Forces on Cranks
Andre Jute
Wouldn't it be more in keeping with the actual forces a crank has to
resolve to lighten/decorate it on its top and bottom surface rather
than at the front and the back?
****
Something about what has been called the "vanity" machining/forging on
bicycle cranks bothers me.
Consider a crank in action. At the pedal end there are two directions
of force on the crank, a circular motion roughly in the plane of the
crank (if we ignore the angling on the crank to clear the gubbins),
plus an offset twisting moment to the outside on the pedal, which is
at right angles to the crank. The offset force is stayed by the bottom
bracket end of the crank, and the observed twist will therefore be
larger at the pedal end. From the point of highest twist there is then
an unwinding action as the crank rotates. It seems to me likely that
over a full rotation the force in the up-down plane will be larger
than the twisting force on the crank. Whether at the point in the
rotation where the twisting force is the largest, it is fact larger
than the vertical force in the crank's plane of rotation would depend
on the design of the crank, the force of the pedalist, and the exact
offset of the pedals from the plane of the crank's rotation; we can
abstract from these details because my problem concerns the principle
of force in the crank, not an exact measurement.
Given this description of the forces on a crank, surely it follows
that any lightening (or vanity machining/forging) should be done on
the crank's top and bottom surfaces, not its outside and inside
permanently vertical faces. The tendency for vanity fluting by
machining or stamping on the classic model is towards creating an H-
beam or U-beam crank. In practice, as commonly seen on bicycle cranks,
the beam lies on its side with the connecting web vertical. That's
what bothers me. Shouldn't the two deepest faces be applied in the
vertical plane where they will be able to resolve the most torque,
with the web perpendicular to them? That is exactly the opposite of
the arrangement we invariably see now.
It seems to me that, because of the engineering considerations I have
laid out above, such "vanity" flutes on the vertical face of the crank
can have no structural justification, indeed the opposite applies:
their engineering effect is negative and destructive. Such fluting
merely creates undercuts which won't survive years of flexing without
becoming the locus of a fracture. Lightening machining/forging if
considered necessary should, if I am right, be carried out on the top
and/or bottom face of the crank.
Andre Jute
"The brain of an engineer is a delicate instrument which must be
protected against the unevenness of the ground." -- Wifredo-Pelayo
Ricart Medina
Peter Cole
Jobst has frequently posted on crank failures and causes. Several
pictures here:
Jobst Brandt
>> H-beam or U-beam crank. In practice, as commonly seen on bicycle
>> cranks, the beam lies on its side with the connecting web vertical.
>> That's what bothers me. Shouldn't the two deepest faces be applied
>> in the vertical plane where they will be able to resolve the most
>> torque, with the web perpendicular to them? That is exactly the
>> opposite of the arrangement we invariably see now.
>> It seems to me that, because of the engineering considerations I
>> have laid out above, such "vanity" flutes on the vertical face of
>> the crank can have no structural justification, indeed the opposite
>> applies: their engineering effect is negative and destructive.
>> Such fluting merely creates undercuts which won't survive years of
>> flexing without becoming the locus of a fracture. Lightening
>> machining/forging if considered necessary should, if I am right, be
>> carried out on the top and/or bottom face of the crank.
> Jobst has frequently posted on crank failures and causes. Several
> pictures here:
http://www.pardo.net/bike/pic/fail-001/000.html
The whole crank problem falls apart when the effective forces are
analyzed. Above all, a left hand thread and significant fretting
damage to both cranks at the pedal shaft shoulder indicate why many
cranks break across the "pedal eye" where the pedal is attached.
Beyond that, the torsion, radial (torque) loading and lateral bending
from the center of pressure on the pedal are consistently ignored.
The fretting of the pedal shaft face is the most important one to me
because I broke at least one crank per 10,000 miles for 30 years,
until I modified the interface to emulate the conical face on an
automobile lug nut. I have not had a crank failure in the last 20
years as a result.
Talking to crank manufacturers at InterBike trade show, I am convinced
that few if any have an idea where the forces are and have made no
stress concentration tests. That was brought out by the recent
failure of a Shimano Hollowtech crank right where one would expect it,
there where the crank diverges from the disk of the chainwheel
"spider" that in this design is extremely rigid.
I am amazed when one of these component manufacturers introduces a
reliable design, such as Shimano free-hubs that do not use screw-on
sprockets that warp and become extremely hard to remove... and of
course no screw-on freewheel.
Jobst Brandt
devel...@osoftollc.com
Would the crank arm be a suitable candidate for power (Watts)
measurement through the use of strain gauges and other circuitry?
-Tony
Jobst Brandt
Tony,
Where do all the paragraph breaks and empty lines go on your web
browser? This text gets difficult to follow in the form you post.
You might also delete items that do not affect your posting.
No, the crank is not suitable, as should be apparent by the plurality
of three dimensional bending and torsion it supports as has been
discussed in this thread. Even the BB spindle has complex stresses
that require careful separation to do any power measurements. That
could be done by a leaf lever that transmits only rotational torque,
its flexibility in its thin direction could be ignored.
As you see, no one has made such a device, that would be a boon to
tandem riders if practical because it could show power of each rider
on identical displays on each rider's handlebars to make clear who is
doing what part of propulsion work.
Jobst Brandt
Andre Jute
I've seen those, thanks. I didn't mention Jobst for fear that he would
go into a masochistic ecstacy about the bee in his bonnet about left-
hand threads, and never get around to what I want to discuss, which is
exactly what happened. -- AJ
Andre Jute
* Still Just Me * wrote:
> On Wed, 28 Apr 2010 07:54:33 -0700 (PDT), Andre Jute
> <fiul...@yahoo.com> wrote:
>
> <>
> >It seems to me that, because of the engineering considerations I have
> >laid out above, such "vanity" flutes on the vertical face of the crank
> >can have no structural justification, indeed the opposite applies:
> >their engineering effect is negative and destructive. Such fluting
> >merely creates undercuts which won't survive years of flexing without
> >becoming the locus of a fracture. Lightening machining/forging if
> >considered necessary should, if I am right, be carried out on the top
> >and/or bottom face of the crank.
>
> Maybe The key question would be whether or not the crank actually
> flexes significantly in the direction you suggest. If not, then the
> vanity flutes are irrelevant.
>
> IMHE, the (my vintage steel) frame flexes by large, visible amounts. I
> think the stiffness of the crank is far greater than the frame, based
> on observation with the bike in a trainer.
>
> At the same time, I do see some flex apparently introduced in the
> chainwheels from the cranks when on the road if I start to pedal in a
> poor way, pushing out towards the right when pushing hard. I think
> that's more of a technique issue than an engineering issue.
>
> So, my rough field observation tells me it's not an issue. But, there
> may be laboratory results that further detail. I can say that without
> pushing hard, it's all immaterial. It's only when you really "get on
> it" that it's noticeable.
I'm not viewing this as problem or a concern for my current cranks. I
have steel cranks and they don't appear to be stressed in the least.
But I'm thinking of designing cranks of my own and having them
machined, and then the question of the forces on the cranks comes up.
Not much point in having plain steel cranks cut just to have your own
design of plain steel crank -- I have plain steel cranks already! So
the question of decor/lightening arises, and with the question of
where it will do the least harm, and we're back at forces and vectors.
Andre Jute
The rest is magic hidden in the hub.
For rare hub gear bikes, visit Jute on Bicycles at
http://www.audio-talk.co.uk/fiultra/BICYCLE%20%26%20CYCLING.html
Andre Jute
I just included them all in my analysis above.
> The fretting of the pedal shaft face is the most important one to me
> because I broke at least one crank per 10,000 miles for 30 years,
> until I modified the interface to emulate the conical face on an
> automobile lug nut. I have not had a crank failure in the last 20
> years as a result.
Congratulations. Okay, now that you have that off your chest, dear
Jobst, do you agree with me that if fluting on a crank turns it into
some kind of an H or U sectional shape, the longer sides should be
vertical and the web horizontal (when the pedal is at the quarter to
three position)? In short, do you agree with me that lightening/
decoration is best applied to the top and bottom of the arm rather
than the vertical faces to the outside and the inside of the crank?
Snipped, more whining about component manufacturers that would be
amusingly scurrilous if I were not so pissed off at having my thread
diverted by the swarm of wasps on Jobst's belfry.
Do try to stick to the point, Jobst. I already know what to do about
the pedal/crank interface fretting: you've told us all that. Now I
want to move on to dealing with the other forces.
Andre Jute
Visit Jute on Amps at
http://www.audio-talk.co.uk/fiultra/
"wonderfully well written and reasoned information for the tube audio
constructor"
John Broskie TubeCAD & GlassWare
"an unbelievably comprehensive web site containing vital gems of
wisdom"
Stuart Perry Hi-Fi News & Record Review
Jobst Brandt
http://www.pardo.net/bike/pic/fail-001/000.html
Well you seem to forget fast so repetition is not out of place. Just
look at "H" and "U" shapes in structures and I'll repeat the rule once
more. The torsional stiffness of an element with other than round
cross section is like that of the largest inscribed solid circle. You
can test this by twisting a U-channel in the sheet metal section of
your local hardware store.
Piston rods on IC or steam engines have no significant torsion so they
are "H" shaped for bending in the vertical I-beam direction as I-beams
are. That is where the bicycle folks get confused, thinking pedal
cranks are connecting rods with pure rotational bending... which they
aren't.
Jobst Brandt
Andre Jute
***Given that someone *will* decorate/lighten a bicycle crank, is the
balance of forces on it such that material is less damagingly removed
from the top and bottom faces than from the outer and inner faces, as
happens now?***
"Given that someone will decorate/lighten a bicycle crank" means we're
not interested in the normative case (engineers telling us we can't do
what we already decided to do) but in starting at the managerial
decision already arrived at and moving forward. If you can't (or more
likely won't) help, at least try not to get in the way.
Andre Jute
Check out Andre's recipes at
http://www.audio-talk.co.uk/fiultra/FOOD.html
Frank Krygowski
At first glance, I assumed this post was intended as a parody of 17th
century technical writing - the sort produced before current
engineering vocabulary terms like vertical (vs. "up-down") tangential
(vs. "circular motion") torque (vs. "twisting force") were well
known. Based on that, I skipped the rest, as usual.
Now that I see that others are taking the question somewhat seriously,
the short answer to the question is:
No.
- Frank Krygowski
Frank Krygowski
> The torsional stiffness of an element with other than round
> cross section is like that of the largest inscribed solid circle.
Well, for some value of "like." A solid square bar is about 1.4 times
as stiff in torsion as the solid round bar whose diameter equals the
side of the square. The square is less efficient on a weight basis,
though.
