Bosch Gen 4 Chainring Nut printed Torque of 25 to 30 NM is likely wrong

[vman]

After following Bosch instructions (grease don't loctite) the chain ring lockring nut to 25 to 30 NM and having it fall off 3 times I was pretty annoyed and about to start looking into drilling it out and using safety wire. I even went to the local Trek Dealer and had them torque it down in case my torque wrench was to blame. Most local Trek dealers just state the Torque is 25 to 30 NM (as printed on the part) - when you call Bosch if you are lucky you will get someone who will read an outdated manual and confirm the old torque spec. However, what Bosch consumer support and most Trek folks don't know is a Technical Service Bulletin was issued by Bosch that stated the spec on the lockring is actually 40 NM to 45 NM. When my wife and I dropped both chain rings while riding on vacation, thankfully we found a local mechanic who actually cared about solving the problem and informed me of the change in torque spec - he torqued the lockrings on both our Trek Rails to 45 NM and so far after 300 miles they have not moved.

V

 

[Flupke]

Many thanks for sharing this information. Our bike faced the issue three times already (over 6000km) and purchased the necessary tools to fix it without having to go back to the bike shop. I will however at this stage stick to the official spec: 30 N.m to give it a try (as I am not 100% convinced the shop did carefully monitor it) unless I find the Technical Service Bulletin.
I am however grateful for you to have shared the information and will seek a confirmation from Bosch.

 

[TommyC]

Don’t worry about over tightening it. I did it but it just chewed the ally nut up, not the steel crank of the motor. I guess it’s designed to be the weak point. I bought a few spare lock nuts after that.

 

[Bummers]

Any chance of finding this technical service bulletin? I've had this issue happen to me.

 

[Flupke]

Any chance of finding this technical service bulletin? I've had this issue happen to me.

I actually opened a ticket yesterday with Bosch (contact form for questions related to electrical engines for bike) precisely with that question in order to get their official line. Let's see. I'll share the answer I will get on the forum.

Meanwhile I stick to 30 N.m and if it still become loose than I will try with 30N.m with some Loctite 222 (purple - designed to be removable)

 

[vman]

Many thanks for sharing this information. Our bike faced the issue three times already (over 6000km) and purchased the necessary tools to fix it without having to go back to the bike shop. I will however at this stage stick to the official spec: 30 N.m to give it a try (as I am not 100% convinced the shop did carefully monitor it) unless I find the Technical Service Bulletin.
I am however grateful for you to have shared the information and will seek a confirmation from Bosch.

Sadly Bosch customer service support refuses to give consumes access to the TSB's, they are worthless. My local mechanic did mention that the torque spec on cranks and BB cups is generally 40 to 50 nm for as long as I can remember ====so I am pretty certain bosch got it wrong on the label.

 

[vman]

I am one of those folks that owns several torque wrenches that I calibrate regularly, so I understand you folks are rightfully reluctant to take the information I have posted stating a 40 to 45 nm torque is the likely correct torque for a lockring vs. the printed 25 to 30 nm without trying to do your own research or seeing a copy of a Bosch TSB to rely on.

Of note, when I did speak to Bosch US for what its worth they did inform me that you are not to use Loctite on the lockring – only light grease. Also, while I have all the tools to drill out and wire tie the lockring – we do it all the time in gokarts and race cars, to do so on that small an alloy ring seemed potentially more problematic than helpful so in our case we had two choices: ignore bosch’s don’t use Loctite recommendation or ignore the 25 to 30 nm spec printed on the lockring, because we could not ride the bikes for more than 20 to 30 Km without the chainring going loose using the Bosch spec of 25 to 30 nm while tightened using light grease. The aforementioned is why when a trusted mechanic told us the torque spec was 40 to 45 nm per a dealer level TSB it made sense to us. Fingers crossed two bikes that were dropping chainrings every ride or two have now gone over a dozen rides with no issues.

