953 fillet brazed frame strength....?

[Randall Shimizu]

I noticed some people had built some fillet brazed 953 frames....?  Any idea on how strong the frames are...?

[Andy Newlands]

Hi Randall,

I work for Reynolds Technology Ltd, the company which manufactures 953 stainless steel cycle tubing, and passed your inquiry along to Reynolds Managing Director, Mr. Keith Noronha.  Here is Keith's response:

"Reynolds do not have an EN test frame data for fillet brazed frames (by far, most are made as TIG welded frames which can pass).

 

In my opinion, the problem with EN testing fillet brazed frames is that the braze material build-up  around the joint and surface preparation will be variables that affect the final result. Several have used Fillet Brazage Pro on 953 (lower melting point, reasonable strength) over the last 4-5 years and this seems to work well. As always with stainless and heat-treated materials, avoid over-heating the 953 but we know the annealed strength of 953 is higher than heat-treated Cr-Mo!"

Keith also says, “do a test joint first and check”.

Cheers,

Andy N.

www.torchandfile.com

www.strawberrybicycle.com

www.oregonframebuilders.org

ps.  If you like I am glad to mail you the 2015 Reynolds Price Lest.  Please email your physical mailing address.

On Feb 3, 2015, at 1:02 AM, Randall Shimizu wrote:

I noticed some people had built some fillet brazed 953 frames....?  Any idea on how strong the frames are...?

--
--
You received this message because you are subscribed to the Google
Groups "Framebuilders" group.
 
Searchable archives for this group can be found at http://groups.google.com/group/framebuilders (recent content) and http://search.bikelist.org (older content).
 
To post to this group, send email to frameb...@googlegroups.com
To unsubscribe from this group, send email to
framebuilder...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/framebuilders?hl=en

---
You received this message because you are subscribed to the Google Groups "Framebuilders" group.
To unsubscribe from this group and stop receiving emails from it, send an email to framebuilder...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[M-gineering]

On 2/3/2015 5:41 PM, Andy Newlands wrote:
> Hi Randall,
> I work for Reynolds Technology Ltd, the company which manufactures 953
> stainless steel cycle tubing, and passed your inquiry along to Reynolds
> Managing Director, Mr. Keith Noronha. Here is Keith's response:
>

> "Reynolds do not have an EN test frame data for fillet brazed frames.....

Would a standard sized 531 frame pass an EN test?
How many 531 frames have been used succesfully through the ages?

;)

--
mvg

Marten Gerritsen
Kiel Windeweer
Netherlands

[Megan Dean]

I have fillet brazed a handful of KVA frames and not had any issues. 

I set up two different friends on their bikes on 2011 before I offered it up to outside customers. Both are happy with the ride and have put them through the ringer for me. 

On Feb 3, 2015, at 1:02 AM, Randall Shimizu <randall....@gmail.com> wrote:

I noticed some people had built some fillet brazed 953 frames....?  Any idea on how strong the frames are...?

[Randall Shimizu]

The problem I see with filet brazing is that you relying on the brazing material. Both silver solder or brass or soft materials.

[Mark Bulgier]

Randall Shimizu wrote:

Ø  The problem I see with filet brazing is that you relying on the brazing material. Both silver solder or brass or soft materials.

 

Randall,

 

When you say you “see” a problem, you mean that metaphorically, right?  I’m willing to bet you have never seen an actual problem with weak fillets with your actual eyes.  I was a framebuilder for over 20 years and I never saw one.  Anyone else here ever seen a problem with a weak fillet?

 

I think you are armchair theorizing based on near zero experience, yet you have no qualms about disparaging the fillet brazed frames made by others, in a public forum.

 

For the record, fillet-brazed joints (also called “bronze welded”) are almost always stronger than the tubing, to where if you stress it to failure, it will be the tube that fails, not the fillet.  (I had to add the “almost” there to be scientifically valid, even though I have never seen or heard of a single counter-example, myself.)

