Most flexible steel seatstays and fork blades?

[satanas]

In June I did a frame building course in the UK, and now have a few projects in the pipeline. One of these is a rando frame, which I'd like to be as comfortable (and flexible) as possible.

It seems like small diameter stays tend to have thicker walls than some of the larger ones, so I'm wondering if the two things will cancel out. I have thinwall S3 stays at home, but the Columbus SL double taper ones are smaller - will they flex more? Or is there a better option? I've noticed that Rob English uses very small seatstays, but no idea what tubing it is, presumably straight gauge CrMo of some sort.

And I've also been wondering if there are any currently available fork blades likely to flex more than Columbus SL. It'd be useful if they fitted a 27x20 crown if so, but not essential.

Thanks,
Stephen

[Alex Wetmore]

Around 5 years ago Rob English was using 3/8" straight gauge tubing to make those pencil thin stays. I think it is 0.035", but I might have that detail wrong. I chatted with him about it at one of the Portland-area shows.

It's a triangle, so I wouldn't look to get a lot of suspension out of the rear of the bike. I'd pay attention to good fork blades instead. Hahn Rossman got a nice set of Columbus blades drawn that are much longer than normal so that you can tailor them to the ride. Framebuilders Supply sells them in the US:
http://framebuildersupply.com/collections/fork-blades/products/columbus-cromor-rando-fork-blades-28x19-oval-9-wall-length-450

alex
________________________________________
From: frameb...@googlegroups.com <frameb...@googlegroups.com> on behalf of satanas <nsc.e...@gmail.com>
Sent: Sunday, September 6, 2015 1:06 PM
To: Framebuilders
Subject: [Frame] Most flexible steel seatstays and fork blades?

--
--
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.

[Johan Larsson]

The Kaisei Toei Special have the thinnest and longest lower section fork blades I've seen, and I would assume they are more flexible than even the long version of the Columbus SL. (Correct me if I'm wrong.) They have a different shape than "normal" blades though, it's not 27x20 but more elongated.

Johan Larsson,
Sweden

[Stephen Poole]

Thanks Alex (and Hahn)!

If I use those fork blades and trim only from the top there should be significantly more flex low down, which is what I'm after. :-)

Re seat stays: There might not be a lot of suspension there, but different stays can have noticeable effects. I have two Tony Oliver frames with effectively the same tubing (531 SL and 531 Pro), apart from the seatstays. The first is an audax frame, and has longer chainstays and wheelbase but 1/2" x 0.9mm seatstays, which Tony wanted to use due to brazed-on brakes. The other is a road frame, with a shorter top tube, etc, but with Reynolds double taper seatstays, with a wall of 0.5/0.55mm (different places show different specs). The road frame is more comfortable over bumps, even with 23 versus 28mm tyres, on the same wheels.

This year I took a Soma GR overseas and neither the standard fork blades nor the seat stays appear to flex to any degree. The top tube twists significantly though, especially with front panniers.

Frame flex might not be as big a factor as low pressure tyres, but IMHO the GR with 42mm tyres at 35-40 psi was no more comfortable most of the time than my 1991 Specialized CF frame with 23mm tyres at 100psi. And then there's the handling <shudder>.

Later,

Stephen

[CjellMoney]

If you are considering tubes themselves as flexible or not, diameter is the dominant parameter. Thickness plays a distant second. 

I 100% agree with Alex, if you are looking to add 'compliance' the rear triangle is not your best area to go about it. Litespeed used to add a moderate concave curve to their stays but it's my hunch that it was more marketable than it was compliant. I too have chatted with Rob about his stays, (I would guess it's a conversation that he's had once or twice), and he is of the same camp, unless it's buckling your ss are simply not moving. Bend them a bit, make them thin, and make them of ti, maybe a bit more but true compliance (measurable defection) comes from beam bending like a small diameter thicker seatpost, or fork blades. How heavily you plan to load your bike is you biggest consideration on where to build in your squish. Watch a VERY flexy fork, I had a Alan (lugged ALuminum ANodised) road bike a while back and that fork flexed like crazy, brake shudder was not confidence inspiring. Not a fun bike.

All that being said, check out Sean Burn's stuff at Oddity Cycles, he makes some bikes with trippy stays that miss the seat cluster all together and connect mid downtube. I have ridden these bikes and there is easily perceptible movement at the rear axle. 