To visualize torque capacity, relative stress levels and stress
directions of a non-circular torsion member, google "membrane analogy
torsion" or "soap bubble analogy torsion."
- Frank Krygowski
carl...@comcast.net
Dear Frank,
Just to make sure that I'm following you, the square cross-section
covering a circle like this . . .
http://i43.tinypic.com/6r7zog.jpg
. . . is stiffer in torsion because of the extra material at the
corners.
But if you melt the square bar and recast it as a circle, it becomes
even stiffer than the original bar because the extra material is
evenly distributed?
Maybe a dumb question, but would a triangle encompassing a circle be
even stiffer in torsion than a square encompassing the same circle,
while a pentagram would be less stiff?
That is, I'm wondering if the triangle is the stiffest and things
gradually decline with more sides until a circle is approximated, or
if something about the triangle makes it less stiff in torsion than
the square.
triangle stiffer than squware in torsion as inset circle
square 1.4 times as stiff in torsion as inset circle
pentagram between 1.4 and 1.0 times as stiff in torsion
hexagram less stiff than pentagram, stiffer than circle
. . . and so on, adding more and more sides to reach a circle
circle 1.0 stiff in torsion
Cheers,
Carl Fogel
Sergio Moretti
> On Wed, 28 Apr 2010 19:20:29 -0700 (PDT), Frank Krygowski
>
Yes, the triangle in the problem you state is stiffest. It also has
the largest cross-sectional area and is the least efficient shape.
- Sergio Moretti
thirty-six
This is old hat.
thirty-six
wrote:
> On Wed, 28 Apr 2010 07:54:33 -0700 (PDT), Andre Jute
>
>
> <>
>
> >It seems to me that, because of the engineering considerations I have
> >laid out above, such "vanity" flutes on the vertical face of the crank
> >can have no structural justification, indeed the opposite applies:
> >their engineering effect is negative and destructive. Such fluting
> >merely creates undercuts which won't survive years of flexing without
> >becoming the locus of a fracture. Lightening machining/forging if
> >considered necessary should, if I am right, be carried out on the top
> >and/or bottom face of the crank.
>
> flexes significantly in the direction you suggest. If not, then the
> vanity flutes are irrelevant.
>
> IMHE, the (my vintage steel) frame flexes by large, visible amounts. I
> think the stiffness of the crank is far greater than the frame, based
> on observation with the bike in a trainer.
>
> At the same time, I do see some flex apparently introduced in the
> chainwheels from the cranks when on the road if I start to pedal in a
> poor way, pushing out towards the right when pushing hard. I think
> that's more of a technique issue than an engineering issue.
>
> So, my rough field observation tells me it's not an issue. But, there
> may be laboratory results that further detail. I can say that without
> pushing hard, it's all immaterial. It's only when you really "get on
> it" that it's noticeable.
and ride like an urangutan.
thirty-six
Make the cross section of the crank circular. You are limited
anatomically how thick you can make the crankss, that's all. Cost may
mean you use less material, this can be put forward as being
'lightweight' and has been a good sales point for 'racer types' for
over a century.
thirty-six
> The whole crank problem falls apart
What an eloquent way of saying you need some castor oil.
> when the effective forces are
> analyzed. Above all, a left hand thread and significant fretting
> damage to both cranks at the pedal shaft shoulder indicate why many
> cranks break across the "pedal eye" where the pedal is attached.
Not on steel cranks and not on any alloy cranks I've used. I pedalled
awkward enough to bend a wheel built exactly to your methods with
exactly recommended equipment, I didn't bother with your fuss over
cranks except to lightly grease the tapers in 1988 of which I've
always since done, but with oil not mechanics grease. Everything done
up tight.#
> Beyond that, the torsion, radial (torque) loading and lateral bending
> from the center of pressure on the pedal are consistently ignored.
Go and weigh your leg in the cycling position and see how many
pounds. This is all the weight a crank needs to carry for a
proficient cyclist who rides flattish roads. The tangential force
applied to a crank from a pedal may be much higher but rarely exceeds
twice this value for a proficient cyclist not racing, hill climbing or
training for such.
> The fretting of the pedal shaft face is the most important one to me
> because I broke at least one crank per 10,000 miles for 30 years,
Urangutan
> until I modified the interface to emulate the conical face on an
> automobile lug nut. I have not had a crank failure in the last 20
> years as a result.
You mean since you stopped using Campagnolo Super Record cranks!
>
> Talking to crank manufacturers at InterBike trade show, I am convinced
> that few if any have an idea where the forces are and have made no
> stress concentration tests. That was brought out by the recent
> failure of a Shimano Hollowtech crank right where one would expect it,
> there where the crank diverges from the disk of the chainwheel
> "spider" that in this design is extremely rigid.
All with the aid of computer design, he, he.
>
> I am amazed when one of these component manufacturers introduces a
> reliable design, such as Shimano free-hubs that do not use screw-on
> sprockets that warp and become extremely hard to remove... and of
> course no screw-on freewheel.
Black trucks?
thirty-six
>
> Jobst has frequently posted on crank failures and causes. Several
> pictures here:
>
> http://www.pXXXXXXXXXXXXXX00.html
Cut to preserve the innocent.
Psst, he's a bit cranky.
thirty-six
> Where do all the paragraph breaks and empty lines go on your web
> browser?
Down the sink.
> This text gets difficult to follow in the form you post.
That'll be your eye problems again.
> You might also delete items that do not affect your posting.
Calm down.
>
> No, the crank is not suitable, as should be apparent by the plurality
> of three dimensional bending and torsion it supports as has been
> discussed in this thread. Even the BB spindle has complex stresses
> that require careful separation to do any power measurements. That
> could be done by a leaf lever that transmits only rotational torque,
> its flexibility in its thin direction could be ignored.
>
> As you see, no one has made such a device, that would be a boon to
> tandem riders if practical because it could show power of each rider
> on identical displays on each rider's handlebars to make clear who is
> doing what part of propulsion work.
>
> Jobst Brandt
Er, Bourlet's pedal as described in Bicycles and tricycles gives both
magnitude and direction of pedal force. As there is so other source
of productive energy input through a crank, the device was already
with us over a century ago,
thirty-six
> Congratulations. Okay, now that you have that off your chest, dear
> Jobst, do you agree with me
Huh, why do you want agreement from oddball?
> that if fluting on a crank turns it into
> some kind of an H or U sectional shape, the longer sides should be
> vertical and the web horizontal (when the pedal is at the quarter to
> three position)? In short, do you agree with me that lightening/
> decoration is best applied to the top and bottom of the arm rather
> than the vertical faces to the outside and the inside of the crank?
Make em round and stick polyethylene edges on the forward rotating
face for aerodynamics when you use that monster gear chasing trucks at
60mph plus.
>
All the different shapes you may have seen in alloy cranks are not
only due to conflicting hypothesis about which loading is most
important to any one particular rider, but also the hollow you see in
H-section cranks is historically due to the forging process in
improvement of grain in the forging. This same technique is used in
strong and light fighting knives and swords. Remember that bicycle
manufacturers rose from the ashes of redundant cutlers in some cases
and the use of such a technique of forging for crank manufacture is
hardly surprising to have arisen. Whether they managed to get the
orientation theoretically correct is irrelevant, the best alloy
cranks, being this type, would only be of real interest to racers.
Aerodynamic constraints as well as the dimensional space available
would mean that making the impression facing fore/aft a bit silly.
thirty-six
The rods are forged, the shape is relevant to the forging process.
thirty-six
Use pre-gummed 'stickers', they come in cereal packets aimed at
children.
thirty-six
MMMMmmm. Ha ha ahh ahaha.
thirty-six
Too seriously. I see that if decoration is wanted it can be applied
to any part of the crank as long as it's big enough to make that
decoration viable. Paint is a good choice.
andre...@aol.com
In my puter, the text was very easy to follow. The question appeared
by itself as a completely separated post.
Andre Jute
Andre Jute
That's okay, Franki Shavelegs. I'm not writing for the more arid-
minded fascists like you. Far be it from me to support by my language
your attempts to regiment everyone and every creative instinct.
>Based on that, I skipped the rest, as usual.
Par for the course, Kreepy Krygo offering opinions without reading the
question. Below we'll see with what dire result.
> Now that I see that others are taking the question somewhat seriously,
Poor old Kreey is so lacking in initiative, he has to wait for the
street corner gang to run past before he runs after them, shouting,
"Wait for me, I'm your leader!" And then he heads in the wrong
direction!
> the short answer to the question is:
>
> No.
Wrong answer, Franki-boy, especially from a "professor" of
engineering, even if at some jumped-up tech. You should have checked
what everyone else said: they said "yes". In fact, I already said
"yes". That's what a "professor" of engineering gets when he's a) bog-
ignorant and b) lets his personal dislikes of people influence his
"engineering" judgements and c) snaps out his "answer" without reading
the question and d) puts personalities before the facts.
> - Frank Krygowski
You're a dead loss to engineering -- and to cycling -- poor old
Franki. And now you've been caught out lying on professional
engineering matters for personal reasons. The alternative is that
you're an ignorant peasant. Choose only one.
Andre Jute
Krygo, he say, "Any old number is good number."
Andre Jute
> On Wed, 28 Apr 2010 19:20:29 -0700 (PDT), Frank Krygowski
>
"Inscribed", gee.
You might make your trick question more interesting still, dear Carl,
by noting that there are two ways to inscribe a triangle on a square,
by diagonally halving the square or by inscribing lines from two
corners to the middle of the opposite face and cutting away the
smaller triangles to the sides. Now, which of the three sizes of
triangles so formed will be the stiffest, which will be stiffest in
relation to material used, and which will be most efficient on other
parameters, say aerodynamically?
Andre Jute
Wide awake this time of the afternoon
Andre Jute
> This is old hat.
Depends whether the brim of the hat section is welded to something how
stiff it is. If open, the brim could gain considerable stiffness from
a small return of its own. -- AJ
thirty-six
As to pedal cranks:
Unless the grooves have been formed because of the forging process,
they will have an overall negative effect on the crank function.
The aerodynamics of the con-rods are also in consideration for fast
engines. The typical forging pattern is so superior for its cost that
it is rarely altered.
Welding fabricated cranks is a viable option. Cook Bros, I think is
one brand. Can be done in aluminium alloy as well as steel.
Andre Jute
This is the best point made in the responses in this thread so far. I
in fact thought of a hollow section but I don't fancy welded-on ends
for the BB and pedal mountings, and to stop the tube after drilling or
drawing, so the thing will have to be split lengthwise and then glued
(Tune round section hollow alloy cranks are glued lengthwise) or
welded together again. I'm very keen to have it made as one piece.
