rms13

Still troubleshooting different creaks and noises from my new build. I feel like I could tighten down my stem a little more but this is the first full carbon fork I've had and I'm worried about damaging it if I tighten too much. Is that something I should even worry about using t handle allen wrench or can I torque it as hard as I can safely?

Zef

it only needs to be tight enough that it does not move in use. Torquing the bolts tighter may lead to a cracked steerer tube. While proper torque may vary from manufacturer to manufacture, without a specific recommendation from the company that made your stem and/or fork, I would use 5nm as a good ball park for proper torque.

Pick up an inexpensive beam type torque wrench designed for use with 1/4inch sockets and also pick up some hex/allen key sockets (if you don't already have these tools). Carbon forks are an expensive thing to play "guess how tight" with.

-j

rpenmanparker 

The common upper limit for most steerers is a very generous 6 Nm. Usually the only direction you can get is from the stem maker when it is really the steerer (or handlebar) that is the delicate item. I can tell you that you need nowhere near 6 Nm of torque on stem bolts to firmly fix either the stem to the steerer or the bars to the stem. 2 Nm is likely plenty. 4 Nm is for sure enough.

FBinNY 

If the stem is well made, with a round ID of correct diameter (all mid to upper level stems are) then it' impossible to damage the steerer by over-tightening. The torque limits are usually about the bolt or threads in the stem, or may be boiler plate for hose who insist on a number.

In understanding the implications, consider crushing an egg. If you palm it and squeeze evenly, the shell is amazingly strong. OTOH if you squeeze unevenly and distort the shape it cracks at very low force. It's the shape which works like an arch shape, or a bicycle rim and converts a radial compression load to one running along the cicumference. Like a rim or egg, the steerer won't crack under compression.

What is harmful to steerers is tension from an over expanded compression plug, or distortion or ovalization from a cheap stem not bored to the right diameter, or more usually not exactly round.

NOW- how to cure the creak

No promises here, nut often steerers are distorted into a barrel shape by over expansion of the plug. Try this. Set plug no tighter than necessary to hold against the top cap screw. Traction can be improved by painting the plug's grip surface with grit paint (fine grit in nail polish), or carbon assembly paste (IMO- the plugs should have been so treated by the makers).

Otherwise, follow this sequence. adjust headset, tighten stem reasonably tight, remove top cap, and loosen the plug slightly or remove, tighten stem to spec. retighten the plug firm and install top cap.

This sequence ensures that the stem is tightened against a cylindrical steerer rather than a barrel shape which allows motion.

Al1943

Don't know about 2 nm but I can confirm that you should not go over 4 nm. It's easy to crush a cf steer tube, don't ask how I know.

FBinNY 

I don't doubt that it's possible to crush steerers, or at least it was before fork and stem makers got serious about holding tolerance. But there's no reliable correlation between torque on the fastener(s) and crushing force on the steerer. That relationship depends on the helix angle of the fastener's threads, which varies stem to stem. 4nm on stem A is very different from 4nm on stem B.

rms13

This is my first full carbon fork so I'm used to round tubes and star nuts. Apparently this Easton EC90 steerer has two flat edges on it and round the rest of the way. It has an SI integration compression bolt. The top of the steerer looks slightly uneven. Not sure if that matters. I have 30mm of spacers above the stem that I will want to cut off so I'm not too worried about the top of the steerer

FBinNY 

I still suspect that your issue is because of barrel shaped distortion of the steerer. Essentially the stem is clamping on the sides of the barrel and free to rock (slightly). after adjusting and tightening, try removing the expander, or a least loosening it materially, and see if it's possible to take some more on the clamp screws. If so, that was the issue, and relaxing the expander should have solved it.

BTW- not all creaking is stem/steerer. Sometimes the steerer itself flexes under load causing some movement of the bearings, so don't get tunnel vision. If nothing you do solves the problem, open your mind to other possibilities.

Al1943

I don't mean to sound argumentative but the cf steer tube that I broke was a 1 1/8" Colnago Carbon Star that retails for $850. The stem is a two bolt Syntace. This happened 10 years ago as I was building up the bike (I still ride with a new fork). The only torque specs I could find were shown on the stem, 8 nm. Using a 1/4" torque wrench I alternated tightening the two bolts. The steer tube broke just before reaching 8 nm. Since that time I've read everything I've been able to find on the subject of tightening stems on cf components and the only information I've seen is that one manufacturer has a torque tool that releases a ratchet at 5 nm. By experimentation I've concluded that 5 nm is more than enough.

FBinNY 

You're not being argumentative. This is what happened. I suspect, but can't possibly know that the stem was off spec. I say this because it's nea impossible to crush a cylinder with even pressure.

