Chain-Suck :
the OVERALL picture
see Main Article : Chain-Suck in a NUTSHELL

by Jonathan Levy

For the summary version of this article, see ... Chain-Suck in a NUTSHELL
For detailed field and workshop testing, see ... the DETAILED Investigation

Chain-suck becomes THE major issue when you experience it. Pedalling action is brought to an abrupt binding stop, and you fall over into the dirt ; your chain seems to be gripped in a steel vice. It can be tenaciously difficult to eliminate chain-suck during a bicycle ride, as it keeps recurring once initiated. So it can bring your trip or race to a complete standstill, and you walk up the hills pushing your bike.

It usually occurs when the chain and cogs get muddy or wet and gritty and is worse on steep uphill slopes or when otherwise pedalling with force. But simply washing mud and grit from the chain and cogs often won't get rid of the problem.

Chain-suck is a failure of the chain to disengage from the bottom teeth of a front chain-ring; instead the chain is carried up and around the full circumference of the ring, to wind back onto itself, and to jam between the chain-ring and chain-stays.

Pedalling action is brought to a sudden jammed stop, often under the high force of a hard pedal downstroke. The chain can become permanently twisted, teeth can be damaged, and chain-stays of aluminium or carbon-fibre can also be damaged by the chain being wedged hard against them.

The most frequent incidence of chain-suck is on small chain-rings ("granny-rings") of triple-ring mountain bikes, but it also occurs on larger rings, and with road bikes. Two main forms of chain suck occur, namely :

  • 1-ring suck : the chain hangs-up on a single ring (usually the "granny-ring")

  • 2-ring suck : the chain hangs-up across two rings when changing gears (usually from the middle ring to the "granny")

Other circumstances which are NOT chain-suck, even though they might appear to be so, can also result in the chain being dragged into the chain-stay and cause severe damage :

  • The chain hangs-up, jammed in the narrow space between chain-rings, caused by poor design or manufacture of the spacing between chain-rings. [Resolved by filling the gap between problem rings or by installing narrower spacers, to prevent the chain getting trapped by the offending space]

  • The chain is thrown off the chain-rings onto the chain-stay by poor derailleur adjustment and then jams between the chain-rings and chain-stay.
    [Resolved by better adjustments]

2-ring suck can often be temporarily alleviated by a short backpedalling action, which frees the chain enough to allow the gear change-over to be completed (provided the chain has not yet jammed badly in the chain-stays); thereafter normal pedalling can be resumed. Backpedalling for 1-ring suck can free the chain similarly (if not jammed), but in this case, chain-suck usually recurs immediately that forceful pedalling is resumed, making this form of suck more tenacious.

The basis of this chain-suck investigation was a substantial and careful programme of testing, initially in the
field, and later also in the workshop. The programme covered 1-ring suck directly, as described in the above links.

Thereafter, a further large amount of testing was undertaken to check issues raised by respondents to the initial website articles. The fundamental findings did not change, but their application was extended to provide more detail for 1-ring suck. Furthermore, most of the issues identified were later found also to have bearing on 2-ring suck, the investigation process for which is described below.

Subsequent to writing the initial articles about 1-ring suck, I was also able to evaluate and on occasions observe, 2-ring suck which affected biking colleagues, acquaintances, and respondents to both the website articles and my postings to the Bicycle Science forum. Some respondents provided their own valuable insights, with particular thanks to Chris Juden of the CTC. Understanding about 1-ring suck, and extension of this to the issues relevant to 2-ring suck, enabled me to provide assistance to cyclists in alleviating this related problem.

A formal testing programme for 2-ring suck was not undertaken. However, the understanding brought about by finding out what worked and what didn't work for cyclists experiencing this problem, why this was the case, and discussions with them relying on their observations which I guided, constituted an informal testing process ; this, together with my own direct observations, also allowed insight to differences which could occur in the underlying mechanisms of 2-ring suck.

Most of the issues causing or contributing to 1-ring suck, can also cause or contribute to 2-ring suck. The basic mechanism of 2-ring suck is similar to 1-ring suck, but the underlying componentry problems driving it can be somewhat different, and the details of its mechanism are a bit more complex.

