The DIY Butcher
Bearing Swap - 2015 Juliana Roubion 1st Gen. (SC Bronson)
I'd like to think I approached this task with more humility, after botching both the swap on my Yeti SB150 and then, bursting with confidence from that failure, doing even worse with the Santa Cruz Tallboy. I realize in retrospect that I was pretty sure the mishaps I encountered in these two outings taught me enough to think a little more critically and to be a little more careful. Unfortunately, I think I was so proud of this approach that I was once again over-confident. Or that's what I'm blaming for my three bodged removal efforts. In the end I was rescued but if I'd gone back and read the comments beneath my my Tallboy article, I would have remembered the suggestion by Hbar and avoided the stress that comes with shitting the bed and only removing part of a bearing. Three times.
Before I begin, I should confess that I have been a bad mountain bike husband. While my wife's bike got a complete overhaul perhaps 3 years ago, I haven't been attentive enough in the interim. Beyond that, she was actually overdue for a new bike, and while I had planned to make that happen, I waited too long and things were so backed up by then, nobody would even take my order. My only hope was to completely rebuild the Roubion and make it run better than new. She's been riding the bike since 2015, and she rides a lot. I had already checked to see that her drivetrain, which is perhaps 3-years old, was completely pooched, but until last weekend it was, remarkably, not skipping. And it was only a 10-42 10-spd. She can turn that gear up our local climbing trail without any trouble, but the possibility of a bailout gear for long rides has been mentioned.
The worst bits however were her brakes and her bearings, which were both original. The XTR brakesets of that generation, I later learned, were subject to warranty replacement for some engineering defect. In my experience, they didn't hold a bleed, wandered like a bloodhound on a scent, and recently it became almost impossible to push fluid through the front line. They were awful. As for the bearings, I'd noticed some play in the rear end recently that I'd mistakenly diagnosed as a loose hub, but it turned out to be the lower link bearings. I subsequently learned, while I was researching the bearing swap, that this generation of Roubion or Bronson uses angular contact bearings, so I could have simply tightened the collets to solve this problem. The bike's handling might have been a little imprecise as a result of this neglect. The bearing swap was step one in my effort to become a better MTB hubby, but the drivetrain and brakes will be replaced shortly as well, with actual working examples.
I ordered replacement bearings through the Santa Cruz website and they arrived in around a month, which I consider a victory in the time of COVID. This service was absolutely free, including shipping, and high quality Enduro Bearings, identical to the originals, were sent. The four bearings in the lower link are the aforementioned angular contact bearings, with no shields on the inside so grease can be added using the Zerk fittings and the supplied Santa Cruz injector. I chose to remove these bearings first. Fate, instead, chose to piss on these plans.
As I inspected the bearings in the lower link, I was surprised to find them spinning quite freely, and relatively smoothly, They weren't perfect, but if I'd tightened the collets and injected grease, they probably would have worked fine for another couple of years, which is even more impressive when you consider they sit below the bottom bracket of the bike and we live in a rain forest. Beyond the shield on the bearings themselves, Santa Cruz added an additional rubber seal and an aluminum cap to keep muck out. I would have been less surprised if the bearings had fallen apart upon disassembly that I was to find them working so well.
In my previous article, I called the bearing removal tool that involves an expandable collet, and expanding collet tool, but I've since learned it's actually called a blind hole puller, which is a much better name. It's also what I decided to use to remove the bearings in the lower link. While this plan wasn't entirely flawed, but my execution was. And I could have made the process very easy if I'd listened to the suggestion I mentioned above, and heated up the link before beginning the removal. Instead, I inserted the blind puller, but didn't fully expand the collet, and successfully extracted everything except the outer race, which remained in place as if cemented there. I didn't panic immediately however, it seemed to happen even faster than that. What if I'd ruined the entire frame? Was that link even available? I vainly tried to find a solution but there was no way to get a tool in from the other side, because of the narrow diameter between the bearings, designed to push the grease outward toward the open ends of the bearings, and there wasn't room to pry it out either.
