<lou.holt...@usenet.nl> wrote:
>Op 18-8-2012 13:10, John B. schreef:
>> On Fri, 17 Aug 2012 22:59:29 -0700 (PDT), DirtRoadie
>> <DirtRoa...@aol.com> wrote:
>>> On Aug 17, 5:46 pm, John B. <johnbsloc...@gmail.com> wrote:
>>>> On Fri, 17 Aug 2012 06:22:42 -0700 (PDT), DirtRoadie
>>>> <DirtRoa...@aol.com> wrote:
>>>>> On Aug 17, 5:24 am, John B. <johnbsloc...@gmail.com> wrote:
>>>>>> On Thu, 16 Aug 2012 19:36:41 -0700 (PDT), jspace...@linuxquestions.net
>>>>>> wrote:
>>>>>>> How safe is it to ride with a cracked carbon fibre road handlebar?
>>>>>>> This evening I went for a ride and ran over a small bump in the road while I was up out of the saddle with some of my weight on the tops of my bars. I heard a cracking sound and felt a splinter in my finger.
>>>>>>> I stopped to take a look and discovered that I had a small crack on the right underside of my FSA K-Wing handlebar. It's not around the stem. It appears to be around the opening where the brake cables go into.
>>>>>>> The bar itself is around 6 years old. The bump I ran over wasn't that harsh, so maybe it was a combination of my weight on it and age/fatigue of the bar.
>>>>>>> Anyways, is the handlebar still safe to ride on? It doesn't flex when I put weight on it, but I don't want it to snap off completely while I'm out riding and putting stress on it.
>>>>>>> Thanks,
>>>>>>> J. Spaceman
>>>>>> A carbon fiber reinforced component is immensely strong.... as long as
>>>>>> it maintains its structural integrity, but once cracked its strength
>>>>>> drops dramatically. I would suggest that you either replace or repair
>>>>>> the item
>>>>>> Contrary to what seems to be popular belief reinforced composites can
>>>>>> be repaired and returned to original, or even higher, strength. Below
>>>>>> is the Web Site of a German company that specializes in the work and
>>>>>> I'm sure that there must be firms in the U.S. that do the same sort of
>>>>>> work.
>>>> You are correct. He says that because of critical dimensional
>>>> requirements he doesn't repair handlebars and forks.
>>>> Cheers,
>>>> John B.
>>> Those are also structures that carry cantilevered loads. In contrast
>>> to frames and rims.
>>> See also
>>> http://polytube-cycles.de/en/services.html >>> "Note:
>>> The repair of damaged rims, forks, stems and other components is
>>> currently not possible"
>>> DR
>> He may not have the capability to repair forks, or stems; or more
>> likely he may not want to fool with small stuff that he can't make a
>> decent profit on, but I can tell you that building or repairing
>> composite items that carry a "cantilevered loads" is certainly
>> possible - I've done it myself and repairing composite is about as
>> easy a repair as you can find. Far easier then repairing a wooden
>> cantilevered beam, in fact.
>> They have been building, and repairing fiberglass items for 70 years
>> or more. I think that the first glass fiber reinforced boat dates back
>> to 193-something. It is hardly an esoteric art.
>> Cheers,
>> John B.
>Are you comparing a boat to a fork/stem safety wise or stress wise? When >a fork breaks suddenly there is a great chance you end up at the dentist >for a major repair. A decent fork costs 200-300 euro from a respected >manufacturer. That is a lot cheaper than the dentist bill.
And when a boat breaks you drown. Which is better? A trip to the
dentist or dying?
And, I might add, the boat building came long before the composite
bikes. The bikes are built with technology developed by the boat
builders.
>People are so ignorant. Buddy of mine crashed and the handlebar was 30 >degrees out of wack. In a rush he wanted to straighten without loosen >the bolts of his stem. Geezzzz.... When I told him to check the CF >steerer at home he asked "Why?"
>Lou, don't try to repair forks, stems, seatposts. Even discussing it...
> >Of course, there are significant differences between (say) a bike fork
> >and a golf club shaft - or a fly rod, or a violin bow, or many other
> >objects available in carbon fiber. I think the loads seen by bike
> >components are much more variable (both from rider to rider, and from
> >time to time under a given rider) and thus, less well understood.
> I'm not so sure.
> Columbus, certainly, constructed a device to measure and record strain
> at various points on the frame, in real time, and brags that it
> allowed them to design tubes with not only varying shapes and size but
> thickness and butt dimensions.
> At the same time the power output of the motor is far better
> understood. I'll bet that every top rider in the world has, by now
> been tested for power output both instantaneous and sustained.
I'm sure that there have been lots of riders tested for power output. But I don't think that's the source of the major loads on a bike frame. Rather, the major loads come from things like hitting potholes, jumping the bike (for those who do that), unplanned events like falls, and perhaps even mishandling the bike when it's not being ridden - like today, when my wife and I nearly dropped our tandem when we tried to load it onto the roof rack on our car. :-(
Furthermore, I think a lot of those loads are far worse for, um, larger amateur riders who ride with less skill. A 120 pound pro probably handles potholes with a lot more finesse than a 230 pound novice who just bought the "best" (i.e. most expensive) bike the bike shop had - so he could lose the weight he gained by working long hours to afford fancy toys.
> There isn't much left, is there. A 110 lb rider with a power output of
> 3 HP on a 15 lb. vehicle with the following dimensions and a vertical
> and horizontal CG of X and Y, mounted on wheels and tires with a
> compliance of Z traveling at a velocity of V on a slope of a specified
> angle with known air density and a measured drag, in a district with a
> known gravitational attraction approaching a curve of known angle,
> radius and length.
> Certainly it can be calculated.
I think it would be easier to calculate if the 3 HP came from a smooth running motor than from a human being. For example, one human might do that in a high gear at 60 rpm, rocking the bike mightily while yanking on the bars, while another might do it in a lower gear at 130 rpm with the bike running much straighter. Motors are more predictable.
> >And at the same time designers try to resist relatively unknown loads,
> >they try to minimize weight. Building a carbon fork or frame that
> >matched the weight of an aluminum or steel one would be much less of a
> >trick. Trying to shave every gram makes the job much more difficult.
> Unknown loads? Not hardly. I've been out of school for over 50 years
> now and I'll bet I can still do (if I still find my references books)
> a proper stress analysis.
Yet there are people who take the trouble to do FEA analyses of bike frames, something that's not necessary for simple stress situations.
Even FEA doesn't necessarily give a suitable picture. With all FEA, the loads and restraints make a big difference in the results, and I'm not sure the common assumptions are really accurate. For example, it's common to assume the rear dropouts are fixed in both translation and rotation (as at http://engr.bd.psu.edu/davej/classes/fea1_hw5b.html). But if they're actually clamped to a somewhat flexible axle in a somewhat flexible wheel on a tire that moves laterally as the bike rolls, is a totally fixed dropout really accurate? I'm not so sure. Thus even FEA gives only an approximation.
> Back in the mid 1960's I was peripherally involved in a project to
> measure the load on an F-4 tail hook when it grabbed the arresting
> gear. At that time the strain gage was a piece of plastic abut 1/4"
> wide and 3/4" long, bonded to the hook shank. Certainly measuring
> technology has gotten more sophisticated in the past 50 years.
Well, strain gages are still the same, although they've long been available in much smaller sizes than that. They're a very mature technology.
Thing is, a tailhook is probably a relatively simple problem. It's probably what's called a "two force member." It supports only tensile loads, except for some brief rotational inertia effects. It's a problem suitable for a beginning class in stress analysis.
And BTW, assuming the tailhook is essentially a rod of uniform cross section, a large strain gage is perfectly appropriate, since the stress would be expected to be very uniform, i.e. no steep stress gradient. An area like the bottom bracket or head tube of a bike has stresses that vary tremendously over small distances. See http://www.designworldonline.com/mountain-bike-company-uses-fea-to-de... So a 1/16" strain gage would be more appropriate, there, and even that only reports the average (not peak) stress over its length. Yet fatigue resistance depends on the peak stress, not any average stress.