1-ring suck has WORN "GRANNY-RING"
teeth as its MAIN underlying cause, where "pockets" form in the tooth pressure faces. Burrs on the sides of the teeth, thickened teeth, or grit jammed between the tooth sides and the chain possibly contribute to chain-suck to a lesser degree but are not normally the main cause, despite popular belief. Chain-suck usually initiates under high pedal forces when the chain gets MUDDY
(or wet and gritty). A NEW (or "Low-Stretch") CHAIN
used on a worn "granny-ring" makes matters much worse. The mechanisms by which these issues cause chain-suck are explained in the next section.

2-ring suck can also be precipitated by the above issues, but it its main underlying cause is usually a mismatch in tooth alignment across adjacent chain-rings, and this can be sufficient on its own to initiate the problem. The number of teeth on adjacent rings are necessarily different. As a result teeth on adjacent chain-rings may not be optimally aligned (in a rotational sense) for crossover of the chain. This may be due to design/manufacture of the rings, but it can also arise from other causes. This type of chain-suck is more likely to be worsened by burrs on the tooth sides and thickened teeth because, during a gear-change, the chain is "bent" sideways and is at an angle to both chain-rings; again this contributory factor is not normally the main cause.

Other causes of chain-suck are also reported, but are usually less common. They are often for obvious reasons such as accident damage to chain-rings (bent rings), to teeth (bent teeth), or to the chain (twisted or tight links). Poor chain maintenance, especially after riding in a lot of mud, can also lead to tight chain links. Design/manufacture problems with the drive-train can also cause it, but these problems can be more difficult to identify.

A substantial mythology exists about what causes chain-suck. The issues below are frequently cited as causes or possiblities. Neither my testing programme, nor careful trial and evaluation of alternative solutions in the field, nor consideration whether they have any potential for sensible mechanisms, has supported them as causes. I don't consider them to affect chain-suck in any meaningful way for the commonly occuring chain-suck experiences :

The "granny-ring" is inherently more vulnerable to retention of the chain (ie chain-suck), than larger rings. Its chain load is very high, because of the pedal-arm's leverage on the small diameter ring, and because it is usually used on steep hills needing high pedal forces. This high load is borne by fewer teeth compared with the larger rings; each tooth thus bears a greater share of the chain load. In some circumstances high load can be transferred even to the last teeth at the bottom of the "granny-ring".

When there is high loading on the bottom teeth, higher friction on the tooth pressure-faces and within the chain's rotating parts will result, thus making it more difficult for the chain links to rotate, or the rollers to roll and slide off the bottom teeth. Again it is noted that this friction is in-line with the chain and its tension loading (not sideways across the teeth by a jamming/wedging action). The high load also leads to faster wear of the tooth pressure-faces. Additionally, the small diameter of the ring causes a greater upward pull angle at the bottom links, thereby making it more difficult for the bottom links to drop down quite as easily. By themselves, these inherent factors are insufficient to initiate chain-suck but they do enable the following issues to cause it when in sufficient combination.

TOOTH WEAR results in "hooking" / indentation / "pockets" in the pressure faces of the teeth. This is the MAIN cause of CHAIN-SUCK. It results in two effects both of which promote chain-suck, namely increases in :

Burrs or ridges on the tooth sides and thickened teeth can also occur, but despite popular belief this does not normally cause chain-suck, although it might contribute to a lesser degree. Once teeth are adversely shaped by wear (or any other cause), the following factors can initiate chain-suck.

HIGH FRICTION occurs when MUD or GRIT is introduced, because fine particles and water work into the chain's moving parts, get further crushed by chain loading, and rapidly strip out all chain lubrication. If there is also high tooth loading, then high in-line friction retaining forces will be generated. Under such frictional and mechanical resistance, the links cannot rotate, and the chain rollers cannot slide or roll to free themselves from the bottom of the "granny-ring". In fact, the whole section of chain contacting the "granny-ring" teeth gets so bound by the friction generated by its tension loading that it becomes temporarily semi-rigid along its length. When combined with a degree of mismatch between chain and tooth pitch (see below), even the bottom most tooth can be under load, and then friction will contribute to resisting disengagement of the chain-rollers. It is important to bear in mind that not all muds are equal ; different muds, grits, clays, sands, etc, can have very different frictional characteristics, so some will precipitate chain-suck readily while others may not cause problems. The Chain-Suck-TEST described in this article can be used to evaluate a soil's ability to precipitate chain-suck by friction.