I see that some of you are in the UK or EU, it will be interesting to see if Bosch gives you folks a different answer than the one we received in the U.S. where the person on the consumer phone is a customer service agent, not an engineer or technical person. In the U.S. Bosch will not release technical data to consumers - no manuals, no TSB, otherwise I would have just posted the TSB.

V

 

[Flupke]

Many thanks for the details.

30 N.Mm is clearly mentioned not only on the label but in all official service manuals

You can find those (in their French version) on the web site of the reputated Velolab in Belgium for example Manuels technique BOSCH eBike

Now the speed assistance limit is max 20 mph = 32 km/h in the US while we in EU are limited at 25 km/h so there is maybe an inflation on torque recommendations (joke )

 

[TommyC]

I just did mine up ‘something’ Nm with some loctite. Stop overthinking it!

 

[vman]

Does the service manual have an update date? If it is the original service manual before a TSB it will have the incorrect torque number. See below, for example this is a TSB by E Thirteen (one of the largest OEM crank providers) that clearly states not to use the O-ring and they claim they have Bosch internal clearance to do so, but it contradicts the service manual you just pulled. I am done running down this rabbit hole - either you add more torque or you effectively glue the thing with loctite - what I know is that at 30 nm the two I have on our bikes will come off everytime so its either torque it to 40 nm or put on loctite and effectively glue it in place.

V



Note - i

t also tells you to use the rubber O-ring which the E thirteen folkds

 

[Durrti]

Orbea also says 30nm

 

[Mikerb]

I have always used 30nm and never had any problems. I find it unbelieveable that an ally nut with so few threads would need or even survive 50nm!! I also wonder how many have a torque wrench for anti clockwise threads!

I believe the issue of chainrings coming loose for folk starts with the original installation being poor, and is followed by people then re using the lock nut after it has come loose...only for it to come loose a gain even if correctly installed this time. The reason I would suggest is that the minimal amount of thread on the locknut has been damaged. It is also the case that the initial installation may have involved too much torque (power tool used??).

With a bolt/machine screw, exceeding the engineering book torque risks shearing the bolt at its weakest point......where the thread initiates a fter the head of the bolt.

With a nut, it is different. The correct torque will apply the maximum friction between the threads of the nut and the threads of what it is fixed to. The correct torque is only marginally below the shear point and will involve some "spreading of the thread shoulders. So re using that nut may or may not be possible depending on how much spread occured in the threads.

Either way for the sake of c £10 the safest course of action is to always use a fresh lock nut.

Yes, blue loctite will also work but it depends how damaged the nut is......and it will make it more difficult to remove if needed.

 

[vman]

I have always used 30nm and never had any problems. I find it unbelieveable that an ally nut with so few threads would need or even survive 50nm!! I also wonder how many have a torque wrench for anti clockwise threads!

I believe the issue of chainrings coming loose for folk starts with the original installation being poor, and is followed by people then re using the lock nut after it has come loose...only for it to come loose a gain even if correctly installed this time. The reason I would suggest is that the minimal amount of thread on the locknut has been damaged. It is also the case that the initial installation may have involved too much torque (power tool used??).

With a bolt/machine screw, exceeding the engineering book torque risks shearing the bolt at its weakest point......where the thread initiates a fter the head of the bolt.

With a nut, it is different. The correct torque will apply the maximum friction between the threads of the nut and the threads of what it is fixed to. The correct torque is only marginally below the shear point and will involve some "spreading of the thread shoulders. So re using that nut may or may not be possible depending on how much spread occured in the threads.

Either way for the sake of c £10 the safest course of action is to always use a fresh lock nut.

Yes, blue loctite will also work but it depends how damaged the nut is......and it will make it more difficult to remove if needed.

Used a park tools torque wrench calibrated to support both forward and reverse torque for my first two installs. Bike shop used a snap on tools torque wrench. Each time set to 30NM.
No visible damage to threads and lock ring held torque smoothly.

Glad some folks are not having problems with the lock ring but to assume those that are must be ham fisted idiots is not very helpful.