But how about this for a production process for a round almost-one-
piece crank: Take a bloc of steel, forge or machine a crankshaped
blank. Drill through end of BB barbell lengthwise to almost at pedal
end. Drilling a straight passage will leave thicker walls nearer BB
end. Stop hole at BB end with fine-threaded bolt just long enough to
go from outside to a little way into the now hollow arm. Machine now
hollow-shafted crank blank further, finishing up with a barbell shape,
small bulb at pedal end, bigger bulb at BB end, circular shaft tapered
from thick at BB end to thinner at pedal end. Now tap one end for
pedal and machine other end (right through center of stopping bolt)
for square taper.
A blacksmith way of making this crank would be to start with thickwall
hollow tube, fold over the ends repeatedly until he arrives at a
suitable block of solid metal at each end. Then beat the ends round,
machine the pedal threads and square taper, and polish with fine grit
and elbow grease, then apply black chrome. Ettore Bugatti made bent
hollow-centre solid-ended axles like that, with the added twist that
he started with solid metal and did the gun drilling in his own works.
Andre Jute
I think you will find the controlling element in Carl's riddle is the
word 'inset' (now wait for Krygo to complain that it should be
"inscribed" or even "included"). The triangle you posit is one made
with all the material of the square rod, not inscribed on it. Still,
Carl has confused his own riddle by talking about melting down and
recasting the square rod in the same post as he defines the problem as
one of insets.
Anyhow, all this Rod Science is for Railroad Minds and Other
Librarians. Those of us who live in the 21st century rather than the
19th, do Tube Science, with less material disposed edgewise (just for
you, Franki Shavelegs, "peripherally"). Some of us have even passed
through the stage of triangular small tubes which we called
Spaceframes, and to the ultimate largest size of tube for the job,
which we call Monocoque, because the cock crowed only once, that being
more efficient in an Engineering Religion.
Andre Jute
Who says the cargo cult is dead?
Andre Jute
> On 28 Apr, 19:25, Jobst Brandt <jbra...@sonic.net> wrote:
> > Beyond that, the torsion, radial (torque) loading and lateral bending
> > from the center of pressure on the pedal are consistently ignored.
>
> Go and weigh your leg in the cycling position and see how many
> pounds. This is all the weight a crank needs to carry for a
> proficient cyclist who rides flattish roads. The tangential force
> applied to a crank from a pedal may be much higher but rarely exceeds
> twice this value for a proficient cyclist not racing, hill climbing or
> training for such.
What is the source of these rather particular estimates, Trevor?
And what would the higher-end values (racing, hill climbing, an
awkward cyclist) amount to in terms of so many leg weights or whatever
other measure you prefer.
What was that movie called with the cute robot whose tagline was,
"Must have more data. Data! Data! Must have more data!" So far this
thread is thick on evasion, handwaving and abuse and thin on useful
data.
Andre Jute
"The brain of an engineer is a delicate instrument which must be
protected against the unevenness of the ground." -- Wifredo-Pelayo
Ricart Medina
Schwalbe BIg Apples on one's bike also help to stabilize the brain. --
Andre Jute
Andre Jute
> On 28 Apr, 22:16, Andre Jute <fiult...@yahoo.com> wrote:
>
> > Congratulations. Okay, now that you have that off your chest, dear
> > Jobst, do you agree with me
>
> Huh, why do you want agreement from oddball?
Because Jobst is running around waffling to avoid admitting that I got
it right. He's taken in Krygowski, who in consequence gave the wrong
answer to a simple engineering question! Lovel-ly!
> > that if fluting on a crank turns it into
> > some kind of an H or U sectional shape, the longer sides should be
> > vertical and the web horizontal (when the pedal is at the quarter to
> > three position)? In short, do you agree with me that lightening/
> > decoration is best applied to the top and bottom of the arm rather
> > than the vertical faces to the outside and the inside of the crank?
>
> Make em round and stick polyethylene edges on the forward rotating
> face for aerodynamics when you use that monster gear chasing trucks at
> 60mph plus.
You flatter me, Trevor. Where shall I send the brown envelope?
> All the different shapes you may have seen in alloy cranks are not
> only due to conflicting hypothesis about which loading is most
> important to any one particular rider, but also the hollow you see in
> H-section cranks is historically due to the forging process in
> improvement of grain in the forging. This same technique is used in
> strong and light fighting knives and swords. Remember that bicycle
> manufacturers rose from the ashes of redundant cutlers in some cases
> and the use of such a technique of forging for crank manufacture is
> hardly surprising to have arisen.
Okay, crank manufacturers get it wrong now because they always got it
wrong, because they are the descendents of horse fettlers. Makes
sense.
>Whether they managed to get the
> orientation theoretically correct is irrelevant, the best alloy
> cranks, being this type, would only be of real interest to racers.
Well, Jobst for one is convinced that it matters, though I must say
that anyone who expresses surprise that a crank which had ascended 600
vertical miles of Alps and then broke, as is true of one of the
examples in that oft-quoted site of broken cranks Jobst drools over,
is a cheapskate and an idiot who should have replaced his cranks long,
long since as routine maintenance.
> Aerodynamic constraints as well as the dimensional space available
> would mean that making the impression facing fore/aft a bit silly.
Imagination, dear boy, imagination. That imagination is in short
supply among the roadies (the pressure of all that excess of blood to
the brain because they have their arses in the air for hours every
day!) is no reason for me to emulate them; on the contrary.
Andre Jute
Visit Jute on Thisthatandtheother
http://www.audio-talk.co.uk/fiultra/OTHER%20MATTERS%20ARISING.html
Frank Krygowski
>
>
> Dear Frank,
>
> Just to make sure that I'm following you, the square cross-section
> covering a circle like this . . .
> http://i43.tinypic.com/6r7zog.jpg
>
> . . . is stiffer in torsion because of the extra material at the
> corners.
Yes.
>
> But if you melt the square bar and recast it as a circle...
IOW, keep the same cross sectional area (and weight)...
> ... it becomes
> even stiffer than the original bar because the extra material is
> evenly distributed?
Yes, the round bar of equivalent weight will be a little stiffer in
torsion than the square bar. About 13% stiffer.
>
> Maybe a dumb question, but would a triangle encompassing a circle...
OK, so now we're not talking about equivalent area (or weight) any
more.
> be
> even stiffer in torsion than a square encompassing the same circle,...
The triangle encompassing the circle has about twice the torsional
stiffness of the circle. It has less than twice the weight.
> while a pentagram would be less stiff?
Less than the triangle.
The general idea is this: For a member in torsion, the material at
the centroid of the cross section isn't working for you. It's just
adding weight. Material further out along a radius does more to
resist torsion. But it's not a simple relationship. The material at
the very tip of that triangular cross section is essentially
unstressed, i.e. not working for you.
But we've been talking stiffness here, which probably isn't the most
important issue. I think it's more accurate to say you want maximum
strength per unit weight, with still adequate stiffness. And you want
that strength for withstanding both torsion and bending. That won't
come from a triangle cross section. A hollow shape would be the
ultimate, I think. (I know that years ago, there were some hollow
steel cranks, brazed or welded up as assemblies.)
It's a complicated problem because the forces and the orientation of
the crank varies so much around the circle, plus you get various
unusual loading conditions (like, say, hitting a pothole while
standing on the lower pedal while coasting). You want a shape that's
optimum for the full range of loading.
While I understand Jobst's frustration at some of the designers and
their designs (including the pedal eye problem), I imagine that most
large firms are using design tools (software, etc.) that are more
capable than - say - the imaginings of pseudonymous Irish fantasy
writers!
- Frank Krygowski
N8N
To be consistent with your subject line, the whole thing would need to
be machine-turned after all other machining steps were finished.
NTTAWTT.
nate
(former owner of a '56 Studebaker Golden Hawk, complete with faux-
engine turned dashboard, which was one of the nicest features of the
whole car...)
AMuzi
>> Jobst has frequently posted on crank failures and causes. Several
>> pictures here:
>> http://www.pXXXXXXXXXXXXXX00.html
thirty-six wrote:
> Cut to preserve the innocent.
> Psst, he's a bit cranky.
Also linked from here:
http://www.yellowjersey.org/fail.html
notably "crank failure porn shot":
http://www.yellowjersey.org/photosfromthepast/DEADCRNX.JPG
--
Andrew Muzi
<www.yellowjersey.org/>
Open every day since 1 April, 1971
thirty-six
> On Apr 29, 11:08 am, thirty-six <thirty-...@live.co.uk> wrote:
> > Make the cross section of the crank circular. You are limited
> > anatomically how thick you can make the crankss, that's all. Cost may
> > mean you use less material, this can be put forward as being
> > 'lightweight' and has been a good sales point for 'racer types' for
> > over a century.
>
> This is the best point made in the responses in this thread so far.
It took me best part of a minute to think and compose. I have
considered the problem before. You may be able to use the steel
chainstays from a roadster or mountain bike. Get your smith to forge
weld the pedal axles to the ends and to pierce out (in) a taper at the
crank bearing to fit the regular ISO taper for alu cranks. The smith
will be able to correct alignment following welding much easier and
harden and temper the finished assembly after making everything smooth
by hammer or file. Give a specification for acceptable surface finish
irregularities, remember a finish can be obtained by filling and
painting.
> in fact thought of a hollow section but I don't fancy welded-on ends
> for the BB and pedal mountings, and to stop the tube after drilling or
> drawing, so the thing will have to be split lengthwise and then glued
> (Tune round section hollow alloy cranks are glued lengthwise) or
> welded together again. I'm very keen to have it made as one piece.
Flippin 'eck!
>
> But how about this for a production process for a round almost-one-
> piece crank: Take a bloc of steel, forge or machine a crankshaped
> blank.
FAIL , you're already worse off than a Chinese steel crank if you
machine it. Get this machining crap outa your head, its for
decoration only, not structural.
> Drill through end of BB barbell lengthwise to almost at pedal
> end. Drilling a straight passage will leave thicker walls nearer BB
> end. Stop hole at BB end with fine-threaded bolt just long enough to
> go from outside to a little way into the now hollow arm. Machine now
> hollow-shafted crank blank further, finishing up with a barbell shape,
> small bulb at pedal end, bigger bulb at BB end, circular shaft tapered
> from thick at BB end to thinner at pedal end. Now tap one end for
> pedal and machine other end (right through center of stopping bolt)
> for square taper.
>
> A blacksmith way of making this crank would be to start with thickwall
> hollow tube,
No, he starts with a billet and pierces it. He then hammers it out,
slowly drawing the length of the tube around a former. Unless your
smith has an auto hammer the costs arn't feasible even as a hobby. Or
are they?
