Ghrumpy

For torque transmission, it could be a round taper, this is how a lot of machine tool attachments work. They are held on by friction alone. But mills and drills don't require timing. You could have a keyway and a key I suppose. But even in press fits like that, the key isn't primarily transferring the torque, the friction is.
I can think of two reasons for the square shape. One, that's the easiest way to time the cranks 180° from each other. It's a relatively easy machine operation to make four sides on a round bar and clock them accurately. Two, it will reduce the chance of the relative motion that causes fretting.
I can also think of two reasons for the taper. One is to make repeated removal and installation easier. The other is because of the dissimilar metals used for axles and arms. A taper allows relatively easy management of the deformation of the softer crank by the harder axle.
These seem like pretty basic engineering solutions well suited to the requirements. Stronglight figured this out and had them on the market in the 1930s.

You don't really even need a tapered end. Viscount made a square BB with no taper. Straight up press fit. The idea being that since the Viscount BB had cartridge bearings, it would not need service anywhere nearly as often as a cup-type BB, so fewer removal/installation cycles would be needed. In theory anyway.

Modern splined interfaces are better suited to the larger diameter and hollow axles that are used than a square taper would be. As with square ends, the taper they have is to make repeated removal and installation easier. And of course, advances in technology have made their more complicated machining easier and cheaper.
I never said lubricating reduced the material strength. I never denied that dry mounting doesn't increase the likelihood of fretting. As you said, I won't explain the same thing for the 10th time.

Bike Gremlin

Yes, for round attachments, like drill bits, I agree that friction is what holds it in place. Attach a 170 mm crank to a drill bit, step on it, and see if it stays in place. Square taper attachment doesn't rely on friction for torque transfer.

Taper is necessary in order to assure there is no play and minimize fretting. It takes some elastic deformation of the crank to fit snugly and prevent any play. If there's no preload and elastic deformation, it would move under load, as it gets elastically deformed.

And how did that design fare?

Modern as Octalink were a poor design, prone to a crank coming loose if one stands on the pedals with the right foot (pedal) forward often. No preload caused the reverse torque (reverse to the pedaling torque) that happens in this position to move the cranks a bit, unscrewing the bolts, bit by bit. Shimano tried to address the problem by making the splines "deeper" in the Octalink 2 revision. Without eliminating the problem that had caused cranks loosening.

Hollowtech II has split left crank, that is then tightened with bolts against the axle. This creates preload. The design has other problems, but the lack of preload and crank loosening due to it are solved at least. On to the next iteration... or not.

You said that lubing increases the risk of plastic deformation? How?

If the material strength isn't changed.
If the tightening torque is the same.
If the contact area isn't any smaller with lubed mounting.

Where does the "extra" force that results in plastic deformation (when lubed, and not when dry?) come from?

pdlamb

Once the mounting bolt is torqued tight, it should not loosen (and doesn't, in my experience). I like to grease the bolt likely before installation, hoping that'll help when it's time to remove it.

gearbasher

You know, I almost feel guilty about posting the page from the Campy manual.


Almost.

Now I'll just add some more fuel to the fire.

If you look at Park tool's site, they recommend dry tapers. Shimano's crank install instructions call for greasing the cups and spindle for their hollowtech, but for square taper they say only to grease the cup threads. Barnett's manual also recommends dry tapers. So, you have the two largest component manufacturers, probably the largest bicycle tool manufacturer and a major repair manual all saying dry tapers. I think I'll continue to not grease mine.

Bike Gremlin

Yes, all great authorities. But, just like I don't consider everything a stupid man says to be incorrect, just because he's made mistakes (I do question it a bit more), I also don't consider everything an expert says as gospel. It does hold much more weight, but it's also open for questioning IMO.