 

[Mikerb]

Used a park tools torque wrench calibrated to support both forward and reverse torque for my first two installs. Bike shop used a snap on tools torque wrench. Each time set to 30NM.
No visible damage to threads and lock ring held torque smoothly.

Glad some folks are not having problems with the lock ring but to assume those that are must be ham fisted idiots is not very helpful.

The only difference between locknuts that come loose and those that do not is how the lock nut is installed and for most riders they of course have no involvement in that given that for most they have never touched the lock nut since buying the bike. So I was referring in the main to the original bike build.
Re "no visible damage to the threads"........spreading of the thread is not like a cross thread damage etc it is not something you can readilly see.

Pleased you are using good tools but your post focused on my input, does nothing to help others experiencing these problems other than to add c 15/20 nm to the torque specified on the nut based on some information gained which nobody else thus far has been a ble to confirm............and the purpose of the thread is help those suffering the problem by suggesting solutions. A lock nut that come loose whilst the bike is continued to be used even for a short time is highly likely to damage the lock nut. My advise is.........dont re use it.

 

[vman]

The only difference between locknuts that come loose and those that do not is how the lock nut is installed and for most riders they of course have no involvement in that given that for most they have never touched the lock nut since buying the bike. So I was referring in the main to the original bike build.
Re "no visible damage to the threads"........spreading of the thread is not like a cross thread damage etc it is not something you can readilly see.

Pleased you are using good tools but your post focused on my input, does nothing to help others experiencing these problems other than to add c 15/20 nm to the torque specified on the nut based on some information gained which nobody else thus far has been a ble to confirm............and the purpose of the thread is help those suffering the problem by suggesting solutions. A lock nut that come loose whilst the bike is continued to be used even for a short time is highly likely to damage the lock nut. My advise is.........dont re use it.

The issue I had was that your post implied the problem was more likely to happen due to install error vs. a problem with the torque spec. I agree that when possible replacing a bolt or lockring that has been exposed to any load it was not designed for after coming loose is good practice, but many would argue if the threads look clean and the bolt torques up smoothly to spec that is sufficient evidence it has not suffered damage. If you want to defend the 30 nm spec because bosch has not published on the internet a consumer available TSB that I can point to that is your choice, I will go by what has worked which is 40 nm and light grease.

We had two bikes that suffered constant lockring issues until both of them were moved from 30 to 40 nm. I waited months before even posting this because I knew it would cause a bit of s storm ...my goal is to help folks out that are having the same issues we have experience with our bikes. I think there are 3 paths forward - 1)Your lockring comes off and you put it back on with threadlocker 2) You replace the lockring with a new one and lightly grease it to 30 nm and hope it does not happen again 3) you lightly grease it an put it up to 40 nm.



V

 

[TommyC]

Pretty sure I read elsewhere the lock nuts shouldn’t be reused.

 

[bazxa]

Have refitted mine at the spec torque, original chain ring and o-ring, no problem yet. I would like to know the correct torque too.

 

[Tony4wd]

My Cube Stereo 120 already has blue threadlock on it from new. With the nut being so thin I'd rather stick to the lower torque value with threadlock.
I guess you could also use higher torque but put a new nut on each time.

 

[RipGroove]

Having the o-ring fitted helps too, on my previous bike the chainring came loose and I found it never had the o-ring fitted so I ordered one and fitted it. IIRC the lock nut has a left hand thread and my torque wrenches won't work that way so I put one in a vice to get a feel for what 30nm felt like in right hand mode then went by feel on the lock nut after putting a smear of red grease on it. It never came loose again after that.

Definitely be interested in seeing what the correct torque actually is supposed to be.

 

[vman]

Neither of our bikes came with o rings - E Thirteen's TSB advises against using one and has a allowance for Bosch that states you are still in spec if you don't use one, but that is an interesting variable.

We have had to abandon the SRAM direct mount chainrings - just can't find them and use a Race Face spider with 104 BCD standard ring that too might change what level torque will actually hold the lockring.

Use case may have something to do with lockrings coming undone as the load and chain angle on the drivetrain may change when riding steep hills all the time - which is our use case (typical rides involve 1,000 to 2,000 meters of climbing some on very steep grades.