 

And fillet is spelled with two ells.  Filets are braised, not brazed.  I don’t usually call out typos, but I’m feeling grumpy at the moment…

 

Mark Bulgier

Seattle

 

 

[M-gineering]

On 2/6/2015 9:23 AM, Mark Bulgier wrote:
> Randall Shimizu wrote:
>
> ØThe problem I see with filet brazing is that you relying on the brazing

Yeah that statement was a bit fishy ;).

Although I'm not a great fan of fat fillets in low melting silver, I
think there is too much movement going on in the joint (due to thermal
contraction at the stage where the almost solidified fillet has very low
strenght) which upsets the bond

[Randall Shimizu]

I have built frames before, but have not built a fillet brazed frame.  I would like to see some studies if any that have been done.

On Tuesday, February 3, 2015 at 1:02:11 AM UTC-8, Randall Shimizu wrote:

[Andrew R Stewart]

Randall- Well then I have a suggestion. Why don’t you fund such a study. There are a number of very qualified builders and testing labs who would be quite willing to be paid for their work in brazing together and then running the appropriate tests. I don’t know the cost to do all this but I can guess it’s substantial.

 

In real life there have been, at least, less formal comparisons between lugged, tigged and filleted frames before. The one that Bicycling mag did about 25 years ago comes to mind. They both compared three frames, each made with one of the joining methods, as well as sent out samples to be examined by some one who knew their stuff. IIRC the three frames rode so similarly that no conclusion was had. The test samples each failed in their own way, the location of the HAZ determined where along the tube the buckle occurred, but each actual joint didn’t fail.

 

I have done basic destruction attempts in my own shop and find that a properly filleted joint won’t see a failure at the fillet. UNLESS the fillet is too small (like only sweated and not built up), the filler was cooked (so it boiled off some of it’s components) or the miter was really (really) bad a filleted joint will out last the tube during the yanking around that I did. I have had silvered filleted (56% not Fillet Pro or Nickel Silver) break at the joint but then 56% isn’t meant for this joining method so I don’t count that as a failure as such

 

Now none of this involved Stainless Steel frame tubes but some did use SS plate.

 

I, too, would like to read a well designed and done testing. I look forward to your results. Andy. 

 

Andrew R Stewart
Rochester, NY USA

[Charles Hobbs]

Mr. Bulgiers' response is just the opening I need to ask a very dumb question that has been eating at me for a long time.  Is "fillet" pronounced like you are at the gas station; fill it up, the steak house; filet mignon, or simply fill let.

Dumb question which reveals that everything I know on the topic I have read, yet another problem of the self taught, no mentor noob.

Charles Hobbs

From the middle of nowhere where the only welders work on oilfield rigs

[M-gineering]

On 2/6/2015 6:39 PM, Charles Hobbs wrote:
> Mr. Bulgiers' response is just the opening I need to ask a very dumb
> question that has been eating at me for a long time. Is "fillet"

> pronounced like you are at the gas station; /fill it/
\
nothing faux french about it, no need to braze em with your pinky in the
air ;), /fill it/ it is

[M-gineering]

On 2/6/2015 3:57 PM, Andrew R Stewart wrote:
> Randall- Well then I have a suggestion. Why don’t you fund such a study.
> There are a number of very qualified builders and testing labs who would
> be quite willing to be paid for their work in brazing together and then
> running the appropriate tests. I don’t know the cost to do all this but
> I can guess it’s substantial.

deciding about what constitutes a realistic test which mimics real world
conditions will be the tough one. Meanwhile lots of customers pay me to
'test' fillet brazed frames by cycling up and down to China and so
forth, and so far they've been strong enough ;)

Fillet brazing is old hat, used for suspension components of racing
cars, motor cycle frames, subframes etc. No reason you couldn't fillet
braze 953, provided you find a combination of flux and rod which suits
you and the material

[VeryRedBike]

There has been a study... like M is saying, the 100's of thousands of bikes on the road made with this technique.  