> fold over the ends repeatedly until he arrives at a
> suitable block of solid metal at each end. Then beat the ends round,
> machine the pedal threads and square taper, and polish with fine grit
> and elbow grease, then apply black chrome. Ettore Bugatti made bent
> hollow-centre solid-ended axles like that, with the added twist that
> he started with solid metal and did the gun drilling in his own works.
He'd not seen a smith pierce a hole then.
Tim McNamara
<48cb5c21-03b4-400c...@o11g2000yqj.googlegroups.com>,
Frank Krygowski <frkr...@gmail.com> wrote:
> But we've been talking stiffness here, which probably isn't the most
> important issue. I think it's more accurate to say you want maximum
> strength per unit weight, with still adequate stiffness. And you
> want that strength for withstanding both torsion and bending. That
> won't come from a triangle cross section. A hollow shape would be
> the ultimate, I think. (I know that years ago, there were some
> hollow steel cranks, brazed or welded up as assemblies.)
Those broke, too.
This conversation has focused mainly on shape, it seems, but material
choice must also play a significant role.
How about an "isotruss"?
thirty-six
> On Apr 29, 11:25 am, thirty-six <thirty-...@live.co.uk> wrote:
>
> > On 28 Apr, 19:25, Jobst Brandt <jbra...@sonic.net> wrote:
> > > Beyond that, the torsion, radial (torque) loading and lateral bending
> > > from the center of pressure on the pedal are consistently ignored.
>
> > Go and weigh your leg in the cycling position and see how many
> > pounds. This is all the weight a crank needs to carry for a
> > proficient cyclist who rides flattish roads. The tangential force
> > applied to a crank from a pedal may be much higher but rarely exceeds
> > twice this value for a proficient cyclist not racing, hill climbing or
> > training for such.
>
> What is the source of these rather particular estimates, Trevor?
Totally unbiased and accurate guesstimation based on own experience.
It requires the particular application of an easy and light, round
pedalling technique.
>
> And what would the higher-end values (racing, hill climbing, an
> awkward cyclist) amount to in terms of so many leg weights or whatever
> other measure you prefer.
That's going back some. For hill climbing urang-utans and sprinters
who overgear then the forces may be double the body weight on the end
of the crank tangentially to it and probably up to 1.1/2 times bodt
weight as a bending load with the pedal at the bottom. On top of this
force you also have to factor in shock loading originating at the
wheel contact, although I think of this will be irrelevant if the
rider does not lock out his knees.
>
> What was that movie called with the cute robot whose tagline was,
> "Must have more data. Data! Data! Must have more data!" So far this
> thread is thick on evasion, handwaving and abuse and thin on useful
> data.
That's because few have a clue and those that do, realise that the
economics of creating the perfect crank far outweigh any benefits the
crank may bring over whatever basic racing crank is on offer, which is
probably in the region of sixty pounds a pair now. Your smith'd
cranks are gonna cost ya 800 spondoolies I reckon, all finished.
thirty-six
> In article
> <48cb5c21-03b4-400c-a1a9-05a2b519f...@o11g2000yqj.googlegroups.com>,
> Frank Krygowski <frkry...@gmail.com> wrote:
>
> > But we've been talking stiffness here, which probably isn't the most
> > important issue. I think it's more accurate to say you want maximum
> > strength per unit weight, with still adequate stiffness. And you
> > want that strength for withstanding both torsion and bending. That
> > won't come from a triangle cross section. A hollow shape would be
> > the ultimate, I think. (I know that years ago, there were some
> > hollow steel cranks, brazed or welded up as assemblies.)
>
> Those broke, too.
Probably due to "computer controlled heat treatment", rrelevant for a
complex for such as a pedal crank. A smith can watch the colour of
oxide form and draw the temper in different amounts for different
areas of the crank. The automated heat treatment relies upon the
design to correctly aid in the treatment. A smith uses his eyes and
experience.
>
> This conversation has focused mainly on shape, it seems, but material
> choice must also play a significant role.
So who has used wood and what where the problems?
>
> How about an "isotruss"?
How about chain oil and road dirt?
>
> http://www.isotruss.org/
Andre Jute
The Ettore ones were plain steel in a very high polish. They were on
cars intended for people who kept a few Krygowskis or whatever local
peasants they had chained in a(n inspection pit) in the garage to
polish plain polished steel daily. Modern owners have mostly had them
nickel-plated, I imagine.
But a machine-turned finish isn't a bad idea, if you can find a
craftsman who understands that if he makes it too regular, it will
look fake, like the "machine-turned" finish in some mid-century
American cars which was both too fine and too regular to be the real
thing.
> nate
>
> (former owner of a '56 Studebaker Golden Hawk, complete with faux-
> engine turned dashboard, which was one of the nicest features of the
> whole car...)
Sorry about that! But it's true: if it were more amateurish, it woulda
looked more real.
Andre Jute
Reformed petrol head
Car-free since 1992
Greener than thou!
Andre Jute
> On 29 Apr, 15:43, Andre Jute <fiult...@yahoo.com> wrote:
> > I'm very keen to have it made as one piece.
>
> Flippin 'eck!
>
> > But how about this for a production process for a round almost-one-
> > piece crank: Take a bloc of steel, forge or machine a crankshaped
> > blank.
>
> FAIL , you're already worse off than a Chinese steel crank if you
> machine it. Get this machining crap outa your head, its for
> decoration only, not structural.
Ever hear of grain-oriented billet stock, Trevor? From everything I
can gather here, the forces on bicycle cranks are nowhere near the
explosive (heh-heh) level of forces on automobile cranks. Hell, you
can buy machined billet alloy round cranks off the shelf, made as
halves and glued together by Tune in Germany; those Tunes are very
highly reputed.
Andre Jute
Feed a tree today, produce more CO2!
Michael Press
<48cb5c21-03b4-400c...@o11g2000yqj.googlegroups.com>,
Frank Krygowski <frkr...@gmail.com> wrote:
Thank you for the write up, Frank.
The designers have the tools, but do they type in _all_
the forces induced by the pedal?
The interested reader might consult Landau & Lifschitz,
Theory_of_Elasticity, pp 68-75. They derive the theory
of torsion for rods, and work out the torsional
rigidity for various rods.
rod with cross section circle with radius R:
(1/2).mu.pi.R^4.
rod with cross section ellipse with semi-axes a,b:
mu.pi.a^3.b^3/(a^2+b^2).
rod with cross section equilateral triangle with side a:
sqrt{3}.mu.a^4/80
long thin plate width d, thickness h, h << d:
(1/3).mu.d.h^3
cylindrical pipe with radii R,r.
(1/2).mu.pi.(R^4-r^4).
thin wall pipe with perimeter L, area S, thickness h:
4.mu.h.S^2/L.
If the pipe is cut along a generator it becomes a long
thin plate and the stiffness becomes (1/3).mu.L.h^3.
mu is the shear modulus.
Also they give formulae for the change in shape of the cross section.
--
Michael Press
Michael Press
Tim McNamara <tim...@bitstream.net> wrote:
Never mind chain ring tattoos--these babies scarify.
--
Michael Press
thirty-six
> On Apr 29, 7:02 pm, thirty-six <thirty-...@live.co.uk> wrote:
>
> > On 29 Apr, 15:43, Andre Jute <fiult...@yahoo.com> wrote:
> > > I'm very keen to have it made as one piece.
>
> > Flippin 'eck!
>
> > > But how about this for a production process for a round almost-one-
> > > piece crank: Take a bloc of steel, forge or machine a crankshaped
> > > blank.
>
> > FAIL , you're already worse off than a Chinese steel crank if you
> > machine it. Get this machining crap outa your head, its for
> > decoration only, not structural.
>
> Ever hear of grain-oriented billet stock, Trevor? From everything I
> can gather here, the forces on bicycle cranks are nowhere near the
> explosive (heh-heh) level of forces on automobile cranks.
Stop being piss ignorant. Typical automobile engines have torque
levels below a cyclist. They have power because of the speed they can
cycle at, typically around 7000-9000rpm in a modern road car and more
in a racing engine (18thousand or so for a bike).
Hell, you
> can buy machined billet alloy round cranks off the shelf, made as
> halves and glued together by Tune in Germany; those Tunes are very
> highly reputed.
and what about the ends where the crank turns?
Just shove a BMX crankset in there, you'll need a bigger shell.
>
thirty-six
> Thank you for the write up, Frank.
>
> The designers have the tools, but do they type in _all_
> the forces induced by the pedal?
Who's going to decide when and what data to collect, where and who
from, etc?
Who's gona pay for it all? What if they correct the wrong data and
its all wasted, eh?
Andre Jute
>
> But we've been talking stiffness here, which probably isn't the most
> important issue. I think it's more accurate to say you want maximum
> strength per unit weight, with still adequate stiffness. And you want
> that strength for withstanding both torsion and bending.
>
I've taken this profoundly silly statement up in the thread "The
engineering ignorance of Frank Krygowski" at
http://groups.google.ie/group/rec.bicycles.tech/browse_thread/thread/bc713fefc5c9e3e5?hl=en#
Andre Jute
> In article
> <48cb5c21-03b4-400c-a1a9-05a2b519f...@o11g2000yqj.googlegroups.com>,
Krygowski certainly won't. He thinks that when the component is strong
enough, it will also be stiff enough. The truth is the other way
round, that when the component is stiff enough to resist all scheduled
inputs, it is generally over-designed purely from the viewpoint of
strength. That Krygo doesn't know this bespeaks a complete lack of
real-life engineering experience. Don't ride any bike with components
designed by Krygo unless you fancy non-elective plastic surgery -- or
worse.
Andre Jute
“We must get rid of the Medieval Warm Period.” -- Jonathan Overpeck,
climate "scientist", IPCC supervising lead writer (i.e. spin doctor)
Andre Jute
"professor" of engineering such as you claim to be publicly got a
simple engineering problem wrong when everyone else got it right. When
can we look forward to your explanation?
Not holding my breath.
Andre Jute
Krygo, he say, "Any old number is good number."
carl...@comcast.net
Dear Frank,
Thanks for going over things--the membrane stuff was well beyond me:
http://en.wikipedia.org/wiki/Membrane_analogy
Cheers,
Carl Fogel
Frank Krygowski
>
> Thank you for the write up, Frank.
>
> The designers have the tools, but do they type in _all_
> the forces induced by the pedal?
>
> The interested reader might consult Landau & Lifschitz,
> Theory_of_Elasticity, pp 68-75. They derive the theory
> of torsion for rods, and work out the torsional
> rigidity for various rods.
>
> rod with cross section circle with radius R:
>
These days, any company can afford to buy a finite element analysis
program that will run on a personal computer. In fact, there are many
packages that are free and open source. They take knowledge and some
skill to use correctly, but they're far more reliable than the sort of
simple approximations allowed by such formulas.