> A modern racing bicycle is essentially two triangles and calculating
> the strain on a triangle is a very, very, old technique.
If it were a pinned truss, or something approximating a pinned truss, it would be easy. But it's not. If were so easy, nobody would even bother to do FEA.
> Joe Riel wrote:
>> DirtRoadie<DirtRoa...@aol.com> writes:
>>>>>>> Agreed. I had half the bars break off next to the clamp near the
>>>>>>> end of
>>>>>>> a longish road race. 10 km to go and I finished with the break, even
>>>>>>> chased down an escapee coming in to town. I didn't crash.
>>>>>> So you rode 10 km with half the handlebar broken off? Did you lash
>>>>>> the
>>>>>> broken-off half to your top tube, toss it jauntily over your
>>>>>> shoulder,
>>>>>> or just carry it loosely in your hand as you chased down that other
>>>>>> rider? I really am curious about details, James.
>>>> Oh. I missed the part where James said he whipped out the tape and
>>>> taped the bar down while he rode in the race. My fault.
>>> Yes please try to pay attention. The guy that I described in a
>>> criterium rode in essentially the same manner as shown the video,
>>> sans tape, just holding bars together on one side with a hand.
>>> So do you or don't you retract your silly "zero chance of catching
>>> yourself" comment?
>> Ah, but did the bar break suddenly? What is the typical failure mode?
>> I've broken handlebars near the stem and had to ride home with half a
>> bar. The initial fracture was on the lower half, the separation occurred
>> when pulling up on the bars from a stop. Didn't crash.
> I was talking about _suddenly_ breaking. And whether or not it's
> deserved, carbon fiber parts have a reputation for breaking suddenly -
> or at least, with much less warning than metal.
The Al bar that broke while I was racing did not give minutes of warning. During the race I had been pulling on them out of the seat up some short but steep hills. When it actually broke was just as I finished a turn on a long slight down hill at about 50 km/h. I moved to the right to let the bunch through and while in the drops noticed the bars felt odd for a moment, then the right side just dropped away. That moment when the bars felt odd was enough for me to react and take weight off the RHS. That moment lasted for all of a second - maybe 2.
> I've never broken a handlebar, but I've replaced two sets (both
> aluminum) when I got worried about their extended life. But that friend
> of mine I mentioned got no warning at all with his aluminum bars. He
> said the break occurred when a light turned green and he began
> accelerating hard across the intersection, pulling up on the bars.
> Instead of making it across the road, he took a ride in an ambulance.
> FWIW, he ran a bike shop, and he hung the bars above the counter as a
> "don't let this happen to you" display.
My mate who broke a CF bar recently had it snap as he was accelerating across an intersection. He didn't fall, but when I questioned him on it yesterday said he was lucky not to fall because there was plenty of cars and he may well have been run over had he fallen.
The fact remains that a crash is not as inevitable as you made out, though all agree a crash is quite possible and the bars should be replaced before starting a ride.
> Are you comparing a boat to a fork/stem safety wise or stress wise? When
> a fork breaks suddenly there is a great chance you end up at the dentist
> for a major repair. A decent fork costs 200-300 euro from a respected
> manufacturer. That is a lot cheaper than the dentist bill.
> On Sat, 18 Aug 2012 16:14:00 -0400, Frank Krygowski
> <frkrygowREM...@gEEmail.com> wrote:
>> John B. wrote:
>>> ...
>>> But I've always looked at the carbon bicycle business with a certain
>>> amount of awe. Here is a bunch of guys building bicycle frames out of
>>> a material that has been used for golf club shafts ( a very similar
>>> practice to building tubes for bicycle frames) since 1973 - nearly 40
>>> years ago, and talking as though they have just invented gravity.
>>> (And a bunch of guys on the other side of the counter paying premium
>>> for 40 year old technology :-)
>> Of course, there are significant differences between (say) a bike fork
>> and a golf club shaft - or a fly rod, or a violin bow, or many other
>> objects available in carbon fiber. I think the loads seen by bike
>> components are much more variable (both from rider to rider, and from
>> time to time under a given rider) and thus, less well understood.
> I'm not so sure.
> Columbus, certainly, constructed a device to measure and record strain
> at various points on the frame, in real time, and brags that it
> allowed them to design tubes with not only varying shapes and size but
> thickness and butt dimensions.
> At the same time the power output of the motor is far better
> understood. I'll bet that every top rider in the world has, by now
> been tested for power output both instantaneous and sustained.
> Wind resistance? Been tested and re-tested. Even the material in Lance
> Armstrong's shirt was designed with the data from wind tunnel testing.
> There isn't much left, is there. A 110 lb rider with a power output of
> 3 HP on a 15 lb. vehicle with the following dimensions and a vertical
> and horizontal CG of X and Y, mounted on wheels and tires with a
> compliance of Z traveling at a velocity of V on a slope of a specified
> angle with known air density and a measured drag, in a district with a
> known gravitational attraction approaching a curve of known angle,
> radius and length.
> <lou.holt...@usenet.nl> wrote:
> >Op 18-8-2012 13:10, John B. schreef:
> >> On Fri, 17 Aug 2012 22:59:29 -0700 (PDT), DirtRoadie
> >> <DirtRoa...@aol.com> wrote:
> >>> On Aug 17, 5:46 pm, John B. <johnbsloc...@gmail.com> wrote:
> >>>> On Fri, 17 Aug 2012 06:22:42 -0700 (PDT), DirtRoadie
> >>>> <DirtRoa...@aol.com> wrote:
> >>>>> On Aug 17, 5:24 am, John B. <johnbsloc...@gmail.com> wrote:
> >>>>>> On Thu, 16 Aug 2012 19:36:41 -0700 (PDT), jspace...@linuxquestions.net
> >>>>>> wrote:
> >>>>>>> How safe is it to ride with a cracked carbon fibre road handlebar?
> >>>>>>> This evening I went for a ride and ran over a small bump in the road while I was up out of the saddle with some of my weight on the tops of my bars. I heard a cracking sound and felt a splinter in my finger.
> >>>>>>> I stopped to take a look and discovered that I had a small crack on the right underside of my FSA K-Wing handlebar. It's not around the stem. It appears to be around the opening where the brake cables go into.
> >>>>>>> The bar itself is around 6 years old. The bump I ran over wasn't that harsh, so maybe it was a combination of my weight on it and age/fatigue of the bar.
> >>>>>>> Anyways, is the handlebar still safe to ride on? It doesn't flex when I put weight on it, but I don't want it to snap off completely while I'm out riding and putting stress on it.
> >>>>>>> Thanks,
> >>>>>>> J. Spaceman
> >>>>>> A carbon fiber reinforced component is immensely strong.... as long as
> >>>>>> it maintains its structural integrity, but once cracked its strength
> >>>>>> drops dramatically. I would suggest that you either replace or repair
> >>>>>> the item
> >>>>>> Contrary to what seems to be popular belief reinforced composites can
> >>>>>> be repaired and returned to original, or even higher, strength. Below
> >>>>>> is the Web Site of a German company that specializes in the work and
> >>>>>> I'm sure that there must be firms in the U.S. that do the same sort of
> >>>>>> work.
> >>>> You are correct. He says that because of critical dimensional
> >>>> requirements he doesn't repair handlebars and forks.
> >>>> Cheers,
> >>>> John B.
> >>> Those are also structures that carry cantilevered loads. In contrast
> >>> to frames and rims.
> >>> See also
> >>>http://polytube-cycles.de/en/services.html > >>> "Note:
> >>> The repair of damaged rims, forks, stems and other components is
> >>> currently not possible"
> >>> DR
> >> He may not have the capability to repair forks, or stems; or more
> >> likely he may not want to fool with small stuff that he can't make a
> >> decent profit on, but I can tell you that building or repairing
> >> composite items that carry a "cantilevered loads" is certainly
> >> possible - I've done it myself and repairing composite is about as
> >> easy a repair as you can find. Far easier then repairing a wooden
> >> cantilevered beam, in fact.
> >> They have been building, and repairing fiberglass items for 70 years
> >> or more. I think that the first glass fiber reinforced boat dates back
> >> to 193-something. It is hardly an esoteric art.
