and AR tuning, these two configurations are identical; despite Dave Weagle and Trek each holding separate patents for these configurations. The reason this is possible, is that these patents don’t actually claim the main performance benefit of this configuration (being able to tune AS and AR independently) but rather they claim different specific details about how the shock is mounted and/or actuated (which has nothing to do with AS and AR!). With these configurations, the brake link and control link have a dual purpose, in that they control the position of the brake caliper, and also actuate the shock. Typically, the brake link is quite long, and the control link is quite short (opposite of floating caliper configuration) which allows for more interesting tuning of the AR curve. Specifically, AR can be made to increase throughout suspension travel. Since the axle path is still governed by a single pivot, the AS characteristics are exactly the same as what can be achieved with a single pivot design.
• When designing an axle concentric pivot system, the designer should start with deciding what AS value they want at sag, then the main pivot must lie anywhere along that particular AS line. The shape of the overall AS curve can then be tuned by moving the pivot to various positions along that line. Closer to the rear will have a steeper decrease in AS, and closer to the front of the bike will have a flatter (or increasing) AS curve.
• Once the main pivot location is established, the AR curve can be independently tuned by altering the dimensions of the brake link and control link.
• A shorter control link will make the IC move more rapidly, which can be used to generate an increasing AR curve. While a longer control link will produce a decreasing AR curve.

4-bar (Horst Link):

• Horst Link is a 4-bar linkage where the brake caliper and rear axle are mounted on the “4th bar”, which is connected to the front triangle by a pair of links (upper and lower).
• The lower link is quite long, and connects to the 4th bar quite close to the rear axle, and the upper link is relatively short.
• While Horst Link is a true 4-bar linkage, from the point of AS and AR tuning, Horst Link offers very similar traits to the Split Pivot / Trek ABP configurations described above.
• Since the axle is mounted on the 4th bar, its arc is non-circular. However, due to the closeness of the lower pivot to the rear axle, the axle path is very close to being circular. This means that AS characteristics are very similar to what can be achieved by a single pivot. In fact, we would suggest that there is no significant advantage in AS tunability with this configuration over a single pivot.
• AR can be tuned by changing the location/length of the upper link. Again, because of the closeness of the lower pivot to the rear axle, this has a very small effect on the AS curve. Because of this, the Horst Link configuration allows semi-independent tuning of AS and AR.
• Horst Link has been hugely popular, and has featured on many different bikes. The most likely reason for this is that it offered the ability to semi-independently tune AS and AR, with a package that could be manufactured economically (i.e. without having an axle concentric pivot).
• Specialized have used Horst Link on their bikes for many years as “FSR”. Sometimes the suspension configuration gets a reputation for poor pedalling performance. This is not a characteristic of the Horst Link configuration, rather it’s the way that Specialized have chosen to implement it.
• The process for designing a Horst Link system is very similar to that of a Split Pivot / ABP configuration. The main difference is that when tuning the AR, there will be a small effect on the AS, so it may be necessary to go back and adjust the position of the main pivot slightly before returning back to final tuning of the AR.

4-bar (Short Links):