If you want to know about the relative strengths of different construction techniques, folks will be glad to answer that kind of question... but jumping in claiming something is a problem on a REALLY tried and true technique (really, any of them... lugs, tig, fillet, etc) is a little like walking into an autobody and saying "I don't think sparkplugs really work"

You'll find folks react strongly to this.  I've worked at bike shops (selling, wrenching, managing) for nearly a decade (cue eyerolling from folks who have me beat by half a century ;-) and I can't count the number of times someone has "explained" to me how this part that's been functioning fine with minimum changes for 30 years doesn't really work, or that construction method is a deathtrap even though it's industry standard and failures are nearly unheard of... and it gets under your skin.

I'm not trying to pile on... I just wanted to point at a hot-button that some folks might not be aware of.  My advice: if folks with a lot of knowledge are doing something, and you think you see a big problem... ask why it works anyway.  Give them the benefit of the doubt... and years of experience.  You'll get piled on a lot less and you'll probably get a lot more interesting info.

[Jon Norstog]

Getting back to the original question, why use 953 for a mountain bike?  If you take advantage of the material's super-strength to build an ultra light frame, it will probably be a flexy POS.  And if you build for a reasonable combination of stiffness and compliance, you can use chrome moly.  It will cut easier, go together better, and you'll have a pile of money left over.

my 2 cents.

jn

"Thursday"

[Mark Bulgier]

Randall Shimizu wrote:

Ø  I have built frames before, but have not built a fillet brazed frame.  I would like to see some studies if any that have been done.

 

It’s a valid question. I don’t have any studies to point to, for several reasons:

1.       I’m not an academic researcher or scholar; I don’t know how to search for published studies.  Maybe someone else here can.

2.       Studies of these and other joining methods probably have been done in engineering schools, but not exactly about our usage (bicycle frames), making them perhaps less valuable than real-world experience.

3.       Bike companies have done lots of studies but the results are proprietary, not published publicly.

 

The bike magazine that made and tested three frames (I forget who that was) doesn’t really count as a scientific study because of a number of problems with the protocol.  Too small a sample size, no way to correct for bias (did the framebuilder or the test rider have a preference?), etc.

 

Along the same lines (anecdotal/unscientific studies with too small a sample size), I’ve done a few myself, like welding/brazing samples and breaking them in the vise with a cheaterbar. 

 

Or like when I turned my fillet-brazed 1933 Excelsior into a mountain bike by adding derailleur gears and better brakes, in about 1978.  The previous owner told me he rode it nearly every day in the ‘30s and ‘40s, and then less often in the’50s,  ‘60s and ‘70s. So it was not exactly a new bike when I started abusing it.   I raced it (informally; there was no NORBA back then) and won some.  I weighed almost 200 lb and was a decent amateur racer, so I definitely stress a bike more than most people. I competed in Observed Trials meets.  (Yes it was a very bad Trials bike, but it was all I had.)  I jumped it on the BMX jumps, did big dropoffs including once high enough to buckle both wheels, and crashed it hard enough to bend the fork back, which was not easy because it was a true truss fork*. 

 

* I just looked for “truss fork” in google images and found hundreds of pictures of fake truss forks, where the struts out in front are curved to look cute.  (Can a curved rope transmit tension?  Not at all until it straightens out.) Often they even leave out the compression member between the forward strut and the crown, making them 100% decorative, extra weight that does nothing.   To be a real truss, the small-diameter tubes out in front have to be straight line segments because they need to transmit tension, and there must be a compression strut between the crown and the tension strut.  My Excelsior fork, as I said, is a true truss, which makes it very strong against bending backward.  When I crashed it hard enough to bend the fork, it was the plate at the top of the headset, where the forward struts tie in, that snapped.  It was maybe 1/8” thick mild steel.   When it snapped, the forward strut couldn’t work in tension anymore.  The oval fork blade tubes and the steerer bent, but nothing happened to the small fillets on the frame.

 

Those old American bikes like the Excelsior used small diameter, thick-wall tubes, so they are not prone to buckling failure like a thinwall tube.  The small tube diameter makes for a smaller fillet “footprint”, and on my Excelsior the fillet size is rather small, smaller than most any modern example I can think of.  (Bike frame builders sometimes make overly-large fillets for aesthetic reasons, but I almost never see anyone flirting with the lower boundary of a too-small fillet.)  So this frame is nearly a worst-case scenario for fillet strength.