> >> Cheers,
> >> John B.
> >Are you comparing a boat to a fork/stem safety wise or stress wise? When
> >a fork breaks suddenly there is a great chance you end up at the dentist
> >for a major repair. A decent fork costs 200-300 euro from a respected
> >manufacturer. That is a lot cheaper than the dentist bill.
> And when a boat breaks you drown. Which is better? A trip to the
> dentist or dying?
> And, I might add, the boat building came long before the composite
> bikes. The bikes are built with technology developed by the boat
> builders.
i think you will find, if you care to look, that the current
technology in carbon bicycle frames evolved from the work on monocoque
sports cars, most notably Lotus, who IIRC were offering 3/4 monocoque
carbon-fibre chassis from around 1979
> >People are so ignorant. Buddy of mine crashed and the handlebar was 30
> >degrees out of wack. In a rush he wanted to straighten without loosen
> >the bolts of his stem. Geezzzz.... When I told him to check the CF
> >steerer at home he asked "Why?"
> >Lou, don't try to repair forks, stems, seatposts. Even discussing it...
> > >Of course, there are significant differences between (say) a bike fork
> > >and a golf club shaft - or a fly rod, or a violin bow, or many other
> > >objects available in carbon fiber. I think the loads seen by bike
> > >components are much more variable (both from rider to rider, and from
> > >time to time under a given rider) and thus, less well understood.
> > I'm not so sure.
> > Columbus, certainly, constructed a device to measure and record strain
> > at various points on the frame, in real time, and brags that it
> > allowed them to design tubes with not only varying shapes and size but
> > thickness and butt dimensions.
> > At the same time the power output of the motor is far better
> > understood. I'll bet that every top rider in the world has, by now
> > been tested for power output both instantaneous and sustained.
> I'm sure that there have been lots of riders tested for power output. But I don't think that's the source of the major loads on a bike frame. Rather, the major loads come from things like hitting potholes, jumping the bike (for those who do that), unplanned events like falls, and perhaps even mishandling the bike when it's not being ridden - like today, when my wife and I nearly dropped our tandem when we tried to load it onto the roof rack on our car. :-(
> Furthermore, I think a lot of those loads are far worse for, um, larger amateur riders who ride with less skill. A 120 pound pro probably handles potholes with a lot more finesse than a 230 pound novice who just bought the "best" (i.e. most expensive) bike the bike shop had - so he could lose the weight he gained by working long hours to afford fancy toys.
Poor wheels will ruin both a terrifc or mediocre frameset. A really
good pair of wheels will usually transform a mediocre frame (with a
fork swap) to something outstanding, with the right pair of legs. The
difference between an average frame and a great frame is 1/10 the
difference (wet finger in the air test) between optimised wheels and
any old pair of sloppily built wheels.
Trouble is that many medium and small production frames are grossly
overbuilt due to the ecpectation that they are going to be ridden by
some fat-arsed wannabe with sloppy wheels.
> > There isn't much left, is there. A 110 lb rider with a power output of
> > 3 HP on a 15 lb. vehicle with the following dimensions and a vertical
> > and horizontal CG of X and Y, mounted on wheels and tires with a
> > compliance of Z traveling at a velocity of V on a slope of a specified
> > angle with known air density and a measured drag, in a district with a
> > known gravitational attraction approaching a curve of known angle,
> > radius and length.
> > Certainly it can be calculated.
> I think it would be easier to calculate if the 3 HP came from a smooth running motor than from a human being. For example, one human might do that in a high gear at 60 rpm, rocking the bike mightily while yanking on the bars, while another might do it in a lower gear at 130 rpm with the bike running much straighter. Motors are more predictable.
> > >And at the same time designers try to resist relatively unknown loads,
> > >they try to minimize weight. Building a carbon fork or frame that
> > >matched the weight of an aluminum or steel one would be much less of a
> > >trick. Trying to shave every gram makes the job much more difficult.
> > Unknown loads? Not hardly. I've been out of school for over 50 years
> > now and I'll bet I can still do (if I still find my references books)
> > a proper stress analysis.
> Yet there are people who take the trouble to do FEA analyses of bike frames, something that's not necessary for simple stress situations.
> Even FEA doesn't necessarily give a suitable picture. With all FEA, the loads and restraints make a big difference in the results, and I'm not sure the common assumptions are really accurate. For example, it's common to assume the rear dropouts are fixed in both translation and rotation (as athttp://engr.bd.psu.edu/davej/classes/fea1_hw5b.html). But if they're actually clamped to a somewhat flexible axle in a somewhat flexible wheel on a tire that moves laterally as the bike rolls, is a totally fixed dropout really accurate? I'm not so sure. Thus even FEA gives only an approximation.
> > Back in the mid 1960's I was peripherally involved in a project to
> > measure the load on an F-4 tail hook when it grabbed the arresting
> > gear. At that time the strain gage was a piece of plastic abut 1/4"
> > wide and 3/4" long, bonded to the hook shank. Certainly measuring
> > technology has gotten more sophisticated in the past 50 years.
> Well, strain gages are still the same, although they've long been available in much smaller sizes than that. They're a very mature technology.
> Thing is, a tailhook is probably a relatively simple problem. It's probably what's called a "two force member." It supports only tensile loads, except for some brief rotational inertia effects. It's a problem suitable for a beginning class in stress analysis.
> And BTW, assuming the tailhook is essentially a rod of uniform cross section, a large strain gage is perfectly appropriate, since the stress would be expected to be very uniform, i.e. no steep stress gradient. An area like the bottom bracket or head tube of a bike has stresses that vary tremendously over small distances. Seehttp://www.designworldonline.com/mountain-bike-company-uses-fea-to-de...
> So a 1/16" strain gage would be more appropriate, there, and even that only reports the average (not peak) stress over its length. Yet fatigue resistance depends on the peak stress, not any average stress.
> > A modern racing bicycle is essentially two triangles and calculating
> > the strain on a triangle is a very, very, old technique.
> If it were a pinned truss, or something approximating a pinned truss, it would be easy. But it's not. If were so easy, nobody would even bother to do FEA.
<thirty-...@live.co.uk> wrote:
>On Aug 19, 3:24 am, John B. <johnbsloc...@gmail.com> wrote:
>> On Sat, 18 Aug 2012 13:34:41 +0200, Lou Holtman
>> <lou.holt...@usenet.nl> wrote:
>> >Op 18-8-2012 13:10, John B. schreef:
>> >> On Fri, 17 Aug 2012 22:59:29 -0700 (PDT), DirtRoadie
>> >> <DirtRoa...@aol.com> wrote:
>> >>> On Aug 17, 5:46 pm, John B. <johnbsloc...@gmail.com> wrote:
>> >>>> On Fri, 17 Aug 2012 06:22:42 -0700 (PDT), DirtRoadie
>> >>>> <DirtRoa...@aol.com> wrote:
>> >>>>> On Aug 17, 5:24 am, John B. <johnbsloc...@gmail.com> wrote:
>> >>>>>> On Thu, 16 Aug 2012 19:36:41 -0700 (PDT), jspace...@linuxquestions.net
>> >>>>>> wrote:
>> >>>>>>> How safe is it to ride with a cracked carbon fibre road handlebar?
>> >>>>>>> This evening I went for a ride and ran over a small bump in the road while I was up out of the saddle with some of my weight on the tops of my bars. I heard a cracking sound and felt a splinter in my finger.
>> >>>>>>> I stopped to take a look and discovered that I had a small crack on the right underside of my FSA K-Wing handlebar. It's not around the stem. It appears to be around the opening where the brake cables go into.
>> >>>>>>> The bar itself is around 6 years old. The bump I ran over wasn't that harsh, so maybe it was a combination of my weight on it and age/fatigue of the bar.
>> >>>>>>> Anyways, is the handlebar still safe to ride on? It doesn't flex when I put weight on it, but I don't want it to snap off completely while I'm out riding and putting stress on it.