I'm sticking with the anti sieze and am yet to get a customer complaining about cranks loosening. Put a lot of thought into this (obviously ), just like most other things I do - since I want it all done as good as it can be done (bit of OCD, but does more good than harm, at least for the customers, in my lines of work ). Did all the testing I could (without a lab - and am planing to do some lab tasting in a few months, but probably not the square tapers - yet). Heavy riders with great mileages, especially the ones riding all year long are an invaluable source of data - I fix those people's bikes for free, just under one condition: "give me feedback, whether it's good, or bad". This works exceptionally well for bearing greases (the winter riders), wheel building, but also for the square taper mounting using a "radical" method.

I'm yet to hear/read an explanation that holds water for why, in this case, a dry mount is superior. The arguments I could read here, against lubing, boil down to:

1. I've always done it so, no problems - from this data one can only conclude the method is good enough for the subject, not that it's superior - a good analogy would be claiming that disk brakes aren't better than hub brakes, since I've always stopped in time with hub brakes.

2. This authority says so (without explanation why). This sounds religious to me.

3. Lubing will damage the cranks - without explaining the mechanics of how, compared to a lubed method. Argument that, with the same tightening force, the cranks would somehow get damaged if lubed contradicts the laws of physics/mechanics.

4. Cranks will more likely fall off? If I understood right. This is not probable, since when lubed they will climb as high as they can "up" the taper, and the preload of the mounting bolt will keep both the crank and the bolt in place, prevent them from loosening. Dry mounting is more likely to result in the bolts and the cranks loosening - in case the friction prevents the cranks from "climbing" the splines high enough at the given bolt tightening torque.

SquidPuppet

I have been on a bit of a grease (bearing use specifically) experiment binge for a couple years now, mostly (but not all) for entertainment purposes. I'd be very interested in reading your findings and opinions, if you care to share.

trail_monkey

This is what I started using in my hubs a couple years ago. It is good stuff. Waterproof marine grade wheel bearing grease and it’s not as thick as standard automotive grease so it’s great for cold weather too. I am sticking with it. Have also used it for caged ball headsets and loose ball bottom brackets as well. Just damn good stuff.

https://shop.advanceautoparts.com/p/...E&gclsrc=aw.ds

Bike Gremlin

It boils down to - anything is good enough for the relatively low load and low speed bicycle bearing. If it all works out, I'd get to do some lab test to check that in a few months.

Long(er) story is on my site:
Bicycle bearing greases - explained - Bike Gremlin - Bicycles

SquidPuppet

Thanks. I had already read that, but didn't know it was yours.

Ghrumpy

Clearly you are an intelligent person who thinks about things a lot, but your counter-examples are too often reductio ad absurdum or straw men, or you argue against points I didn't make. Not a good way to prove a point, IMO. But I'll speak to your assertion anyway.

First off, there's no real debate as to whether a round taper fit transfers or resists torque. It seems like a pretty basic engineering principle, and (not that it matters) I have seen it happen. Most drill bits mount into a collet or a chuck, and those are often mounted onto the motor shaft with a taper fit, a Morse taper being one example. They don't come off very easily. They are preloaded by the force of the bit onto the workpiece, instead of a bolt, but they don't slip off without a prybar or some other forcible method. You can't twist them off, IME.

Another common round taper fit is an automobile ball joint. It is preloaded by the nut because they have to withstand a lot of vibration and force. But they do not just fall out when you take the nut off. And they can't be turned out either.

I would be willing to bet my dollar against your penny that a round taper crank could be designed to withstand human pedaling forces and not slip. It may not even be much different looking from a square taper. With proper preload, it would stay in place. I am not qualified to design such a piece and frankly I don't feel much need to because of the above. It's a pointless argument to prove because it's too impractical to spend any research on. But if you have a penny burning a hole in your pocket....

Almost.
Preload is necessary to prevent play and minimize fretting. That preload can be provided by pressing the arm to the taper, or because of tolerance fit, as in the case of the Viscount crankarm and BB.
Taper exists to make repeated removal and installation more convenient, and less dependent on difficult manufacturing.
Square exists to clock the cranks easily and repeatedly, and because it's simple manufacturing operation.

