This is Part 1 of a series of posts about replacing the fork bearings on a TiLite Aero wheelchair.
In this post, I only describe the “teardown” part of the process. That is, getting the forks off the chair. The removal and replacement of the bearings is for Part 2.
If you didn’t realize this repair might involve a complicated “teardown” step, and you were thinking of doing this repair yourself, then this post has done its job.
On this particular chair I ran into a worst-case scenario: The steel fork bearings had rusted solidly to the steel axles that they spin around. This stops you from removing the forks from the chair, which you need to do, in order to get to the fork bearings. Your choices are a) replace a few hundred dollars of wheelchair hardware, or b) break the bearings free from the steel axle that the bearing races are rusted to.
This step took several rounds of heating the axles with a propane torch, spraying with lube, then pounding and prying until the rusted-on bearings broke loose.
Edit: You can see an alternative way to beat on the axle in this reference. There, the user removed the fork axle from its fitting first (i.e., took the fork axle off the wheelchair, fork and all), then beat the axle out of the fork. That’s arguably a smarter approach than what I did. At the minimum, it shows that I’m not the only one have the problem of fork bearings that rusted solidly to the axle they sit on.
Other than spending a couple of hours doing that, the repair went smoothly.
The only practical takeaway is that before you buy new bearings, bearing puller, bearing press, and so on — first try to remove your forks from the wheelchair.
If they come off readily — once you have removed any retaining hardware — move on to the next post, where I talk about options for replacement bearings, in some detail.
But if the forks don’t come off, even with a bit of lubricant and some gentle persuasion, then ponder just how hard you are willing to hammer on a wheelchair.
My lesson is that, even thought this repair eventually succeeded, I got lucky. With those fork bearing races rusted to the axle, it could just as easily have ended up with an unusable wheelchair, and a few hundred dollars plus a wait for replacement forks and fork axle assemblies.
N.B., I don’t use a wheelchair. I did this repair for a friend who does. I’m writing it up for benefit of anyone thinking about doing a similar wheelchair repair themselves.
Fork bearing replacement: What are we talking about?
Image source: Spinlife.com. The chair shown is a newer version of what I’m working on, but the part more-or-less identical to what’s circled above.
This post is about replacing the fork bearings on this wheelchair.
As shown above, the forks hold the front (caster) wheels of the chair. The fork bearings — two in each fork, inside the area circled in red — allow the forks to rotate freely. Each fork rotates around a little stubby downward pointing axle that is solidly attached to the chair frame.
The free rotation of the forks is what allows the user to steer the chair. And conversely, if those bearings lock up, the chair becomes unusable.
Source: Southwestmedical.com
Here’s the “bearing stack”, laid out, after I (finally) removed it from the fork axle.
I should clarify that “fork bearings” are not bearings specially manufactured to be used in the forks. The “fork bearings” are the generic general-purpose small bearings that happen to have been used on this fork. This chair, for example, uses the exact same generic bearings in both the hubs of the spoked wheels, and in the forks.
A bit more detail
As I went through this repair, I noted that the top bearing (of the two in each fork) takes most of the abuse. Both from water, grit, and grime. And from the downward load of the weight of chair and user..
First, the bearing housing — the round recess in the fork, where the bearings fit — is not sealed at the top. There is, by design, a little open gap between the outer rim of the cap (left, picture above) and fork (right, picture above). That’s because those two surfaces have to rotate, relative to one another. Water or dirt that gets into the housing through that gap immediately lands on the top (left-hand, above) bearing.
Separately, all the vertical force from the chair through the fork to the ground is transmitted as an axial (or lateral) loading, on the center race of the top bearing. The wheelchair frame bears solidly on the center of the cap. The center of the cap, in turn, bears solidly on the inner bearing race of the top bearing. Meanwhile, the outer race of that bearing is pinned in place by being tightly friction-fit within the fork, and by a c-clip that locks it in place, within the fork.
In effect, the entire time these are in use, the weight of user plus chair is trying to punch the center out of the top bearing. The weight is borne along the axis of that round bearing.
I look into this a bit in Part 2, regarding allowable axial loadings for thin steel bearings. And, FWIW, I guesstimate that a safe axial load on these bearings — one that won’t prematurely grind them to dust — is around 185 pounds, for each fork. Given that almost all the users weight is borne by the spoked wheel (and borne as a radial load on the spoked wheel hub bearings), a combined allowable 370 pound load on the two fork bearings combines seems far more than adequate.
I should also point out that the center of the top bearing is unsupported. The c-clip spaces the top and bottom bearings apart, just a bit. This is why the entire downward load on the forks is borne by just the top one, of the two bearings in the fork.
Between the potential for water intrusion via the un-sealed end cap, and side-loading the bearing, I was unsurprised to find that the top bearing in each pair was more worn than the bottom. And that the bearing that gave me so much trouble — being rusted solidly to the fork axle — was the top bearing.
Removing the fork-retaining hardware was the easy part.
It was easy for this particular chair. That doesn’t mean it will be easy on all chairs.
I’m not going to document this part, except to assure you that, on this chair, once you remove any retaining hardware (in this case, the washer and locknut that threads onto the end of the fork axle), the forks should just slide right off the axle.
If they don’t, you might want to give them a spritz of some lubricant such as Liquid Wrench, WD-40, or just light oil. Pull and bang on the fork a bit. It’s a pretty tough piece of metal. See if you can’t knock it loose with a little percussive maintenance.