If you want a comfy bike I would make clearance for big supple tires, just me. It's fun to push it a bit when you build your own stuff, please do so and report back on your findings.

Cjell

[Mark Bulgier]

CjellMoney wrote:

If you are considering tubes themselves as flexible or not, diameter is the dominant parameter. Thickness plays a distant second. 

I disagree, if you're talking about straight seatstays in compression.  For seatstays of equal length and same material (same modulus of inertia), their stiffness depends only on their cross-sectional area – essentially, their weight, given same material (thus density).  Heavier seatstays will be stiffer in compression regardless of diameter.  This is for compression within the bounds of stresses where the stays remain in a straight line, a classic column.  Long thin columns (such as a wire) will not remain straight under even low compressive stress, but I think normal seatstays pretty much do remain straight within the range of normal riding stresses.

Now, this is not very important since even light seatstays probably don't have much strain (shortening) in compression.  Not enough to feel anyway.  And they probably do deviate from a straight line a little, to where the diameter does matter.  But I still think those effects are minor.  I'd love to see some measurements to confirm or deny my suspicions, but I don't know of any.

Now curved (raked) seatstays might be a different story -- probably are. They should allow the dropout to move closer to the seat cluster when hitting bumps, for a small amount of suspension travel.  Allow too much though, and something will crack in fatigue, unless you add pivots.  But I suspect there's hope for a curved seatstay that adds a tiny amount of comfort, while not shortening the fatigue endurance of the frame too much.

I've never built any frames like that though, even though the concept has been around almost as long as there have been bicycles, because I think the comfort effect is bound to be so small compared to what you can get from saddles and tires.  Again I welcome any empirical evidence.  Maybe someone has done such measurements, but I suspect the results are most likely proprietary, not published.  And even if a company that makes curved seatstays published numbers to back up their design, I'd be suspicious of data being cooked to sell more units.

Mark Bulgier
Seattle

[M-gineering]

On 9/8/2015 9:50 AM, Mark Bulgier wrote:

>
> I've never built any frames like that though, even though the concept
> has been around almost as long as there have been bicycles, because I
> think the comfort effect is bound to be so small compared to what you
> can get from saddles and tires. Again I welcome any empirical
> evidence. Maybe someone has done such measurements, but I suspect the
> results are most likely proprietary, not published. And even if a
> company that makes curved seatstays published numbers to back up their
> design, I'd be suspicious of data being cooked to sell more units.

If you can compress a bent seatstay, you can also extend it. So it
wouldn't take much to seriously affect the alignment of the frame (with
one DO going up and the other down and twisting the rearwheel) and get
lots of rearsteering if you enter a curve. So I guess it either doesn't
much for comfort or screws up your handling

--
mvg

Marten Gerritsen
Kiel Windeweer
Netherlands

[Truls Erik Johnsen]

Compression (creating "vertical compliance") in seatstays under standard load (IE rider only) AND given that they are straight and set in a standard rear triangle with a rear wheel fixed in both dropouts, are negligible (at the best!). Do a search for bicycle frame fea and you'll see some calculations/imaging. Or just do a mental excersise of trying to compress a straight 12mm or 14mm tube linearly without bending it. No. Not possible. Regardless of wall thickness. And IF you where able to do it (by brute strength), it would bend and collapse. And as we all know, that seldom (luckily) happens on a bicycle. But then again, there could be other reasons for choosing different OD's, wall thicknesses and alloys for stays. But to gain "vertical compliance"? I don't think so..

Truls

[Hahn Rossman]

Stephan
I'm really curious about your GR and how it was set up. Can you post a picture of how it was at PBP?  Ideally it would have you on it riding,but also just a side view.
I talked a lot with Victor at cycles alex singer about bike fit and the differences between our bikes before the event. I'm probably going to do some more testing before I decide if what we were discussing makes any difference other than aesthetic. 
I also think that re: seat stay compliance the seat post plays a bigger part in comfort or lack of it and is usually overlooked. Can you switch posts/saddles in your Tony frames?
For the record TT has some nice double taper seat stays that are thin and small diameter. With the compass brakes (brazed on) I haven't had any problems, but the bosses are so close to the bridge compared to cantilever bikes.

