Mechanics and Fixing Any Questions Answered

I'll admit that also have me stumped, only way I can diagnosed is to have the bike with me.

I may have an italian GXP at work, so I'll find it and measure it so you can compare it.

The other quick solution is to try an Italian Shimano hollowtech bottom bracket plus adapter on the NDS to run the GXP spindle.

It’s not a groove as to denote a left hand thread?

I think you were on the right track, because GXP bearings are handed, but Italian threads aren't, it's useful to have the drive side cup clearly identified.

fixed axle length

...ish. GXP is usually tight well before it captures the NDS bearing. I bet it would work OK 90% of the time if you put a 2mm spacer between the shoulder on the axle and the NDS bearing 🙂

Ah yeah I guess so.

Would that mean the chainline would be different depending which bb you were using though? Probably not by a significant enough amount to be meaningful.

Are you saying that 2x DS cups might have been supplied?

DS and NDS are the same threading for Italian BB no?

Same threading but as Tester points out, for GXP there's a 22mm id bearing on the nds and a 24mm id one on the ds. The groove (which is only in the ds cup) would denote the 24mm id bearing.

Yeah, same threading, but slightly different bearing size.

Don't GXP BBs have the same bearing in each side, with a little lipped sleeve in the left one?

My gravel bike keeps developing some play in the headset after rides - I've tried everything I can think of but no success. Any ideas? So far I have tried different expander plugs (longer, more textured), ensured there's enough space above the stem to preload, torqued everything, tried different stems... at a bit of a loss now!

torqued everything

Including the preload screw? I think there's some benefit in giving that 5Nm after you've tightened the steerer clamp screws of the stem.

Will give that a shot, thanks. Torqued the expander plug to recommended 5nm, stem is all torqued appropriately, hadn't really occurred to me that the top cap / preload would contribute but worth a try!

hadn't really occurred to me that the top cap / preload would contribute but worth a try!

If you don't do it, the preload screw only has about 1-2Nm, which isn't enough to stop it rattling loose. Once that's loose, the stem walks up the steerer to take up the space. That's the hypothesis anyway, and I've had good luck with it.

How about the crown race? On properly? Sometimes it slowly works it's way down if not on properly. Or even the bearings in properly.

Me again with the 4700 crank questions - just received the crank and think the chainring teeth are shot (shame on me for not realising earlier). No worry I thinks, just find a replacement set given the number of 5800 and 6800 spares there must be. Then I notice that Shimano says 5800 rings aren't compatible with the 4700 cranks, even though BCD and spider design seems to be identical and every other bit seems to be interchangeable; what gives?

All of those cranks have unique sculpting where the arms meet the big chainring. Some combinations will physically fit (but look wrong), but Shimano doesn’t endorse anything but an exact match.

Have you actually found original 5800 or 6800 chainrings for reasonable money? I couldn’t last time I looked. Your best bet might be flat rings from Ali Express etc.

Shims on top of the split ring at the top bearing. Make sure the outer lip of the bearing cover isn't contacting the frame before contacting the split ring.

Make sure the outer lip of the bearing cover isn't contacting the frame before contacting the split ring.

I'm guessing he's not an idiot, you can feel that when you waggle the bars about on the assembly bench.

crown race? On properly? Sometimes it slowly works it's way down if not on properly. Or even the bearings in properly.

That might open up the clearances once, maybe twice, but not repeatedly as implied by the OP

Next time it happens, check if any space has opened up between the head of the plug and the top of the steerer. That's still the most likely cause.

Patient searching of the classifieds here often turns up quite a lot given how quickly people (myself included) decide they need longer/shorter cranks or larger/smaller rings and then get rid of the old stuff! We're all for the circular economy here on LFGSS.

But just to be sure, does anyone know if 5800/6800 rings will physically fit 4700 cranks, and if they sit flat am I ok to assume that's all that's needed and a mismatch in the 3d part isn't going to be structurally relevant? Annoys the hell out of me that cranks that all look (on the surface) to be identical have mm-level differences ): On that note, although the 5800 cranks don't fit perfectly with the 6800 rings, Shimano indicates those can be swapped? Wondering if it's the 10-11 speed difference.. although those shouldn't affect chainring widths or BCDs?

Yes. On multiple bikes. Which was why I replaced all of them with hex bolts back before TCR4 or something. Probably posted about it on here. You should pay more attention. :P

As to the fix. Think one bike I just used pliers and gave no fucks the other bike needed to be drilled out.

Y-8JZ05000
Y8JZ98010

You're holding the button down too long.

1 billabong 2 billabong 3 billabong should be enough.
(I memorised the timing so I could do it while riding)

check if any space has opened up between the head of the plug and the top of the steerer

If it's moving, the space would be closing up

I had this with my Kinesis 4SD during ultras. I moved to using longer, knurled expander bungs from aftermarket Italian supplier I think (Deda?) and I torqued everything properly (probably a little bit over because it was a metal steerer and fuck it), including the top bolt after the side bolts were done.