effect of chainstay stiffness on overall BB stiffness? (beyond a certain point)

[wallymann]

wondering if having really phat chainstays matters to perceived lateral stiffness @ the BB, versus say a massive diameter down-tube at the BB.

of course larger diameter tubes are stiffer (chainstays are no different) but the BB shell is working with the triangulated pair of chainstays, so i'd think the diameter of each chainstays wouldnt matter too much once you achieve a "sufficient" stiffness (say the equivalent of SL chainstays, for the sake of discussion) you're better off investing additional material beyond the "enough" stiffness chainstay elsewhere such as making the downtube diameter larger @ the BB to achieve more perceive lateral BB stiffness.

i googled and found a post by Kirk Frameworks that makes assertions that are completely contrary to my intuition, so i'm curious what other builders might have to say on the matter.

here post from Kirk, which references a thread at the old serotta forum that i cant seem to locate here in paceline.

http://www.kirkframeworks.com/blog/2009/03/18/chainstaydrivetrain-thoughts/

[wallymann]

http://forums.thepaceline.net/showpost.php?p=221046&postcount=17

the information provided by "chrisroph" is attributed to tom kellogg:

the most efficient way to stiffen up a BB is to use torsionally stiffer chain stays and dropouts. I say most efficient because the down tube and seat tube support the BB as well, but increasing BB stiffness using "beefier" seat tubes and down tubes adds more mass for a similar result or as much mass for a smaller result. The other advantage to using the chain stays is that they have no negative effect on vertical compliance. Cake and eat it too...

maybe those slingshot guys were onto something...design a frame with the phattest BigLegEmma chainstays (for a stiff/efficient BB) with a comparably phat top-tube (to ensure a stiff/confident front-end) and then a bit of stainless-steel cable for the downtube (to keep things light) and complete s stiff-but-light frame design. ;-)

[cmg]

i fall into the group that thinks fat chainstays stiffens the bottom bracket. the short piece of tubing that links both chainstays together in front of the rear tire also does some stiffening as well. look at a Merckx built out of MAX tubing. the chainstays are massive. i don't like frames where the chainstays are the same diameter from the BB to the dropouts (original Litespeeds). also chainstay lenght effects stiffness.

[carpediemracing]

Wouldn't both be important?

The modern carbon bikes all have relatively massive down tubes as well as massive chainstays.

I've never ridden a bike with massive chainstays that didn't have a massive down tube (unless you count the original Cannondale as having a less-than-massive down tube).

I've also never ridden a bike with a massive downtube that didn't have reasonably beefy chainstays.

I personally want both, and I seemed to prefer bikes with larger than smaller stays.

[shovelhd]

I use to race a custom Ted Wojcik in the 80's. It was built of 753 except it had SP track stays, both seat and chain. The bottom bracket lugs were also reinforced. It made for a pretty stiff bottom bracket for a 753 frame. The little ring clearance was super tight, even after massaging the stay.

http://i331.photobucket.com/albums/l453/shovelfl/ebay%20jul%202010/IMG_0627.jpg

http://i331.photobucket.com/albums/l453/shovelfl/ebay%20jul%202010/IMG_0630.jpg

http://i331.photobucket.com/albums/l453/shovelfl/ebay%20jul%202010/IMG_0629.jpg

[Chance]

..........
maybe those slingshot guys were onto something...design a frame with the phattest BigLegEmma chainstays (for a stiff/efficient BB) with a comparably phat top-tube (to ensure a stiff/confident front-end) and then a bit of stainless-steel cable for the downtube (to keep things light) and complete s stiff-but-light frame design. ;-)

The frame then just flexes in a different manner. Since the cable can't carry torsional loads it has to be transferred through top tube and seat tube. There is a reason these bikes are no longer made. Lots of ideas may seem good on the surface but if analyzed properly and comprehensively they really don't work well.

[Chance]

Wouldn't both be important?

.........

Exactly. And for two different reasons.

BBKT flex from side to side while pedaling hard is a function of two very different primary loads. While both chainstays and downtube resist flex of both, each is more directly connected with each type of load.

[cmg]

interesting article, differing opinions from 2 builders.

"First I’ll say something that you won’t hear from many builders…….curving or bending the stays into any configuration other than straight will make the stays flex more. There are very few absolutes in life but this is one of them. Curved stays flex more than the equivalent stay without the bend. They won’t flex a lot more……just a little. It depends on the type of bend and the degree of bend. One thing is for sure, the bent stay can never be stiffer due to its bend. The biggest reason to bend a stay is to allow for better clearances with tire and chainring. This is a bigger deal on mountain bikes than road but true nonetheless."

The Serotta 2003 cataloque says that stiffeness is increased.

opinions? i tend to believe bending adds stiffness, changes the direction stress flows through the material.

[wallymann]

The frame then just flexes in a different manner. Since the cable can't carry torsional loads it has to be transferred through top tube and seat tube. There is a reason these bikes are no longer made. Lots of ideas may seem good on the surface but if analyzed properly and comprehensively they really don't work well.