Let's just say nobody followed them.
Doesn't mean it didn't "work." It's mostly that it rendered the BB and crank incompatible with every other BB and crank on the market. And when the company went under, it left people who had such a bike in a tough spot.

Octalink I seems to have fared about as well as Viscount's flat square.

But whatever, I don't care to discuss those interfaces' flaws here. I'll restate my point: splined interfaces are better suited to the larger diameter tubular axles that have become the new standard, and modern manufacturing makes them possible and affordable.

The first and third points have nothing to do with anything I've proposed. I don't know why you bring them up. Torque is the only thing that I have brought up.

Let's start with the proposition that, as I'm sure you'll agree, the average mechanic, whether professional or not, is more likely to overtighten a fastener than undertighten it. Most crank wrenches are (deliberately) not long enough that this is a problem. But most people of average strength could probably manage to do it anyway.
If you accept that proposition, then the rest is just math, counting screw threads..

Lubrication of a fastener and/or a press fit reduces the amount of friction between the parts, and therefore the amount of force needed to get them to their correct preload setting. In the case of the crankarm, the torque on the bolt is what applies the force to the taper fit. (The taper could be pressed on another way, and the bolt set afterwards, which is basically what happens if you install with a steel bolt, remove it, and replace it with a titanium one, torqued to its spec. The titanium bolt won't press the crank on any farther.)
If a lubed bolt is therefore set with the amount of torque recommended for a dry bolt, it will push the crankarm farther onto the spindle than a dry bolt, because it will engage farther into the threaded hole in the spindle before it reached its proper torque setting. If a lubed bolt is used to press a lubed taper, that effect is compounded by the reduced friction between the spindle and crankarm, so the taper also will slide farther onto the spindle at the same time that the bolt goes farther into the spindle before it reaches its torque spec. This increases the crank's taper hole deformation.

The first two or three or even ten times, it may not be enough to cause a problem, it it's a tenth of a turn each time. But after twenty times or more, it might start to be a problem. With a "tighter is better" mechanic behind the wrench, it might happen sooner.

So, the "extra force" comes from the fact that most people don't reduce the torque needed to adjust for the grease on the bolt, or the taper. Most people don't even use a torque wrench.

As I said previously, if you measure your torque, and adjust the bolt torque downward to accommodate the lower friction of a lubed bolt and taper, it probably won't ever be a problem. But you damn well better know what you're doing.

Bike Gremlin

Could be a language thing (English not being my native), but I'm not doing it intentionally, don't even think this accusation is legit/true. But won't argue - whatever you feel/think. Sorry I came across that way.

I generally agree. With an axle diameter large enough compared to the crank length, round interface could be held in place by friction alone, without exceeding the plastic deformation limit force with preload. Bicycle square taper axle has a relatively small diameter compared to the crank length and expected loads, so it's made square - for that reason and for an easier crank alignment.

For all I know, the Viscount design did not fare well at all.
The other points were answerd in the previous paragraph.

Octalink was a problematic design, with a (well?) known problem caused by poor design.

The Hollowtech II splined design deals with the fretting damage problem of square taper and axle fatigue braking at the crank attachment area.

You mentioned modern designs and no need for a taper. That's why.

I agree that, given the same tightening torque, crank arms will move further. However:
All the manufacturers recommend lubing the tightening bolt threads, yes? So, for the same tightening torque, we have the same load on the crank pushing it up the tapered axle. The only variation is whether the axle-crank interface is lubed or not.

If a crank is not lubed, it is pushed by the same force dry - allowing more scratching/fretting damage during mounting. If the interface is lubed, then most of that is eliminated. The crank will go as high as it can, not stop earlier due to friction (so it can creep up later under load, unloading the bolt), not having any play, therefore minimizing the chances of any fretting, or coming loose.

A certain force that isn't enough to make plastic deformation of a dry interface will not make plastic deformation on a lubed one. If there were a sharp edge near the top part of the axle taper, and moving the crank a bit more further would make it come to the edge, then your argument would make sense, but the taper is evenly sloped/angled.