Many, if not most, commercial packages will let you produce a solid
model of a proposed part, then do stress analysis. Once the model is
built, mesh generation is nearly (maybe not perfectly) automatic.
Once the mesh is generated, you can run analysis of any load you
choose. If you don't like the resulting stress levels, you can
redesign before you ever cut metal. It's far less expensive than the
old days, when every stage of the prototyping required metalworking
time plus physical testing.
Once the model is verified, it can (with proper software) be used to
generate CNC code to machine positive prototypes, or negative dies for
casting or forging. If preferred, rapid prototypes can also be
produced in plastic or even metal. Other, more specialized FEA
software can be used to help ensure the proper flow of metal in dies
and molds.
I know one engineer whose company does all the above for low-
production replacement parts for humans - i.e. artificial hips and the
like. I assume companies like Shimano and Campagnolo use such
processes to design and manufacture their products. It's likely that
small companies producing boutique CNC parts don't go this route, but
in those cases, I think you don't get what you pay for.
Again, these processes have now replaced a lot of approximations via
simple stress formulas, at least for production parts. (They're not
necessary for simple structural parts in non-weight-critical
applications.) They also replace Jutish attempts at visualization -
although any competent design engineer will cut out a lot of
nonsensical Jutish speculation just by his understanding of stress
distributions.
- Frank Krygowski
Frank Krygowski
>
>
> Dear Frank,
>
> Thanks for going over things--the membrane stuff was well beyond me:
> http://en.wikipedia.org/wiki/Membrane_analogy
Using it as a visualization tool isn't too tough. Here it is,
briefly:
Most students catch on quickly to the fact that in a solid circular
bar in torsion, the shear stress is maximum at the surface. The exact
center has zero shear stress - which means it's just dead weight.
Shear stress is proportional to the distance from the center - but
only for a round section.
Lots of people extend that principle wrongly to other shapes. For
example, in a previous job, I worked with a Brit who had been an
engineer for Lockheed. He described a huge fatigue testing rig that
used hexagonal bars as torsion elements. He said the rig's hex bars
failed in fatigue, and it baffled them because inspection showed the
fatigue cracks originated in the center of the flats, not at the
extreme (max radius) corners.
But the membrane analogy accurately predicts the max stress is in the
center of each flat. The extreme corners see zero shear stress.
Here's the analogy: To mentally model the torsion stress on a cross
section, imagine forming a wire outline of the shape. For that rig,
it would be a wire hexagon. Put a handle on it. Now imagine dipping
it in a soap bubble solution, just like a kid's bubble wand, and
imagine applying air pressure to one side, but only enough to bulge
the membrane, not blow a bubble.
The slope of the membrane is exactly analogous to the torsional
stress. Higher slope = higher stress and that occurs at the center of
each hexagonal flat. The angled wire pieces sort of reinforce the
bubble near the corners, giving it zero slope (i.e. zero stress) at
the exact corners. This visualization trick works for _any_ shape.
Another aspect is torque capacity, for a given max shear stress. The
torque capacity, or strength, is proportional to the volume of the
bubble. Thin shapes - like, for example, an I-beam - have little
bubble volume, i.e. little torque capacity.
To model a hollow shape, imagine the missing hollow area in the center
being represented by a flat but weightless stiff surface - say,
floating a piece of cardboard on your bubble before applying
pressure. Using this detail, you can compare a 1" solid bar to a tube
with 1" OD and 3/4" ID. The 3/4" disk representing the ID will be
lifted up by the air pressure, but remain flat. The resulting bubble
volume (or torque capacity) will be only a little less than that of
the solid 1" bar; but the weight will be a lot less. A tube is thus
not quite as strong as a solid bar, but far stronger per unit weight.
Without the membrane analogy, visualizing torsion stresses is tricky
for non-circular shapes. Once you catch on to it, it becomes easy to
make reasonable judgments and avoid a lot of tedious trial and error.
(Or outright foolishness.)
- Frank Krygowski
carl...@comcast.net
Dear Frank,
Alas, you made the fatal mistake of assuming that I could correctly
imagine how a soap bubble looks on a square frame!
For some reason, my imaginary soap bubble had impossibly steep slopes
at the corners, sort of like an M.C. Escher drawing.
Fortunately, I didn't have to make a square bubble-blower.
The internet provided a nice illustration of how the soap film bulges
and straightened me out:
http://emweb.unl.edu/Mechanics-Pages/J-Elder/Image120.gif
The wire diagram makes it clear that the steepest slopes are in the
middle of the flats, right where your UK engineer's hexagonal bars
failed in your interesting illustration.
Now your clear explanation makes sense to an intelligence meaner than
you imagined--thanks.
Cheers,
Carl Fogel
Michael Press
<76cd3bd7-ffe6-4e2d...@q32g2000yqb.googlegroups.com>,
Frank Krygowski <frkr...@gmail.com> wrote:
Does not answer the question.
Do they type in the bending moment on the crank
out of the plane of the chain wheel? Do they
read the answer?
--
Michael Press
thirty-six
> Does not answer the question.
> Do they type in the bending moment on the crank
> out of the plane of the chain wheel? Do they
> read the answer?
>
That's a rude question. Listen to the marketing department, you know
it makes sense.
Andre Jute
Make sure you get the direction of the force causing the stress too
out of Krygo. He already got it wrong once in this thread
http://groups.google.ie/group/rec.bicycles.tech/msg/f32fa4cc4569b898?hl=en&dmode=source
and hasn't yet recanted despite being twice prodded
http://groups.google.ie/group/rec.bicycles.tech/msg/0a2d4dd0ac79af23?hl=en&dmode=source
Thanks for trying to get real information, Michael. I wish you luck
with getting sense out of Krygowski.
Andre Jute
Be nice to run into a self-declared "expert" who knows what he's
talking about. Even once.
Andre Jute
explanation of your gross error on a simple engineering matter. Don't
you claim to be a "professor" of engineering? Is this how you teach
"engineering"? Even more distressing to parents who pay fees for you
to "teach" their children, do you allow your personal reactions to the
children to influence the marking of their papers, as you allow your
personal prejudices on RBT to drive you into lying (the only
alternative to ignorance) on professional matters?
Andre Jute
Relentless rigour -- Gaius Germanicus Caesar
Frank Krygowski
> In article
> <76cd3bd7-ffe6-4e2d-89b0-7eb8cc4ab...@q32g2000yqb.googlegroups.com>,
In the packages I've used, if I were designing a crank, I'd be looking
at a mesh model of the crank. It would be critical to properly
represent the boundary conditions (say, attachment to the crank axle)
and the loads; otherwise, results would be erroneous.
I could either apply a force and a couple at the location of the pedal
threads, or I could model the pedal axle as an extension to the crank
and apply a force at the appropriate place on that pedal axle. Either
strategy could be quickly repeated for force and couple in any
direction, once the mesh model was created. That would allow study of
the complete 360 degrees of crank rotation.
"Reading the answer" would consist partly of looking at the color
coded stress levels shown on the model, and reading information out of
resulting tables. You also get to see deflections, if you're
interested. (Or if you're a person who Jutishly confuses stiffness and
strength. But realistically, those folks wouldn't get far enough to
use the software.)
Here's a site that has some screenshots of their software. http://www.algor.com/
Again, I'd be shocked if Shimano, Campagnolo, and even Chinese or
Taiwanese manufacturers, weren't using the processes I described.
- Frank Krygowski
Simon Lewis
>
> In the packages I've used, if I were designing a crank, I'd be looking
> at a mesh model of the crank. It would be critical to properly
> represent the boundary conditions (say, attachment to the crank axle)
> and the loads; otherwise, results would be erroneous.
Wowa! Rocket science eat your heart out. "It's important to represent
the stress areas or the results would be wrong".
Who wudda thunk it? duh....
>
> I could either apply a force and a couple at the location of the pedal
> threads, or I could model the pedal axle as an extension to the crank
> and apply a force at the appropriate place on that pedal axle. Either
> strategy could be quickly repeated for force and couple in any
> direction, once the mesh model was created. That would allow study of
> the complete 360 degrees of crank rotation.
All hot air and displaying nothing more than you have a vague idea that
a green 3d wire mesh would rotate on the screen while you rub your chin.
>
> "Reading the answer" would consist partly of looking at the color
> coded stress levels shown on the model, and reading information out of
> resulting tables. You also get to see deflections, if you're
So lets get this straight : you would look at the RESULTS to see the
results ..
> interested. (Or if you're a person who Jutishly confuses stiffness and
> strength. But realistically, those folks wouldn't get far enough to
> use the software.)
>
> Here's a site that has some screenshots of their software. http://www.algor.com/
>
> Again, I'd be shocked if Shimano, Campagnolo, and even Chinese or
> Taiwanese manufacturers, weren't using the processes I described.
>
> - Frank Krygowski
Maybe you could advise them. You sure seem to know an awful lot
about it.
Seriously Frank, who do you think you are kidding with this airy fairy
nonsense?
Tim McNamara
Simon Lewis <simonle...@gmail.com> wrote:
Seriously, Simon, who do you think you're impressing with your fact-free
critique?
By all means, if you are competent in this field then post your own
analysis and procedures. If you are not competent in this field, well,
then, we've already discussed the Kruger-Dunning effect.
Michael Press
<d1072aed-3af3-4c34...@t21g2000yqg.googlegroups.com>,
Frank Krygowski <frkr...@gmail.com> wrote:
Am I?
> But realistically, those folks wouldn't get far enough to
> use the software.)
>
> Here's a site that has some screenshots of their software. http://www.algor.com/
>
> Again, I'd be shocked if Shimano, Campagnolo, and even Chinese or
> Taiwanese manufacturers, weren't using the processes I described.
The recent hollow crank from Shimano is a jungle of stress risers.
New Al cranks continue put all the stiffness in the direction
it is least needed: in the plane of the chain wheel.
--
Michael Press
Jobst Brandt
>>>>> (1/2).mu.pi.R^4. [etc.]...
> Am I?
>> Here's a site that has some screenshots of their software:
>> Again, I'd be shocked if Shimano, Campagnolo, and even Chinese or
>> Taiwanese manufacturers, weren't using the processes I described.
> The recent hollow crank from Shimano is a jungle of stress risers.
> New Al cranks continue put all the stiffness in the direction it is
> least needed: in the plane of the chain wheel.
That is the proof that they don't and are incapable of properly
analyzing the stress on cranks, as I have said for years now. They
also don't understand why they use left hand threads or they would
stop doing that and also not load threads with dynamic radial loads.
No FEA will help them understand this because it is a dynamics and
tribological problem, not a stress problem.