On this chair, removing the 14mm retaining lock nut was ridiculously easy. It wasn’t torqued down. Only upon re-assembly did I realize that nut is only supposed to be snug enough to prevent the fork from rattling. If you torque it down hard, it locks the fork in place.
The upshot is that when you go to remove it, you’ll find that the retaining nut is not very tight. And, accordingly, when you re-assemble that, you’ll only tighten that nut up enough to prevent the fork from rattling when you bang on it with the palm of your hand. The retaining nut should be lose enough that the forks spin freely and do not bind on the retaining nut.
But, I’ve tried this on one other chair, and the very different fork-retaining hardware totally defeated me. So the ease or difficulty of this step will depend on how the chair in question was engineered.
In any case, assuming you can get any fork-retaining hardware off, nothing should prevent the fork from sliding off the axle. If you are so fortunate as to have those forks slip right off the axles, skip on to Part 2, next post, discussion of bearings.
So, I was about four hours into the first five minutes of the repair …
… when it dawned on me that this was a lot harder than most internet discussion let on.
As it turns out, this titanium wheelchair uses steel bearings on a steel axle, with just about zero clearance between the inner bearing race and the axle that passes through it.
Under this sort of zero-clearance metal-on-metal fit, even a smidge of rust can stick those two parts together hard. So, while the forks would still rotate (some), they wouldn’t slide off the axles, on account of the rusted-in-place bearings.
It is difficult to convey to just exactly how screwed your are, if this happens. No “normal” methods for dealing with rusty metal equipment can be applied. There’s no way to grab onto the bearing to pull it. There’s nothing solid to leverage off. The geometry of the fork prevents using a standard bearing puller to pop the entire assembly (fork and bearings) off the axle.
And you don’t want to take a sledgehammer to it.
Well, that’s not actually true. I dearly wanted to take a sledgehammer to it after, say, the first two hours of trying to coax it off nicely. What I mean is, if you are rational, you will refrain from beating the snot out of that stuck fork, for fear of breaking some expensive titanium parts, or worse, damaging the chair frame. Fully acknowledging that it would be satisfying to do so.
If there is any good way to deal with this situation, it was not obvious to me. I’m no master mechanic, but I’ve worked on more than a few old cars, bicycles, and appliances. The absolute intractability of this situation was a new one on me. I’m willing to bet that the people who designed this chair did not plan for the bearings to rust in place.
Luckily, my thousands of hours of watching rust-belt auto mechanics on YouTube finally paid a dividend. I ended up taking a propane torch to the axle and bearings. I’d seen it work for car mechanics in the salt belt, I figured it was worth a shot here. A few cycles of heating, lubing, beating and prying finally broke it the last stuck bearing loose.
I also ended up using two knife blades as wedges, sliding the edges between fork and fork cap (parts 2 and 11 in the diagram above). Then pounding them in. That, to try to get the rusted fork bearing to slide along the axle.
That highly dubious method, plus the more standard heating/lubing/pounding, eventually broke the bearing loose. I dinged up one cap badly — contrast the two caps in the photo above. But that’s a relatively cheap part to replace. And works mechanically, as-is.
And there was literally nothing else to leverage off of. That I could figure out.
In summary, for this repair, the instruction to “slide the forks off the frame” turned into three hours of escalating violence on those stuck fork bearings. Which, by luck and patience, culminated in the rusted bearings breaking loose. Instead of my breaking the forks.
A mere stuck fork is not the worst case scenario.
The moral of this story is that if you can’t get both forks off the chair, with a little gentle persuasion, ponder whether or not you want to leave this repair to a professional.
I did eventually free the forks. Patience and a low value of your time are keys to success. Along with luck. And having (e.g.) a torch on hand.
But, in the interest of full disclosure, even though I succeeded, I should mention that:
One possible outcome is that the bearings actually break, which means you’ll end up having to cut (grind) the inner bearing race off the axle. Ideally without messing up either the axle or the bearing cap. I was already planning how I would use a cutting wheel on a Dremel tool, when the bearing finally moved. If you don’t have a small grinder such as a Dremel tool, grinding those hard steel bearing races off the steel axle would a challenge.
There’s nothing good to anchor to or pry on, in trying to get that fork off. No safe place to put a pry bar, no place where a standard bearing puller would stay in place, and so on. As noted above, in the end, out of desperation, I used knives to wedge apart the fork and bearing cap (parts 2 and 11 in the diagram), fully realizing the cap was nowhere near strong enough to take that undamaged.
C’est la guerre. In the end, dinging up that small replaceable part was a small price to pay for getting that bearing off.
Finally, you will pass a point of no return, if that bearing is well and truly stuck. You’ll end up at a point where you have a broken, unusable bearing that’s still stuck to the axle. For example, the rubber seals and grease of the bearing won’t survive being heated by a propane torch.
At that point, you kinda haveta keep going, no matter what. Or give up and purchase a few hundred dollars in replacement fork and fork axle.
So, if you try this, and the fork is well and truly stuck, decide whether or not to forge ahead with your eyes open.
Conclusion to Part 1
Ultimately, the moral of the story is that rust never sleeps.
You might think, hey, the bearing is a cheap disposable part, what difference does it make if I let it really wear out, rather than replace it on a regular schedule? It’s going into the trash either way.
Now you know. Rust never sleeps. Three hours. Propane torch. Repeated rounds. And then it was iffy.
If your forks are already stuck, just be aware of what you might be getting into.
But equally, when’s the last time you changed those fork bearings?
I’m willing to say that waiting until the bearings are obvious worn — until the forks get hard to pivot — is too long.
Forks off? Go on to part 2, a discussion of wheelchair bearings.
