Welding Vs. Brazing
Michael Zenner
I'm looking into starting a small framebuilding setup, and had a few
questions to lay before all of the experts that I know read this group.
First off, I've heard varying things about the pros and cons of using
brazing as a joining technique vs. TIG welding. My understanding is that
TIG welding is a less expensive method of joining, but that a skillfully
worked lugged and brazed joint can be just as light, and, because the metal
is heated to a lower temperature, there is less fatigue at the joint, and
therefore the joint is stronger (Thi assumes proper technique all around,
as well as a high quality investment cast lug).
Any comments from builder types out there?
On a related topic: Would anyone who knows care to share with me any info
regarding the relative costs involved in getting set up with the equipment to
do either type of construction? That is, exactly what kind of oxy/acet. torch
does one need to make a _small_ production run of lugged frames, and what a
would one expect to need to pay for such equipment vs. one of the newer TIG
welders that can run off of house current? Is the cost difference something
that is noticible at a very small volume of frames produced, or is that mainly
a concern if you want to be the next Specialized?
On a less-related topic: Why can't you braze titanium? Or is it just that
nobody does?
Thanks greatly,
-Michael Zenner
Gary Helfrich
: First off, I've heard varying things about the pros and cons of using
: brazing as a joining technique vs. TIG welding. My understanding is that
: TIG welding is a less expensive method of joining, but that a skillfully
: worked lugged and brazed joint can be just as light, and, because the metal
: is heated to a lower temperature, there is less fatigue at the joint, and
: therefore the joint is stronger (Thi assumes proper technique all around,
: as well as a high quality investment cast lug).
The lugged (or fillet brazed) joint is far heavier than the tig welded
joint. Remember that I am speaking of the weight of the joint itself,
not the entire frame. While the theoretical differences in strength
between these joining methods can be argued until the cows come home, I
am unaware of any documented evidence that one method has a lower failure
rate in the field. Brazing equipment is far less expensive (see below)
and is an easier skill to acquire.
: On a related topic: Would anyone who knows care to share with me any info
: regarding the relative costs involved in getting set up with the equipment to
: do either type of construction? That is, exactly what kind of oxy/acet. torch
: does one need to make a _small_ production run of lugged frames, and what a
: would one expect to need to pay for such equipment vs. one of the newer TIG
: welders that can run off of house current? Is the cost difference something
: that is noticible at a very small volume of frames produced, or is that mainly
: a concern if you want to be the next Specialized?
Good brazing equipment will set you back about $400. A tig setup with a
Miller Maxstar 91 (runs on 120v) is about $2500. A used brazing setup
can be had for even less, but I doubt that you will find any state of the
art inverter tig machines used at this point. This makes the difference
in price even more dramatic. Remember that joining the tubes is only a
small part of the framebuilding picture. If you chose to use snips and
files to miter the tubes, the cost is not that great, but a used milling
machine and lathe can cost the better part of $10,000. Jigs can range
from a sheet of plywood to a Bike Machinery for $9,000.
One other point about welding: It is far faster than brazing and imposes
no limits on your design the way lugs will.
: On a less-related topic: Why can't you braze titanium? Or is it just that
: nobody does?
You can braze titanium in a vacuum furnace. The question is why would
you want to? Welded construction has a proven record of reliability.
Gary Helfrich
Arctos Machine
John Olsen
>
>Michael Zenner (mze...@agora.rdrop.com) wrote:
>
>: First off, I've heard varying things about the pros and cons of using
>: brazing as a joining technique vs. TIG welding.
While the theoretical differences in strength
>between these joining methods can be argued until the cows come home, I
>am unaware of any documented evidence that one method has a lower failure
>rate in the field.
>Gary Helfrich
>Arctos Machine
Hey, Gary, Cycling Science published a test years ago comparing the fatigue
resistance of lugged vs. fillet brazed vs. TIG welded sample joints.
Seemed like a good test at the time (I don't remember who did it), and it
showed fatigue life to stack up with lugged joints best, fillet joints
second, and TIG joints 3rd. The author thought TIG's poor performance
might have something to do with the small but abrupt heat affected zone
being entirely located so close to the high local stresses of the joint.
<flames here>
In the field, they seem to all work if the frame is designed half way well
and the welding is competent. I would add that brazing is a lot more
peaceful, doesn't require very dark goggles, and smells worse, not to
mention the dreaded Black Booger Syndrome. TIG welding's major advantage
in my opinion (at least on friendly old steel) is the total absence of
cleanup time. A fillet brazed bike can take hours and hours of filing if
you are as mediocre a brazer as I am. A TIG'd bike doesn't need cleanup.
I'd do TIG if I weren't so cheap. Plus, of course, TIG welds steel, al,
and titanium, and brazing anything but steel is... unusual.
Olsen
Gary Helfrich
: Hey, Gary, Cycling Science published a test years ago comparing the fatigue
: resistance of lugged vs. fillet brazed vs. TIG welded sample joints.
: Seemed like a good test at the time (I don't remember who did it), and it
: showed fatigue life to stack up with lugged joints best, fillet joints
: second, and TIG joints 3rd. The author thought TIG's poor performance
: might have something to do with the small but abrupt heat affected zone
: being entirely located so close to the high local stresses of the joint.
I was the one who did the welded samples for that test. The results were
interesting, but did not correlate to real life failure rates.
Specifically, Fat City was making both brazed and tig frames at that
time, using the same tubing, and if anything, the brazed bike failure
rate was slightly higher. Who knows?
Gary Helfrich
Arctos Machine
Jim Papadopoulos
> I was the one who did the welded samples for that test. The results were
> interesting, but did not correlate to real life failure rates.