A WORN or "STRETCHED" CHAIN exerts less load or even no load on teeth at the bottom of the "granny-ring" [This "stretch" is not due to lengthening of the chain's link-plates, but arises from wear of the internal pins and swaged split-bushings]. This slack in the individual links accumulates as the links move away from the top of the ring to progressively start freeing the rollers from the worn teeth, both in terms of tooth load and engagement geometry. The degree to which the chain does or doesn't "suck", depends on how "stretched" it is relative to the tooth spacing at the contact points where it is able to take actual load from the teeth (if at all). Worn chains UNLOAD bottom teeth ...& make chain-suck less likely. But, a major drawback of a "stretched" chain is that it overloads the top teeth in turn, causing accelerated metal failure and wear to the tooth-faces.

A NEW or "LOW-STRETCH" CHAIN, by contrast, has its tight fit against the tooth pressure-faces extended farther towards to the bottom of the chain-ring. This occurs because its link-lengths (chain-pitch) are correct, are thus slightly shorter than a "stretched" chain ; they match the fundamental tooth spacing (tooth-pitch) more accurately. The chain load is spread more evenly between teeth and accelerated tooth wear is avoided, but the pressure-faces of the teeth at the bottom of the "granny-ring" now bear a share of the already high load. If sufficiently deep "pockets" are worn into the tooth pressure faces, then retention of the chain due to mechanical resistance alone can be sufficient to prevent disengagement and cause chain-suck. As already noted, increased load also generates increased frictional resistance to disengagement which will combine with the mechanical resistance to further bind the chain. New chains LOAD bottom teeth ...& make chain-suck very likely (eg STX-ring).

Further effects occur, interaction of MUD, CHAIN-STRETCH, & TOOTH-PITCH, and also interaction of TOOTH-WEAR & TOOTH-PITCH, are as follows :

In both the above two cases, the increase in effective tooth-pitch is the flip-side of a "low-stretch" chain (small chain-pitch), and has the same effect. This mismatch in pitch will worsen the interaction of a new or "low-stretch" chain bearing on worn tooth faces. Additional chain load will transfer to the bottom teeth. Overloading bottom teeth further increases mechanical resistance to disengagement from "pockets" in the teeth, and also increases friction within the chain and between chain-rollers and worn teeth. This is a significant additional effect (which also accelerates tooth wear). Increased tooth-pitch together with new chains OVERLOAD bottom teeth ...& make chain-suck almost certain. (eg LX-ring).

Thus TOOTH-WEAR, LOW-STRETCH CHAINS, and HIGH FRICTION cause the chain-ring's bottom teeth to be loaded, and resist chain disengagement. The rear-derailleur spring cannot provide enough tension nor has the chain enough self-weight to disengage from the bottom teeth of the chain-ring in these circumstances. The links are carried around and up the rear of the chain-ring under load .... the chain "sucks" !! This is the primary mechanism of chain-suck.

Wear of the tooth pressure-faces remains the biggest issue at all times.

see chain-suck MECHANISM

Pedalling force, is the final precipitator of repetetive incidents of chain-suck once a sufficient mix of the fundamental factors is present. This occurs on steep hills, or elsewhere where substantial force is applied. The reason why it rarely occurs away from steeper gradients is that on flatter sections, even when high pedalling force is used, the "granny" ring is rarely in use, and that is the most vulnerable chain-ring. It has also been known to occur with the middle ring, but more rarely.

The way in which a chain interacts with the shape of a worn tooth is progressive. As increasing force is applied, so the contact point of the chain-roller "elevates" progressively up ramps worn into the lower/middle portions of a tooth's pressure-face, thereby transferring load to the bottom teeth in an accelerating manner.