>> >>>>>>> Thanks,
>> >>>>>>> J. Spaceman
>> >>>>>> A carbon fiber reinforced component is immensely strong.... as long as
>> >>>>>> it maintains its structural integrity, but once cracked its strength
>> >>>>>> drops dramatically. I would suggest that you either replace or repair
>> >>>>>> the item
>> >>>>>> Contrary to what seems to be popular belief reinforced composites can
>> >>>>>> be repaired and returned to original, or even higher, strength. Below
>> >>>>>> is the Web Site of a German company that specializes in the work and
>> >>>>>> I'm sure that there must be firms in the U.S. that do the same sort of
>> >>>>>> work.
>> >>>> You are correct. He says that because of critical dimensional
>> >>>> requirements he doesn't repair handlebars and forks.
>> >>>> Cheers,
>> >>>> John B.
>> >>> Those are also structures that carry cantilevered loads. In contrast
>> >>> to frames and rims.
>> >>> See also
>> >>>http://polytube-cycles.de/en/services.html >> >>> "Note:
>> >>> The repair of damaged rims, forks, stems and other components is
>> >>> currently not possible"
>> >>> DR
>> >> He may not have the capability to repair forks, or stems; or more
>> >> likely he may not want to fool with small stuff that he can't make a
>> >> decent profit on, but I can tell you that building or repairing
>> >> composite items that carry a "cantilevered loads" is certainly
>> >> possible - I've done it myself and repairing composite is about as
>> >> easy a repair as you can find. Far easier then repairing a wooden
>> >> cantilevered beam, in fact.
>> >> They have been building, and repairing fiberglass items for 70 years
>> >> or more. I think that the first glass fiber reinforced boat dates back
>> >> to 193-something. It is hardly an esoteric art.
>> >> Cheers,
>> >> John B.
>> >Are you comparing a boat to a fork/stem safety wise or stress wise? When
>> >a fork breaks suddenly there is a great chance you end up at the dentist
>> >for a major repair. A decent fork costs 200-300 euro from a respected
>> >manufacturer. That is a lot cheaper than the dentist bill.
>> And when a boat breaks you drown. Which is better? A trip to the
>> dentist or dying?
>> And, I might add, the boat building came long before the composite
>> bikes. The bikes are built with technology developed by the boat
>> builders.
>i think you will find, if you care to look, that the current
>technology in carbon bicycle frames evolved from the work on monocoque
>sports cars, most notably Lotus, who IIRC were offering 3/4 monocoque
>carbon-fibre chassis from around 1979
So? What is the difference between a fiberglass hull and a sports car
body? Look at the shape of the very early Porche's :-) But more
seriously, making a reinforced plastic structure is basically just a
matter of saturating a fabric with activated resin. Fiber orientation,
SCRIMP and all the rest are simply modifications of the basic
technique.
>> >People are so ignorant. Buddy of mine crashed and the handlebar was 30
>> >degrees out of wack. In a rush he wanted to straighten without loosen
>> >the bolts of his stem. Geezzzz.... When I told him to check the CF
>> >steerer at home he asked "Why?"
>> >Lou, don't try to repair forks, stems, seatposts. Even discussing it...
<thirty-...@live.co.uk> wrote:
>On Aug 19, 4:57 am, frkry...@gmail.com wrote:
>> On Saturday, August 18, 2012 9:49:55 PM UTC-4, John B. wrote:
>> > On Sat, 18 Aug 2012 16:14:00 -0400, Frank Krygowski
>> > >Of course, there are significant differences between (say) a bike fork
>> > >and a golf club shaft - or a fly rod, or a violin bow, or many other
>> > >objects available in carbon fiber. I think the loads seen by bike
>> > >components are much more variable (both from rider to rider, and from
>> > >time to time under a given rider) and thus, less well understood.
>> > I'm not so sure.
>> > Columbus, certainly, constructed a device to measure and record strain
>> > at various points on the frame, in real time, and brags that it
>> > allowed them to design tubes with not only varying shapes and size but
>> > thickness and butt dimensions.
>> > At the same time the power output of the motor is far better
>> > understood. I'll bet that every top rider in the world has, by now
>> > been tested for power output both instantaneous and sustained.
>> I'm sure that there have been lots of riders tested for power output. But I don't think that's the source of the major loads on a bike frame. Rather, the major loads come from things like hitting potholes, jumping the bike (for those who do that), unplanned events like falls, and perhaps even mishandling the bike when it's not being ridden - like today, when my wife and I nearly dropped our tandem when we tried to load it onto the roof rack on our car. :-(
And of course only bicycles have hit a pot hole so calculating the
stress, is a totally new science?
>> Furthermore, I think a lot of those loads are far worse for, um, larger amateur riders who ride with less skill. A 120 pound pro probably handles potholes with a lot more finesse than a 230 pound novice who just bought the "best" (i.e. most expensive) bike the bike shop had - so he could lose the weight he gained by working long hours to afford fancy toys.
What difference does it make? A body of X weight hitting a pot hole is
a doable calculation.
>Poor wheels will ruin both a terrifc or mediocre frameset. A really
>good pair of wheels will usually transform a mediocre frame (with a
>fork swap) to something outstanding, with the right pair of legs. The
>difference between an average frame and a great frame is 1/10 the
>difference (wet finger in the air test) between optimised wheels and
>any old pair of sloppily built wheels.
>Trouble is that many medium and small production frames are grossly
>overbuilt due to the ecpectation that they are going to be ridden by
>some fat-arsed wannabe with sloppy wheels.
>> > There isn't much left, is there. A 110 lb rider with a power output of
>> > 3 HP on a 15 lb. vehicle with the following dimensions and a vertical
>> > and horizontal CG of X and Y, mounted on wheels and tires with a
>> > compliance of Z traveling at a velocity of V on a slope of a specified
>> > angle with known air density and a measured drag, in a district with a
>> > known gravitational attraction approaching a curve of known angle,
>> > radius and length.
>> > Certainly it can be calculated.
>> I think it would be easier to calculate if the 3 HP came from a smooth running motor than from a human being. For example, one human might do that in a high gear at 60 rpm, rocking the bike mightily while yanking on the bars, while another might do it in a lower gear at 130 rpm with the bike running much straighter. Motors are more predictable.
I wonder why? After all an internal combustion, or a steam, engine for
that matter, does not have a constant connecting rod loading and
engineers have been calculating those loads for a while now. And
(gulp) even doing it at different RPM's in difference directions
calculating both the loading in both tension and compression.
>> > >And at the same time designers try to resist relatively unknown loads,
>> > >they try to minimize weight. Building a carbon fork or frame that
>> > >matched the weight of an aluminum or steel one would be much less of a
>> > >trick. Trying to shave every gram makes the job much more difficult.
>> > Unknown loads? Not hardly. I've been out of school for over 50 years
>> > now and I'll bet I can still do (if I still find my references books)
>> > a proper stress analysis.
>> Yet there are people who take the trouble to do FEA analyses of bike frames, something that's not necessary for simple stress situations.
>> Even FEA doesn't necessarily give a suitable picture. With all FEA, the loads and restraints make a big difference in the results, and I'm not sure the common assumptions are really accurate. For example, it's common to assume the rear dropouts are fixed in both translation and rotation (as athttp://engr.bd.psu.edu/davej/classes/fea1_hw5b.html). But if they're actually clamped to a somewhat flexible axle in a somewhat flexible wheel on a tire that moves laterally as the bike rolls, is a totally fixed dropout really accurate? I'm not so sure. Thus even FEA gives only an approximation.
>> > Back in the mid 1960's I was peripherally involved in a project to
>> > measure the load on an F-4 tail hook when it grabbed the arresting
>> > gear. At that time the strain gage was a piece of plastic abut 1/4"
>> > wide and 3/4" long, bonded to the hook shank. Certainly measuring
>> > technology has gotten more sophisticated in the past 50 years.
>> Well, strain gages are still the same, although they've long been available in much smaller sizes than that. They're a very mature technology.
>> Thing is, a tailhook is probably a relatively simple problem. It's probably what's called a "two force member." It supports only tensile loads, except for some brief rotational inertia effects. It's a problem suitable for a beginning class in stress analysis.