Try putting some old crank (on an old axle) 10 times - lubed. Mount, dismount - not re-tightening after riding. You are not very likely to damage the interface. The weak link in that "chain" are the bolt threads. If you re-tighten after several rides, and do so a few times, it's a different story, but that is not what we're discussing, is it?

gearbasher

I hate beating a dead horse, but I feel I need to respond.
First of all, I never said my method was superior. I just posted a page out of Campy's user manual. You're the one who "couldn't believe" they would say that. More or less saying they were wrong.

I felt I needed to defend my reasoning by stating that it always worked fine for me. And, your argument was that your method always worked for you. Adding that you have a larger amount of customer feedback to prove your point. Well, don't you think that Campagnolo, with dealers and distributors all over the world and pro teams using their product, have an even larger consumer feedback network to base their recommendations. You have to look at their warranty and liability issues, also. In fact, it sounds to me like you're either a shop owner or a shop mechanic. Your claim at being OCD alone should make you follow manufacturers recommendations just for liability issues alone.

Let me give you two example from experience.

1) A friend of mine had just bought a new Italian steel frame (I will not mention the make). Two months after he purchased it, we were on a large group ride with a fast pace. He was forced into a pot hole. It was no larger than anything I've been though. When he hit it, the top tube pulled out of the head tube lug. He fell and thankfully was not injured. When looking at the frame you could easily tell it was not brazed correctly. It looked as if the was only about a 3 mm circular area on either side of the tube that was actually brazed. When he contacted the manufacturer and explained what happened. Their response was: "our frames are designed for paved roads." He wasn't sure if it was a language barrier or not, but they refused to honor the warranty. He eventually got a new frame from the dealer. Whether the dealer took a loss or not, I don't know.

2) Back in the day, I always rode a 42/52 front combo. When Campy went to 9/10 speed systems the front combo went to 39/53. Too big of a jump for me, so I use an aftermarket 50 tooth large chain ring. When Campy came out with the compact 34/50 crank and the compact front derailleur, I wondered if the compact derailleur would work better on my 39/50 set-up. Instead of experimenting, I emailed Campy. Their response was, "our systems are designed to work with only our components. We cannot guarantee anything to work with other parts."

That is the mindset of companies when it comes to warranties and liability.

Now say, God forbid, someone gets injured due to a loose or broken crank. The first thing the company will say is: "Was it installed following our instructions?"

I just emailed Campagnolo. Let's see what they say.

Plus, no offense to you, the late Sheldon Brown and Jobst Brandt, but I will go by the manufacturer's recommendation over anyone who has a web page or posts on a forum.

BTW: in my search for different manufacturer's recommendations. Phil Wood recommends lightly oiling the tapers. And, if I were to use a Phil Wood BB, I probably would follow that recommendation.

As far as points 3 and 4. I had no comments about that, so I have nothing to say about it.

Well, I guess it's time for me to exit stage left. @Ghrumpy your on your own.

Bike Gremlin

Yes, I think it's a bad piece of advice.

I've provided theoretical claims why lubing is better IMO, adding that it's been tested in practice as well.

I understand - you don't want to risk loosing a warranty claim, so would stick to recommendations.
No(t) arguing that. Just, like I've said, the instructions recommend a procedure that isn't optimal, IMO.

BTW, I've found a few other - smaller - companies recommending lubrication (greasing in those cases). Will try googling it. Could be a warranty/liability thing. Just guessing: could be something an non-insignificant percentage of people do to mess up when mounting lubed, and that doesn't happen as often when mounting dry. Though, if doing it properly, I'm yet to hear an explanation why a lubed method would not be better.