There tends to be a fairly sudden onset of chain-suck. With increasing pedal force, not only is the absolute chain load on each tooth increased, but also an increasing proportion of the total load gets transferred to the bottom teeth. The steep or "hooked" part of the teeth occurs on all teeth at approximately the same "elevation" which is reached at a fairly specific chain load ; thus mechanical resistance to disengagement starts coming into effect quite suddenly. And finally, there is an effect which amplifies the load just as rollers try to move past the "hooks", and must be overcome in order to allow the roller to finally disengage ; it makes it suddenly much harder to open the final chain-suck "gate" to get free.

see LARGER mechanism image

For 2-ring suck, the manner of derailing a chain needs to be considered. Modern manufacturers try to ensure (by careful profiling/shortening/ramping of specific teeth), that the leading and trailing stages of chain departure from a ring (respectively when the chain is at the top and bottom of the ring), occur only during the low-load portions of the crank cycle (ie when the crank is approximately vertical). They also chamfer tooth sides to encourage the process of cross-over at the desired locations on the rings. Additionally, they try to ensure a suitable rotational alignment of teeth on adjacent rings, along the zones of intended chain derailment. And finally, the teeth of the more critical dispatching ring (ie middle-ring) have different in-plane widths/shapes at critical locations in the cross-over zone, to allow the teeth on the receiving ring to actively take up the chain load in order to remove this load from the dispatching ring.

If the manufacturer(s) of crank, spider, and chain-rings get both the design and the manufacture right, the teeth of the dispatching chain-ring should not be subject to chain load after enough of the receiving ring teeth have come fully into mesh with the chain and are taking load. The receiving teeth should be able to "pull" the chain very slightly forward to unload the dispatching teeth. If, for some reason, the dispatching teeth are still under load at the bottom of the ring it will be more difficult for the chain to disengage from these teeth and hence also more difficult for final dispatch of the chain to complete. Older and simpler chain-ring designs were more susceptible to 2-ring suck than modern ones which have a range of improvements in tooth design to ensure that the chain transfers at locations where, in a limited zone, there is no adverse mismatch in cross-over length.

The problems pertinent to 1-ring suck also play a role with 2-ring suck. However, as explained above, there can be an alternative or additional source of mismatch of tooth/chain pitch with 2-ring suck ; it is across chain-rings instead of on a single chain-ring because adjacent rings have different diameters and numbers of teeth which results in an inherent mismatch. During chain cross-over, an adverse mismatch can result, for example, in 3.00" of chain links needing to get taken up by 3.05" of space between the pressure-faces of receiving and dispatching teeth. This will load the pressure-face of the dispatching tooth of the chain-ring (and correspondingly unload the receiving teeth), because the dispatching tooth leads in the crank cycle. It will remain loaded until it also becomes the disengagement tooth at the bottom of the ring. This tooth now needs to both disengage longitudinally from the chain and dispatch it laterally ; under load this is more difficult. Also, if for some reason initial derailment did not occur at the desired location, then the actual transfer zone might have a significant mismatch in cross-over length, while the desired location would have had a good match ; furthermore, the location for final dispatch might inadvertently occur during a high-load portion of the crank cycle. Both these effects make disengagement even more likely to fail, in which case dispatch will also fail, and 2-ring suck then becomes likely. If disengagement and dispatch is further impeded by the increased mismatch resulting from a new (or "low" stretch) chain, by higher frictional resistance (mud), and by higher mechanical resistance (wear "pockets" in the dispatching tooth), then the chance of 2-ring suck is further increased. Once chain mismatch has loaded a bottom tooth, the mechanism is much the same as 1-ring suck.

Chain-Bridge : A more extreme mechanism unique to 2-ring suck can also occur if the mismatch in cross-over distance is more extreme (say at 3.15" for 3.00" of chain length). The chain not only loads the pressure-face of the dispatching tooth, but also loads the trailing-face on the receiving tooth. This is sometimes called a "Chain-Bridge". The chain can then become "locked" between the dispatching tooth from which it won't disengage, and the receiving tooth upon which it is "wedged" down by the full tension loading in the chain. Not only is all tension loading in the chain borne by the dispatching tooth, but it is also amplified by the wedging action onto the noted pressure and trailing tooth-faces ; this can cause slight elastic deformation, or deeper plastically deformed dimples in these teeth, which in turn resists disengagement of the chain. Without any dimpling, the load alone is not as likely to cause failure of disengagement. Elastic deformation is not permanent, does not "wear" the teeth, and once the teeth are unloaded will disappear ; plastic dimples will, of course, remain because they constitute permanent failure of the tooth material. When this whole mechanism occurs, the receiving teeth now have no chance to "pull" the chain slightly forward to unload the dispatching tooth. Again, a "low" stretch chain, high friction, and previous tooth-face indentations (dimples) will make matters worse. This form of suck will occur more readily and be more extreme in effect if the normal precaution is not taken of easing off on pedal pressure during gear changes.