> Specifically, Fat City was making both brazed and tig frames at that
> time, using the same tubing, and if anything, the brazed bike failure
> rate was slightly higher. Who knows?
WELL, now I get to ask -- were the ensuing critical letters correct or not?
One had charged that no filler was used..... another that (?) there was no
postweld stress relief..... I don't remember what else.
I recall being very unconvinced by that particular article.
JP
John Olsen
>
>Jim Papadopoulos (jj...@cornell.edu) wrote:
>: In article <3i0dpk$4...@crl.crl.com>, lap...@crl.com (Gary Helfrich) wrote:
>
>: > I was the one who did the welded samples for that test.
Gary, Jim: That article may have been flawed, but if so it deserves to
be redone in a better way. Any suggestions, specifics? We should redo
it one way or the other. I know Bicycling would be interested. Thoughts?
Olsen
Gary Helfrich
Yeah, if you want to do it, I'll do the welding again. This time it
might be interesting to compare similar weight assemblies, rather than
using the same tubing with with each attachment method. After all,
strength/weight is what we are interested in. It would be easy to
figure how much additional weight is involved in lugged and brazed
construction. This would tell you how much thicker you could make the
tubing in a welded bike, and not increase the weight. Wall thickness on
all three samples would be adjusted to yield identical projected frame
weights.
It would also be interesting to test for more than just endurance limit
of the joint. Short cycle fatigue, yield strength, and impact testing
would all be useful. Since you are going to be performing the tests,
lets keep the list as long as possible ;-).
Gary Helfrich
Arctos Machine
Jim Papadopoulos
jol...@techcenter.paccar.com (John Olsen) wrote:
> Gary, Jim: That article may have been flawed, but if so it deserves to
> be redone in a better way. Any suggestions, specifics? We should redo
> it one way or the other. I know Bicycling would be interested. Thoughts?
I don't remember all my questions and reservations.
The first concern was to have 'best practice' joints of all types --
probably by boosters of each technology. ..... with precise descriptions
of what was done and authority for calling it 'best practice'.
Then, the loading should possibly be non-reversed bending, sagging-type
moment, so as to fail the un-affected tubing in 50,000 cycles.
I would not try for constant weight, but simply measure the weight of each
joint.
If you mimic an upper head joint effectively, maybe some gusset or
mending-plate devotees would also submit samples for evaluation?
I might have time available on my tester, late Spring?
JP
Gary Helfrich
: I would not try for constant weight, but simply measure the weight of each
: joint.
Strength to weight ratio is an extremely important issue in bike frame
design. Interpolated data from using identical tubes is unlikely to be
accurate. Heat affected zones, quench rate, and several other factors
make accurate interpolation more time consuming than compensating the
tubing weight in the samples. The most accurate test would to build
entire front triangles of identical weight, and test the entire assembly,
not one isolated joint. This is the way the big boys do it (Trek, GT, et
al), so we sould try to be at least as thorough as they are.
Gary Helfrich
Arctos Machine
Matt Bushore
|> NNTP-Posting-Host: crl10.crl.com
|> X-Newsreader: TIN [version 1.2 PL2]
Why must they be of equal weight? My understanding of the purpose
of this test (which is admittingly incomplete) is to measure the strength
to weight ratio.
If you create an assembly to a specific wieght, you may be falling
prey to non linear effects of the joining processes. Samples that are considered
to be representative of the process, to me, would provide better indicators
of the process efficiency.
Perhaps the test should also be done at several different target
weights as well?
To put it a different, less(?) clouded way, if you build at specified
weight, the variables that you are testing would involve joint strength at
that weight. If you build at a weight typical of the process, you would
get information more indicative of process effeciency.
If you build at several weights, statistically, the data will be much,
much better, becuase you will be able to determine if the strength is
determined by the weight, or the joining method.
Holy crap-I actually remember something from a Stats class about
confounded effects.
If I hadn't sold my stats book, I would review it. The gist being
that if the test is conducted so as to control certian variables that may
have an effect on the outcome, you cannot say that main effects that you are
attempting to measure are statistically significant.
I am going back to class before I get smacked with a wet trout.
Matt Bushore
Wannabe statician Poor Speller
Snappy Dresser Groovy Guy
John Olsen
Since you are going to be performing the tests,
>lets keep the list as long as possible ;-).
>
>Gary Helfrich
>Arctos Machine
>
Gary:
Thanks. You are a natural Managing Editor. ;-(
Olsen
John Olsen
>I don't remember all my questions and reservations.
>The first concern was to have 'best practice' joints of all types --
>probably by boosters of each technology. ..... with precise descriptions
>of what was done and authority for calling it 'best practice'.
Jim- would you be willing to look up the article and see what it was that
bugged you? The Best Practice thing was my reservation...
>
>Then, the loading should possibly be non-reversed bending, sagging-type
>moment, so as to fail the un-affected tubing in 50,000 cycles.
>
>I would not try for constant weight, but simply measure the weight of each
>joint.
>
>If you mimic an upper head joint effectively, maybe some gusset or
>mending-plate devotees would also submit samples for evaluation?
>
>I might have time available on my tester, late Spring?
>
>JP
Sounds great! We could get masters of each art form to do each type of
joint using agreed upon materials, with each sample made by the same
person. If Gary could do the TIG sample, maybe Glenn Ericson would do the
lugged and fillet jointed samples. We could cut a bunch of sample tube
sets identically to fit whatever test rig you come up with.
Should we test aluminum and titanium samples as well? They would obviously
be tigged but they would give a comparative life that would be interesting.
?
Olsen
John Olsen
>Yeah, if you want to do it, I'll do the welding again.
Great. Jim P. has a test machine he can let us use.