--
Jobst Brandt
Andre Jute
Here's an example of Frank Krygowski's "engineering":
***
Andre Jute: "...vanity" flutes on the vertical face of the crank can
have no structural justification... Lightening machining/forging if
considered necessary should ... be carried out on the top and/or
bottom face of the crank."
Frank Krygowski: "No."
***
You really can't expect much except posturing from a self-declared
"engineer" who can make a silly error like that.
thirty-six
> Frank Krygowski <frkry...@gmail.com> wrote:
That's probably because they used the cheapest method of manufacture
to incorporate the designers features for which the marketing
department can sell at the highest price..
thirty-six
None of the above. I'm certain they have capable engineers, they also
have a marketing department. The marketing department sell NEW,
BETTER and BEST for each new year. The production facilities are
unable to manufacture to the standards the design requires at the cost
wanted in order to maximise profits. They basically sell you an idea
and then give you (s)crap.
Frank Krygowski
> In article
> <d1072aed-3af3-4c34-8e5a-97410a2f0...@t21g2000yqg.googlegroups.com>,
> Frank Krygowski <frkry...@gmail.com> wrote:
>
> > "Reading the answer" would consist partly of looking at the color
> > coded stress levels shown on the model, and reading information out of
> > resulting tables. You also get to see deflections, if you're
> > interested. (Or if you're a person who Jutishly confuses stiffness and
> > strength.
>
> Am I?
I doubt that very much.
The "you" in the paragraph I wrote was intended to mean the person
using the software. Sorry if that wasn't clear.
And as I said, anyone who gained competence with FEA software would be
way beyond Jute's level of understanding. My guess is that almost all
RBT readers understand the difference between stiffness and strength.
- Frank Krygowski
Michael Press
<016502bb-5a87-4e44...@n15g2000yqf.googlegroups.com>,
Frank Krygowski <frkr...@gmail.com> wrote:
My main point is that manufacturers do no even look for
strain out of the plane of the chain wheel. No tool,
however well made, will get an engineer to look at
something he does not want to see. If they did see it
then they would make cranks to take up the strain at
the bottom of the stroke. Notice that broken cranks
show the fracture growing in a direction that supports my claim.
Chalo uses cranks made from steel tubes, and they pass the test.
--
Michael Press
Frank Krygowski
> Frank Krygowski <frkry...@gmail.com> wrote:
>
>
>
> > On Apr 30, 4:16 pm, Michael Press <rub...@pacbell.net> wrote:
> > > In article
> > > <d1072aed-3af3-4c34-8e5a-97410a2f0...@t21g2000yqg.googlegroups.com>,
> > > Frank Krygowski <frkry...@gmail.com> wrote:
>
> > > > "Reading the answer" would consist partly of looking at the color
> > > > coded stress levels shown on the model, and reading information out of
> > > > resulting tables. You also get to see deflections, if you're
> > > > interested. (Or if you're a person who Jutishly confuses stiffness and
> > > > strength.
>
> > > Am I?
>
> > I doubt that very much.
>
> > The "you" in the paragraph I wrote was intended to mean the person
> > using the software. Sorry if that wasn't clear.
>
> > And as I said, anyone who gained competence with FEA software would be
> > way beyond Jute's level of understanding. My guess is that almost all
> > RBT readers understand the difference between stiffness and strength.
>
> My main point is that manufacturers do no even look for
> strain out of the plane of the chain wheel. No tool,
> however well made, will get an engineer to look at
> something he does not want to see. If they did see it
> then they would make cranks to take up the strain at
> the bottom of the stroke.
Did you mean strain, or stress? They're different.
- Frank Krygowski
Andre Jute
> then they would make cranks to take up the strain at
> the bottom of the stroke.
You misspoke, strain for stress. Strain would be the elongation caused
by the load expressed as a fraction of the original unit length. I'm
sure you know this, and everyone else understood you correctly, but
the wretched Krygowski will soon be foaming fascistically at the mouth
about correct terminology.
Andre Jute
A little inaccuracy sometimes saves tons of explanation. --H.H.Munro
("Saki")(1870-1916)
Visit Jute on Amps at http://members.lycos.co.uk/fiultra/
"wonderfully well written and reasoned information
for the tube audio constructor"
John Broskie TubeCAD & GlassWare
"an unbelievably comprehensive web site
containing vital gems of wisdom"
Stuart Perry Hi-Fi News & Record Review
Andre Jute
>
> Did you mean strain, or stress? They're different.
>
> - Frank Krygowski
Surely, since you insist with such prissy-mouthed fervour on precise
terminology, Kreepy Krygo, you mean *unit* strain.
LOL.
Michael Press
<c3963b38-9fca-4c0d...@a34g2000yqn.googlegroups.com>,
Frank Krygowski <frkr...@gmail.com> wrote:
I meant strain.
--
Michael Press
Michael Press
<cca7efe8-c058-4499...@a27g2000prj.googlegroups.com>,
Andre Jute <fiul...@yahoo.com> wrote:
> On May 1, 5:01 am, Michael Press <rub...@pacbell.net> wrote:
>
> > then they would make cranks to take up the strain at
> > the bottom of the stroke.
>
> You misspoke, strain for stress. Strain would be the elongation caused
> by the load expressed as a fraction of the original unit length. I'm
> sure you know this, and everyone else understood you correctly, but
> the wretched Krygowski will soon be foaming fascistically at the mouth
> about correct terminology.
I cited Landau and Lifschitz in this thread.
Do you think I have not studied it?
--
Michael Press
Jay Beattie
I know I can look it up, but would you be so kind as to explain stress
versus strain for us non-engineers. Also, wouldn't you expect the
engineering department of a Campagnolo or Shimano (as compared to a
one-off CNC operation) to be aware of all the stresses and strains
seen by a crank? I know that design or marketing departments can
override good engineering, but it is not like the engineers at Shimano
or Campagnolo know less than those here on RBT. -- Jay Beattie.
David Scheidt
:I know I can look it up, but would you be so kind as to explain stress
:versus strain for us non-engineers. Also, wouldn't you expect the
Stress is force; strain is deformation as a result of that force.
:engineering department of a Campagnolo or Shimano (as compared to a
:one-off CNC operation) to be aware of all the stresses and strains
:seen by a crank? I know that design or marketing departments can
:override good engineering, but it is not like the engineers at Shimano
:or Campagnolo know less than those here on RBT. -- Jay Beattie.
You'd think so.
--
This is a randomly numbered sig.
Andre Jute
> Andre Jute <fiult...@yahoo.com> wrote:
>
> > On May 1, 5:01 am, Michael Press <rub...@pacbell.net> wrote:
>
> > > then they would make cranks to take up the strain at
> > > the bottom of the stroke.
>
> > You misspoke, strain for stress. Strain would be the elongation caused
> > by the load expressed as a fraction of the original unit length. I'm
> > sure you know this, and everyone else understood you correctly, but
> > the wretched Krygowski will soon be foaming fascistically at the mouth
> > about correct terminology.
>
> I cited Landau and Lifschitz in this thread.
> Do you think I have not studied it?
>
> --
> Michael Press
All right, humble apologies for mistaking your intention.
Still, why, practically, would you wish to work with the strain? After
all, Young's Modulus is usually given in tables and Poisson's ratio
too, and they're far more useful than knowing strain by itself, for
instance between E and Poisson you can calculate the shear modulus of
the material.
I mean, I understand intellectual curiosity, but this is a tech
conference, applied basic physics rather than original physics... This
thread is full of the instinctive luddite reaction of the Neanderthals
like Krygowski and his hangers-on to any attempt at innovation.
Andre Jute
Visit Andre's recipes:
http://www.audio-talk.co.uk/fiultra/FOOD.html
Frank Krygowski
> Jay Beattie <jbeat...@lindsayhart.com> wrote:
>
> :I know I can look it up, but would you be so kind as to explain stress
> :versus strain for us non-engineers. Also, wouldn't you expect the
>
> Stress is force; strain is deformation as a result of that force.
More specifically, stress is force per unit area. Force would be
measured in pounds, while stress is measured in pounds per square
inch.
Strain is deformation divided by original length. Deformation would
be measured in inches, while strain is measured in inches per inch.
(Actually, microinches per inch would be more typical, because strain
values are so small for metals.)
> :engineering department of a Campagnolo or Shimano (as compared to a
> :one-off CNC operation) to be aware of all the stresses and strains
> :seen by a crank? I know that design or marketing departments can
> :override good engineering, but it is not like the engineers at Shimano
> :or Campagnolo know less than those here on RBT. -- Jay Beattie.
>
> You'd think so.
Yes, you would think so.
What started this dispute was a fiction writer pretending his
imagination is more accurate than the skills, design tools and
experience of all the engineers who had been designing cranks for the
past 100 years. It's a ridiculous idea.
Very few manufacturers put fiction writers in charge of mechanical
design. Only a few people fail to understand why.
- Frank Krygowski
Michael Press
<ededa442-5785-4e84...@k25g2000prh.googlegroups.com>,
Andre Jute <fiul...@yahoo.com> wrote:
> On May 1, 4:58 pm, Michael Press <rub...@pacbell.net> wrote:
> > In article
> > <cca7efe8-c058-4499-8beb-a8e801e04...@a27g2000prj.googlegroups.com>,
> > Andre Jute <fiult...@yahoo.com> wrote:
> >
> > > On May 1, 5:01 am, Michael Press <rub...@pacbell.net> wrote:
> >
> > > > then they would make cranks to take up the strain at
> > > > the bottom of the stroke.
> >
> > > You misspoke, strain for stress. Strain would be the elongation caused
> > > by the load expressed as a fraction of the original unit length. I'm
> > > sure you know this, and everyone else understood you correctly, but
> > > the wretched Krygowski will soon be foaming fascistically at the mouth
> > > about correct terminology.
> >
> > I cited Landau and Lifschitz in this thread.
> > Do you think I have not studied it?
>
> All right, humble apologies for mistaking your intention.
>
> Still, why, practically, would you wish to work with the strain? After
> all, Young's Modulus is usually given in tables and Poisson's ratio
> too, and they're far more useful than knowing strain by itself, for
> instance between E and Poisson you can calculate the shear modulus of
> the material.
>
> I mean, I understand intellectual curiosity, but this is a tech
> conference, applied basic physics rather than original physics... This
> thread is full of the instinctive luddite reaction of the Neanderthals
> like Krygowski and his hangers-on to any attempt at innovation.
When the strain is large the Al alloy crank fatigues faster.
Also riders can notice when a crank flexes more.
--
Michael Press
thirty-six
> Also riders can notice when a crank flexes more.
Only a pair or tyres give more than the cranks do.
Andre Jute
STRAIN = (length extension caused by load/length before loading)
STRESS = (load/(width x thickness)) lbf/in^2
which is a measure of the load compared to the amount of material
carrying the load.