Because mismatch in cross-over length is not constant around the circumference of adjacent rings, some locations of initiating a gear-change will be worse than others for precipitating 2-ring suck (which can explain why some bikers might be more subject to this problem than others - individual habit might mean gear changing at a particular pedal position). This is also the reason why older chain-ring sets gave more trouble than modern ones for this type of chain-suck ; the older types did not promote chain transfer in advantageous locations as effectively. It is also why certain ratios of adjacent tooth-count combinations are worse than others ; achieving a reasonable length of transfer zone with a favourable cross-over distance can be inherently more difficult to arrange with some chain-ring ratios.

In fact the "Chain-Bridge" effect can occur even for well designed and manufactured chain-rings if pedal pressure is not eased off, and gear-shifts are done under high pedalling loads. The chain may then be forced to shift at an unsuitable location.

A mismatch in the cross-over tooth/chain pitch can be caused by :

.... and finally : Indentation and "hooking" of dispatching teeth which occurs during the process of wear/failure can make mechanical resistance to disengagement the final precipitating cause once the dispatching tooth is loaded, thus being virtually identical in effect to 1-ring suck.


Tooth "wear" is a combination of two main contributors :

Metal failure of the teeth deforms them plastically with pressure-face indentation, tooth widening and side-burrs ; significant tooth deformation can arise from a single instance of over-stressing. By contrast, abrasive wear takes time.

"Granny-Rings" can wear suprisingly fast under moderate-heavy usage, if not made of very tough material and if their teeth are not machined accurately to a suitable profile. Substantial deformation failure of die-pressed powder-alloy tooth-faces has been known to arise from a single very steep trail, even though an equivalent amount of abrasive wear could not occur during such limited usage.

Furthermore, total wear does not need to be at all prominent for chain-suck to start. Wear may be noticeable only as a slight steepening of the pressure faces of the teeth (in appearance, like an ocean wave before it starts to break). Once one understands that in-line friction, within the chain and at the tooth/roller contact, can easily become high enough to initiate chain-suck, it becomes clear that wear does not have to become so severe as to make mechanical resistance by deeply "hooked" teeth the sole mechanism.

Normal die-stamped steel rings are inaccurate and often not tough. Aluminium and die-pressed powder alloy particularly (even special heat-treated alloys), and many normal steels, are notoriously soft compared to stainless steel or titanium; ....but die-stamped rings are the most widely available, and are convenient and fairly cheap to replace. However, tough accurate rings are available from some manufacturers at a higher cost - and usually last many times the life of softer materials.

There is a surprising incidence of chain-suck with new or almost-new drive-trains, and with newly installed drive-train components. This would seem to contradict the key thesis of this article that "wear" to the chain-rings is the main cause of chain-suck. However, there are explicable technical reasons why this can occur ; they are usually cause for rejecting the new components as faulty or unsuitable !

Problems with new equipment which are explicable by the "triangle-model", can arise because of the following :

  1. Very rapid failure of new tooth faces in unsuitably soft material (eg aluminium, die-pressed powder-alloys, and many steels) has already been explained. It both worsens the tooth profile and increases its effective pitch by forming indentations at elevated positions on the tooth-face. This can occur on first usage, if high pedalling forces are applied.

  2. New-tooth profiles which are good for normal power transfer, but unintentionally poorly shaped for chain-suck (whether by design or by manufacturing error) allow friction to initiate chain-suck at an early stage of wear, and possibly even upon first use if they are really poor. Some tooth profile shapes are also more vulnerable to failure of the pressure-faces because the roller loads are not well spread, and localised over-stressing occurs; the inaccuracy of die-stamped rings also promotes this.

  3. A small manufacturing error in the effective pitch-circle-diameter (pcd) of a new compact "granny-ring", can cause real problems. As little as 0.0175" general increase in pcd, will increase the tooth-pitch by 0.0025"=0.065mm which amounts to +0.5%. This is enough to cause the problem. Increase in the effective tooth-pitch caused by wear was mentioned in point 1 above. But, anything leading to an increase in the effective tooth-pitch and mismatch with the chain-pitch, can have the same effect of overloading bottom teeth, as discussed already. It just needs to be combined with a sufficient degree of the other factors in the chain-suck triangle to cause problems.