I know how good you are at TIG, would you feel comfy doing the other two?
If not, maybe Glenn Ericson does the brazing and filleting? Up to you.
This time it
>might be interesting to compare similar weight assemblies, rather than
>using the same tubing with with each attachment method. After all,
>strength/weight is what we are interested in. It would be easy to
>figure how much additional weight is involved in lugged and brazed
>construction. This would tell you how much thicker you could make the
>tubing in a welded bike, and not increase the weight. Wall thickness on
>all three samples would be adjusted to yield identical projected frame
>weights.
This seems like a fair approach, but I wonder if we could get frame tubing
in fine enough increments to really equalize the frame wts? The differences
in joint wt are small, nes pas? Plus, do manufacturers really tune the
thicknesses that closely? I don't know.
>
>It would also be interesting to test for more than just endurance limit
>of the joint. Short cycle fatigue, yield strength, and impact testing
>would all be useful. Since you are going to be performing the tests,
>lets keep the list as long as possible ;-).
>
>Gary Helfrich
>Arctos Machine
>
I certainly see the virtues in all of those tests, but I don't immediately
see how to do the impact test. Shouldn't be too hard to rig a tester.
Charpy-style swinging weight?
Olsen
Pilato
School of Mines. They have a setup where the will try to determine
which is stronger, brazing or TIG. If you guys are interested my e-mail
adress is ppi...@mines.colorado.edu
pasquale
Gary Helfrich
: A question for Gary Helfrich and others knowledgable in the field. Why
: does just about everybody use torches to heat the joints in quality
: bicycle frames? Couldn't an induction brazing setup deliver the necessary
: power to just the right place, thus getting a similar heat affected zone as
: welding?
Some large manufacturers do use induction brazing on some joints. Areas
with great dissimilarities in section, or complex part geometry, such as
the stay/dropout joints do not lend themselves to accurate induction
heating. The HAZ is slightly smaller in the case of induction heating,
but not still does not approach the size of a good welded joint. Remember
that the HAZ must be at least as large as the lug itself, which is about
40mm away from the joint. A 100mm HAZ is not uncommon. Good tig welding
will have a HAZ in the 25mm range.
Gary Helfrich
Arctos Machine
Daryel M. Bush
> [...] brazing anything but steel is... unusual.
Why? Besides the fact that people just aren't
doing it of course. Is it "too heavy" for ti
bikes or something? ---Daryel
Gary Helfrich
Heavy in the pocketbook. Brazing alloys with good mechanical properties
for titanium require brazing in a vacuum. Kinda makes fillet a bit
tough. There are brazing alloys that work (sort of) in air with
incredibly toxic flux systems, but the joints are low in strength and
ductility.
Gary Helfrich
Arctos Machine
Keith Bontrager
Gary,
Maybe it's time to discuss the implications of HAZ location and
size. I see this running off as a smaller HAZ is better without
a rational basis for this. We'll all have to buy lazers and Ebeams
to stay cool if that unfortunate event comes to pass eh?
As I see it:
Any method of joining except adhesive techniques creates a
significant HAZ. ie loss of material mechanical properties.
There are "features" in the HAZ that are complex. The existance
of the HAZ can be roughly demonstrated from oxides on the surface
of the material sometimes (depending on process stuff), but this
is not very useful. The thermal gradient and thermal cycling
that occurs in the welding or brazing cycle at each point in the
HAZ is complicated, and only with a knowledge of these details
could one begin to understand the metallurgical changes in the
base metal.
In many load cases the stress on the material in a joint is higher
closer to the joint. There are also typically abrupt changes in
the section of the joint as one gets closer to the weld. For these
reasons it is not necessarily a good thing to have a small HAZ
very close to the weld area. A "metallurgical notch" in this area
reduces the fatigue strength of the joint in the simplest case.
Small HAZs are easier to manage w/respect to reinforcement methods.
This is one of the reasons welding is useful. There are others.
We'll see what kind of mood you are in today.
KB
Thomas H. Kunich
Keith Bontrager <rich...@cruzio.com> wrote:
>Maybe it's time to discuss the implications of HAZ location and
>size. I see this running off as a smaller HAZ is better without
>a rational basis for this. We'll all have to buy lazers and Ebeams
>to stay cool if that unfortunate event comes to pass eh?
Keith, isn't all this stuff about brazing and welding just so much
salesmanship? We all know that a smaller HAZ makes tubing with shorter
butting practical. This allows the tubing to be lighter. OK
But the difference in weight in, for all intents and purposes, insignificant.
The short butts and the decreased weight is so slight that it only gives
the PR folks something to throw in the faces of the aluminum and carbon
frmae crowds.
Aluminum, titanium and carbon fiber will probably pretty much disappear
in one of their frequent popularity cycles simply because steel is
so much more cost effective and feels so lively.
But let's not lose sight of the fact that brazing a lugged frame bike
is a perfectly acceptable method of fastening things together and
welding of any sort requires far more skill on super thin tubes for
not much return in improvements (OK, you can change the design just
by adjusting the jigs rather than making new lugs.)
Gary Helfrich
: In article <3ivifd$d...@crl9.crl.com>, lap...@crl.com (Gary Helfrich) writes:
: > Some large manufacturers do use induction brazing on some joints. Areas
: > with great dissimilarities in section, or complex part geometry, such as
: > the stay/dropout joints do not lend themselves to accurate induction
: > heating. The HAZ is slightly smaller in the case of induction heating,
: > but not still does not approach the size of a good welded joint. Remember
: > that the HAZ must be at least as large as the lug itself, which is about
: > 40mm away from the joint. A 100mm HAZ is not uncommon. Good tig welding
: > will have a HAZ in the 25mm range.