[Jim G]

On Monday, September 7, 2015 at 10:24:02 PM UTC-7, satanas wrote:

This year I took a Soma GR overseas and neither the standard fork blades nor the seat stays appear to flex to any degree. The top tube twists significantly though, especially with front panniers.

Interested to hear more about this -- it sounds a lot like a characteristic of my old Kogswell 700C P/R frame, which shimmied/speed-wobbled like a mofo.  While the frame felt lively, at times it felt like the front and rear wheels were headed in opposite directions.

Another random comment on seatstays:  I just built a rando frame with 16mm seat stays, and built it up with V-brakes (I had them on hand leftover from an old cross bike).  If I squeeze the rear brake hard, I can visibly see the stays bow outward.  I thought 16mm stays would be stout enough,  but now I'm wondering if I should've used 19mm.

-Jim G

[Alex Wetmore]

From: frameb...@googlegroups.com <frameb...@googlegroups.com> on behalf of Jim G <yoj...@gmail.com>

> Another random comment on seatstays:  I just built a rando frame with 16mm seat stays, and built it up with 

> V-brakes (I had them on hand leftover from an old cross bike).  If I squeeze the rear brake hard, I can visibly 

> see the stays bow outward.  I thought 16mm stays would be stout enough,  but now I'm wondering if I 

> should've used 19mm.

I built two nearly identical touring frames (using canti studs) where the only real tube difference is in the seatstays.  The newer one has double tapered stays from True Temper (HOXPLATSS3) that are 19mm at the center and they provide much more stiffness to the brakes.  The older uses the basic Verus 16mm stays (VERSS1).  The double taper stays don't look quite as nice, but I'd use them again for a bike with cantis.

For any bike with limited tire clearance (near 40mm and fenders) I'd just use centerpulls and then you can use the skinniest seatstays you can find.  I built my touring bikes to fit 55mm tires and really need cantilevers to fit the tires.

alex

[Johan Larsson]

I think the comment on diameter vs. wall thickness was regarding flexing/bending tubes in general, and fork blades in this case.

Anyone that doesn't think Kaisei has the most flexible fork blades? They are thinner a longer way up than any other blades I've seen.

Johan Larsson,
Sweden

[Jason Henkle]

Just a data point to add to the seat stay conversation. I do believe Dave Kirk has stated that his Terraplane seat stays afford a mm or so of movement. Pretty sure a write-up of his can be found on his blog or in his "smoked out" thread on Vsalon.

Jason

--

[satanas]

Thanks everyone! I've been offline for a few days as I got home after Europe, then was ill for a week. :-(

I've tried to address people's comments below.

Cjell: I owned ALANs at various times in the 1980s and really liked them, even the most flexible ones. Yes, the forks did flex fore and aft noticeably, but this did not have any adverse effects on handling; the bikes tracked dead straight IME. Rumor had it that ALAN cut the dropout slots after the forks were bonded together. Whatever the reason, the frames I rode were all very directionally stable, way more so than steel frames of the time. Now, I know that any visible flex is anathema to some people, but in no way did I find those frames or forks a problem. In fact, what I'd like to do is try to recreate the same ride in steel, but it may not be possible.

Had a quick look at pics of Oddity Cycles, and to me the stays look sort of like mixte midstays, but connected partway along the down tube, or to a more normal sized top tube, rather than having full length twin lats. Small diameter, non-triangulated seatstays are becoming more common, for instance on current Giant Defy frames, and ISTR an article in Bicycle Guide about a Tango (tandem builder) single bike which used similarly low-angle seatstays in an attempt to increase ride comfort; BG thought it worked.

I agree that supple tyres can only help, but at least in the case of the GR, they are not enough by themselves. YMMV; the GR is the only bike I've ever owned or ridden for any length of time that I've actively hated. Unfortunately by the time I realised this I was overseas, and buying a new frame and wheelset wasn't a viable option.

Mark Bulgier said: I disagree, if you're talking about straight seatstays in compression.  For seatstays of equal length and same material (same modulus of inertia), their stiffness depends only on their cross-sectional area – essentially, their weight, given same material (thus density).  Heavier seatstays will be stiffer in compression regardless of diameter.  This is for compression within the bounds of stresses where the stays remain in a straight line, a classic column.  Long thin columns (such as a wire) will not remain straight under even low compressive stress, but I think normal seatstays pretty much do remain straight within the range of normal riding stresses.

