Shimano Crankset recall: bonded Hollowtech ones

[fossyant]

You only get replacement cranks IF they are showing signs of de-lamination (i.e. falling to bits).

 

[wafter]

The 105 cranks of this vintage (5800) aren't glued at all. I believe the square section of the crank arm is a one piece extrusion and the end is forged over to close it off. Presumably the 'spider' bit of the crank is solid, where in 6800 the two piece design allows for more of the crank to be hollow.

Tiagra and below were solid with the back side of the arm scalloped out

I'm affected - I have a set marked MC. I'm sure mine will be fine as whilst I'm heavy and relatively powerful, I pretty much never stand, so surely the peak force going through the cranks will be low compared to many.

I'll still get mine checked, as this will ensure that if I have a problem after they've been checked, Shimano will almost certainly have a big problem on their hands.

In some ways this doesn't feel very well thought out. Surely any Shimano dealer who inspects and passes any of these cranks is exposing themselves to an awful lot of risk?

Thanks, but there is an overlap in availabilty and format - the recall affects Ultegra 6800 (2013-2017) and R8000 (2017-2021), plus Dura Ace 9000 (2012-2016) and 9100 (2016-2021) apparently manufactured before July 2019. By contrast 105 5800 ran from 2014-2018, and R7000 from 2018 to 2023... Sauce - Wikipedia

While as you say the 5800 crankset is different (and arguably nicer), the R7000 has ostensibly the same construction as the affected cranks above and there is an overlap in its availability with the recall period for the other two. Maybe when they were developing the R7000 the process changed and was then later applied to the two higher-end groupsets, since the recall suggests these would be OK after July 2019...

You might want to do a bit of research on yours - there's a lot of stuff on the net and I think some were suggesting it was a corrosion issue in wet / humid climates that was causing the adhesive / bonded surfaces to fail; so it's not just about outright loading.

I agree about the liability thing - would be interested to know how they plan to inspect them / what their criteria are, too...

 

[PapaZita]

So, if they’re inspected once and found to be OK, presumably there then needs to be a plan for periodic re-inspection?

 

[DCLane]

Be quick peeps: supplies of replacements are severely limited. There's probably 2 million+ of these 'recall' chainsets in use.
Once you've had the inspection and confirmation it's toast, are you going to climb on that chainset?
Have a standby solution in place.

If mine fail then it's retro bikes ...

Not that I expect them all to fail, but some may

 

[lazybloke]

The 105 cranks of this vintage (5800) aren't glued at all. I believe the square section of the crank arm is a one piece extrusion and the end is forged over to close it off. Presumably the 'spider' bit of the crank is solid, where in 6800 the two piece design allows for more of the crank to be hollow.

If I was listening properly to the Hambini video, he recommended 105 because the parts were "welded" instead of bonded.
Not sure which version of 105 that was.

 

[faster]

Thanks, but there is an overlap in availabilty and format - the recall affects Ultegra 6800 (2013-2017) and R8000 (2017-2021), plus Dura Ace 9000 (2012-2016) and 9100 (2016-2021) apparently manufactured before July 2019. By contrast 105 5800 ran from 2014-2018, and R7000 from 2018 to 2023... Sauce - Wikipedia

While as you say the 5800 crankset is different (and arguably nicer), the R7000 has ostensibly the same construction as the affected cranks above and there is an overlap in its availability with the recall period for the other two. Maybe when they were developing the R7000 the process changed and was then later applied to the two higher-end groupsets, since the recall suggests these would be OK after July 2019...

You might want to do a bit of research on yours - there's a lot of stuff on the net and I think some were suggesting it was a corrosion issue in wet / humid climates that was causing the adhesive / bonded surfaces to fail; so it's not just about outright loading.

I agree about the liability thing - would be interested to know how they plan to inspect them / what their criteria are, too...

Thanks for that, i'd assumed R7000 was later than that.

I've had a quick Google and I found it difficult to confirm how R7000 is constructed, but it looks like they don't often fail.

Also from googling, looks like I misremembered how 5800 is made. The hollow bit of the crank arm is cast around a former (rather than being extruded) before the end is forged over. As such, the spider must be mostly solid - otherwise they'd never get the former out.

 

[wafter]

Thanks for that, i'd assumed R7000 was later than that.

I've had a quick Google and I found it difficult to confirm how R7000 is constructed, but it looks like they don't often fail.

Also from googling, looks like I misremembered how 5800 is made. The hollow bit of the crank arm is cast around a former (rather than being extruded) before the end is forged over. As such, the spider must be mostly solid - otherwise they'd never get the former out.

No worries and cheers - I did a bit of research when this first started being talked about and couldn't find any evidence of R7000 failures. I think I did find some stuff on the differences / construction but the particulars escape me now.

Good news for R5800 owners too

 

[Bollo]

Reckon I might have two bikes in the frame, TT bike and turbo/CC tourer. Away now but will check when I’m back.

Hang on a minute....you've got friends?

 

[Ming the Merciless]

Shimano don't have a good record on cranksets. First Octalink and now this.

Got some XTR Hollowtech from 2004 that are still going strong.

 

[Solocle]

I had one fall apart and warranty denied as "crash damage". Probably a good time to chase up Shimano.

 

[Deleted member 35268]

 

[Jameshow]

I had one fall apart and warranty denied as "crash damage". Probably a good time to chase up Shimano.

View attachment 707390

View attachment 707391

Did you crash?

 

[Solocle]

Did you crash?

I had had a crash prior, but the bike was fully checked out and repaired professionally (on the responsible party's insurance), months prior. It wasn't a heavy impact, it was a slide. So I could just see both denying responsibility.

The scrape on the crank which is probably what led to that conclusion actually is from my shoe repeatedly rubbing against it...

 

[Ajax Bay]

Definitely shoe rub

 

[wafter]

Hambini's on it again; and absolutely on-point as usual (swearyness-averse look away now):

Really lays bare what an enormous, disgraceful situation this is.

If I was listening properly to the Hambini video, he recommended 105 because the parts were "welded" instead of bonded.
Not sure which version of 105 that was.

Yes, he confirms this in the video above. I'd assume it's R7000 as it's the same form factor; but evidently just a different bonding / fusing method. Nice short comment in the vid about the cost of perpetually chasing light weight.

For reference masses of the affected (complete) chainsets below courtesy of Jamesthebikeguy on youtube:
105 R7000 - "welded" - 705g
Ultegra R8000 - bonded - 685g
Dura-Ace R9100 - bonded - 629g

So, looking at the two extremes of 105 and Dura-Ace, you're looking at a saving of all of around 10% in exchange for a massive increase in both cost and the liklihood of coming back from your ride with a face like a slab of corned beef...