: >
: > Gary Helfrich
: > Arctos Machine
: Gary,
: Maybe it's time to discuss the implications of HAZ location and
: size. I see this running off as a smaller HAZ is better without
: a rational basis for this. We'll all have to buy lazers and Ebeams
: to stay cool if that unfortunate event comes to pass eh?
Am I running off about HAZ size? All I did was offer some observations on
what I have observed as typical HAZ's. I did not offer a judgment about
the merits of HAZ size. If you wish to lecture on the subject, enjoy
yourself!
<lecture ommitted>
: We'll see what kind of mood you are in today.
Is this the royal we? I'm flattered that you care so much.
Robert Horvatich
This is an interesting discussion arguing over what variables are going
to be held constant. Much debate over lugged vs. tig welded frames hinges
on that tig welding is lighter but supposedly less durable. Isn't this
test suppose to reveal these differences. I don't think welding up
samples may be the best approach to take to demonstrate this. The results
would yield information that would be very difficult to correlate to
factory built frames. Most readers of the results of this proposed test
would undoubtedly try to fit their own conclusion and assume it to be
universally applicable. Failing all along to include other variables
that were controlled in your test but not in the real world. I think
this was displayed in the shortcomings of the first iteration of this test.
What I am getting at ultimately is that other factors along the manufacturing
process are going to effect the strength and fatigue characteristics of
a frame. Lets say your very controlled test concludes that TIG welding
is much stronger than lugged joining. The consumer is going to go out
and buy a TIG welded frame assuming it is stronger without realizing the
other factors involved in the equation (eg. design, heat treatment, surface
prep before welding, ect...)
We can test until we are blue in the face to try to understand the
gazzilion and one variables. Admittedly a very complicated and expensive
process. I think it would be more useful to compare frames that people
would compare when buying a bike. Take what the manufacturer offers off
the floor and subject them to standardized tests. This would include
all the variables that effect frame strength that are present in the
manufacturing process. Undoubtedly, we will find TIG welded frame A
to be better than lugged frame B. I am sure just the opposite
would be true with lugged frame C better than poor design of TIG
frame D. Get my point? I think this would be far more valuable to
the reader than a very specific controlled test.
I am currently thinking of all the standards that have been
established over the years for the automotive industry. A car magazine
can report that in the standard 35 mph crash barrier test, the driver
will be exposed to certain G loads. It doesn't matter if the car is
a unibody or chassis frame construction, it matters if the driver is
going to get out unharmed. These tests give the reader a way to evaluate
and compare whole products, not just one aspect of it.
I think standardized tests can give us some insight to the relative
durability and strength of bikes as whole products. This information
would be invaluable to those heavy riders that tend to break bikes
frequently. We then could honestly say bike A is better than bike B
according to these standards. We may quantitatively find differences
in materials and designs if correlations exist in test results. Gosh,
it would probably make r.b.tech a lot quieter place with its many
strong opinions based on limited knowledge. ;)
Food for thought on further discussions to a highly controlled test.
Sorry about the length here, I guess this is something that I think many
people could really benefit from.
Rob
email: |"You can't take life too seriously,
rho...@tbd161.tbd.ford.com | you don't get out alive." Bugs Bunny
John Olsen
>
>On 24 Feb 1995, Gary Helfrich wrote:
>
>> Jim Papadopoulos (jj...@cornell.edu) wrote:
>>
>> : I would not try for constant weight, but simply measure the weight of each
>> : joint.
>>
I don't think welding up
>samples may be the best approach to take to demonstrate this. The results
>would yield information that would be very difficult to correlate to
>factory built frames. Most readers of the results of this proposed test
>would undoubtedly try to fit their own conclusion and assume it to be
>universally applicable. Failing all along to include other variables
>that were controlled in your test but not in the real world. I think
>this was displayed in the shortcomings of the first iteration of this test.
>
Robert-
I agree that there are lots of manufacturing variables that come into the
real world, but we are talking about a comparison between best-case samples
of each type of joint. This relatively simple test would give a useful
nugget of information. One thing at a time.
The variability and sensitivity of the different types of joining methods
is a real problem, but it is a different problem. One could go out and
buy random samples of a whole bunch of different frames and run standardized
tests on them, but even then you would be open to criticism because of
individual variability. How many Trek 950 frames do you need to test to
understand the statistics of fatigue failure of this one model? 5? 10?
The test soon becomes very complicated.
In our handlebar test, we tested 3 samples as a compromise and saw a lot of
variation. What a customer would really like to know is "what frame will
give me a sufficient fatigue life under MY loading history, and what
frame will have the minimum number of outliers with very short lives?"
This is a challenge, nes pas?
Olsen
Gary Helfrich
: Keith, isn't all this stuff about brazing and welding just so much
: salesmanship? We all know that a smaller HAZ makes tubing with shorter
: butting practical. This allows the tubing to be lighter. OK
: But the difference in weight in, for all intents and purposes, insignificant.
: The short butts and the decreased weight is so slight that it only gives
: the PR folks something to throw in the faces of the aluminum and carbon
: frmae crowds.
A set of lugs can weigh more than the toptube in your frame. If this is
insignificant to you, fine. The weight of the brazing alloy is similar
to the weight of the welding wire used in tig welding. At Fat City, we
built both welded and fillet brazed frames using the same tubeset. The
fillet brazed frames averaged 400-450 grams heavier. This weight savings
also resulted in a bike that was less expensive to manufacture, and had a
lower return rate. I would not discount this as just PR.