Note that strain is a dimensionless ratio and stress is the same in
any material, so that in order not to overstress less capable
materials larger dimensions must be chosen.
Simple, eh?
More useful is the Modulus of Elasticity, which relates stress and
strain:
E = (Stress/Strain) lbf/in^2
which is essential for discussing Life After Mild Steel.
Adapted from DESIGNING AND BUILDING SPECIAL CARS by Andre Jute,
published by BT Batsford, London and Robert Bentley, Boston
>Also, wouldn't you expect the
> engineering department of a Campagnolo or Shimano (as compared to a
> one-off CNC operation) to be aware of all the stresses and strains
> seen by a crank?
Of course you would. It is only on RBT that the little losers know
better than everyone else.
> I know that design or marketing departments can
> override good engineering, but it is not like the engineers at Shimano
> or Campagnolo know less than those here on RBT. -- Jay Beattie.
That aesthetic design and marketing departments can "override good
engineering" is a street myth spread by people who aren't good enough
to get jobs at places like Shimano and Campagnolo. It just doesn't
happen. What is far more likely is that trying to make a component to
a price point causes restrictions on production methods, for instance
having to use a casting instead of a forging, but within the price
point the engineer in charge will still make his choice, still be in
charge of the engineering.
If every product were made perfectly, the same clowns who whine that
the engineers weren't given a free hand would be the first to whine
that the price is outrageous.
Andre Jute
The IPCC -- longest hand job in the history of mass hysteria -- has
now lasted almost twice as long as the Third Reich
Andre Jute
> What started this dispute
Dispute? I don't see a dispute. I see a fellow who tells us he's a
"professor of engineering called Frank Krygowski" making a gross
engineering error:
Here, let's make it real simple for you, Krygo:
***
Andre Jute: "...vanity" flutes on the vertical face of the crank can
have no structural justification... Lightening machining/forging if
considered necessary should ... be carried out on the top and/or
bottom face of the crank."
Frank Krygowski: "No."
***
It is real simple. Do you stand by that erroneous statement,
Krygowski? Yes or no.
>was a fiction writer pretending his
> imagination is more accurate
Accurate? You're projection your obsessions on me, Kreepy. I hire a
lathe operator to be accurate.
>than the skills, design tools and
> experience of all the engineers who had been designing cranks for the
> past 100 years.
Really? You're over-sensitive, Kreepy. All I did was ask a question,
copied out above. If you're too dumb to understand sentences with
subsidiary clauses, that's your problem, not ours.
>It's a ridiculous idea.
Is it? What have you ever designed, Frank Krygowski? You're an ivory
tower warrior. I'm not. At least some of what I've designed and built
has been published in peer reviewed books, whereas I fail to find a
single book by you. Just who the fuck do you think you are, sonny?
(That leaves aside the ludicrously fascist attitude of this creep
Krygowski that only approved diploma-holders are entitled to
imagination...)
> Very few manufacturers put fiction writers in charge of mechanical
> design.
No one would put an idiot like Krygowski in charge of anything except
a classroom. Those who can, do, those who can't, teach.
>Only a few people fail to understand why.
Pretty obvious to anyone who comes into contact with Krygowski.
Andre Jute
Never more brutal than he has to be -- Nelson Mandela
Michael Press
<1538ad0f-95d0-4d9b...@a2g2000prd.googlegroups.com>,
Jay Beattie <jbea...@lindsayhart.com> wrote:
Very roughly speaking, strain is change of shape of a
body. It is a dimensionless quantity. It measures the
change in distance between two close, small bits
divided by the original distance between the two small
bits. To specify the strain at any point we need to
give how much the point moves in each of three
dimensions relative to other bits around it. Relative
to another bit, one bit can move along the line
connecting them, and can move perpendicular to the line
connecting them (shear). After a bit of work it is
shown that six numbers suffice at each point in the
body to specify the change in shape. They forms a three
by three symmetric tensor.
Stress gives the forces acting on a small bit of the
body, roughly speaking. Roughly because the dimensions
of stress is the same as pressure: force/area. Again,
it takes a symmetric tensor to specify the stress; six
numbers. Consider a small facet in the body. The force
acting on the plane can be resolved into a component of
force perpendicular to the plane and two components in
the plane. The two latter forces are shear forces.
Divide the forces by the area of the small facet to get
elements of the stress.
As for the question about Shimano and Campagnolo
engineers, Jobst (and now I) are asking the same
question. Welcome aboard.
--
Michael Press
Frank Krygowski
>
> As for the question about Shimano and Campagnolo
> engineers, Jobst (and now I) are asking the same
> question.
I don't know how many of the crank failure photos are of designs that
predate the use of FEA. But in any case:
I think Jobst is right about his conical taper interface for pedals
and cranks. Its use would reduce failures. The cyclical lateral load
on typical pedal threads isn't very good design, for the reasons he's
described.
OTOH, it's probably hard to convince (say) Shimano or Campagnolo to
improve the pedal attachment, partly because the existing design looks
just fine to most people. (How often have you heard _anybody_
complain about it?) Also, the current failure rate is extremely low.
Obviously, it hasn't caught the attention of the Consumer Product
Safety Commission, and those folks sometimes jump on a company for
quite rare failures.
This discussion group includes some people who are quite sophisticated
technically, and some people who put their equipment to pretty extreme
tests, compared to the typical cyclist. Jobst mashes up mountains in
high gears and does high mileage. Chalo outweighs two average
cyclists. That gives both of those guys plenty of failures to
examine, and the opportunity to suggest improvements.
All this is quite different from what's happening with Jute (AKA
McCoy). I've seen no hint he's addressing any real mode of crank
failure, and even if he were, he's approaching it as an overconfident
tyro.
Bike design has evolved for over 100 years, and the improvements have
been tremendous. But for every Jobst or Chalo inventing real
improvement, there's a Jute proposing things like
http://forums.mtbr.com/showthread.php?t=501193
or even
http://tinyurl.com/bre9xn
- Frank Krygowski
Andre Jute
Frank Krygowski wrote:
> All this is quite different from what's happening with Jute (AKA
> McCoy).
Poor old Franki "Shavelegs" Krygowski. Shown up as an engineering
ignoramus, he now claims to know my own name better than I do...
> I've seen no hint he's addressing any real mode of crank
> failure,
Who cares shit what you've "seen", Krygo. I don't ask the permission
of a jumped-up peasant like you to do anything at all. And your claim
that I'm "addressing any real mode of crank failure" in my new crank
design is a lie. Furthermore, you implication that "modes of crank
failure" only come into "real" existence (LOL!) when you approve of
them is a grandiose delusion. Do you also imagine your name is
Napoleon?
>and even if he were, he's approaching it as an overconfident
> tyro.
That's a lie too, but even if it were true, what is of your business,
Krygo? I don't need your permission, you fascist control freak. Even
if some misguided legislator were to make a law to put you in charge,
I'm accustomed to stepping in the faces of your kind. This is my
thread and you've maliciously made it odious.
> Bike design has evolved for over 100 years, and the improvements have
> been tremendous.
Aw, sheet, more platitudes from this useless clown Krygowski. Do you
want him, Lewis? We'll pay you to take him away.
>But for every Jobst or Chalo inventing real
> improvement, there's a Jute proposing things like
> http://forums.mtbr.com/showthread.php?t=501193
This is a lie. I didn't propose a circular bike, though if I'd known
it would give poor old Kreepy palpitations, I would've.
> or even
> http://tinyurl.com/bre9xn
This is a lie too, which arises because that fool Krygowski is too
insensitive and uncultured to grasp that, despite the way I speak, my
engineering is Calvinist, aimed at simplification.
For those few for whom the slimeball Krygowski's antics, and those of
his scummy claque, have not yet ruined the thread, and for the record,
my question, far from the monstrous edifice of fanciful motives and
inventions the obsessed maniac Frank Krygowski has now built on it,
was only this:
***Considering all the forces on it, would it not be less damaging in
an engineering sense to cut lightening and/or decorative patterns on
the upper and lower faces of the crank than on the vertical faces?***
Why that should send Kreepy Krygo off foaming at the mouth, God alone
knows. Or perhaps it is his memory of making a fool of himself over it
by claiming the answer is "No" which of course proves conclusively
that Frank Krygowski is an an engineering ignoramus.
Meanwhile I'm having fun putting down the wretched, undeserving little
man. Us good guys deserve an easy target every now and again.
Andre Jute
Visit Jute on Amps at
http://www.audio-talk.co.uk/fiultra/
Tim McNamara
<f77e31b5-1a2e-4dda...@s2g2000yqa.googlegroups.com>,
Frank Krygowski <frkr...@gmail.com> wrote:
> On May 1, 3:09�pm, Michael Press <rub...@pacbell.net> wrote:
> >
> > As for the question about Shimano and Campagnolo engineers, Jobst
> > (and now I) are asking the same question.
>
> I don't know how many of the crank failure photos are of designs that
> predate the use of FEA. But in any case:
>
> I think Jobst is right about his conical taper interface for pedals
> and cranks. Its use would reduce failures. The cyclical lateral
> load on typical pedal threads isn't very good design, for the reasons
> he's described.
>
> OTOH, it's probably hard to convince (say) Shimano or Campagnolo to
> improve the pedal attachment, partly because the existing design
> looks just fine to most people. (How often have you heard _anybody_
> complain about it?)
Outside of r.b.t., you mean?
Googling will find failure reports and sometimes analysis. Here's one
online museum of broken bike bits, which was probably linked already in
the thread.:
Jay Beattie
> In article
> <f77e31b5-1a2e-4dda-8d4e-8829170f7...@s2g2000yqa.googlegroups.com>,
>
By the way, would FEA make any difference as to the pedal eye
failures? It's not a matter of the material being insufficient, it's
cracks caused by gouging, at least as I understand it. I've broken a
half dozen cranks or more, mostly Campy -- none at the pedal eye --
but undoubtedly due to some stress riser and lots of load cycles.
These seem to be more machining or material flaws than inadequate
engineering, viz., bad load calculations and material selection. --
Jay Beattie.
Michael Press
<9c4c7f9e-f452-46a3...@g5g2000pre.googlegroups.com>,
A threaded fastener should only be loaded along its
axis notwithstanding all designs you see that do not
abide by this rule: bottom bracket threads, for
instance. The fastening bolt holding a square taper
crank on the spindle is used properly. A threaded
fastener should be torqued in sufficiently so as to
prevent unscrewing. Screwing a pedal into the crank
then standing on the end of the peda puts stress on
threads they are not designed for. Jobst's angular
collet between the pedal and the crank works to
redirect transverse force on the threads into axial
force that the threads are designed for.