So Mark, are you saying that cross sectional area is all that counts, and section modulus is irrelevant for seatstays??? If so I may as well use the S3 stays I have now, as they're very light. As to curved stays, I agree that those selling them have reasons to slant any data. They may still do something though.

m-gineering: Seatstay flex causing handling problems: I doubt this is matters in reality, unless one is perhaps Mark Cavendish or similar. Some of recent carbon frames use stays that are very thin indeed, but I haven't heard any reports about dodgy handling. And I'm light and not terribly big so for me it's unlikely to be a problem anyway.

Hahnster: The 55cm GR wasn't (IMO) set up particularly strangely; the position was made to echo the Tony Oliver audax bike I've had since 1982. Now that might be a bit non-standard as I have longer than average femurs and arms, so the seat is right back on a very offset post, and the bike has a (nominally) 120mm 3T stem - actual horizontal length C-C is more like 112mm.The original fork was out of spec, with the actual measured head angle being 72 arather than 72.5 degrees, and the offset being 74mm not 69mm. These were measured with a digital angle finder and an Anvil fork jig respectively. The fork used for PBP had Columbus SL blades and 60mm offset (yielding 45mm trail instaed of 31mm), unfortunately not with a particularly tight bend as a suitable former was not available. Still, the bike handled better/mpore predictably than stock with every load condition: barbag only (GB22 plus Nitto rack + Gillles Berthoud decaleur modified to reduce the forward reach), barbag plus Revelate saddlebag, these plus 28 litre front panniers on a Tubus Tara rack. It *still* wanted to shimmy occasionally without the front panniers, and did so on a twisty wet descent on the last night of PBP; I am not a fan of low trail. I'd ideally change everything about the GR(!), including: seat angle 71-ish not 73.5, BB ~8mm lower, chainstays 430-440mm not 415, more trail (less offset), shorter seat tube and longer post for more flex, longer top tube (more reach and shorter stem), thinner wall tubing everywhere but with a 28.6 top tube to prevent the head tube twisting with a load (I like BB flex but despise instability). Oh yes, and no canti brakes! (I'm also not convinced that 700c wouldn't be better for PBP, or faster at least.)

Re tube flex from brazed on brakes (Jim G): whilst this can be annoying, it's not the end of the world and is usually easily fixed by using a brake booster. I'd rather do that than have a super rough ride. My old MTB with Ritchey Logic forks felt much better with a booster plate, and in fact I had to change to Shimano cantis from DC 987s as the latter couldn't be made not to squeal, due apparently to the way the pads were attached to the arms.

Later,

Stephen

[Thomas]

Stephen...
My unFolding bike has (essentially) no seat-stays
www.kokoPedli.com (embryo web-site with pictures)
and the only time I've had any problem was when I put a 1300 watt
electric motor through the drivechain it pulled the rear-end sideways a
tiny bit everytime I accelerated suddenly, and the chainring sometimes
bit the chainstay a tiny bit...
The seat-stays are aircraftspruce streamline 4130 cro-mo .035" with
a listed "equivalent round tension" of a 7/8 tube and an "equivalent
tubing compression" of 7/16th (Also, with small wheels, they are
about 80% as long as on most standard bikes.)
I have a firm elastomeer between the seat-tube and the chain-stays
a couple inches up/out from the B.B.


Also, if memory serves me correctly, a bike called, I believe, Merlin
was available years ago with a small elastomeer between the seat-cluster
and the seat-stays at the very top ... with NO pivot at the
chain-stay - B.B. joint... I wonder whatever became of those bikes
and that design ??


