The importance of frame stiffness.

[waterlaz]

I'm trying to understand which parts of the frame one would want to make stiffer.
Particularly, which is more important: a stiff down tube or a stiff seat tube?
I've spent a lot of time googling on the subject, but most of the info out there is just marketing wank.

[mmoore]

you want it all to balance.

[Andrew R Stewart]

No engineer but here's my thoughts. I suspect the seat tube's strength is more important and the down tube's stiffness is more important.

This is from seeing more seat tubes with cracks over the years then down tubes have had. I believe that the seat tube sees more bending stress and the down tube sees more torsional stress.

But I'm welcome to correction and look forward to others with greater understanding give their views. I also agree with Matt's answer, in that using tubes with radically different characteristics in the same frame is a waste of something. Andy.

[steve garro]

Stiffness & strength are two different things


- Garro.

[jclay]

I'm trying to understand which parts of the frame one would want to make stiffer.
Particularly, which is more important: a stiff down tube or a stiff seat tube?
I've spent a lot of time googling on the subject, but most of the info out there is just marketing wank.

That's impossible to answer as posed. Firstly, stiffer relative to what reference frame, materials, diameters and wall thicknesses? Secondly, what is the intended use? Road, off-road, heavy/light rider? Thirdly, which do you subscribe to? The notion that stiffer is always faster or that some amount of flex, short of stiffest, is more efficient? That sort of information needs to be known before any educated guesses can be made.

[Mark Walberg]

.....which do you subscribe to? The notion that stiffer is always faster or that some amount of flex, short of stiffest, is more efficient? .....

Well, that leaves it wide open, doesn't it ???

[waterlaz]

That's impossible to answer as posed.

Sure it is.

Firstly, stiffer relative to what reference frame, materials, diameters and wall thicknesses?

If this helps consider a classic steel road frame.

Secondly, what is the intended use? Road, off-road, heavy/light rider?

If you have an answer for at least one of the mentioned intended uses I would be happy to hear it. I am more interested in road though

Thirdly, which do you subscribe to? The notion that stiffer is always faster or that some amount of flex, short of stiffest, is more efficient? That sort of information needs to be known before any educated guesses can be made.

I consider this part of my question. I don't know if stiffness is all that important, but for simplicity's sake let us suppose that stiffer is always better.

[waterlaz]

you want it all to balance.

I suspect that a classic road frame's seat tube and down tube have almost arbitrary diameters of 28.6mm. At least I'm sure that those dimensions did not come from any kind of stiffness analyses.

[waterlaz]

No engineer but here's my thoughts. I suspect the seat tube's strength is more important and the down tube's stiffness is more important.

This is from seeing more seat tubes with cracks over the years then down tubes have had. I believe that the seat tube sees more bending stress and the down tube sees more torsional stress.

But I'm welcome to correction and look forward to others with greater understanding give their views. I also agree with Matt's answer, in that using tubes with radically different characteristics in the same frame is a waste of something. Andy.

It is interesting that if there is a crack near the bottom bracket it is usually at the seat tube.

[jclay]

Sure it is.

If this helps consider a classic steel road frame.

If you have an answer for at least one of the mentioned intended uses I would be happy to hear it. I am more interested in road though

I consider this part of my question. I don't know if stiffness is all that important, but for simplicity's sake let us suppose that stiffer is always better.

If you assume stiffer is always better and a 969 road frame, normal old-school tubes, is your reference point then you'd want 10/7/10 tubes in conventional diameters. Or you could go with OS tubes of 969 wall thickness and get even more stiffness. I wouldn't want that for myself but it seems to me to be the logical conclusion of your parameters.

But I happen to think that some measure of flexibility is beneficial to propulsion and fatigue reduction. How flexible to go is difficult to define; I'd use Jan Heine's articles for guidance.

Is there some particular frame related problem you've had?

[fastupslowdown]

I had a an aluminum frame some years back a Trek 1200. I felt beat up when ever I rode it. Very stiff. Transmitted everything to my body. I ride longer and faster on my C-59 which is not known as a super-stiff bike. Also enjoy my Titanium Firefly. You don't want the bike to feel like a wet noodle, with stiffness more is not necessarily better despite the marketing shpiel which suggests otherwise. -Mike G

[jclay]

I suspect that a classic road frame's seat tube and down tube have almost arbitrary diameters of 28.6mm. At least I'm sure that those dimensions did not come from any kind of stiffness analyses.

I'd imagine the diameters started out that way and corresponded to standard industrial tubing/pipe sizes that were a reasonable mechanical fit. I doubt there was very much quantative analysis in the early days (though we might be surprised...it was the mechanical age and a lot of engineering analysis was applied to machinery) but based on snippets of reading and some email conversations I had with Olivier Csuka some years ago it's apparent that some builders and riders were investigating it qualitatively, hence the variety of wall thicknesses available. Klein was certainly thinking about it when they went to OS aluminum tubes in, what, the '70s? The industry reduced the TT OD for some reason; it would have been simpler to make the tre-tubi all the same size so some other line of thought was the driver; weight and flexibility? Strain energy was certainly a recognized reality in general industry and technical arenas.