--
Michael Press
Jay Beattie
> In article
> <9c4c7f9e-f452-46a3-99fd-7bf401bfc...@g5g2000pre.googlegroups.com>,
> Jay Beattie <jbeat...@lindsayhart.com> wrote:
>
>
>
>
>
> > On May 1, 3:04 pm, Tim McNamara <tim...@bitstream.net> wrote:
> >> In article
> >> <f77e31b5-1a2e-4dda-8d4e-8829170f7...@s2g2000yqa.googlegroups.com>,
> >> Frank Krygowski <frkry...@gmail.com> wrote:
> >>> On May 1, 3:09 pm, Michael Press <rub...@pacbell.net> wrote:
>
> >>>> As for the question about Shimano and Campagnolo engineers, Jobst
> >>>> (and now I) are asking the same question.
>
> >>> I don't know how many of the crank failure photos are of designs that
> >>> predate the use of FEA. But in any case:
>
> >>> I think Jobst is right about his conical taper interface for pedals
> >>> and cranks. Its use would reduce failures. The cyclical lateral
> >>> load on typical pedal threads isn't very good design, for the reasons
> >>> he's described.
>
> >>> OTOH, it's probably hard to convince (say) Shimano or Campagnolo to
> >>> improve the pedal attachment, partly because the existing design
> >>> looks just fine to most people. (How often have you heard _anybody_
> >>> complain about it?)
>
> >> Outside of r.b.t., you mean?
>
> >> Googling will find failure reports and sometimes analysis. Here's one
> >> online museum of broken bike bits, which was probably linked already in
> >> the thread.:
>
>
> > By the way, would FEA make any difference as to the pedal eye
> > failures? It's not a matter of the material being insufficient, it's
> > cracks caused by gouging, at least as I understand it. I've broken a
> > half dozen cranks or more, mostly Campy -- none at the pedal eye --
> > but undoubtedly due to some stress riser and lots of load cycles.
> > These seem to be more machining or material flaws than inadequate
> > engineering, viz., bad load calculations and material selection. --
>
> A threaded fastener should only be loaded along its
> axis notwithstanding all designs you see that do not
> abide by this rule: bottom bracket threads, for
> instance. The fastening bolt holding a square taper
> crank on the spindle is used properly. A threaded
> fastener should be torqued in sufficiently so as to
> prevent unscrewing. Screwing a pedal into the crank
> then standing on the end of the peda puts stress on
> threads they are not designed for. Jobst's angular
> collet between the pedal and the crank works to
> redirect transverse force on the threads into axial
> force that the threads are designed for.
>
failure? I just don't see many modern cranks that haved failed at the
pedal eye. Jobst's experience was with NR cranks, although he did his
collet job on a mid-80s Dura Ace, IIRC. -- Jay Beattie.
Michael Press
<34fc7fb4-bd95-4199...@j36g2000prj.googlegroups.com>,
Jay Beattie <jbea...@lindsayhart.com> wrote:
Some cranks failure pictures show a crack between the
pedal eye and the outside of the crank.
> I just don't see many modern cranks that haved failed at the
> pedal eye. Jobst's experience was with NR cranks, although he did his
> collet job on a mid-80s Dura Ace, IIRC.
Pedaling force on the pedal spindle makes it move
transversely in the crank threads slightly. The
direction of the movement rotates in the crank threads.
This makes the pedal spindle walk or precess in the
threads. When the left side threads are normal helix,
the pedal walks out of the threads; hence opposite
handed threads in the left and right cranks. This is
the biggest consequence. Another consequence is that
the threads wear away; gall.
--
Michael Press
Tim McNamara
<9c4c7f9e-f452-46a3...@g5g2000pre.googlegroups.com>,
Jay Beattie <jbea...@lindsayhart.com> wrote:
Gouging can be an instigator of failure, but TA used to provide flat
washers between the shoulder of the pedal and the crank to prevent
gouging. It didn't prevent pedal eye failures, however.
The pedal rocks in the threads with the current design, transmitting
radial loads to the threads. Jobst's machined collet prevents this.
All that being said, however, I've never (yet) broken a crank in this
manner and I have run the gamut from 250 lbs 20 years ago to 195 10-15
years ago (and racing at that time) to 210-215 these days. However, I
tend to climb seated and I have only short climbs of 400-500 feet
vertical gain around here. But the crank on my most-used bike has been
in service since late 1993 (an "old school" Ritchey 100/74 with no sharp
edges, flat surfaces, cold forged, etc).
thirty-six
> A threaded fastener should only be loaded along its
> axis notwithstanding all designs you see that do not
> abide by this rule: bottom bracket threads, for
> instance. The fastening bolt holding a square taper
> crank on the spindle is used properly.
Bending of the bolt is not eliminated here. The ideal you seek may be
found with cylinder head bolts.
> A threaded
> fastener should be torqued in sufficiently so as to
> prevent unscrewing.
The torque required is proportional to the desired clamping load,
fastener retention is a seperate issue.
>Screwing a pedal into the crank
> then standing on the end of the peda puts stress on
> threads they are not designed for.
A 9/16" pedal axle does not overload axle threads.
> Jobst's angular
> collet between the pedal and the crank works to
> redirect transverse force on the threads into axial
> force that the threads are designed for.
Total bull.
Tighten a pedal, meant for aluminium cranks so that the boss is
adequately clamped on to the face of the crank and there is no crank
failure.
thirty-six
>
> > > By the way, would FEA make any difference as to the pedal eye
> > > failures? It's not a matter of the material being insufficient, it's
> > > cracks caused by gouging, at least as I understand it. I've broken a
> > > half dozen cranks or more, mostly Campy -- none at the pedal eye --
> > > but undoubtedly due to some stress riser and lots of load cycles.
> > > These seem to be more machining or material flaws than inadequate
> > > engineering, viz., bad load calculations and material selection. --
>
> > A threaded fastener should only be loaded along its
> > axis notwithstanding all designs you see that do not
> > abide by this rule: bottom bracket threads, for
> > instance. The fastening bolt holding a square taper
> > crank on the spindle is used properly. A threaded
> > fastener should be torqued in sufficiently so as to
> > prevent unscrewing. Screwing a pedal into the crank
> > then standing on the end of the peda puts stress on
> > threads they are not designed for. Jobst's angular
> > collet between the pedal and the crank works to
> > redirect transverse force on the threads into axial
> > force that the threads are designed for.
>
> Very interesting, So is thread failure causing cracking and crank
> failure? I just don't see many modern cranks that haved failed at the
> pedal eye. Jobst's experience was with NR cranks, although he did his
> collet job on a mid-80s Dura Ace, IIRC. -- Jay Beattie.
Record cranks which were known for failure, they being pared down to
acceptable failure by the manufacturer. It was likely the use of
improper pedal threads and inadequate tightening which caused most
failures.
thirty-six
Total bullshit. A pedal thread which moves is because there is
insufficient pressure between pedal boss and crank face to support the
load. This is remedied by tightening it up. Oil helps in getting it
tight enough. Tight enough is for the rider to stand on a 14" pedal
spanner, not doing a hand job on some poxy 6" adjustable.
thirty-six
>
> Gouging can be an instigator of failure,
Unlikely, crank materials are not on the whole of sufficient hardness
to be affected by notch sensitivity.
> but TA used to provide flat
> washers between the shoulder of the pedal and the crank to prevent
> gouging. It didn't prevent pedal eye failures, however.
:-)
>
> The pedal rocks in the threads with the current design, transmitting
> radial loads to the threads. Jobst's machined collet prevents this.
Dont be ridiculous, with the axle boss clamped solid, the thread stays
stuck.
>
> All that being said, however, I've never (yet) broken a crank in this
> manner and I have run the gamut from 250 lbs 20 years ago to 195 10-15
> years ago (and racing at that time) to 210-215 these days. However, I
> tend to climb seated and I have only short climbs of 400-500 feet
> vertical gain around here. But the crank on my most-used bike has been
> in service since late 1993 (an "old school" Ritchey 100/74 with no sharp
> edges, flat surfaces, cold forged, etc).
Your experience is common for those using good pedal axles tightly.
This includes urang-utans.
Jay Beattie
It would be nice to see some experimental work on this -- or at least
a survey of failures on modern cranks. I have looked at a lot of
crank threads -- just yesterday, as a matter of fact, when I replaced
the spindles on my recalled Keos. I have never seen thread galling.
I'm going to rat through all my junk cranks downstairs to look at the
threads.
I can say that I cracked a threaded insert on an FSA carbon fiber ISIS
drive crank (non-warranty and no offer of a discount for a replacement
-- POS). That insert probably broke because of the forces described
above -- along with the fact that it was a thin, aluminum piece
embedded in dissimilar material. I bet that type of failure is going
to be a problem with all full carbon (as apposed to carbon wrap over
aluminum) cranks. This crank was the only piece of superlight crap I
bought when building my current racing bike five or so years ago. It
was cheap-ish for CF. The BB also broke. My confidence in FSA is at
about 10-12%. -- Jay Beattie.
thirty-six
I don't see what it would bring new to the arena, using sufficient
clamping force minimimises side-loading in the threaded area as with
all bolts and screws. There is nothing unique about a pedal axle
thread, make it tight to fit the rider. The rider should stand on a
14" spanner after lubricating the thread and boss. The pedal boss has
to be tight to prevent movement of it, the thread is therefore taken
care of. End of story.
> I have looked at a lot of
> crank threads -- just yesterday, as a matter of fact, when I replaced
> the spindles on my recalled Keos. I have never seen thread galling.
> I'm going to rat through all my junk cranks downstairs to look at the
> threads.
>
> I can say that I cracked a threaded insert on an FSA carbon fiber ISIS
Unsuitable for intended use.
> drive crank (non-warranty and no offer of a discount for a replacement
> -- POS). That insert probably broke because of the forces described
> above -- along with the fact that it was a thin, aluminum piece
> embedded in dissimilar material. I bet that type of failure is going
> to be a problem with all full carbon (as apposed to carbon wrap over
> aluminum) cranks. This crank was the only piece of superlight crap I
> bought when building my current racing bike five or so years ago. It
> was cheap-ish for CF. The BB also broke. My confidence in FSA is at
> about 10-12%. -- Jay Beattie.
If they are racing, not touring or training cranks then you can expect
failures, that's what comes of inadequately realised designs. The
blame does not lie with FSA, but rather with who miss-sold you the
cranks. If of course, your requirements were not discussed and you
bought on an image, you should be totally satisfied that the image
held up until you actually used the product.
Jay Beattie
Yes, they did look cool, and I got this big FSA decal that is now on
my re-painted (powder blue) 1950s Electrolux canister vacuum cleaner
that we affectionately call "Sparky." -- Jay Beattie.