Edit: one after a quick search, not sure if it's valid. They didn't state if they greased the Campy one, or not. Might do some testing of my own when I find the time:
https://janheine.wordpress.com/2012/...not-to-grease/

davidad

From: Jobst Brandt
My cranks get loose, quite quickly too; over about 10 miles or so from being solid to flopping about in the breeze. Any suggestions?
One or both of the cranks are ruined!* Once ridden in the "floppy" mode, the tapered square bore of the crank has been deformed and can no longer be secured on a spindle. Install and properly tighten new cranks on the spindle after greasing the tapered square ends of the spindle. Proper tightness should be achieved with a torque wrench or by a skilled hand.
The admonition not to grease the spindle finds life mainly in the bicycle trade. When I discussed the "dry assembly" rule with crank manufacturers, I discovered that they had warranty claims from customers who split cranks. However, cranks cannot be split by overtightening them. This can be proven by attempting to do so. An M8x1 bolt is not strong enough to split a major-brand crank.
Failure from "over-tightening" is caused by repeated re-tightening of properly installed cranks. In use, an aluminum crank squirms on its taper, and because the retaining bolt prevents it from moving off the taper, it elbows itself away from the bolt and up the taper ever so slightly. The resulting loss of preload, after hard riding, can be detected by how easily the bolt can be turned.
Loss of crank bolt preload is greater on left than right cranks, because left cranks transmit torque and bending simultaneously while right cranks transmit these forces separately. The left crank transmits driving torque through the spindle to the right crank and chainwheel while the right crank drives the chainwheel directly. Besides that, the right crank transmits torque to the spindle only when standing on both pedals. Doing this with the right foot forward (goofy footed) is the only time the spindle transmits reverse torque.
Mechanics, unaware of why crank bolts lose preload (and commensurate crank tightening), have re-tightened bolts until cranks split. No warnings against re-tightening properly installed cranks are evident although it is here where the warning should be directed rather than at lubrication.

Ghrumpy

This is great.

Now don't anyone go claiming victory. Jobst seems to be agreeing with everyone. (FWIW this discussion isn't about winning or losing for me. As far as I'm concerned, there's not a right and wrong way to do this. Just different ways that have their pros and cons.)

But I will point out that Brandt's main point is the same as mine: proper installation torque is more important than lubrication. He and I differ on taper lubrication because Brandt seemed to be a mechanical idealist, while I am more inclined to be practical.

Ideally, lubing the taper is better, BUT ONLY IF BOLT TORQUE IS MEASURED. And if a manufacturer recommends a dry taper and/or bolt, lubing the taper will necessitate a corresponding decrease in the bolt installation torque. If these are not considered and done, the chance of overtightening is higher.
Practically speaking, a dry taper reduces the chance of someone overtightening the crank bolt because of the higher torque required, while acknowledging the risk of fretting damage.

There are tradeoffs. Consider them, do not just accept a "right" way of doing things because some authority said so. Greasing tapers because Jobst said so is no more "right" than degreasing them because Campagnolo said so.
If you're an idealist then, go ahead and anti-seize your tapers. But leave your peanut butter wrench in the toolbox, and don't install without a torque wrench and a guide, and the correct compensations.

Bike Gremlin

This is simply incorrect.

Manufacturers recommend lubing the bolt (most of them, anyway). Lubed bolt at X Nm of torque gives the same force - no matter whether the crank/spindle interface is lubed, or not. If that force is enough to cause damage, it is more likely to damage a non-lubed interface, than a lubed one.

Ghrumpy

I do not dispute that it gives the same force. Force is force.

A lubed fastener will thread in farther than a dry one to get to the same torque value. Do you agree?

Feel free to test this.

But you're changing your tune a little. Earlier on you were talking about fretting, which does not happen upon installation. It happens from micro-motions between installed parts in use, not from installation forces. Which is the problem you want to discuss: damage from fretting, or "damage" from installation?

wschruba

You do realize that his first language is not English (nor, I would assume, his second), and it could easily bea mistranslation/misunderstanding?

Of course, if you'd rather rage into the void, don't let me stop you.