Are you investigating frame flexibility/stiffness in the context of Jan Hein and other work vis-a-vis optimal characteristics as it relates to propulsion efficiency, rider fatigue and frame durability? You won't find concord on the subject here....but the conversation could get interesting. If you haven't already done so the search term "planing" will return some earlier conversations that may be of interest.

The little bit I build is generally in 969. I like the little bit of insurance against dings and rough use that it offers relative to thinner stuff and I don't miss (not powerful enough) any performance advantage that thinner walls might provide. Just for talking, if I was 6' and 200# I might use a 10/7/10 DT, still in conventional diameters.

[Paul Jacobs]

Butted steel tubes were invented by Reynolds in England in 1897. I imagine that they were used to working in inches and fractions of an inch rather than decimals of an inch. 28.6 mm = 1 1/8 inches, 25.4 mm = 1 inch and 31.75 mm = 1 1/4 inches.
The current Reynolds site has some useful information about strength and stiffness (see Steve Garro's post above). http://www.reynoldstechnology.biz/wp...ys_extract.pdf

For steel frames Tony Oliver's book, although published 26 years ago, contains a lot of information which is still useful to the steel frame designer and builder. You can still find used copies on ebay and elsewhere. Touring Bikes, A Practical Guide by Tony Oliver - RARE Sought after copy. (I have no connection with this ebay seller, I just attached the link for illustration purposes).

[Andrew R Stewart]

John's post an hour ago sums up my understanding of the beginning of "classic" tube diameters. But I will say that there are many examples of deviation from this formula a hundred, or so, years ago. I had the luck to work in a shop whose owners collected pre WWI bikes. WE, the staff, would overhaul a couple each winter. I was repeatedly impressed with what these bikes consisted of. A 1.75" wide sew up with steel cranks and inch pitch chain weighing 23 lbs having anatomical saddles and compact frame design of OS diameters. (The dual sprung Columbias with shaft drive and the tandem with the timing chain running down the center of the frame stand out too). In this interweb world it's easy to think that the search results that show on the first page is everything. Andy.

[Asteroid]

Somewhat off-topic - but since you wrote about Reynolds tube diameters - kindly let me know how we end up with differing seatpost sizes. Italian, French, British, etc. using 531 back in the day, you'd see anything from 25.0 to 27.4. This never made sense to me.

[e-RICHIE]

Somewhat off-topic - but since you wrote about Reynolds tube diameters - kindly let me know how we end up with differing seatpost sizes. Italian, French, British, etc. using 531 back in the day, you'd see anything from 25.0 to 27.4. This never made sense to me.

They are different diameters and also each came in a range of wall thicknesses.
Pretty simple really.
531 was a material, not an indicator of what spec the tube was.
PS Re names and anonymity: this please: http://www.velocipedesalon.com/forum...ons-43597.html

[Mark Kelly]

Brighton Graph

Data from:

"Parametric Finite Element Analysis of Steel Bicycle Frames: The Influence of Tube Selection on Frame Stiffness" Covill et al

Procedia Engineering Volume 112, 2015, Pages 34-39

[jclay]

Data from:

"Parametric Finite Element Analysis of Steel Bicycle Frames: The Influence of Tube Selection on Frame Stiffness" Covill et al

Procedia Engineering Volume 112, 2015, Pages 34-39

Neat stuff Mark. Interestingly the author uses the term "highest performing" as opposed to "stiffest". I wonder what he intended or thought, or possibly conflated? The reference section looks to be a good place for interested folks to investigate. Here is another paper from the same site: http://www.sciencedirect.com/science...77705815014216

[Mark Kelly]

They define highest performing as:

lateral displacement in out of plane load case divided by vertical displacement in in-plane load case

[Paul Jacobs]

Neat stuff Mark. Interestingly the author uses the term "highest performing" as opposed to "stiffest". I wonder what he intended or thought, or possibly conflated? The reference section looks to be a good place for interested folks to investigate. Here is another paper from the same site: http://www.sciencedirect.com/science...77705815014216

I think those links both go to the same paper. I thought the paper was confusing as it claimed to compare different tube sets from different manufacturers but it did not specify which tubes were used. They say" Clearly for the out of plane load case, the dominant tube is the down tube whereby the lateral displacement can vary by up to 71% simply by changing the down tube from Reynolds 853 (lowest stiffness) to Columbus Life (highest stiffness), while in the vertical load case the seat stays can have a significant effect on the vertical compliance of the frame with only a relatively small change in sectional profile compared with other tubes. " There are several possible Reynolds 853 and Columbus Life tubes which could be used as downtubes. If you stick to 31.75 alone, Reynolds produce 11 different 853 tubes with wall thicknesses varying from .65/.45/.65 to .9/.6/.9 and variations in the length of the butts. Similarly there is a range of possibilities in the Columbus Life series.