DUB crank removal problems


I’m having imense trouble removing my DUB GXP cranks off a BMC Fourstroke (2019).
The bolt that should be tightend to 54Nm is just impossible to remove.

So far we’ve tried:

  • Putting in a normal 8mm hex wrench, putting a 1.5m long tube over this wrench and pulling down, while blocking the non-driveside pedal with some blocks of wood. Then 1 person hold down the bike, while the other pushes down on the tube/wrench. The thing doesn’t budge.
  • We also tried attaching an impact wrench, but this just starts to wear out the inside of the crank.

The crank has been removed before. But the last time it was installed, it was done by a bike shop.

Any tips would be greatly appreciated.

Are you turning it anticlockwise?

This is a shot of a GXP Stages crank. The original crank arm is similar.
I made the mistake of turning it the wrong way due to the “tight” arrow on the retainer cap. The head of the bolt sheared off without damaging the spindle.

Yes, I tried to turn anti-clockwise.
This is a normal threaded bolt, right? “Left” or anti-clockwise should be loosening the bolt?

My crank doesn’t have any singnage on the retainer cap. Just “54Nm” and an arrow pointing clockwise.

But i’m happy to hear that the bolt broke before the crank spindle did, because i was starting to worry that we were causing damage that would make it impossible to ever remove the crank.

PS: can post some pictures tomorrow

I’m running GXP on a 2018 model. I’m pretty sure the new DUB turns anticlockwise to loosen as well.
I guess the spindle was much stronger than the bolt in my case. The spindle on my cranks is steel vs the aluminium bolt.
$25 for a new bolt was a cheap lesson.

It looks like Sram went to an aluminium spindle when they moved from GXP to DUB. Some penetrating oil might go along way if you can get it in between the threads.

Best of luck getting it off cleanly.

I think it’s a fairly well documented issue, I’ve seen videos of shops using massive pieces of tubing for leverage to get it off.

Heat, but with a carbon arm you’ll need to be very careful. I wrestled off a set of seized SRAM cranks earlier this year, but I had to resort to a torch. These were older and aluminum though. Same principle applies but the obvious caveat is that you can permanently damage the crank arm so an open flame type of heat is probably a last resort. A heat gun might work and less likely to damage the crank, but you’ll still need to take care. Try the penetrating oils and leverage first.

Unfortunately this is a fairly common issue. But it’s also solvable - here’s what I do. Take your pedals off, get a toe strap, loop it through the pedal threads in the opposite crank arm, wrap it twice around the crank arm and chainstay, and tighten that sucker down. Then get the longest 8mm you have - anything longer than a regular L-key should do - and insert it into the extractor bolt so it’s level. Put the bike on the ground, and stand on the 8mm. It’ll make a loud bang when it comes loose. If you don’t have a toe strap you can brace the end of the crank arm on a block of wood. Just be careful, there’s a lot of force involved here.

A lot of this problem is caused by galling on the threads with aluminum-aluminum contact. A zinc-based anti-seize can help (the regular copper stuff actually doesn’t work very well on aluminum), but I think the best solution is to get a steel bolt. SRAM does make them (part 11.6118.066.000), and your local shop should be able to order them if they’re in stock.

Good luck!

1 Like

I take the pedal off the drive side, slide a piece of old innertube over the crank arm to protect it), then slide a piece of cheater bar (1-1/4" pipe, snug fit), then use a long 1/2" drive breaker bar with 8mm hex socket to break (loosen) the torque. I squeeze the breaker bar and crank arm together, and it loosens pretty easily. Usually a mallet is still needed to persuade the crank arm off. This seems to work on XX1 cranks. When I put it back together, lots of grease, and I torque it to less than the maximum value recommended.

Thanks for the tips everyone.

What we tried to do, is put the non-driveside crank and pedal on a car jack to support it.
Then heated up the crank with a heat gun (not super hot, but couldn’t touch the crank anymore).
1 person holds down the bike, while we use a 1.5m wrench to leverage it loose.

EDIT: also sprayed some penatrating oil inside the spindle that I left overnight.

But this still didn’t get the bolt loose.
While the inside of the bolt keeps looking worse.

I’m gonna hand it over to my bike shop to handle this.
Too afraid to damage the bike (that’s why i didn’t try putting toe wraps around the frame).

Picture of the bolt/crank:

1 Like

The method described in this video is what has worked for me with my crankset:

Short version: Remove the cap covering the bolt using lockring pliers. Clean everything. Spray tons of lubricating/degreasing oil there, hopefully so it penetrates the threads. Slap a bunch of grease on top of the bolt. Add the covering cap back. Hope the friction in the threads and the friction of the bolt on the covering cap is reduced enough that you can muscle the damn crankarm off.