Gary Helfrich
Arctos Machine
rich...@cruzio.com
I agree in a very narrow sense. The discussion of the advantages
and disadvantages of the different joining process is generally
insufficiently technically accurate to be of any use. This is especially
true w/respect to short butted tubes and welding. I've never bought
into the logic of that :>
There is a weight difference, and it is small. But don't discount
the advantages of any legitimate weight reduction. It's hard to
reduce the weight of a highly stressed structure without reducing
it's reliability. I don't feel that frames built with this style
of tube can be made sufficiently durable to sell to the general
public (unless you do the whole thing through a post office box).
I don't think aluminum or ti will go away. Each material has properties
that, when thoroughly understood, will make good bike frames. Steel
is cheaper in the lowest grades, and it gets closer to the other
materials as the design, material, and manufacturing process is
refined to compete with the other materials. It would seem that
aluminum is relatively cheap these days judging by the number
of bikes that are available at mid prices. Hell, Specialized is
even giving one away to the lucky net freak just for sending them
some Email (I did!).
Of course, I wouldn't mind if the other disappeared for awhile.
It would make my life easier for a bit anyway.
I'm not sure I agree with your use of the term "lively", and
I'd refer you to the long and tedious postings between Jobst
and Jim P if you want to find out more about this. Not me.
Not now. No way. Save your electrons.
And with respect to lugged frames:
(Excuse the frankness of this bit. I expect to get some flames.
I will crush every one of the authors that want to go beyond
name calling though, so, when you compose your flames, use lots
of four letter words and stay away from the tech talk or risk
being exposed for what you are :> )
Lugs are great for arty road bikes. You can carve little patterns
into them. You can spend your life filing clean square edges
all the way around them. You can perfect your brazing skills
to the point that there is little or no clean up labor required
after you're done. That's about it. They've no useful application
on off road bikes if you are igoing to ride it hard.
The same can be said for fillet brazed frames unless a substantial
internal reinforcement is used. On a road bike a strong racer can
get quite a few miles in on one before it breaks. They can crash it
and get away with it as long as they don't run into anything in
the process. But, for severe use, neither one has much going for it
except heritage.
Welding is not much better off unless there is something done to
improve the fatigue strength of the joints. The folks that
realize this use rienforcements on the joints. If the joints
are designed properly and executed well the welded and reinforced
joint will out perform the others by a significant factor.
Strength to weight, minimum weight possible, you name the
design protocol, welded and reinforced will be the best.
I figure the guys who were blue skying about doing the joint
tests will go off on this one. You guys should work it into
your plans in my humble opinion. Gary, do you have a humble
opinion on the matter? (Do you have a humble opinion - period?)
Back to Mr. Kunich's post.
It isn't that hard to weld thin tubes together. That was a bit
of misinformation that will go away in a while. It takes a fair
amount of skill to connect thin tubes together no matter what
process you use. The welders in our shop make more than I do.
Welding vs brazing is not an issue for off road bikes. The real
question is how well a given reinforcement scheme works on a welded
frame - all materials included - only matters in severe use, which
happens quite a bit in the dirt.
There can
be a discussion of the differences these processes yield in
the construction of road frames, but only in the sense that art
and tradition can take precedence over structural concerns.
Adios,
KB
bikes
John Olsen
>
>Daryel M. Bush (dmb...@tntech.edu) wrote:
>: jol...@techcenter.paccar.com (John Olsen) writes:
>: > [...] brazing anything but steel is... unusual.
>
>: Why?
>
>ductility.
>
>Gary Helfrich
Gary, do you know what Pino used to use when he brazed titanium? One of
the incredibly toxic options, perhaps?
People sell aluminum brazing kits, but I know nothing about the strength
and durability of the joints. I see them at every motorcycle and car
show. Do you know anything about this?
Olsen
Keith Bontrager
This is primarily for Jobst.
KB
Gary Helfrich
: Gary, do you know what Pino used to use when he brazed titanium? One of
: the incredibly toxic options, perhaps?
: People sell aluminum brazing kits, but I know nothing about the strength
: and durability of the joints. I see them at every motorcycle and car
: show. Do you know anything about this?
Pino, according to second hand information, used a vacuum furnace. I'd
tend to believe the story, considering what a fanatic he was.
Gary Helfrich
Arctos Machine
Gary Helfrich
: I figure the guys who were blue skying about doing the joint
: tests will go off on this one. You guys should work it into
: your plans in my humble opinion. Gary, do you have a humble
: opinion on the matter? (Do you have a humble opinion - period?)
I think Keith overdid his Jobst medication this morning.
OK.. Prove that your bikes have lower failure rates than other similar
high quality tig welded steel bikes. Publish some actual return rates.
I doubt that Ibis, for example, has a significantly higher failure rate
than Bontrager. Oh right, you have your demographic studies that prove
that only yuppies ride Ibis, and those aggro blue collar types ride your
bikes, so of course any comparison is invalid. We all know that
professional types never ride hard, and just buy bikes that look good on
the roof rack of their Explorer. :-).
Gary Helfrich
Arctos Machine
John Olsen
>
>In article <tomkD50...@netcom.com>, to...@netcom.com (Thomas H. Kunich) writes:
>> In article <D4zMp...@cruzio.com>,
>> Keith Bontrager <rich...@cruzio.com> wrote:
>>
If the joints
>are designed properly and executed well the welded and reinforced
>joint will out perform the others by a significant factor.
>Strength to weight, minimum weight possible, you name the
>design protocol, welded and reinforced will be the best.
You may be right. Maybe we should include this construction in
any testing we do. Notice: This constitutes "going off".
>I figure the guys who were blue skying about doing the joint
>tests will go off on this one.
See above.
>You guys should work it into
>your plans in my humble opinion. Gary, do you have a humble
>opinion on the matter? (Do you have a humble opinion - period?)