Thomas Seaman: noMadic

> Had a quick look at pics of Oddity Cycles, and to me the stays look
> sort of like mixte midstays, but connected partway along the down
> tube, or to a more normal sized top tube, rather than having full
> length twin lats. Small diameter, non-triangulated seatstays are
> becoming more common, for instance on current Giant Defy frames, and
> ISTR an article in Bicycle Guide about a Tango (tandem builder) single
> bike which used similarly low-angle seatstays in an attempt to
> increase ride comfort; BG thought it worked.
>

> So Mark, are you saying that cross sectional area is all that counts,
> and section modulus is irrelevant for seatstays??? If so I may as well
> use the S3 stays I have now, as they're very light. As to curved
> stays, I agree that those selling them have reasons to slant any data.
> They may still do something though.
>
> m-gineering: Seatstay flex causing handling problems: I doubt this is
> matters in reality, unless one is perhaps Mark Cavendish or similar.
> Some of recent carbon frames use stays that are very thin indeed, but
> I haven't heard any reports about dodgy handling. And I'm light and
> not terribly big so for me it's unlikely to be a problem anyway.
>

> Later,
> Stephen

[Mark Bulgier]

satanas wrote:

> So Mark, are you saying that cross sectional area is all that counts, and section modulus is irrelevant for seatstays???

I meant only in that one dimension, compression.  Which is minuscule, not worth talking about in a bike frame rear triangle. 
Even for compression, it only holds within the bounds of straight stays that act as columns.  Once they get too skinny compared to their length to stay straight when compressed, and they bend sideways (buckle), then we're not talking about compression anymore.  Compression in a long wire or rope is like the sound of one hand clapping.  Diameter affects how much compressive stress the column can take without buckling, but I'm pretty sure all normal seatstays are well within the safe range, nowhere near the point of buckling.

There are real world bending loads on seatstays where the diameter matters, for instance from braking (especially cantilevers) and from rear racks.  On a track bike (no brakes or racks) I think the seatstays could be made very light indeed.  But then, weight being so unimportant on track bikes, we might as well make them able to withstand crashing, or being thrown in the trunk of a car.

Due to the way the seatpost sticks out, the seat cluster has a bending load on it that is shared by the top tube, seat tube and seatstays.  But the seatstays being the smallest in diameter, their contribution to resisting the bending from the seatpost is probably minuscule.  I don't make the seatstays larger diameter on compact or sloping-tt frames.  In fact, resisting this bending from the seatpost is not a high priority to me, I like it to flex there.  Don't overdo it; take some care to keep it within safe bounds, and eliminate stress risers that will become fatigue cracks.  But that is a place where some noticeable amount of attenuation of bumps can happen.  More in the seatpost itself, but also in the bending of the frame tubes near the seat cluster.  I think -- I don't have strain measurements or FEA to back that up, just thinking about it, and some literal seat-of-the-pants feeling.

satanas wrote:

> m-gineering: Seatstay flex causing handling problems: I doubt this is matters in reality,

At the risk of putting words in Marten's mouth, I think he meant that only if you had enough seatstay flex to add appreciable amount of suspension or comfort.  I think Marten agrees that you can't really get much suspension from the seatstays, or at least it isn't done on currently offered frames. And if it were done, it could cause problems.

Mark Bulgier
Seattle

[Gary Young]

I'm curious what people here think of this recent study, which purports to give guidance to framebuilders on choosing steel tubes that are most vertically compliant and most laterally stiff:

http://www.sciencedirect.com/science/article/pii/S1877705815014216

(link to free pdf download at the top of the page).

I'll admit that I'm prejudiced against it -- mostly on the basis of discussions on rec.bicycle.tech suggesting that most differences in vertical compliance are dwarfed by the compliance of tires, saddles and handlebar tape. But this study doesn't do much to overturn my prejudices. If I'm reading the chart on page 36 right, there was only 0.02 mm difference in displacement between the least and most compliant tubes.

[Matt Wilkinson]

Thanks for posting this study Gary.  I'm not an engineer, so I might find it harder than some to unravel those findings.  Could soneone please explain, were the wall thicknesses in the fea test comparable ( I understand that butt lengths and transitions might vary). I am surprised at the difference in stiffnesses for different tubes.  I always thought that steel tubes had mostly similar stiffness traits due to the Young's Modulus not changing much.  Therefore, is this paper not really more a tube profile study than a study of the mechanical properties of a selection of tubesets?

Like I said, I'm not an engineer.

Matt

London, UK

Sent from Samsung Mobile

--

[M-gineering]

I don't think much of their literature list

Given the fact that all steel has the same Youngs modulus, and that
there are a great number of tube shapes and dimensions sold as 853,
spirit etc. all talk as saying one brand was stiffer than some other is
pretty meaningless. I don't see an appendix where they identify which
tubes they used.