Great article. Thanks for sharing.
And to think I was looking at changing my GXP (2018) over to DUB. I think I’ll stick to the steel spindle with aluminium bolt thanks. Paired with the Wheels Manufacturing thread together BB, the longevity has been good to date.

I get the feeling the bike shop is going to destroy the OP’s bike cranks anyway. I’d be interested to know if it’s the same bike shop that over tightened it without anti seize compound in the first place?

Every time I try to take the Dub crank off my MTB I end up having to stand one foot on an 18" ratchet driver and the other foot on the driveside pedal and shift my entire bodyweight up and down a little (locking out the suspension). I’m sure it’s dangerous but it seems to work.

Actually the one thing I find even harder is trying to remove a cassette from an XD drive freehub :hot_face:

Being the person on the other side of the counter here (I don’t work at the OP’s bike shop, to be clear) I’d like to clarify something quickly.

The crank bolt probably wasn’t over-tightened or applied without anti-seize. There is galling on the threads. This is something that happens with fine-pitch fasteners made of soft metals under high load. The regular anti-seize that most bike shops have doesn’t prevent it in this case, and even if they do have the proper stuff they would need to start over with a new bolt because the surface texture of the old one is damaged. These bolts are often installed with grease at the factory to who-knows-what torque, and should really be replaced during the bike build, which is crap and not really a reasonable expectation on the bike shop’s end. This is why I generally suggest the steel bolts for replacement.

That all being said, it may well still be possible to remove that bolt cleanly. But it is extremely important to use a very high quality hex bit (which is not something most home mechanics have) at these torques in order to prevent damage. And even if the bolt is damaged it it still possible to remove the bolt without destroying the crank. I’ve had to do it a few times. It’s time-consuming and therefore expensive, but probably cheaper than a new crank.

So what are valid solutions to prevent this from happening in the future?

  • Steel Bolt
  • New alloy bolt w/ zinc-based anti-seize and slow installation

More importantly though, these are not valid solutions, and anyone who suggests them probably shouldn’t be working on your bike:

  • Lower installation torque
  • Reinstalling the damaged alloy bolt w/ grease

I wish I had insight into the design decisions that went into this product and why this ended up being the crank fixing solution that SRAM settled on. I’m sure there is a reason, the engineers are a smart bunch. But being the guy doing the service, the best I can do is try to accurately identify root causes and work to mitigate them. DUB cranks, like many modern parts, are a good and robust product. But they also have their quirks and you must be careful to work around those in order to get the most out of them.


For whatever it’s worth, here is my process:

I use a long (longer than 1 foot), heavy-duty torque wrench (one-way, so I’m not damaging the torque function). I rotate the crane so that the crank arm is in line with the chain stay. I rotate the torque wrench so that it is also in line with the chain stay. I then simultaneously pull up with my right arm on the crank arm (I am right handed), and push down on the torque wrench with my left hand. This way I am simultaneously pushing/pulling in a counter-clockwise direction.

While it is initially very tight, I hit a point where the bolt “breaks loose”, and then I can proceed to loosen normally.

Is there a service technical service bulletin indicating to not follow the published instructions? SRAM clearly call out grease in the pictograms, and never any “zinc-based anti-seize” compound.

There is not. Unfortunately, grease does not seem to be sufficient in most cases with these, particularly on used bolts that already have surface damage. If it was, I doubt nearly every DUB crank that comes into the shop would require a breaker bar to remove.

If you’re curious, the next time you remove one of these stuck bolts give it a thorough cleaning and look at the threads under a microscope (or a good magnifier if you have better eyes than me). If you see surface roughness, damage to the crests of the threads, or sections where the anodizing is rubbed through, the grease did not do its job.

1 Like

One thing I intend to try going forward is using two bolts when installing these cranks - removing the extractor cap and using a steel bolt to install and compress the whole assembly, then removing it and replacing it with the aluminum bolt to fix everything. My theory is that by using the bolt to draw the assembly together, the threads are spending more time under stress while rotating, which increases friction and the likelihood of galling. So maybe having the steel bolt do that job and then letting the aluminum one do the more static job of fastening would be an improvement?

The threads are under the highest stress at the peak torque spec, not before it. I see what you’re getting at but I think you might be over selling the thread damage prior to hitting the factory torque spec. Those threads are getting damaged on removal, not installation. I’m assuming clean threads to begin with of course.

I see what you’re saying, but I don’t think it’s the absolute stress on the material that’s causing the damage - I think it’s the friction on the faces of the threads. So my train of thought is that anything to reduce the that time that those faces are moving and under load will create less issues, though of course you’re right that starting with clean threads (internally and externally, plus unworn surface coatings) gives the lowest chance for damage.

What makes you say that damage to the threads is occurring during removal? It seems to me that the problem is what causes that excessively high breakaway torque, which is what makes removal so difficult but I’m all ears for other ideas.