Gee, do YOU, KB? Gary's posts tend to be rather nice in contrast to your
own.
rich...@cruzio.com
> rich...@cruzio.com wrote:
>
> : I figure the guys who were blue skying about doing the joint
> : tests will go off on this one. You guys should work it into
> : your plans in my humble opinion. Gary, do you have a humble
> : opinion on the matter? (Do you have a humble opinion - period?)
>
What I meant here was that you were designing an experiment that would
take at least a zillion dollars and a zillion man hours to execute, and
would be tough to understand when you were done. It's easy to get
carried away with the design of experiments (blue skying) but this
generally gets sorted out when you get to the actual work. My comments
on your lack of humility are evident to everyone but you.
> I think Keith overdid his Jobst medication this morning.
>
> OK.. Prove that your bikes have lower failure rates than other similar
> high quality tig welded steel bikes. Publish some actual return rates.
> I doubt that Ibis, for example, has a significantly higher failure rate
> than Bontrager. Oh right, you have your demographic studies that prove
> that only yuppies ride Ibis, and those aggro blue collar types ride your
> bikes, so of course any comparison is invalid. We all know that
> professional types never ride hard, and just buy bikes that look good on
> the roof rack of their Explorer. :-).
>
> Gary Helfrich
> Arctos Machine
>
>
>
I think that the issue is not return rates. The issue is joint strength.
You are confused. Demographics and use patterns are just the complexity
you would avoid if you executed direct tests. We have done this. This is
what we do for a living. At our expense. We don't share the results of
this with the industry. What I suggested was that if you were interested
we would submit samples for the tests, samples that would not have any
information about manufacturing methods or materials, and we would then
see how they fared compared to the other joining methods. Having done the
work (in a somewhat different fashion than you guys were cooking up) I
am confident of the outcome.
Oh yeah, the studies we do on the competition's return rates shows that
more ti frames are crunched in accidents when they are still fastened
to the roof racks of the owner's Explorer. Steel bikes have a much
lower rate of return from this cause, and I am proud to announce that
we have only repaired one bike damaged in this way. You should conduct
more of this sort of research yourself! :>
(For anyone as out there as Gary, this was a stab at humor. I admit
that humor is not my strong suit, but then, I make my living designing
bikes, so its no big deal). :>
Adios (call me when you guys are ready for the samples!)
KB 408 427 2121
Jim Papadopoulos
> I figure the guys who were blue skying about doing the joint
> tests will go off on this one. You guys should work it into
> your plans in my humble opinion.
This is too brief for me to 'get' -- do you have the sense that we were
avoiding matters of reinforcement?
My view is that EVERY technique used in frame construction is fair game,
including lugged, fillet brazed, TIG/unreinforced, TIG/reinforced, etc.
But it's true, I was kind of hazy about this list.
Jim
John Olsen
>
>rich...@cruzio.com wrote:
>
(Do you have a humble opinion - period?)
>
>I think Keith overdid his Jobst medication this morning.
>
>professional types never ride hard, and just buy bikes that look good on
>the roof rack of their Explorer. :-).
>
>Gary Helfrich
>Arctos Machine
I drive a Rodeo, but I DO crash hard. Does that count?
Olsen
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
DaveO
> of bikes that are available at mid prices. Hell, Specialized is
> even giving one away to the lucky net freak just for sending them
> some Email (I did!).
>
Huh?
Sh*t. This law school thing can really cause one to lose touch.
Will someone be kind enough to supply some details?
TIA.
DaveO
db...@columbia.edu
Rick Horwitz
lap...@crl.com (Gary Helfrich) wrote:
>Heavy in the pocketbook. Brazing alloys with good
>mechanical properties for titanium require brazing in a
>vacuum. Kinda makes fillet a bit tough. There are
>brazing alloys that work (sort of) in air with
>incredibly toxic flux systems, but the joints are low
>in strength and ductility.
>
>Gary Helfrich
>Arctos Machine
Intersting! I have read some interesting stuff in my metallurgy
about Ti. However, no mention of brazing. It did mention
about a TIG process that required a pure Ox free environment
that required more than gas shielding. However, the book is a
bit outdated (1983). I have seen some frames from Merlin made
exclusively from titanium. Oh, how sweet! I have never seen
welds and fillets that looked so nice on any bike or machine.
Just to think that the welder is confined to wear a special
breathing apparatus while working in a tent! Is this really how
they do it or is there a modern science that has simplified
this process?
Respectfully,
Rick
**********************************************************
Rickey Horwitz Email: rhor...@hooked.net
Practical Voice: (408)578-5753
Innovations
"Where technology and innovation meet the road!"
***********************************************************
Scot Nicol
> Oh yeah, the studies we do on the competition's return rates shows that
> more ti frames are crunched in accidents when they are still fastened
> to the roof racks of the owner's Explorer. Steel bikes have a much
> lower rate of return from this cause, and I am proud to announce that
> we have only repaired one bike damaged in this way. You should conduct
> more of this sort of research yourself! :>
>
> (For anyone as out there as Gary, this was a stab at humor. I admit
> that humor is not my strong suit, but then, I make my living designing
> bikes, so its no big deal). :>
>
> Adios (call me when you guys are ready for the samples!)
>
> KB 408 427 2121
We're tied, Keith. We've repaired one steel bike that slammed a roof
It was a tandem. It was on a Toyota.
But we haven't seen any of the Ti bikes come back yet. I guess those
docs and lawyers have really tall garages.
Scot Nicol
Ibis
Michael Rowell
In the words of the *immoral* oops, *immortal* Mr. Rodney King...
