primoz-resman profile

With bearings you can pull them out and measure them, unless it's a custom part (looking at DT Swiss and the 1526 bearing in the 1501 gen2 wheelset...). Custom parts are another thing. I really don't like Shimano brakes since if anything goes wrong, you replace a caliper or a complete lever. Well don't even want to try them, mostly because of their lack of modulation and wandering bite point, but this serviceability is a big part of it all. Sram is better in this regard. But not great, they've dropped the ball when it comes to suspension spare parts compared to a few years ago. But at least they have service manuals!

I hear mine when I have earphones in on the way up and ride over technical stuff - sounds like the adjustor on the back of the head rattles a bit.

Asked where I cna find replacement velcro pads to fit the liner, same deal, got a bunch of them in the mail.

Think outside the box, don't use a planetary reduction system (that is easily shifted back and forth) to add another block of range. Use it to fill up the gaps in the range caused by wider spacing of the sprockets in the rear.

So... 'How to modify an 11spd cassette to throw away the big ring and get what is essentially an 11-36T 10spd cassette'?

Pinion?

Anywho, the Hammerschmidt problem was that the chainring was very small and nobody made a bike to work with that, so the antisquat values were off...

Instead of eating lunch I ran some more excel numbers again.

7 cogs on the cassette (10-13-17-22-28-36-47) with a 0,88 reduction in the hub give a 534 range and an average jump between gears 13,75 %. With the largest jump just over 15 % and the lowest jump just over 12 %. Standard deviation of jumps (how far each jump goes from the average) is 0,7 %, while the same value for Sram is 2,3 % for the 10-50 cassette, 3,2 % for the 10-52T cassette (that 52T jump kills it...) and 2,1 % for Shimano (so Shimano made a 'smoother' cassette with more range than Sram's 10-50T cassette).

Graphs: https://i.imgur.com/EiYlCx6.png

While there are a few good points in this, manual changing of the chainrings in the front is something that would not fly for the masses. The other no-no is the big jumps between gears. 10T cassettes have the jumps just a bit larger than the 11T smallest cog cassettes (because 11-13 is less than 20 % of a jump, while 10-12 is in fact a 20 % jump), Sram's 52T cassette is stupid in this regard (a 24 % jump...) and if anything, smaller jumps would be beneficial. Old cassettes (9spd and 10spd) weren't any different in this regard on MTBs as we've more or less just added cogs on the larger side and kept the smaller cogs mostly the same (except the 10T part and all, but the differences there are small).

There is a way to have your cake and eat it too, mostly. Unsprung weight would be a small negative (but yeah, when you have 1+ kg tyres, a few 100 grams of inserts and the like, does it matter THAT much) and you'd more or less need electronic shifting, but, you would have only 6 or 7 cogs, better chainline, all the range we have now or more, smooth steps between gears and no weird machinations with the chain and sprockets.

How to get there? Pair a Classified internal hub with a wide range, large step cassette, but use the internal hub for the inbetween gears. So you shift the hub every time you shift, not like we used the front derailleur.

The cassette with 6 cogs would be a 10-14-19-26-35-49 with a 0,85 reduction ratio in the planetary hub, giving the inbetweener gears of effectively 11,8-16,52-22,42-20,68-41,3-57,82. That would give a whopping 578 % of range, which could be tightened down to a smaller range and give even smaller steps between gear. With the numbers above the steps are between 15 and 20 % in all but one gear change, as shown here (compared to Shimano and Sram 12-speed drivetrains): https://i.imgur.com/9w9JyaT.png

Going to 7 cogs could of course improve things even further. But yeah, electronic would be the way to go, at least as an adapter box with an actuator driving the cable for a standard derailleur (ala Archer components setup).

This is what I was talking about, I didn't know it exists already. Makes sense though. And yeah, expensive as hell, as expected.

I'd try to make it single sided only though for that true blind bearing removal capability (think about RM Slayer chainstay pivot bearings for example).

One point here is that RWC tools are designed to pull bearings out of mounts where you can get behind a bearing. If you have two bearings with a spacer pressed into your frame, into a hub or somewhere similar, heck, if you have two bearings in your chainstay pivot, the RWC kit wont help you.

It's designed to push the bearing out using a block on the back side, that is pulled out into a cup, covering the bearing hole. If you have a blind hole, a spacer with another bearing on the other side, etc., you can't install said block. The easiest way is to use a bearing puller (an expanding collet grabbing the bearing with a slide hammer to knock it out), but a screw type puller on the same principle could also be used. I'm not sure anyone sells those though, it also sounds quite expensive to make.

I really need a lathe in my life...

EDIT: slide hammer puller work great, but ideally you're pulling the bearing out from a heavy part. Like an engine block, electric motor housing, transmission casing, etc. With bike frames, we're talking about light parts. You can't really clamp them into anything (you'll bend the frame easily by just clamping it into a vice or at least when hammering away), so someone has to hold the frame or swingarm. Usually it's hard to do it by someone on their own. Using a screw-type blind bearing puller would mean there would still be a cup sitting on the frame right around the bearing, pressing nicely where you have the most material. This means a quick, easy, quiet bearing removal.