Can the bearing in a Camso bogey wheel be replaced on its own?

It depends which system, and Camso's own two ATV pages say different things. The X4S page describes standard and replaceable sealed bearings; the T4S page describes a moulded bearing inside the wheel. On the bench both turn out to be true of the hardware they describe: some single-bearing quad wheels have the bearing injected into the wheel and come apart no further, so a worn one is a wheel to buy and not a bearing. The double-bearing wheels take two replaceable bearings. Check the system you actually have rather than the other page.

How much side-to-side play means a bogey wheel bearing is finished?

No manufacturer in view publishes a play limit, so there is no figure to hold a wheel against. Working from the bench rather than from a specification: on the large rubber wheels and the wider solid wheels, more than about a quarter of an inch of side-to-side movement has been the point at which both bearings in that wheel get replaced. Release the track tension before judging it — tension pushes the wheel against its play and makes a worn bearing feel tight.

Why will a bogey wheel puller not shift a wheel on a swivel bracket?

Because the swivel bracket on the older side-by-side assemblies carries a rubber bushing, and the puller has nothing solid to react against through it. The method used here instead is to take the whole swivel bracket assembly off the frame, hold it in a vice, and drive the wheel off the shaft — which also gets the work away from everything else the wheel is surrounded by.

Servicing a Bogey Wheel

Camso’s two pages disagree, and both are right

This site has published the contradiction for a while without being able to resolve it. What the parts are called records it against the mid-rollers, and the X4S page ends on it: a bearing you can press out is a bearing job, a bearing moulded into a wheel is a wheel, and nothing published says which one you own. The X4S page describes standard and replaceable sealed bearings. The T4S page, in the same position, describes a molded bearing inside wheels. Same manufacturer, same product family, two sentences that cannot both describe one part.

Taking them apart settles it, and the answer is that both sentences are accurate about different hardware. Some of the single-bearing quad bogey wheels have the bearing injected into the wheel during manufacture. There is no bearing in there to press out and no service operation that ends with the old wheel still on the machine. When it is finished you are buying a wheel. The double-bearing wheels — the X4S arrangement, and the wider wheels on the side-by-side systems — carry two bearings that come out and go back in, with a space between them that matters more than it looks like it should.

The practical consequence is an ordering one, and it is the reason the distinction is worth the paragraph: an owner who diagnoses a worn bearing on an injected-bearing wheel and orders a bearing has ordered the wrong part, and will find out on the day the machine is in pieces. Work out which wheel you have before the order goes in, not after.

One caution on reading your own system’s identity off the serial plate. The quad family in this generation is shared between two product names with only the model year separating them, which is the whole subject of a 6622 serial is ambiguous. Photograph the plate and let a parts desk resolve it rather than matching a prefix by eye.

How much play is too much

Start with what is not available. No manufacturer in view publishes a play limit, a wear limit or an hours-to-replacement figure for any part of these systems — inspecting a used system establishes that absence across the whole of the hardware, and it holds here too. There is no published number to hold a wheel against, and the table below is not one.

What the table is: the thresholds one bench works to, written down so they can be argued with. A reader who wants a specification should ask Camso for one, and if they get an answer this site would like to publish it.

Wheel	How it is checked here	What is done about it here
Single-bearing bogey wheels on the Camso quad systems	Rocked side to side by hand	1/8 to 1/4 in of movement, on its own, has not been treated as a reason to replace the wheel here
The large rubber bogey wheels on the Camso 4S1	Track tension released first, then rocked side to side	More than about 1/4 in: both bearings in that wheel replaced
The wider solid bogey wheels on the side-by-side systems	Pressure taken off that wheel with a bar under the track, then rocked — the track does not have to come off for this one	More than about 1/4 in: both bearings in that wheel replaced

Release the track tension before you judge any of it. This is the part most often skipped and it is the part that makes the check worth doing. A tensioned track pushes the wheel hard against whatever play it has, and a bearing that is well past serviceable will feel tight. Checking with the tension on does not give you a cautious reading; it gives you a wrong one.

On the wider solid wheels the track does not have to come all the way off. Taking the pressure off that one wheel with a bar or a large screwdriver under the track is enough to let it move, which makes this a check worth doing before a riding season rather than one that waits for a changeover.

The large rubber bogey wheels on the 4S1 are the ones to be least relaxed about, and the reason is not bearing cost. A wheel that comes off the shaft under way can take the track with it. That is a consequence out of proportion to the part, and it is why the threshold on that row is applied rather than noted.

Not shot yet. One unit, track tension released, the same wheel shot at rest and held over at the end of its play, from a fixed camera position with a rule in frame. Then the pair that settles the other half of the page: an injected-bearing wheel cut or shown open beside a double-bearing wheel with both bearings out and the cavity between them visible. The page asks a reader to judge a distance by hand and to tell two wheels apart by eye, and gives them nothing to compare either against.