"can't we all just get along." ;)
Now kiss and make up! Ha, ha, ha! Oh yea, its Friday alright.
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
Michael Rowell
Release Tools & Engineering --*-- *- ** @
Sybase ---**-- --* _/\;
77 S. Bedford St. ---*-* -- 7/(*)
Burlington, MA 01803 (*)
michael...@sybase.com www/\/\w www
(617)238-6221 PHONE ~~~~~ \wwwwwwww
(617)238-6327 FAX
'My mama always said, it's better to eat shit that not eat at all'
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
Rick Horwitz
rich...@cruzio.com () wrote:
>I don't think aluminum or ti will go away. Each
>material has properties that, when thoroughly
>understood, will make good bike frames. Steel is
>cheaper in the lowest grades, and it gets closer to the
>other materials as the design, material, and
>manufacturing process is refined to compete with the
>other materials. It would seem that aluminum is
>relatively cheap these days judging by the number of
>bikes that are available at mid prices. Hell,
>Specialized is even giving one away to the lucky net
>freak just for sending them some Email (I did!).
>
>And with respect to lugged frames:
>
bla,bla,bla...
>The same can be said for fillet brazed frames unless a
>substantial internal reinforcement is used. On a road
>bike a strong racer can get quite a few miles in on
>one before it breaks. They can crash it and get away
>with it as long as they don't run into anything in the
>process. But, for severe use, neither one has much
>going for it except heritage.
>
>Brazed fillets.....bla,bla.......
>Welding vs brazing is not an issue for off road bikes.
>The real question is how well a given reinforcement
>scheme works on a welded frame - all materials
>included - only matters in severe use, which happens
>quite a bit in the dirt.
>
I agree with Kieth on all the points he has made. As an
ex-roadie I knew that my columbus lugged frame was very fagile,
but tough where it counted. I don't think it could have lasted
a day or so off-road.
As for using aluminum and the fate of aluminum, I can only say
that aluminum can be easily shaped and machined making it an
excellent choice for any progressive design. I also think that
carbon (graphite/kevlar) have an excellent future in new bike
designs.
Well, thats my .02 worth.
Dog
Horwitz) wrote:
> rich...@cruzio.com () wrote:
> >
> >And with respect to lugged frames:
> >
> bla,bla,bla...
>
> >The same can be said for fillet brazed frames unless a
> >substantial internal reinforcement is used. On a road
> >bike a strong racer can get quite a few miles in on
> >one before it breaks. They can crash it and get away
> >with it as long as they don't run into anything in the
> >process. But, for severe use, neither one has much
> >going for it except heritage.
> >
> >Brazed fillets.....bla,bla.......
>
> >Welding vs brazing is not an issue for off road bikes.
> >The real question is how well a given reinforcement
> >scheme works on a welded frame - all materials
> >included - only matters in severe use, which happens
> >quite a bit in the dirt.
> >
> I agree with Kieth on all the points he has made. As an
> ex-roadie I knew that my columbus lugged frame was very fagile,
> but tough where it counted. I don't think it could have lasted
> a day or so off-road.
>
>
Dissention here:
KB's statement that welding vs. brazing isn't an issue for offroad must be
read bearing in mind that he doesn't make lugged frames. He has
catagoricaly denied the validity of a proven framebuilding technique. If
he doesn't want to build lugged frames, so be it. But, to say that a
lugged frame has no place in the offroad arena is inappropriate.
There are definate advantages to lugged framebuilding. If you are using
silver brazing, you can braze at low temperatures, minimizing distortion
in the tubing and allowing for the use of lighter, thinner-wall tubes.
Well-designed lugs can have all of the cable routing cast into them,
removing the need for brazed-on cable routing (and the distortion that
comes with). This allows for strong, lightweight, *steel* frames. You
can readily build a wide range of frame sizes. You can REPAIR a damaged
frame. This last point seems especially pertinent to the world of offroad
riding where crashing is a common event and where warranty of frames is a
common bone of contention between riders and manufacturers (ATTN: Jim with
broken Lava Dome).
Is there anything that makes a lugged frame inherently weaker than a
welded frame? Is a weld stronger than a lug? Does welding cause more
distortion at critical stress points than brazing?
These questions aren't new and will probably be debated (ad nauseum) for
as long as there are bikes and people passionate about building and riding
them. However, I can't help but think that saying that they don't apply
to the relatively new world of mass produced offroad bikes isn't somehow
marketing-driven.
Cynically yours,
David
Rick Horwitz
A million pardons!
I failed to emphasize that lugged frames have their place.
However, I don't see an application for a superlight weight
frame pressed into action for offroad. Obviously, KG is a TIG
dude. The finest custom built bikes are lugged, but the
technology for producing an offroad lugged frame hasn't evolved
or has it?
Rick Horwitz
Dog
Horwitz) wrote:
>kure...@lmis.loral.com (David Kurensky) wrote:
>
[snipped my brilliant diatribe :)]
>
> A million pardons!
graciously granted
>
> I failed to emphasize that lugged frames have their place.
> However, I don't see an application for a superlight weight
> frame pressed into action for offroad. Obviously, KG is a TIG
> dude. The finest custom built bikes are lugged, but the
> technology for producing an offroad lugged frame hasn't evolved
> or has it?
>
of comfy, responsive steel tubing that will stand up to the repeated
abuses associated with offroad riding. Seems ideal, no?
As for the technology, WARNING! PLUG TO FOLLOW:
Waterford Precision Cycles makes a supercool, lugged MTB.
Check out the April issue of Mountain Bike magazine for a review of the
Waterford 1400. In it, you'll find a discussion of Richard Schwinn's
company and their approach to lugged MTBs.