Daunted, but still dumb enough to continue, I attempted the next bearing, only this time I expanded the puller as much as possible and the bearing popped out nicely. The next two did as well, so it was only the outer race of my first bearing that appeared to be permanently bonded in place. I sent a callout for advice and Deniz Merdano recommended I arm myself with beer and go see Topher, a mechanic who works downstairs at Obsession:Bikes, and who also appears to live down there. Topher was entirely undaunted by the challenge I'd presented, and as he pulled out the blow torch, I remembered the wisdom from Hbar; heat up the link and the bearings will pop right out. Or pop back in. In the end Topher needed another blind puller, in a size I didn't possess, but once that was deployed it did indeed pop right out, and I got to have beers downstairs in a bike shop, which is pretty much the best. Thanks Topher.
Pressing the replacement angular contact bearings into the lower link went smoothly, and I had the right size drifts to allow me to press matching pairs in together while only pressing on the outer races, unlike my successful, but rightly criticized efforts with the SB150. I smeared a little grease in first and then, realizing this was an opportunity, used the grease injector and watched fresh lube coil into the void between bearings where the collet axle would eventually reside.
Next it was time to tackle the upper link, with my newfound know-how. Unfortunately I only used some of it. Instead of placing the link in the oven to bake, or on the rotisserie in the barbeque, or buying a blowtorch,* I figured my "expand the tool fully" strategy would continue working. Instead it failed, and this time I managed to leave two outer races in the link. A quick examination made it clear I could get a tool behind the races and attempt to pry them out. Unfortunately the leverage was being provided by the other edge of the race, meaning I was pressing one side in as I pressed the other out, with a screwdriver. If you'd asked me to honestly assess my prospects of success I think I would have said about 1 in 20, but I patiently levered around one half of the first bearing until it popped out. I checked immediately to see if my lopsided removal had damaged the bearing seat, but it was pristine. My next effort was similarly successful, and I avoided the shame of returning to the Obsession basement, but I also missed out on more beers, laughs and wisdom from the basement dwellers.
*None of these approaches are recommended in Santa Cruz's literature
The larger of the two bearing sets on the upper link were the most pristine of the bunch, and I'll be keeping them as spares. The final small set, which came out easily with a smaller blind puller, were a little rough. They would still turn but their time had come. I cracked open a celebratory stout and revelled in the joy of a successfully completed task. And then I began to panic realizing I had a set of brakes to install and bleed and a drivetrain to bolt on and set up, all with the weekend rapidly approaching.
Reinstalling the links was actually a little tricker than I expected, but I managed to muddle through despite not finding Santa Cruz's detailed instructions until the end. (I used Santa Cruz's website search bar at first and then, after switching to google, found more info). Instead I was using the exploded view document to reference torque numbers. I soon realized this was a flawed strategy. Once I converted the foot-lbs units to Newton metres, I torqued down the collet axles to spec. Without the shock attached it was clear this approach would cause the bearings to bind. Instead I ditched the torque wrench and tightened the axles until the point where there was no play but no binding, like an old school headset. After that I torqued the bolt that holds the wedge in place to the recommended reading, since that will keep things from moving. Santa Cruz recommends greasing the axles and a drizzle of Lockite (I used blue) on every one of these bolts. I will check the axles for play after the first rides to see how they are doing.
While I had the frame disassembled, I used the opportunity to wash, inspect and polish up the frame. As the grime was removed, what I discovered underneath was paint in excellent condition, and virtually all the blemishes I found were so cosmetic they easily buffed out. From the bearings, to the seals, to the fittings and logos and word marks on this 6-year-old, extremely well ridden machine, there was nothing about the finish, quality, engineering and construction that failed to impress me. Santa Cruz builds a hell of a bike.
The company also stands behind their bikes and guarantees spare parts availability for at least 10 years. There are a few gaps in the current inventory, but you can buy a lower link for this bike, with bearings already pressed in, for a mere 53 USD. As luck would have it, and likely COVID, that particular piece is currently out of stock, but you can bet supply will be replenished once the backlog is dealt with. Unfortunately that remains a moving target.