Bike Gremlin

Yes, I agree. If a manufacturer gives torque recommendation for a dry bolt, lubing it does call for a lower torque (a rule of thumb being 25% less torque for a lubed bolt if the torque is specified for a dry mount). However, from my understanding, we were discussing whether it is better to lube the axle when mounting cranks.

Dry installing does for both IMO. Fretting during use, when loaded, and some "extra" damage during the mounting procedure. Lubrication is known to help prevent both. May the expert(s) please correct me for the proper engineering terms.

Ghrumpy

Yes, I understand the language situation, and I'm honestly trying not to take any unfair advantage because of it. No rage on my part, I assure you. Just seeking clarity.

Ghrumpy

Yes, we are. Discussing them separately is a bit of a problem though, because the crank is both pressed on and preloaded by the bolt and its torque setting. So you can't completely separate the one issue from the other. Lubing either the bolt or the taper increases the likelihood of overtightening in the real world where most cranks are not installed with a torque wrench. Lubing both has a cumulative effect; it doubles that chance of overtightening.

How many mechanics in bike shops use a torque wrench to install every crank bolt onto square taper cranks? I'm not saying it doesn't happen. Based on my observation and experience, I'm saying it's probably 1000 without to 1 with, and I think I'm being generous.
That's fine, as long as the mechanic has experience and a sense of when enough is enough. And if you have that deft hand to adjust your tightening force for both lubed bolts and tapers, then super good for you. I think it's rare, and because it's rare I think it's bad general advice to say lubed is "better" without taking into account how things actually work in the rear world. "Better" than what should be clearly specified, and under what specific conditions, for it to be anything but a personal opinion. Again, I'm talking about practical considerations, not ideal ones.

I think you're fine with the terminology. Much better than I would be doing with any other language I can mumble in.

Which is the bigger problem for you? Fretting or the "extra" installation "damage"? Both are worth consideration. But I'd prefer to deal with one at a time, since they do not seem as closely related as bolt tightening and taper deformation.

I have already admitted dry installation increases the chance of fretting. My position has not changed on that. Nor has my position that the chance of damage from fretting on a dry installation is lower than the chance of damage from overtightening on a lubed installation. This is a relative and practical consideration that ties the bolt torque to the plastic deformation of the crankarm taper hole.

I apologize if I'm repeating myself. I don't know if there's anything more I can add to what I've already said.

Bike Gremlin

Yes, but can we assume (I thought we did) that the bolt preload torque is a given? Discuss one variable at a time. Of course, I'd also argue there that lubing the bolt (and going -25% of a given dry tightening torque, when manufacturers recommend dry bolt as well) does help get a more uniform tension/force on each application, but here I'd focus on the axle lubing.

Lubing the bolt does. I agree. Lubing the axle does not. You are using the same force, and the axle/crank interface "suffers" the same force, whether the axle is lubed, or not. With a lubed axle the force is spread more equally, while with a dry mount you get the cranks stuck with friction, until the tightening stress shreads some material off and moves the crank up to the next stuck position - until all the tightening is done.

I understand this. Aside the fact that some manufacturers recommend lubing the bolt, which is the only thing that changes the force for a given tightening torque (axle lubrication playing no part there), yes, if you have no clue what tightening too much is, you are more likely to overtighten with a lubed bolt.

It could be a reason for such recommendations by some manufacturers. However, when defining a better procedure, I'd make it: better one when properly done, since you either didn't understand that, or you are "moving the goal posts" a bit?

It's not too much to expect from home mechanic, or a shop to have some sense of torques, have some decent lube at hand, and proper tools - is it?

As far as dry mounting goes, both are important and go hand in hand. Mounting damage when mounting dry is a thing to consider and is one of the things that prevents the crank from engaging the spindle properly - which later during use attributes to the fretting damage. But yes, we could look at it separately. For cartridge type BBs that don't require annual service, fretting damage is by far no1 consideration. For the cup and cone one - I'd also say fretting, especially for high mileage and heavy riders.


There, now we've finally defined the terms (I hope and believe ) and no more need to repeat yourself. Just provide some arguments.