For info, contact Waterford at:
Waterford Precision Cycles
816 W. Bakke Ave.
Waterford, WI 53185
(414)534-4190
[(hopefully flame-retardant) disclaimer]
I'm not associated with the company in any way other than having good
friends who work there. This has given me a good look at the passion they
feel for their product and our sport and given me the opportunity to ride
one of their fine steeds.
[disclaimer off]
Peace,
David
BTW, there is an interesting article on this subject from Cycling Science
magazine (sorry, don't know what issue) entitled _Comparative Fatigue
Resistance of Methods Used in Joining Bicycle Frame Tubes_ by T.J.
Sutherland et al. In it, they discuss their experiment in which they
compared the failure rate of a lugged, fillet brazed, and TIG welded
join. Their results indicate that brazed lug joins are the most resistant
to fatigue failure followed by brazed fillet and then by butt weld (TIG).
This is by no means a definitive defense of one method over another.
However, their results are interesting and lend credence to the lugged
approach.
BTBTW, none of this is meant to detract from the fact that Bontrager
cycles are real sweet! KB, if you're reading, good luck with the new
venture, seems like a good deal for all interested parties. But that's
another subject, isn't it?
Later
Keith Bontrager
>
>
>
> >KB's statement that welding vs. brazing isn't an issue
> >for offroad must be read bearing in mind that he
> >doesn't make lugged frames. He has catagoricaly
> >denied the validity of a proven framebuilding
> >technique. If he doesn't want to build lugged frames,
> >so be it. But, to say that a lugged frame has no
> >place in the offroad arena is inappropriate.
>
I stand by my original statement, though I think you
missed the point. If you want a durable frame, there
is no issue. Lugs will not work, nor will simple
TIG welding. Reinforcements using TIG welding are
the solution (or perhaps bonding, but I didn't adddress
that).
>
> >There are definate advantages to lugged framebuilding.
> >If you are using silver brazing, you can braze at low
> >temperatures, minimizing distortion in the tubing and
> >allowing for the use of lighter, thinner-wall tubes.
As long as you are wlling to accept the loss of strength.
Minimizing distortion or minimizing the affects of brazing
alloy diffusion into the base metal with silver brazing
does not conserve the strength of a heat treated or cold
worked steel tube. This is simple and indisputable. You
can make nice looking frames that have very little strength,
though they are perfectly ridable.
> >Well-designed lugs can have all of the cable routing
> >cast into them, removing the need for brazed-on cable
> >routing (and the distortion that comes with). This
> >allows for strong, lightweight, *steel* frames. You
> >can readily build a wide range of frame sizes. You can
> >REPAIR a damaged frame. This last point seems
> >especially pertinent to the world of offroad riding
> >where crashing is a common event and where warranty of
> >frames is a common bone of contention between riders
> >and manufacturers (ATTN: Jim with broken Lava Dome).
>
Distortion and mechanical strength, both impact and yield
strength, are not related (unless the builder is using
very primitive techniques). Cable guides and other "braze
on" bits can be attached without compromising the strength
of the final structure if the operation is planned well.
Welded frames can be readily
repaired with very little compromise in the final strength
of the structure and at a cost similar to the repair of a
brazed frame.
>
> >Is there anything that makes a lugged frame inherently
> >weaker than a welded frame? Is a weld stronger than a
> >lug? Does welding cause more distortion at critical
> >stress points than brazing?
> >
The management of the HAZ around a welded joint with reinforcements
makes the joining method inherently stronger (if it is properly
executed) than any other joining method involving high heat. I
do not think that there is any reason to think that this
is wrong from engineering arguments. Do you?
> >These questions aren't new and will probably be debated
> >(ad nauseum) for as long as there are bikes and people
> >passionate about building and riding them. However, I
> >can't help but think that saying that they don't apply
> >to the relatively new world of mass produced offroad
> >bikes isn't somehow marketing-driven.
> >
> >Cynically yours,
> >David
>
There
is no debate about this among those who have studied the
topic. I too am nauseated when I see them come back
around. And, your thoughts about most of it being hype
and marketing jive are on the right track. The problem
for you and others at thge wrong end of marketing is that
you have to be able to understand the arguments in order to
filter out the jive from the facts. In this case, the
argument I am making is based on straightforward engineering
principles and can be verifed. If you put in a little
effort you can sort out the debate once and for all, for
yourself at least. .
KB
>
>
>
>
Steven L. Sheffield
>In article <3k58to$j...@worm.hooked.net>, rhor...@worm.hooked.net (Rick Horwitz) writes:
>>
>> >KB's statement that welding vs. brazing isn't an issue
>> >for offroad must be read bearing in mind that he
>> >doesn't make lugged frames. He has catagoricaly
>> >denied the validity of a proven framebuilding
>> >technique. If he doesn't want to build lugged frames,
>> >so be it. But, to say that a lugged frame has no
>> >place in the offroad arena is inappropriate.
>I stand by my original statement, though I think you
>missed the point. If you want a durable frame, there
>is no issue. Lugs will not work, nor will simple
>TIG welding. Reinforcements using TIG welding are
>the solution (or perhaps bonding, but I didn't adddress
>that).
So a Bridgestone MB-1 has no place off-road then? The MB-1 is a
poorly designed frame?
What about the XO-1? I wonder what John Stamstad has to say about
these frames ...
--
| Steven L. Sheffield (BOB #1765/IBOB #3) Disclaimer? What's that? |
| E-mail: biki...@best.com / biki...@crl.com / biki...@aol.com |
+-------- WWW: http://www.best.com/~bikiebob/cycling.html --------+
| Voice: (415) 296-9893 / Fax: (415) 597-9849 / Ride yer bike! |