>> And BTW, assuming the tailhook is essentially a rod of uniform cross section, a large strain gage is perfectly appropriate, since the stress would be expected to be very uniform, i.e. no steep stress gradient. An area like the bottom bracket or head tube of a bike has stresses that vary tremendously over small distances. Seehttp://www.designworldonline.com/mountain-bike-company-uses-fea-to-de...
>> So a 1/16" strain gage would be more appropriate, there, and even that only reports the average (not peak) stress over its length. Yet fatigue resistance depends on the peak stress, not any average stress.
Frank, I've already told you that Columbus says that they have
instrumented a bicycle frame and used the data to design new frame
tubes with varying thickness and cross section to better resist the
loads. You are getting all concerned about something that is already
accomplished.
As for the tail hook it is not a constant width or cross section so it
is not a single average load. In addition there are several loads
imposed on the hook, sometimes several at the same time.
As for instrumentation, you need a wider experience. Modern race cars
commonly monitor things like the load on a front wheel and the
deflection of the suspension system, in real time, at 250 Km./Hr. The
technology is there, whether a bicycle designer uses it is really just
a decision to do it or not. The hardware is off the shelf.
>> > A modern racing bicycle is essentially two triangles and calculating
>> > the strain on a triangle is a very, very, old technique.
>> If it were a pinned truss, or something approximating a pinned truss, it would be easy. But it's not. If were so easy, nobody would even bother to do FEA.
Nobody said it was simple, or difficult, I only said that it was
doable. If you believe it is too difficult to accomplish then don't
even start.
John B. wrote:
> On Mon, 20 Aug 2012 15:53:31 -0700 (PDT), thirty-six
> <thirty-...@live.co.uk> wrote:
>> On Aug 19, 4:57 am, frkry...@gmail.com wrote:
>>> On Saturday, August 18, 2012 9:49:55 PM UTC-4, John B. wrote:
>>>> A 110 lb rider with a power output of
>>>> 3 HP on a 15 lb. vehicle with the following dimensions and a vertical
>>>> and horizontal CG of X and Y, mounted on wheels and tires with a
>>>> compliance of Z traveling at a velocity of V on a slope of a specified
>>>> angle with known air density and a measured drag, in a district with a
>>>> known gravitational attraction approaching a curve of known angle,
>>>> radius and length.
>>>> Certainly it can be calculated.
>>> I think it would be easier to calculate if the 3 HP came from a smooth running motor than from a human being. For example, one human might do that in a high gear at 60 rpm, rocking the bike mightily while yanking on the bars, while another might do it in a lower gear at 130 rpm with the bike running much straighter. Motors are more predictable.
> I wonder why? After all an internal combustion, or a steam, engine for
> that matter, does not have a constant connecting rod loading and
> engineers have been calculating those loads for a while now. And
> (gulp) even doing it at different RPM's in difference directions
> calculating both the loading in both tension and compression....
> Frank, I've already told you that Columbus says that they have
> instrumented a bicycle frame and used the data to design new frame
> tubes with varying thickness and cross section to better resist the
> loads. You are getting all concerned about something that is already
> accomplished.
You don't seem to understand that Columbus's effort at instrumenting the bike frame argues against your "this is simple" idea. Columbus would not have taken the trouble to instrument the frame if the problem were as simple as you imagine. Likewise, companies wouldn't bother to do FEA analysis of their bike frames if the design process were easy.
Until perhaps 30 years ago, I think bike frame design was based on what might be described as crowdsourced evolution. Designers would look at the work of many other designers, as well as their own, to evaluate their success. If frames often broke at a certain location, a designer would beef up that location. If frames never broke at a certain location, a designer would say "I can save some weight there" and make the metal thinner. Only rarely would a real breakthrough occur - like, for example, Gary Klein's large diameter aluminum tubes.
I doubt anyone since 1890 designed a bike frame using only simple stress formulas and simplifying assumptions, the way (say) plant engineers design (say) fixed mounting brackets. We won't accept the large safety factors (or rather, the extra weight) that method requires.
> As for the tail hook it is not a constant width or cross section so it
> is not a single average load. In addition there are several loads
> imposed on the hook, sometimes several at the same time.
Give us a good drawing or clear close-up photo, and we can discuss in a sub-thread. To me, it looks like a long, pretty simply shaped two force member, with one major tensile load and perhaps some brief rotational inertia loads. I think the loads and stresses are simpler to calculate than those of a bike frame. And as I said, I think a 3/4" strain gage would give perfectly adequate stress information on such a two-force member. On a bike frame bottom bracket area with its extreme stress gradient, that would be far too big to be useful.
> John B. wrote:
>> On Mon, 20 Aug 2012 15:53:31 -0700 (PDT), thirty-six
>> <thirty-...@live.co.uk> wrote:
>>> On Aug 19, 4:57 am, frkry...@gmail.com wrote:
>>>> On Saturday, August 18, 2012 9:49:55 PM UTC-4, John B.
>>>> wrote:
>>>>> A 110 lb rider with a power output of
>>>>> 3 HP on a 15 lb. vehicle with the following dimensions
>>>>> and a vertical
>>>>> and horizontal CG of X and Y, mounted on wheels and
>>>>> tires with a
>>>>> compliance of Z traveling at a velocity of V on a slope
>>>>> of a specified
>>>>> angle with known air density and a measured drag, in a
>>>>> district with a
>>>>> known gravitational attraction approaching a curve of
>>>>> known angle,
>>>>> radius and length.
>>>>> Certainly it can be calculated.
>>>> I think it would be easier to calculate if the 3 HP came
>>>> from a smooth running motor than from a human being.
>>>> For example, one human might do that in a high gear at
>>>> 60 rpm, rocking the bike mightily while yanking on the
>>>> bars, while another might do it in a lower gear at 130
>>>> rpm with the bike running much straighter. Motors are
>>>> more predictable.
>> I wonder why? After all an internal combustion, or a
>> steam, engine for
>> that matter, does not have a constant connecting rod
>> loading and
>> engineers have been calculating those loads for a while
>> now. And
>> (gulp) even doing it at different RPM's in difference
>> directions
>> calculating both the loading in both tension and
>> compression....
>> Frank, I've already told you that Columbus says that they
>> have
>> instrumented a bicycle frame and used the data to design
>> new frame
>> tubes with varying thickness and cross section to better
>> resist the
>> loads. You are getting all concerned about something that
>> is already
>> accomplished.
> You don't seem to understand that Columbus's effort at
> instrumenting the bike frame argues against your "this is
> simple" idea. Columbus would not have taken the trouble to
> instrument the frame if the problem were as simple as you
> imagine. Likewise, companies wouldn't bother to do FEA
> analysis of their bike frames if the design process were easy.
> Until perhaps 30 years ago, I think bike frame design was
> based on what might be described as crowdsourced evolution.
> Designers would look at the work of many other designers, as
> well as their own, to evaluate their success. If frames
> often broke at a certain location, a designer would beef up
> that location. If frames never broke at a certain location,
> a designer would say "I can save some weight there" and make
> the metal thinner. Only rarely would a real breakthrough
> occur - like, for example, Gary Klein's large diameter
> aluminum tubes.
> I doubt anyone since 1890 designed a bike frame using only
> simple stress formulas and simplifying assumptions, the way
> (say) plant engineers design (say) fixed mounting brackets.
> We won't accept the large safety factors (or rather, the
> extra weight) that method requires.
>> As for the tail hook it is not a constant width or cross
>> section so it
>> is not a single average load. In addition there are
>> several loads
>> imposed on the hook, sometimes several at the same time.
> Give us a good drawing or clear close-up photo, and we can
> discuss in a sub-thread. To me, it looks like a long,
> pretty simply shaped two force member, with one major
> tensile load and perhaps some brief rotational inertia
> loads. I think the loads and stresses are simpler to
> calculate than those of a bike frame. And as I said, I
> think a 3/4" strain gage would give perfectly adequate
> stress information on such a two-force member. On a bike
> frame bottom bracket area with its extreme stress gradient,
> that would be far too big to be useful.