In the early days of carbon seat posts, it was easy to crack one if the seat tube was ovalized at the top, or if the clamp was built in a way that allowed the ears to dig in as they tightened. In the end, carbon post makers engaged is some self defense and thickened the tube to resist the distortion that poor fits caused. With time, seat post clamps and stems evolved to ensure even clamping without distortion.

Take a look at the clamp below.


Notice the thin areas where the ears attach to the ring, Under compression load, the ears deflect driving the corners near the slot inward.

Now look at a more recent design



Notice how the ears are built up and extend forward compared to the ring. This allows more clamping force without the ears bending together. This type of design prevents the type of local radial stress at the slot that the older type caused.

You'll see similar changes in stems, with greater stiffening of the transition of ring to ear.

I don't know the design of the Syntace stem you had, so can't say whether it's construction was a factor, but I will say that the early days of carbon posts and steerers were the wild west, and the early adopters were essentially beta testers, and lot's has been learned at their expense.

In any case, I do agree with you about the torque on bolts. This is why I advised the OP solve his issue with finesse rather than brute force. For my part, I don't use torque wrenches, and tighten all hardware to the minimum torque that does the job. When I have a choice, I prefer to reduce clamping force by improving fit or traction.

Camilo

The Ritchey (and other) Torque Key is an excellent tool for stems and many other similar sized bolts on a bike.

I've been using one for several years and what others say above about how much torque is actually needed is absolutely true. The 5nM Torque Key is actually quite a lot of torque, much more than is needed for the stem/steer tube clamp, and more than what is needed for the handlebar if you use assembly paste. But it's easy, safe and sure to just use the Torque Key and be done with it. I also have a regular beam type toque wrench which is not a big deal to use, but the torque key is much easier and therefore more likely to get used and be used accurately.

Al1943

The Syntace stem is a modern "high-end" piece that seems to be very well designed EXCEPT the steer tube fixing bolts should never have been marked "8 nm". https://www.syntace.com/index.cfm?pid=3&pk=1693 I don't doubt that the stem hardware can easily hold 8 nm or more but that is way too much for a carbon steer tube. BTW the steer tube is a round cylinder. "Crushed" was not a good choice of words, the tube actually cracked for about 4 inches.

fietsbob

& I note seat post band clamps for Carbon are offset and angled Campagnolo Seatpost Clamp - Excel Sports

FBinNY 

I stand by my theory, that distortion (ovalizing, or ear ear stress) cracked the steerer, and not excess compression. Of course without the parts to examine and measure precisely, it's just a theory. But I can be very stubborn if I feel I'm right. (doesn't mean I am, just that I think I am)

Soil_Sampler

I like to leave a few millimeters above the stem on carbon forks.

And also like the longer plugs/inserts.

custom carbon fork insert

Al1943

I don't see how it could have been ovalized in this case. But there may well have been some distortion. The top of the stem was about 1/8" above the top of the steer tube and the back side of the tube was not supported behind the upper pinch bolt. The only ear to ear stress was between my two ears and it was pretty significant. I understand what you are saying about the strength of a cylinder but I also know that the steer tube did not crack at all until I applied too much compression, then it all happened in one quick crack. There probably was some longitudinal distortion that I was not aware of until it broke. When I installed the new fork I was careful to position the compression plug to better support the back side of the steer tube and, of course, I set the pinch bolts with less torque. I'm still riding the same stem and have had no more cf problems. At a later date I moved a 5 mm cf spacer from below the stem to above the stem. I have 25 mm below the stem.

FBinNY 

Here's one possible way.

Imagine the effect if the bore is slightly larger than spot on. When tightened the sides press in but since the bore is large the steerer can flex outward front and back (making a circle into an oval). Carbon structures aren't very elastic, and it takes very little ovalization of this sort to cause long lengthwise splits.

Clamp fit on carbon tubes is very critical. In the early days many makers of bars and stems would void the warranty on their bars if another stem was used. This wasn't the typical "warranty void if..." boilerplate. It was a legitimate reaction to questions of tolerance, and the consequences of mismatched curvatures.

BTW- I believe that since the wild west days, makes of carbon posts and steerers have "hardened" the tubes to make them more forgiving of small tolerance mismatches. I know Campagnolo did so with their posts, and I suspect that most followed suit, one way or another.

rpenmanparker 

I agree with you, but isn't your position idealistic at the extreme edges of manufacturing tolerance. I mean steerers and handlebars are routinely crushed by stems being overtightened. In principle you are right, but in practice no stem, bar, or steerer is perfect enough to allow extreme tightening. It doesn't matter where the blame is, the failure occurs. The solution is, as I think you already have stated yourself, to use minimum effective tightness.