• Short link 4-bar configurations offer a significant advantage over single-pivot systems, particularly for the tuning of AS and AR.
• There are various embodiments of this configuration, such as DW-Link, KS Link, LA Link, Switch Link, Maestro, VPP, and probably many others.
• Sometimes one or both links can be replaced by eccentric pivots, which are kinematically identical to pivoting links, but it just allows them to be shorter than the diameter of a pivot bearing.
• Some systems have both links rotating in the same direction, others have them rotating in opposite directions, and others change direction through travel. None of this is kinematically significant, but this feature has been used to characterise the different systems as ‘novel’, thereby allowing patents to be granted to some of them.
• In each configuration, the axle follows a non-circular path, which is determined by the location of the ‘instant centre’ of the wheel carrier link relative to the front triangle.
• Due to the shortness of the upper and lower links (and the longer distance between them), the instant centre can be quite ‘volatile’ and can be made to rapidly change direction throughout suspension travel, if desired.
• The ‘volatility’ of the instant centre can be roughly gauged by looking at the lengths of the upper and lower links, and the distance between the upper and lower pivots:
  • A design with short links separated by a large distance will be very volatile.
  • A design with short links separated by a small distance will be less volatile.
  • A design with long links separated by a large distance will be less volatile.
  • A design with long links separated by a short distance will have very low volatility (and will approach what can be achieved with a single pivot arrangement).
• So a design with ‘low volatility’ doesn’t really make full use of the capabilities of a 4-bar configuration. However, it may be the designer’s goal to have single-pivot like AS and AR characteristics, without having to include a physical pivot at a particular location. So it’s not always a designer’s goal to make the most of a ‘volatile’ system.
• A good suspension designer will be able to take advantage of a more volatile system, by carefully tuning the migration of the instant centre (IC) so that it is a the ‘right’ location at every point in travel.
• For example, AS can be tuned to rapidly change values as suspension travel passes from the ‘pedalling zone’ to the ‘non-pedalling zone’. AS can be made to rapidly decrease after about 70% travel, to reduce pedal kickback on bigger suspension compressions. Note, this overcomes/reduces the compromise to AS tuning described in the single-pivot section.  Also note, in order to make the AS decrease rapidly after about 70% travel, the axle path must arc forwards significantly in the last part of travel. This introduces other compromises with regard to suspension compliance on big hits.
• When designing this type of system, the designer needs to know what they want their overall AS and AR curves to look like.
• The path of the instant centre can only be fully constrained if the designer knows what AS and AR values they want at ALL points in travel.
• Knowing the AS and AR curves will allow the IC to be located at every point in travel, and the IC migration path can be determined.
• The challenging part is now working out what positioning of pivots/links will generate this IC path.
  • This process is called linkage synthesis, and can be done by trial and error (very time-consuming) or more likely can be done using sophisticated computational algorithms.
  • In either case, it’s likely that the 4-bar linkage will not be able to produce the exact IC path desired, and so AS and AR must be compromised somewhere.
  • Also there may be further compromises required if the calculated location of the pivots/links is not spacially convenient, and so AS and AR must be compromised somewhere.
• So, while it’s theoretically possible to independently tune the AS and AR curves independently via the IC migration, the reality is that one or both is usually compromised to make it work with a 4-bar linkage within the spacial constraints of a bike frame.

4-bar (Other):

• There are a few other 4-bar configurations that fall somewhere between Horst Link and Short Link configurations:
  • Ellsworth ICT, Inverted Horst Link, Breezer MLink.
• Based on the discussions above, it can be estimated how useful these systems are, and how a designer might approach working with these systems.

6-bar (e.g. Felt Equilink):

• There are numerous possible configurations that fall under the banner of a 6-bar linkage, so we won’t go into detail about the different types.
• In the same way that short link 4-bar configuration offers kinematic advantages over a single pivot, a 6-bar configuration offers further advantages over a 4-bar.
• The reason for this, is that with a 6-bar linkage, the instant centre (IC) can be made to be even more volatile, and make even more abrupt changes in direction.
• The design process is the same as with a short link 4-bar system, in that the designer needs to know what they want their overall AS and AR curves to look like.
• Knowing both the AS and AR curves allows the IC to be located at every point in travel, and the IC migration path can therefore be determined.
• From there, synthesis of a suitable 6-bar linkage is much more complicated than with a 4-bar linkage, and will certainly require computational algorithms.
• It’s possible that a 6-bar linkage can achieve (or get closer to) a desired IC path, where a 4-bar linkage was not able produce the desired IC path. This would be a good justification for using a 6-bar linkage.
• However, due to the additional links and pivots, and the ‘spacial’ limitations of a bike frame, the 6-bar linkage is unlikely to be a more convenient solution than a 4-bar, so it’s likely that the AS and AR curves are compromised because of this.
• If considering purchasing a bike with a 6-bar linkage, we'd want to see that the use of the 6-bar configuration has produced an improvement in the AS and AR curves that is not possible with a 4-bar system.

Slider/Rail Mechanisms:

• A sliding link is kinematically equivalent to an infinitely long pivoting link, so that it travels in a straight line rather than a circular arc.
• Recently, Yeti has changed design from an eccentric lower link on the 4-bar system, to a sliding lower link. Kinematically, there is very little improvement with this change, so we can only assume that the change has been made to avoid patent infringement that their earlier ‘Switch Link’ design was suffering.
• The design process is the same as with a short link 4-bar system, in that the designer needs to know what they want their overall AS and AR curves to look like.
• Knowing the AS and AR curves, will allow the IC to be located at every point in travel, and the IC migration path can be determined.
• It should be noted that the instant centre of a sliding link is infinitely far away, and perpendicular to the slider.