I'm not an expert but these no longer look like WWII Corsair tail hooks:
>> Are you comparing a boat to a fork/stem safety wise or stress wise? When
>> a fork breaks suddenly there is a great chance you end up at the dentist
>> for a major repair. A decent fork costs 200-300 euro from a respected
>> manufacturer. That is a lot cheaper than the dentist bill.
> John B. wrote:
>> On Mon, 20 Aug 2012 15:53:31 -0700 (PDT), thirty-six
>> <thirty-...@live.co.uk> wrote:
>>> On Aug 19, 4:57 am, frkry...@gmail.com wrote:
>>>> On Saturday, August 18, 2012 9:49:55 PM UTC-4, John B. wrote:
>>>>> A 110 lb rider with a power output of
>>>>> 3 HP on a 15 lb. vehicle with the following dimensions and a vertical
>>>>> and horizontal CG of X and Y, mounted on wheels and tires with a
>>>>> compliance of Z traveling at a velocity of V on a slope of a specified
>>>>> angle with known air density and a measured drag, in a district with a
>>>>> known gravitational attraction approaching a curve of known angle,
>>>>> radius and length.
>>>>> Certainly it can be calculated.
>>>> I think it would be easier to calculate if the 3 HP came from a smooth >>>> running motor than from a human being. For example, one human might >>>> do that in a high gear at 60 rpm, rocking the bike mightily while >>>> yanking on the bars, while another might do it in a lower gear at 130 >>>> rpm with the bike running much straighter. Motors are more >>>> predictable.
>> I wonder why? After all an internal combustion, or a steam, engine for
>> that matter, does not have a constant connecting rod loading and
>> engineers have been calculating those loads for a while now. And
>> (gulp) even doing it at different RPM's in difference directions
>> calculating both the loading in both tension and compression....
>> Frank, I've already told you that Columbus says that they have
>> instrumented a bicycle frame and used the data to design new frame
>> tubes with varying thickness and cross section to better resist the
>> loads. You are getting all concerned about something that is already
>> accomplished.
> You don't seem to understand that Columbus's effort at instrumenting the > bike frame argues against your "this is simple" idea. Columbus would not > have taken the trouble to instrument the frame if the problem were as > simple as you imagine. Likewise, companies wouldn't bother to do FEA > analysis of their bike frames if the design process were easy.
> Until perhaps 30 years ago, I think bike frame design was based on what > might be described as crowdsourced evolution. Designers would look at the > work of many other designers, as well as their own, to evaluate their > success. If frames often broke at a certain location, a designer would > beef up that location. If frames never broke at a certain location, a > designer would say "I can save some weight there" and make the metal > thinner. Only rarely would a real breakthrough occur - like, for example, > Gary Klein's large diameter aluminum tubes.
> I doubt anyone since 1890 designed a bike frame using only simple stress > formulas and simplifying assumptions, the way (say) plant engineers design > (say) fixed mounting brackets. We won't accept the large safety factors > (or rather, the extra weight) that method requires.
>> As for the tail hook it is not a constant width or cross section so it
>> is not a single average load. In addition there are several loads
>> imposed on the hook, sometimes several at the same time.
> Give us a good drawing or clear close-up photo, and we can discuss in a > sub-thread. To me, it looks like a long, pretty simply shaped two force > member, with one major tensile load and perhaps some brief rotational > inertia loads. I think the loads and stresses are simpler to calculate > than those of a bike frame. And as I said, I think a 3/4" strain gage > would give perfectly adequate stress information on such a two-force > member. On a bike frame bottom bracket area with its extreme stress > gradient, that would be far too big to be useful.
> -- > - Frank Krygowski
Frank Krygowski,
There are other loads to be considered: impact when the hook hits the deck, side loading if the plane isn't perpendicular to the wire when it hooks it, and likely more.
Kerry
> Op 20-8-2012 0:15, James schreef:
>> On 18/08/12 21:34, Lou Holtman wrote:
>>> Are you comparing a boat to a fork/stem safety wise or stress wise? When
>>> a fork breaks suddenly there is a great chance you end up at the dentist
>>> for a major repair. A decent fork costs 200-300 euro from a respected
>>> manufacturer. That is a lot cheaper than the dentist bill.
>> Are they decent? I hope so. I put my smile in their hands ;-)
> Al steerer, old technology.
So what? I ride a bicycle with a steel frame, home built wheels and a dynamo. What is decent? Is it chiefly reliability and longevity? How much do you factor in weight? Is it something that hasn't had a large number of failures in the field?
We all use fuzzy logic to determine what is decent in our own mind. Am I to assume that in your mind old tech is not decent?
> > Op 20-8-2012 0:15, James schreef:
> >> On 18/08/12 21:34, Lou Holtman wrote:
> >>> Are you comparing a boat to a fork/stem safety wise or stress wise? When
> >>> a fork breaks suddenly there is a great chance you end up at the dentist
> >>> for a major repair. A decent fork costs 200-300 euro from a respected
> >>> manufacturer. That is a lot cheaper than the dentist bill.
> >> Are they decent? I hope so. I put my smile in their hands ;-)
> > Al steerer, old technology.
> So what? I ride a bicycle with a steel frame, home built wheels and a
> dynamo. What is decent? Is it chiefly reliability and longevity? How
> much do you factor in weight? Is it something that hasn't had a large
> number of failures in the field?
> We all use fuzzy logic to determine what is decent in our own mind. Am
> I to assume that in your mind old tech is not decent?
I hear a lot of different things about carbon steerers -- that modern
designs with thicker tube ends are stronger than aluminum, and that
aluminum presents the risk of disbonding. On the other hand, there
were a number of reported CF steerer failures a few years ago, and
they were perceived as dangerous by the clydesdale set (me). Aluminum
steerers are certainly more tolerant of stem clamp over-tightening. My
forks have an aluminum steerer, but then again, I'm always a few years
behind technology-wise. My brother is similar size and weight, and he
rides full carbon without incident -- so far. Apart from weight, it
kind of hard to tell whether old tech aluminum or new tech CF is
better. From a weight perspective, new tech is clearly better. From
a reliability perspective, old-old tech (steel) is probably the best
-- and retired steel forks can be used for home protection, seal
clubbing, grounding rods, boat anchors and so many other things.
> >>> On Aug 19, 4:57 am, frkry...@gmail.com wrote:
> >>>> On Saturday, August 18, 2012 9:49:55 PM UTC-4, John B. wrote:
> >>>>> A 110 lb rider with a power output of
> >>>>> 3 HP on a 15 lb. vehicle with the following dimensions and a vertical
> >>>>> and horizontal CG of X and Y, mounted on wheels and tires with a
> >>>>> compliance of Z traveling at a velocity of V on a slope of a specified
> >>>>> angle with known air density and a measured drag, in a district with a
> >>>>> known gravitational attraction approaching a curve of known angle,
> >>>>> radius and length.
> >>>>> Certainly it can be calculated.
> >>>> I think it would be easier to calculate if the 3 HP came from a smooth
> >>>> running motor than from a human being. For example, one human might
> >>>> do that in a high gear at 60 rpm, rocking the bike mightily while
> >>>> yanking on the bars, while another might do it in a lower gear at 130
> >>>> rpm with the bike running much straighter. Motors are more
> >>>> predictable.
> >> I wonder why? After all an internal combustion, or a steam, engine for
> >> that matter, does not have a constant connecting rod loading and
> >> engineers have been calculating those loads for a while now. And
> >> (gulp) even doing it at different RPM's in difference directions
> >> calculating both the loading in both tension and compression....
> >> Frank, I've already told you that Columbus says that they have
> >> instrumented a bicycle frame and used the data to design new frame
> >> tubes with varying thickness and cross section to better resist the
> >> loads. You are getting all concerned about something that is already
> >> accomplished.
> > You don't seem to understand that Columbus's effort at instrumenting the
> > bike frame argues against your "this is simple" idea. Columbus would not
> > have taken the trouble to instrument the frame if the problem were as
> > simple as you imagine. Likewise, companies wouldn't bother to do FEA
> > analysis of their bike frames if the design process were easy.