 

On my first fillet-brazed bike, which was only the second frame I ever made (in 1977), I made the mistake of thinking stronger filler would make a stronger frame, so I chose a “nickel silver”, probably Allstate 11.  It made a good frame, but it was harder to file than brass would have been, and my early attempts at fillet brazing did need some filing to look nice.  Since I wasn’t making tiny fillets, I believe there was no advantage to using the stronger filler.  Once the fillet strength exceeds the strength of the tubing, any more joint strength beyond that is highly theoretical.

 

I have heard of cracking or voids inside a fillet if a high silver content filler is used, something about the shrinkage during cooling causing problems while the strength of the material is still low.  I didn’t see that effect myself, when I tested joining Carpenter’s Aermet 100 Alloy (a very high-strength steel) in the 1990s.  We joined samples with silver fillets, brass fillets and TIG, and broke them in the vise, and they were impressively strong either way. But I ended up using silver fillets for the MTB race frame we made for our sponsored racer, a top local amateur who was looking for a pro ride at the time.  He won some races on that frame and never broke it that I know of, and I think I would have heard about it if he did.

 

I have heard that Reynolds gets the steel for 953 from Carpenter, and that 953 is in fact very similar to the older Aermet 100 Alloy.

 

I have tried Fillet Pro from Cycle Design (a high silver content filler) but not on a frame yet, just a pannier rack.  Other builders have made plenty of frames with it though.   I would have no qualms about using it on a 953 frame; I think it might be the perfect rod for that.  I would experiment with fillet size a bit, but I believe you’ll find a fairly small fillet (by normal bike frame standards) will be more than adequate, and there’s little advantage to going with big fillets.  My guess is that big fillets are probably the reason for shrinkage-related cracking during cooling

 

With trad 4130 CrMo frames fillet brazed with brass, too large a fillet is actually a problem, not for the joint strength per se, but for the strength of the tubing in the heat-affected zone (HAZ).  The time-at-temperature is longer when laying it down, and then the extra thermal mass in the big fillet cools too slowly, leaving the adjacent steel too annealed.  Small fillets can be laid down faster, resulting in a smaller HAZ, and the strength of the HAZ is higher too. 

 

I got out of the frame building biz around when air-hardening alloys were coming out so I don’t have much experience with those, but I think you still want to keep the heat input and the HAZ as small as possible.  So small fillets are still probably better, as long as they’re above the minimum size needed to exceed the strength of the tube.

 

Of course a TIG guy will chime in here and ask why not just TIG it, which results in the smallest HAZ.  I don’t know the answer but sometimes the builder or the customer just wants it to be brazed.  I’m not saying brazing is better than TIG, not gonna open that can o’ worms!

 

Mark Bulgier

Seattle

 

[Randall Shimizu]

Not sure where you are getting the hundreds of thousands of fillet brazed frames from. I used to work in bike shop and it fairly rare to see one.  I am not necessarily claiming there is a problem. It's simply that would like to learn more about the implications.

One probable advantage of Tig welding is that requires less material to weld a joint.

[Randall Shimizu]

Tubing diameter is the determining factor in terms of flex. But I do not see why wall 953 cannot be produced in different wall thicknesses to make it more suitable for a MTB frame.

[John Wilson]

Schwinn Town and Country Tandems, '45-'60s, Schwinn Sports Tourer, Super Sports, and Superior until '78. 
Best,

John Wilson

Greensburg, PA

USA

--
--
You received this message because you are subscribed to the Google
Groups "Framebuilders" group.
 
Searchable archives for this group can be found at http://groups.google.com/group/framebuilders (recent content) and http://search.bikelist.org (older content).
 