Fixed Idler (idler mounted to front triangle):

• All the suspension configurations discussed above use a ‘conventional’ drivetrain.
• When using a conventional drivetrain, there is really only a fairly narrow window of possible axle paths that will produce reasonable pedalling performance.
• With all these systems, the axle path is approximately perpendicular to the chain line.
• If the designer wants the axle path to be more rearward, then orientation of the drivetrain needs to be altered so that it will still produce reasonable pedalling performance.
• One possible option is to raise the BB, so that the chainline is again approximately perpendicular to the axle path. However, this is a fairly impractical solution, as it drastically affects the ergonomics of the bike.
• Another option is to include an idler pulley, fixed to the front triangle, so that the chainline is again approximately perpendicular to the axle path.
• Any of the configurations described above could be used to generate a more rearward axle path, therefore a ‘fixed idler’ configuration is possible with each of them. The design process and the pros/cons of each configuration remain the same as if used with a ‘conventional drivetrain’, with the exception that a more rearward axle path allows for higher amounts of anti-squat, with less pedal kickback.

Unified Rear Triangle (URT):

• Trek, Klein
• All of the systems described above have the BB mounted on the front triangle. This is kinematically significant, because the rider’s input forces are all acting on areas that are part of the one body (handlebar (acting on head tube), cranks (acting on BB), and saddle).
• With a URT system, the bottom bracket is mounted on the same structure as the rear wheel. This means that there is no ‘distancing’ between the rear axle and the BB, eliminating the drivetrain’s influence over suspension behaviour. However, it introduces a new influence on the suspension, in that the BB (where the rider’s feet are connected) now moves relative to the front triangle (where the rider’s hands/backside are connected). Now, when a rider exerts pedalling forces between two movable elements of the system (action at the BB, reaction at the bars/saddle), it affects the suspension behaviour. Ultimately, this behaviour can still be described by an AS curve, however it requires a different method for calculating AS, which needs to account for details of the rider position, rider dimensions, and whether they are standing or sitting. At best, this results in an approximation of AS.
• URT enables designers to raise the pivot height (hence create a more rearward axle path) without causing any distancing in the drivetrain. The higher pivot produces a more rearward axle path, but at what cost? With the BB now attached to the rear triangle, it means that the BB also moves rearward and upward as the suspension compresses. Since much of the rider’s weight is carried at the BB, it reduces the effectiveness of the suspension if the BB has to move upwards/rearwards each time the rear wheel encounters a bump. The irony of this system is that it’s quite good at absorbing bumps when seated (with feet off the pedals), however most riders shift to the standing position when riding over rough terrain!

Floating BB:

• These systems are similar to the URT system, in that the BB is not mounted on the front triangle. Instead, it is mounted on a link that moves relative to the front triangle and the wheel carrier link.
• There are two main configurations: Single Pivot axle path (GT I-Drive, GT Independent Drivetrain, GT AOS, Mongoose Freedrive), 4-bar axle path (Lapierre Pendbox).
• For the Single Pivot Axle Path version, the swingarm is mounted to the front axle via a single pivot, and the BB is mounted on one of a pair of links connected between the swingarm and front triangle.
• For the 4-bar version, the swingarm is the 4th bar of a linkage, and the BB is mounted within the lower of the two links.
• Theoretically, 6-bar versions are also possible.
• This configuration can been implemented in a few different ways, for example: 
  • One approach: Some designs (GT AOS, Mongoose) include a significantly rearward axle path, and then design the BB to move significantly rearward/upward to control the distancing between the rear axle and the BB. This approach results in similar behaviour to the URT system described above.
  • Another approach: Other designs have a more conventional axle path, and have the BB make only subtle movement. Perhaps the BB can be made to be fairly stable through the pedalling zone of travel, then begin to accelerate towards the rear axle around 70% travel (thereby causing a downturn in the AS curve, reducing the amount of pedal kickback during big suspension compressions).
• Another possible implementation, would be to have the BB move forward/down as the suspension compresses (and the rear axle moves up/rearwards). Of course, this would introduce large amounts of distancing between the rear axle and BB, which would need to be overcome with the introduction of an idler on the front triangle (or one of the other links). Here at i-track, we extensively explored this configuration back in 2012. Our conclusion was the moving BB does not offer any significant geometric advantages (over a basic rearward axle path design), and it still couldn’t generate ‘sophisticated’ AS curves. 

>>back to top