> > Until perhaps 30 years ago, I think bike frame design was based on what
> > might be described as crowdsourced evolution. Designers would look at the
> > work of many other designers, as well as their own, to evaluate their
> > success. If frames often broke at a certain location, a designer would
> > beef up that location. If frames never broke at a certain location, a
> > designer would say "I can save some weight there" and make the metal
> > thinner. Only rarely would a real breakthrough occur - like, for example,
> > Gary Klein's large diameter aluminum tubes.
> > I doubt anyone since 1890 designed a bike frame using only simple stress
> > formulas and simplifying assumptions, the way (say) plant engineers design
> > (say) fixed mounting brackets. We won't accept the large safety factors
> > (or rather, the extra weight) that method requires.
> >> As for the tail hook it is not a constant width or cross section so it
> >> is not a single average load. In addition there are several loads
> >> imposed on the hook, sometimes several at the same time.
> > Give us a good drawing or clear close-up photo, and we can discuss in a
> > sub-thread. To me, it looks like a long, pretty simply shaped two force
> > member, with one major tensile load and perhaps some brief rotational
> > inertia loads. I think the loads and stresses are simpler to calculate
> > than those of a bike frame. And as I said, I think a 3/4" strain gage
> > would give perfectly adequate stress information on such a two-force
> > member. On a bike frame bottom bracket area with its extreme stress
> > gradient, that would be far too big to be useful.
> > --
> > - Frank Krygowski
> Frank Krygowski,
> There are other loads to be considered: impact when the hook hits the deck,
> side loading if the plane isn't perpendicular to the wire when it hooks it,
> and likely more.
There are always other things to be considered, no matter what the design.* But I maintain that design of a tailhook would be much more straightforward than design of a bike frame, at least if the bike frame was intended to be reasonably competitive regarding weight, stiffness, etc. I've posted links to FEA graphics showing the rather extreme stress gradients around the bottom bracket, head tube and seat cluster. The loads and stresses there are quite difficult to accurately represent and/or calculate. But even in the images Andrew linked, a tailhook sure looks like a two-force member.
* At a Design Education conference I attended long ago, a well-respected professor of mechanical design said he often started his students off by having them design a simple metal shelf that would be bolted to a concrete block wall at a factory work station. It was to protrude about 6" and be about 12" wide, and would hold the worker's adjustable wrench in easy reach, about 2 feet off the floor.
Easy, right? Except anyone who didn't design it to safely support 250 pounds lost serious credit. Because obviously, the worker's going to sit on that shelf the first day it's installed. There are always other things to be considered.
<frkrygowREM...@gEEmail.com> wrote:
>John B. wrote:
>> On Mon, 20 Aug 2012 15:53:31 -0700 (PDT), thirty-six
>> <thirty-...@live.co.uk> wrote:
>>> On Aug 19, 4:57 am, frkry...@gmail.com wrote:
>>>> On Saturday, August 18, 2012 9:49:55 PM UTC-4, John B. wrote:
>>>>> A 110 lb rider with a power output of
>>>>> 3 HP on a 15 lb. vehicle with the following dimensions and a vertical
>>>>> and horizontal CG of X and Y, mounted on wheels and tires with a
>>>>> compliance of Z traveling at a velocity of V on a slope of a specified
>>>>> angle with known air density and a measured drag, in a district with a
>>>>> known gravitational attraction approaching a curve of known angle,
>>>>> radius and length.
>>>>> Certainly it can be calculated.
>>>> I think it would be easier to calculate if the 3 HP came from a smooth running motor than from a human being. For example, one human might do that in a high gear at 60 rpm, rocking the bike mightily while yanking on the bars, while another might do it in a lower gear at 130 rpm with the bike running much straighter. Motors are more predictable.
>> I wonder why? After all an internal combustion, or a steam, engine for
>> that matter, does not have a constant connecting rod loading and
>> engineers have been calculating those loads for a while now. And
>> (gulp) even doing it at different RPM's in difference directions
>> calculating both the loading in both tension and compression....
>> Frank, I've already told you that Columbus says that they have
>> instrumented a bicycle frame and used the data to design new frame
>> tubes with varying thickness and cross section to better resist the
>> loads. You are getting all concerned about something that is already
>> accomplished.
>You don't seem to understand that Columbus's effort at instrumenting the >bike frame argues against your "this is simple" idea. Columbus would >not have taken the trouble to instrument the frame if the problem were >as simple as you imagine. Likewise, companies wouldn't bother to do FEA >analysis of their bike frames if the design process were easy.
Frank, I will say one thing, you are adroit... at squirming about and
changing the subject to one that you might be able to justify.
I didn't say that calculating loads on bicycles was a "simple"
project. I did say that Columbus had done the collection of data and
used this data to design new frame tubes. I did not say that they had
calculated loads and forces on the frame as I assumed that you, being
a PE, would be able to figure out that in order to decide on the size,
shape and thickness of the newly designed tubes they would have had to
know the forces on the frame and therefore had performed the necessary
calculations as part of their design project.
>Until perhaps 30 years ago, I think bike frame design was based on what >might be described as crowdsourced evolution. Designers would look at >the work of many other designers, as well as their own, to evaluate >their success. If frames often broke at a certain location, a designer >would beef up that location. If frames never broke at a certain >location, a designer would say "I can save some weight there" and make >the metal thinner. Only rarely would a real breakthrough occur - like, >for example, Gary Klein's large diameter aluminum tubes.
Well if you call large diameter aluminum tubes a "breakthrough" then
go ahead. I'd rather call it a bowing to the inevitable; that the
strength of aluminum is substantially less then steel.
>I doubt anyone since 1890 designed a bike frame using only simple stress >formulas and simplifying assumptions, the way (say) plant engineers >design (say) fixed mounting brackets. We won't accept the large safety >factors (or rather, the extra weight) that method requires.
I've already said that Frank. I said that in my opinion bike designers
were designing the standard of "Lighter then the other guy's", with a
bit of "Cheap to manufacturer" thrown in as a "fudge factor".
>> As for the tail hook it is not a constant width or cross section so it
>> is not a single average load. In addition there are several loads
>> imposed on the hook, sometimes several at the same time.
>Give us a good drawing or clear close-up photo, and we can discuss in a >sub-thread. To me, it looks like a long, pretty simply shaped two force >member, with one major tensile load and perhaps some brief rotational >inertia loads. I think the loads and stresses are simpler to calculate >than those of a bike frame. And as I said, I think a 3/4" strain gage >would give perfectly adequate stress information on such a two-force >member. On a bike frame bottom bracket area with its extreme stress >gradient, that would be far too big to be useful.
I'm really not interested in arguing the design of the tail hook. I
mentioned the project in partial evidence that instrumentation for
stress analysis is small and light and not a detriment (in case you
used that as an argument) to installing on a bike frame for data
collection.
>"Frank Krygowski" <frkrygowREM...@gEEmail.com> wrote in message >news:k10e64$bm7$1@dont-email.me...
>> Give us a good drawing or clear close-up photo, and we can discuss in a >> sub-thread. To me, it looks like a long, pretty simply shaped two force >> member, with one major tensile load and perhaps some brief rotational >> inertia loads. I think the loads and stresses are simpler to calculate >> than those of a bike frame. And as I said, I think a 3/4" strain gage >> would give perfectly adequate stress information on such a two-force >> member. On a bike frame bottom bracket area with its extreme stress >> gradient, that would be far too big to be useful.
>> -- >> - Frank Krygowski
>Frank Krygowski,
>There are other loads to be considered: impact when the hook hits the deck, >side loading if the plane isn't perpendicular to the wire when it hooks it, >and likely more.
>Kerry
Probably the major loads of interest is the speed and gross weight of
the plane when it hits the wire.
It might be of interest to know that any air force base that hosts
fighters has a arresting cable system installed which originally was
two lengths of ship's anchor chain laid down the sides of the runway
with a cable stretched across the runway connecting the two chains
together. (Very low maintenance device).
> I hear a lot of different things about carbon steerers -- that modern
> designs with thicker tube ends are stronger than aluminum, and that
> aluminum presents the risk of disbonding. On the other hand, there
> were a number of reported CF steerer failures a few years ago,[...]