Food for thought - While I don't have the lockrings to compare their relative size and number of threads between Shimano, Brose and Bosch, but Shimano states a 35 to 45 nm for its lockring and Brose states 50 nm and I have never heard of a Shimano or Brose bike dropping a chainring due to a loose lockring.

So long as we keep riding with a RF spider we are going to stick to at least 40 nm as at 30 nm as mentioned previously we had two bikes with constant lockring failures.

V

 

[vman]

I went into my local Trek factory store today to pickup some spare lockrings, as they still had the 25nm to 30 nm printed on them I asked the service manager if he could provide me anything in writing that the lockring torque specification had changed he confirmed that the new spec is 35 nm to 45 nm, and that the Bosch rep came by the prior week to verbally inform them of the new torque specification as well as make sure they pulled the cranks off new Bosch bikes and retorquing them to the new spec. The store manager looked up the data in one of the wholesaler systems and confirmed they had not changed the spec in writing yet as it still shows 25 to 30 nm, so still nothing in writing to share with you folks. .

Also FWIW - the service manager also confirmed that the lockring is a reusable part, and given that it is alloy felt it would be much more likely to crack then stretch if overstressed so he feels comfortable reusing any lockring that does not have thread damage, any visible cracks, and that continues to hold torque smoothly.

V

 

[RipGroove]

I went into my local Trek factory store today to pickup some spare lockrings, as they still had the 25nm to 30 nm printed on them I asked the service manager if he could provide me anything in writing that the lockring torque specification had changed he confirmed that the new spec is 35 nm to 45 nm, and that the Bosch rep came by the prior week to verbally inform them of the new torque specification as well as make sure they pulled the cranks off new Bosch bikes and retorquing them to the new spec. The store manager looked up the data in one of the wholesaler systems and confirmed they had not changed the spec in writing yet as it still shows 25 to 30 nm, so still nothing in writing to share with you folks. .

Also FWIW - the service manager also confirmed that the lockring is a reusable part, and given that it is alloy felt it would be much more likely to crack then stretch if overstressed so he feels comfortable reusing any lockring that does not have thread damage, any visible cracks, and that continues to hold torque smoothly.

V

Good info, thanks!

 

[slomtbr]

Mine came loose on my first ride.
Rather than investigate the cause of the loosening, I torqued mine to 30Nm, safety wired it, and now just ride my bike.

 

[Peaky Rider]

Why is using loctite such a no no?

 

[vman]

Apparently loctite can sometimes lead to inaccurate readings on your torque wrench, so bosch continues to instruct grease only.

V

 

[RipGroove]

Apparently loctite can sometimes lead to inaccurate readings on your torque wrench, so bosch continues to instruct grease only.

V

This, always use grease when torquing something up to a spec, any debris or glue or anything on the threads will make your torque wrench click prematurely before the thread is actually at the required spec.

 

[Peaky Rider]

Apparently loctite can sometimes lead to inaccurate readings on your torque wrench, so bosch continues to instruct grease only.

V

That's rich seeing as Bosch may "apparently" be printing the wrong torque settings on the lockring in any case.
In thirty years of maintaining my own bikes I have never used a torque wrench, mainly because I have never bothered acquiring one, let alone a left handed one, and I have never had a problem. When I put a new 'bearing seal protection kit' on the Bosch motor on my Rail I think I'll use a bit of Loctite and just tighten the thing up to what feels right.

 

[Tony4wd]

I'll use a bit of Loctite and just tighten the thing up to what feels right

I agree - firm with blue locktite so it can't unscrew. The exact torque figure isn't critical on that component, just needs to be tight enough to stop the chainring moving but not so tight that you damage the threads.
I haven't used purple loctite before but it might be good enough for the locknut too.

 

[Flupke]

I got an official answer from Bosch support: they confirm one should strictly stick to the specified 25-30 N.m which is already a lot by the way.

See attachment.

 

[vman]

Okay so Europe and US Bosch disagree.