To post to this group, send email to frameb...@googlegroups.com
To unsubscribe from this group, send email to
framebuilder...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/framebuilders?hl=en

---
You received this message because you are subscribed to the Google Groups "Framebuilders" group.
To unsubscribe from this group and stop receiving emails from it, send an email to framebuilder...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--
© Copyrighted Material

[Andrew R Stewart]

Randall- Of course 953, or any other material, can be produced in any diameter/wall combo that’s possible (given current production limitations). But once again I go to the bottom line of cost. Who’s going to pay for the stuff. It’s tooling, the payroll to do the drawing and other work, the inevitable rejects that a new production will have and lastly who’s going buy the bikes. Given the market direction that mountain bikes have been following for the last 20 years making a “life time” frame isn’t what the riders are buying. They want the latest tech, they will ride that bike till it needs too much work to be seen as reasonable to do then move up to the then newest tech. Suspension evolves quickly, steerers are growing, braking tech changes, dropper posts?, soon electronic shifting will rule. So what tech do you design and build this 953 frame to? certainly not the one yet to have been thought of.

 

So again all this stuff takes is money. You could contract to buy and then become the world’s distributor of mountain bike focused 953. There would be a few who will beat a path to your door. But I suspect a very few. Andy.

 

Andrew R Stewart
Rochester, NY USA

[David Bohm]

Good one!  Also I would mention the 60k+ Bromptons made each year currently.

 

Dave Bohm

Bohemian Bicycles

 

From: frameb...@googlegroups.com [mailto:frameb...@googlegroups.com] On Behalf Of John Wilson
Sent: Saturday, February 07, 2015 7:29 AM
To: Randall Shimizu
Cc: frameb...@googlegroups.com
Subject: Re: [Frame] 953 fillet brazed frame strength....?

 

Schwinn Town and Country Tandems, '45-'60s, Schwinn Sports Tourer, Super Sports, and Superior until '78. 
Best,

John Wilson

Greensburg, PA

USA

On Saturday, February 7, 2015, Randall Shimizu <randall....@gmail.com> wrote:

--

--
You received this message because you are subscribed to the Google
Groups "Framebuilders" group.
 
Searchable archives for this group can be found at http://groups.google.com/group/framebuilders (recent content) and http://search.bikelist.org (older content).
 
To post to this group, send email to frameb...@googlegroups.com
To unsubscribe from this group, send email to
framebuilder...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/framebuilders?hl=en

---
You received this message because you are subscribed to the Google Groups "Framebuilders" group.
To unsubscribe from this group and stop receiving emails from it, send an email to framebuilder...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--
© Copyrighted Material

[Andrew R Stewart]

Also many thousands of bikes made before lugs were the production norm. An old boss of mine is a collector of, mostly pre WW1 bikes, and none of his bikes that I would service each winter used lugs. As already mentioned when so many examples of a design are used successfully for many decades says that this design has some real merit. Andy.

 

Andrew R Stewart
Rochester, NY USA

To post to this group, send email to javascript:_e(%7B%7D,'cvml','frameb...@googlegroups.com');

To unsubscribe from this group, send email to

javascript:_e(%7B%7D,'cvml','framebuilders%2Bunsu...@googlegroups.com');

For more options, visit this group at
http://groups.google.com/group/framebuilders?hl=en

---
You received this message because you are subscribed to the Google Groups "Framebuilders" group.

To unsubscribe from this group and stop receiving emails from it, send an email to javascript:_e(%7B%7D,'cvml','framebuilders%2Bunsu...@googlegroups.com');.

[Randall Shimizu]

Well Reynolds has said in the past that different 953 tubing sizes were available as I indicated before.

[Andy Newlands]

Hi Randall,

Yes, the various 953 offerings at the present time include:  1 - 25.4mm tube, 4 - 28.6, 8 - 31.75, 5 - 34.9, 1 - 36.4; various teardrop, oval, bi-oval; butted chainstays oval, cranked, ROR, etc; head tubes; 16 & 19 seatstays; fork blades; various sundries.  921 shows more variety including 38.1 double zone butted and DZB bent in both 34.9 and 38.1.  Hope this helps.  Will this allow you to design to the tech not yet to have been thought of?  Perhaps it is up to your imagination?  Have a good weekend.

Cheers,

Andy N.

www.strawberrybicycle.com

www.torchandfile.com

www.oregonframebuilders.org

ps.  Disclaimer:  Torch and File is a USA Reynolds distributor.