Please, "carbon fiber reinforced polymer" or "CFRP". This is "tech", ya know.
-- Tēm ShermĒn - 42.435731°N, 83.985007°W
Post Free or Die!
> Well if you call large diameter aluminum tubes a "breakthrough" then
> go ahead. I'd rather call it a bowing to the inevitable; that the
> strength of aluminum is substantially less then steel.
However, the strength to weight (and stiffness to weight) of cro-moly steel, 6000 and 7000 series aluminium alloys, and titanium/titanium alloys are in the same ballpark.
-- Tēm ShermĒn - 42.435731°N, 83.985007°W
Post Free or Die!
> John B. wrote:
> [...]
>> As for the tail hook it is not a constant width or cross section so it
>> is not a single average load. In addition there are several loads
>> imposed on the hook, sometimes several at the same time.
> Give us a good drawing or clear close-up photo, and we can discuss in a
> sub-thread. To me, it looks like a long, pretty simply shaped two force
> member, with one major tensile load and perhaps some brief rotational
> inertia loads. I think the loads and stresses are simpler to calculate
> than those of a bike frame. And as I said, I think a 3/4" strain gage
> would give perfectly adequate stress information on such a two-force
> member. On a bike frame bottom bracket area with its extreme stress
> gradient, that would be far too big to be useful.
I thought Tailhook involved Navy and Marine officers getting drunk and sexually harassing and assaulting women.
-- Tēm ShermĒn - 42.435731°N, 83.985007°W
Post Free or Die!
>> >>> On Aug 19, 4:57 am, frkry...@gmail.com wrote:
>> >>>> On Saturday, August 18, 2012 9:49:55 PM UTC-4, John B. wrote:
>> >>>>> A 110 lb rider with a power output of
>> >>>>> 3 HP on a 15 lb. vehicle with the following dimensions and a vertical
>> >>>>> and horizontal CG of X and Y, mounted on wheels and tires with a
>> >>>>> compliance of Z traveling at a velocity of V on a slope of a specified
>> >>>>> angle with known air density and a measured drag, in a district with a
>> >>>>> known gravitational attraction approaching a curve of known angle,
>> >>>>> radius and length.
>> >>>>> Certainly it can be calculated.
>> >>>> I think it would be easier to calculate if the 3 HP came from a smooth
>> >>>> running motor than from a human being. For example, one human might
>> >>>> do that in a high gear at 60 rpm, rocking the bike mightily while
>> >>>> yanking on the bars, while another might do it in a lower gear at 130
>> >>>> rpm with the bike running much straighter. Motors are more
>> >>>> predictable.
>> >> I wonder why? After all an internal combustion, or a steam, engine for
>> >> that matter, does not have a constant connecting rod loading and
>> >> engineers have been calculating those loads for a while now. And
>> >> (gulp) even doing it at different RPM's in difference directions
>> >> calculating both the loading in both tension and compression....
>> >> Frank, I've already told you that Columbus says that they have
>> >> instrumented a bicycle frame and used the data to design new frame
>> >> tubes with varying thickness and cross section to better resist the
>> >> loads. You are getting all concerned about something that is already
>> >> accomplished.
>> > You don't seem to understand that Columbus's effort at instrumenting the
>> > bike frame argues against your "this is simple" idea. Columbus would not
>> > have taken the trouble to instrument the frame if the problem were as
>> > simple as you imagine. Likewise, companies wouldn't bother to do FEA
>> > analysis of their bike frames if the design process were easy.
>> > Until perhaps 30 years ago, I think bike frame design was based on what
>> > might be described as crowdsourced evolution. Designers would look at the
>> > work of many other designers, as well as their own, to evaluate their
>> > success. If frames often broke at a certain location, a designer would
>> > beef up that location. If frames never broke at a certain location, a
>> > designer would say "I can save some weight there" and make the metal
>> > thinner. Only rarely would a real breakthrough occur - like, for example,
>> > Gary Klein's large diameter aluminum tubes.
>> > I doubt anyone since 1890 designed a bike frame using only simple stress
>> > formulas and simplifying assumptions, the way (say) plant engineers design
>> > (say) fixed mounting brackets. We won't accept the large safety factors
>> > (or rather, the extra weight) that method requires.
>> >> As for the tail hook it is not a constant width or cross section so it
>> >> is not a single average load. In addition there are several loads
>> >> imposed on the hook, sometimes several at the same time.
>> > Give us a good drawing or clear close-up photo, and we can discuss in a
>> > sub-thread. To me, it looks like a long, pretty simply shaped two force
>> > member, with one major tensile load and perhaps some brief rotational
>> > inertia loads. I think the loads and stresses are simpler to calculate
>> > than those of a bike frame. And as I said, I think a 3/4" strain gage
>> > would give perfectly adequate stress information on such a two-force
>> > member. On a bike frame bottom bracket area with its extreme stress
>> > gradient, that would be far too big to be useful.
>> > --
>> > - Frank Krygowski
>> Frank Krygowski,
>> There are other loads to be considered: impact when the hook hits the deck,
>> side loading if the plane isn't perpendicular to the wire when it hooks it,
>> and likely more.
>There are always other things to be considered, no matter what the design.* But I maintain that design of a tailhook would be much more straightforward than design of a bike frame, at least if the bike frame was intended to be reasonably competitive regarding weight, stiffness, etc. I've posted links to FEA graphics showing the rather extreme stress gradients around the bottom bracket, head tube and seat cluster. The loads and stresses there are quite difficult to accurately represent and/or calculate. But even in the images Andrew linked, a tailhook sure looks like a two-force member.
Frank, why are you rabbiting on about tail hooks for? The discussion
was originally the possibility of repairing carbon frames. Then
digressed to the possibilities of actually analyzing the stresses on a
bicycle frame and no you are off and running chasing tail hooks.
Enough already.
>* At a Design Education conference I attended long ago, a well-respected professor of mechanical design said he often started his students off by having them design a simple metal shelf that would be bolted to a concrete block wall at a factory work station. It was to protrude about 6" and be about 12" wide, and would hold the worker's adjustable wrench in easy reach, about 2 feet off the floor.
>Easy, right? Except anyone who didn't design it to safely support 250 pounds lost serious credit. Because obviously, the worker's going to sit on that shelf the first day it's installed. There are always other things to be considered.
And now it is shelves.... what next? The Mars Mission? Whether the
moon landing was a hoax?
On Wed, 22 Aug 2012 00:58:12 -0500, "Tom $herman (-_-)"
<""twshermanREMOVE\"@THI$southslope.net"> wrote:
>On 8/21/2012 7:57 PM, Jay Beattie wrote:
>> I hear a lot of different things about carbon steerers -- that modern
>> designs with thicker tube ends are stronger than aluminum, and that
>> aluminum presents the risk of disbonding. On the other hand, there
>> were a number of reported CF steerer failures a few years ago,[...]
>Please, "carbon fiber reinforced polymer" or "CFRP". This is "tech", ya >know.
Right. Down to the LBS and say, "I want a carbon fiber reinforced
polymer bike", and the guy says, "We don't have those but we do have
these svelte carbon bikes right over here :-)
On Wed, 22 Aug 2012 01:03:01 -0500, "Tom $herman (-_-)"
<""twshermanREMOVE\"@THI$southslope.net"> wrote:
>On 8/21/2012 10:47 PM, John B. wrote:
>> Well if you call large diameter aluminum tubes a "breakthrough" then
>> go ahead. I'd rather call it a bowing to the inevitable; that the
>> strength of aluminum is substantially less then steel.
>However, the strength to weight (and stiffness to weight) of cro-moly >steel, 6000 and 7000 series aluminium alloys, and titanium/titanium >alloys are in the same ballpark.
So you use large diameter, but thinner, aluminum and viola! You are
lighter and stiffer then steel. Rocket Science it is not.
Apparently been going on for years. I recently read a review of the
Roman soldier's equipment and apparently they changed the helmet from
bronze to steel - lighter and stiffer, i.e., didn't dent so badly :-)