    It is not hard to imagine tooth-pitch errors of this magnitude in problematic production runs or as a statistical percentage of die-stamped rings leaving the factory as new items. How high is this percentage ? Possibly greater than anyone realises. Such problems are impossible to detect visually; even measurement thereof would be difficult because of the complex interaction between the tooth profiles and chain. Unlucky cyclists will suffer chain-suck with these new items if they are also subject to a sufficient degree of other relevant conditions. Being new equipment, the cause of the chain-suck will seem a mystery.

  4. Use of a new (unstretched) chain on a worn drive-train has already been dealt with in this article.

  5. Furthermore, some new chains have been measured to have a chain-pitch slightly smaller than the standard 1" per link-pair (by as much as -0.2%). This may arise from the manufacturers intention to deal with later "stretch" of the chain, or it may arise from the statistical variation in production output. However, it will transfer more load to the bottom teeth worsening any tendency to "suck", especially if combined with an increased tooth-pitch.

  6. The effect on 2-ring suck of a mismatch of tooth/chain pitch across chain-rings has already been dealt with in this article. It can be inherent to the location of teeth and their profiles, and thus occur on new drive-trains ; in this case the whole crank/chain-ring assembly may have a design or manufacturing problem. It is made worse or more frequent by changing gear during high pedal loading.

The "granny-ring" and chain are items subjected to some of the highest loads and stresses of extended duration ; high friction can result from these loads in the presence of mud, or dust (with inadequate lubrication). Errors in the engagement geometry of chain and teeth do not have to be very large to cause problems under such loading and environmental conditions.

One can't assume that a new chain-ring will be exempt from chain-suck, even after a very short duration of usage. The problem can be resolved by a suitable type and properly adjusted setup of crank-set/chain-rings. A proper set of tests is needed to determine the true cause(s) of problems. Problematic equipment is often not rejected by bicycle owners, because they have no method for clearly identifying the key causes of problems.

The Chain-Suck-Triangle Model and the associated Chain-Suck TEST can be used to identify the cause of problems, with new or used equipment.


How will you know if your "granny-ring" is worn enough to cause problems ? No simple measurement is available; a technically knowledgeable bike mechanic might be able to judge its condition. But the most reliable way, is a simple test using only the bike itself. It only takes about 1 hour and involves riding along a suitable route and hand packing the chain with mud/grit in a suitable manner. You don't even need to get dirty ! The test method is provided as a separate document :
Chain-Suck TEST.


The are two basic aims for alleviating chain-suck, namely to ensure that :
a) Tooth pressure-faces are suitably shaped to disengage even if loaded
b) Disengaging bottom teeth are subjected to little or no chain load

  1. Check for the obvious issues such as accident damage to chain-rings (bent), to teeth (bent) and to chains (twisted or tight links), and visible or reported mechanical inadequacies of particular drive-trains arising from poor design, materials, or manufacture. Repair affected components (if feasible), or replace them. Sometimes chain links become tight because of poor maintenance and riding through a lot of mud ; this can cause chain-suck ; the simple solution is to clean the chain thoroughly ; then strongly flex the stiff sections of chain (or stiff individual links) sideways/horizontally, to loosen the links ; and finally lubricate the chain well.

  2. If necessary, use the Chain-Suck TEST to isolate the cause of the problem ; then implement the fix.

  3. The most common solution to chain-suck is to install a new chain-ring :

    • Replace the worn chain-ring (usually the "granny") with a tough accurately machined component. Avoid all aluminium and die-pressed powder-alloys; use an especially tough hard steel (eg an appropriate type of stainless steel), or use titanium. If it is a larger chain-ring (say 30 or more teeth), only then is a good aluminium alloy acceptable.

    • If the existing "granny-ring" is symmetrical, flipping it over to use its unworn trailing tooth faces, can also provide a good solution.

  4. Rotate the fastening position of the chain-ring by one or more bolt-holes. This is a temporary solution which does not always work, but can be carried out on-the-trail, using only a single allen key wrench. It moves those teeth which are usually more highly indented or "hooked", away from the critical high-load zones of the crank's cycle. If the chain-ring has special attachments or individually shaped teeth to ease gear changes, then such rotation can lead to less crisp gear changes. It might resolve 1-ring suck temporarily, but it can worsen (or alleviate) 2-ring suck.