I cracked an ultralight handlebar at about 5 Nm despite a recommendation of 6 Nm for the FSA stem. The replacement bar held up fine and is plenty tight at 2 Nm. My point is that to say it is the fault of the shape of the parts not the tightness is splitting hairs.

FBinNY 

IMO this isn't a matter of torque specs, simply bad design and tolerance. Properly designed and built stems and clamps don't crack the bars, posts or steerers they clamp, poorly made ones do.

To give you one example how a very minor change can make a big difference, consider the problem of broken thinwall carbon posts which was common about 10 or so years back. It turned out that poorly designed seat clamps were digging the ears in at the back (see my earlier post). Rather than redesign the clamp, we simply reversed them, or twisted the around so the slots didn't line up (SOP today). Now, with the slots out of line with each other the ears can't dig in the same way and the failure rate dropped drastically.

There was a learning curve with carbon parts and it took a while for some lessons to take hold, thanks for those unpaid beta testers. Early on, the response was low torque specs, (in some cases, so low that a 10% error meant failure) but these days, stems are bored precisely, ears are made not to bend inward, and most builders use angled slots to prevent longitudinal stress in the tube.

Also the use of various methods to improve traction means that lower clamping forces can provide the required hold. I'm not advocating higher torques on fasteners, and still say that the right torque is the minimum that does the job, but the penalty for exceeding a spec shouldn't be catastrophic failure.

BTW- SOP good engineering practice would be for the fasteners to failure before the mating parts suffer, so I have a strong bias here. Show me a part that failed because of excess torque on the fastener, and I'll always blame poor design or execution.

rpenmanparker 

As I said, I agree with all that. All I meant was there is no such thing as perfect, just better than we know how to measure, rounder steerers, more symmetrical clamps, etc. So we can keep making the parts better, but there will still be a torue-tightness where the post, steerer, or bar will fail. Your squeezing the egg example is a good one to demonstrate this point also. Lots of people can't squeeze evenly enough to keep from breaking the egg. Maybe up to a point, but then the natural variation in strength around the hand or around the egg shell kicks in and whoops, instant mess. In principle it works, but not always in practice. Just sayin' that your conclusion is really the most important part. If you don't use any more torque-tightness than you really need, the success rate will soar. Even mediocre parts will often be okay. And very importantly, the recommended torque limits are almost always much higher than they need to be or much higher than some random, mediocre out-of round component will be able to withstand or higher than some very slightly out-of-round clamp should be pushed to.

FBinNY 

The prevention of failure through good design and execution is not only possible, but done every day. If you accept my egg example, you can see that compressing the tube without distortion will not cause cracking except at very extreme loads. In any case it is possible to design parts that are forgiving of user error. Just as tires can withstand pressures greatly above the rating, parts MUST be able to withstand excess fastener torques of more than 25-50% above the rating. Putting parts with critical, low margin of error, torque specs into a consumer market is both stupid and unnecessary.

The whole world engineers with margins of safety, and with "failsafe" designs that fail in predetermined ways to minimize risk, or expensive consequences of minor user error. As I said earlier SOP with fasteners is for the fastener (the cheapest easiest to replace part) to fail before more critical parts. If it's possible to crack a tube by over-tightening the clamp (any tube/any clamp) somebody didn't sweat details along the way.

As an example, would anybody accept a QR design that gneeded to be tightened to 85% of the tensile strength of the skewer in order to hold a wheel? Would we accept skewers that blew apart without warning if over tightened 15%? Of course not. Yet, we accept expensive carbon parts with stupidly narrow specs for tightening, and fail expensively if those torques aren't met and call it user error.

Good design is possible, the only reason we see bad design is because the market accepts it.

rpenmanparker 

No argument.

basqueonacaad

Hi, all, new to the forum here. I recently bought a used CAAD 10 5 and am a bit of a worrier. Today while adjusting the headset, I removed the fork from the headtube and failed to remove the wheel from the fork...it rolled out of the headtube and the steerer (carbon steerer) hit the ground. I looked at the steerer and there is nothing indicating any damage....im just new to carbon and am a bit neurotic and am hoping for some feedback. Am i worrying WAY too much? I can't imagine it caused any damage due to a minor fall of say, 16 inches....but again, I am a git of a worrier. Thanks in advance

FBinNY 

Yes, you're worrying too much. What happened to you is all too common (possibly in the top 10 newbie mechanic bloopers). The worst consequences are usually lost balls, but since you're not crying about that, I'll assume you have cartridge bearings and put it back together OK, and have nothing to worry about at all.

basqueonacaad

Thanks. I can't even say I was inebriated at the time Sheer clutziness...