Fig. 1. No such photograph exists. Filed as LW-BOGEY-01 on the shot list, which is the whole of what has been done about it.

Getting a seized wheel off

A wheel that has been on a shaft through several wet seasons without the cap coming off can be extremely difficult to move, and the damage usually happens to the thing doing the prying rather than to the wheel. Camso makes a puller for the job. It works on most of these wheels, it works especially well on the old stubborn ones, and it replaces an afternoon of hitting things with a controlled pull.

Take the track off first. The puller needs to sit square on the wheel, and with the belt on there is neither the access to position it nor the room to work it. ATVtracks.net publishes a video of the puller in use which is the clearest record of the sequence; nothing below repeats what is in it.

The assemblies the puller is the wrong tool for

There is one exception worth knowing before you buy the tool, and it is a whole class of assembly rather than an awkward individual wheel: the bogey wheels mounted on the swivel brackets. Those brackets carry a rubber bushing, and a puller reacting through a rubber bushing has nothing solid to pull against. It is not the right tool for that job, and persisting with it is how a bracket gets damaged.

In the serial reference read 2026-09-26, the families this applies to are the side-by-side ones: 6722 (2014–2016) and 6822 (2010–2013), both filed as side-by-side systems and both recorded under Camoplast rather than under the later brand. They are routinely called Camso systems in a shop, which is fair enough — it is the same company — but the badge on the hardware and the name in the parts index are the older one, and that is the name an order has to be placed under. The two Tatou generations is where that naming goes in full.

What is done here instead, in order:

Take the complete swivel bracket assembly off the track frame rather than working on it in place. Everything after this step depends on it.
Hold the bracket securely in a vice. This is the whole advantage of the method: the assembly is held, the work is at bench height, and nothing you are hitting is attached to anything you care about.
Drive the old wheel off the shaft. Controlled, square, and with the bracket supported — which is a different job from persuading the same wheel off while it is still bolted to a machine on a jack.

Oil, between the bearings and under the cap

This is the shortest section and the one most likely to be worth money to a reader, because it costs nothing and it is almost always skipped.

On a double-bearing wheel there is a space of roughly 3/4 in between the two bearings. It will hold oil. Filling it when the wheel goes back together is a few seconds of work, and the reasoning is the same reasoning that makes the bearings a wear item in the first place: these wheels run underwater, in grit, under a load that reverses every time the belt comes over them. Anything that keeps water out of that cavity is working on the failure mode that is actually present.

Put some oil under the wheel cap as well, both when a wheel is reassembled and — on the wheels that are retained with a bolt — about once a year as routine, with the cap off and a gear oil. Some of the quad bogey wheels are pressed on rather than bolted on; a small amount under the cap still reaches the area it needs to reach.

The second effect of the oil under the cap is the one you notice years later. A wheel that was put back wet comes off again. A wheel that was put back dry is the wheel at the top of this page, the one that needs the puller. The maintenance and the removal problem are the same problem, separated by about six seasons.

How much longer the oil makes them last is unmeasured

The practice above is worth doing on the reasoning given — it keeps water out of a cavity that fills with water, and it keeps a press fit from seizing. What it is not is a quantified benefit. No set here has been run against an hour meter with the oil and without it, so the honest form of the claim is that it is cheap, it addresses the failure mode these bearings actually have, and the service life it buys has not been measured by us or published by anybody.

Two wheels on one unit, one serviced with oil between the bearings and one without, run the same season on the same machine and opened at the same meter reading.

What the inside of a cap looks like after a wet season dry, and after a wet season oiled, photographed the same way.

The reading at which an oiled bearing was eventually replaced, recorded as an event rather than an impression.

Whether the easier removal holds on the pressed-on wheels as well as the bolted ones, which is the case the practice is weakest on.

What it would take. One machine, one hour meter, two seasons and the discipline to leave half the wheels alone — which is the hard part, because the whole argument for the practice is that you would not want to leave them alone. What it would produce is one data point on one machine, not a service interval.

What this is not

It is not a maintenance schedule. The schedules, intervals and specified values for these systems are in the manual that shipped with the kit and on the dealer’s maintenance page, which carries the manufacturer’s own per-area schedule. Nothing here replaces either, and where this page and a manual disagree the manual is the one to follow.

It is also one shop’s record rather than a survey. The thresholds in the table are what gets replaced here, on the machines that come through here, in this climate. A reader whose machine lives in dry sand is reading about a different failure rate from a page written where the wheels spend half their life in slush. The method — release the tension, rock the wheel, compare it with the other three on the same unit — transfers. The numbers are a starting point for an argument with a parts desk, not the end of one.

Manufacturer sentences were read 2026-09-26; serial families are derived from the published serial reference. Bench observations are dated where they appear and are one shop's record, not a specification. Where anything here disagrees with the manual that shipped with your system, the manual is the one to use.