  5. Filing Tooth Profiles (see separate document) is a method for the technically proficient to reshape worn pressure faces of the teeth (tooth side-burrs are not the issue) to establish a better tooth profile. This method can also be used on-the-trail. If a new chain is to be installed, this method should not be used as a long-term solution.

  6. For 2-ring suck, the above solutions might resolve the problem. However, alternative or additional steps might be needed to resolve any mismatch in tooth/chain pitch which may exist across adjacent chain-rings ; their rotational alignment to each other may be unsuitable. One or more of the following eight options may be needed :

    The next four options can be used individually or in combination to adjust the rotational alignment of teeth on adjacent rings ; this will improve matters if the change is in the right direction, but will worsen matters if it is not.

    • Install a different "granny" ring and/or middle ring with different tooth shape, and/or also a different manufactured alignment of teeth along its circumference

    • Install a ring with a different number of teeth, to change the tooth-ratio of adjacent rings, and hence the relative alignment of teeth

    • Rotate the fastening position of the chain-ring by one or more bolt-holes, which will change the relative alignment of teeth.

    • Adjust the adjacent granny/middle rings so that the small amount of play which exists in their bolt-holes is used to fasten the rings at that extreme of play which reduces problems

    For a set of new chain-rings, for which the problem just cannot be resolved by any of the above measures :

    • It may be necessary ultimately to get a different make/design for the complete set ; this may also involve a different set of cranks ; it is advisable to get a well matched set, with tooth ratios and tooth shaping/ramping generally known not to suffer this problem, and from a single manufacturer.

  7. For anti chain-suck plates, see the section below.

  8. If you have installed a new "granny-ring" or flipped the ring to use its unworn faces, then replace the chain preferably; definitely if its "stretch" is greater than 0.3%.

  9. If any rear sprockets are worn and the new chain "jumps" over teeth, also replace those sprockets.

  10. Clean and lubricate the drive-train frequently to reduce chain and tooth wear.


Many designs of anti chain-suck plates have been made. Most types do not prevent actual "sucking" of the chain. When they work correctly they do prevent the final gruesome wedging of the chain into the chain-stay, and thereby also forcefully knock the chain free from the teeth to which they are "stuck" by friction, mechanical resistance, and chain-load. One particular design has a more gentle action where the chain rollers are continuously prised free from the bottom teeth of each chain-ring ; it therefore stops "sucking" soon after it initiates.

These devices can prevent 1-ring and 2-ring suck when they work properly.

However, all types of the device need very strong attachment to the chain-stays (eg welded, brazed, or riveted nuts) because they can be subject to high forces, and also require regular fine adjustment to be very close to the chain-ring teeth. Clamps are unreliable for attachment. The average bike shop does not have experience in fitting these plates, and the problems which can occur. Malfunctions are common.

On mountain bikes, the chain-rings are unavoidably subject to abuse. If the chain is allowed to get past the plate (by poor adjustment, by damage to the chain-rings, or by unreliable mounting of the device) then the problem can be worse than if no plate was present. Severe damage can occur to the chain-stay, chain-ring(s), and chain. It can also be extremely difficult, on the trail, to free a chain jammed behind an anti-suck plate.

For the specific case of bikers who are knowledgeable about the issues, maintain their own bikes frequently, and know how to install and adjust the anti-suck plate properly, it may be a suitable solution.

However, anti chain-suck plates are not recommended as a general solution.


When the full CHAIN-SUCK-TRIANGLE grips your bike, there are many circumstances where lubrication won't help.

Cross-country racers or those planning long rides with mud should weigh the modest cost of a suitable new chain-ring against the risk of having their events spoiled by chain-suck.

  Copyright Jonathan Levy, 2000. All rights reserved.
  email Jonathan Levy - must copy email address by hand

For the related articles, follow the links below ...
Chain-Suck : in a NUTSHELL ...comprehensive summary and solutions
Chain-Suck : the OVERALL picture ...details of mechanisms & solutions
Chain-Suck : the DETAILED Investigations ...field & workshop testing
Chain-Suck : the FIELD TEST ...a field testing method for identifying causes
Chain-Suck : Restoring Worn Tooth Pressure-Faces filing them
to the Link Page

: Visitors