FWIW slingshot is still alive and well, selling more stuff every day!

http://www.slingshotbikes.com/

http://www.slingshotbikes.com/images/img_farmboy1.jpg

[christian]

If bottom bracket stiffness is such an issue, why don't builders put a taco gusset between the down tube and seat tube and the chain stays and seat tube?

[wallymann]

http://i331.photobucket.com/albums/l453/shovelfl/ebay%20jul%202010/IMG_0629.jpg
http://i331.photobucket.com/albums/l453/shovelfl/ebay%20jul%202010/IMG_0630.jpg

love the lugwork and thoughtful design considerations on the Wojcik! all those tweaks to make for a stiff BB on an otherwise whippy 753, and those long-long points on the BB lug. nice.

i've always preferred steel bikes whose chainstays try to stay as round as possible. i refuse to own a frame where the stays are explicitly crimped/dimpled for chainring clearance -- it's just wrong. a slight ovalization is all that i'll tolerate!

[shovelhd]

Thanks. Sorry about the dirty pictures (not those kind). I took them right after I took it down from the basement ceiling. I had not ridden a bike in 26 years.

[cmg]

the welded spacer between the chainstays moves the flexing stresses away from the bottom bracket weld points. it acts the same as a taco gussett.

[carpediemracing]

Thanks. Sorry about the dirty pictures (not those kind). I took them right after I took it down from the basement ceiling. I had not ridden a bike in 26 years.

I didn't realize you still had a bike from back then. Very cool.

[shovelhd]

I don't anymore. I had to sell it to fund a modern racing bike. I don't ahve any of my old bikes anymore, unbfortunately. Team Miyata, Team Fuji, Basso Gap.

[carpediemracing]

interesting article, differing opinions from 2 builders.

"First I’ll say something that you won’t hear from many builders…….curving or bending the stays into any configuration other than straight will make the stays flex more. There are very few absolutes in life but this is one of them. Curved stays flex more than the equivalent stay without the bend. They won’t flex a lot more……just a little. It depends on the type of bend and the degree of bend. One thing is for sure, the bent stay can never be stiffer due to its bend. The biggest reason to bend a stay is to allow for better clearances with tire and chainring. This is a bigger deal on mountain bikes than road but true nonetheless."

The Serotta 2003 cataloque says that stiffeness is increased.

opinions? i tend to believe bending adds stiffness, changes the direction stress flows through the material.

I wonder if this logic applies:

With straight chainstays a builder is limited to a certain size tube.

With curved chainstays a builder can use a larger tube and still clear the various obstacles. It would be stiffer.

Bending a tube will work harden it, if my very basic metallurgy knowledge is right. This will stiffen the tube (although it'll also weaken it a bit).

Finally, with the bending/hardening weakening a tube, a bent chainstay probably should have thicker walls, therefore it'll be stiffer.

[Chance]

FWIW slingshot is still alive and well, selling more stuff every day!

http://www.slingshotbikes.com/

http://www.slingshotbikes.com/images/img_farmboy1.jpg

Thanks for update. Thought they had been out of production for some time. Have never personally seen one on the roads so they must sell in very small numbers.

Still seems like a questionable design idea. To me the design seems flawed compared to other options. Patents, if they ever existed, wouldn't apply by now. So if the idea actually worked well wouldn't you think many other manufacturers would be copying it?

In any case not having a downtube to carry most of the bike's twisting (torsional) load means it has to be carried by top and seat tube. Not an efficient path in my opinion.

[Chance]

If bottom bracket stiffness is such an issue, why don't builders put a taco gusset between the down tube and seat tube and the chain stays and seat tube?

Probably because deflection is cumulative. While a gusset helps in reducing flex in the immediate area, the rest of the downtube, seattube, or chainstays that are not supported by gusset will flex the same. Maybe even more than before depending on loads and gussets. For a given amount of added weight, cost, and work, it'd probably be more effective for builders to just make these tubes stiffer their entire length. Would reduce flex more overall.

[David Kirk]

I wonder if this logic applies:

With straight chainstays a builder is limited to a certain size tube.

With curved chainstays a builder can use a larger tube and still clear the various obstacles. It would be stiffer.

Bending a tube will work harden it, if my very basic metallurgy knowledge is right. This will stiffen the tube (although it'll also weaken it a bit).

Finally, with the bending/hardening weakening a tube, a bent chainstay probably should have thicker walls, therefore it'll be stiffer.



I see what you are saying but in my experience it doesn't work this way.

The one place where room is truly limited is between the tire and the chainrings and since these two things fully overlap there is no way curving anything will allow for a much larger horizontal cross section to fit in there.

I say 'horizontal' cross section because one can make the stay oval in cross section with the major axis being vertical. This way you can get more tube into that small spot. Unfortunately an oval tube isn't much stiffer than a road tube when flexed against its minor axis. In other words the tall oval stay will flex as much side to side as a round one will as long as the minor axis of the oval stay is about the same as the diameter of the round stay. And since we can't make more room between the tire and the rings that means that an oval stay with the same wall and minor diameter as a round stay will not be any stiffer is side to side flex. It will be stiffer vertically (not needed) and it will be heavier but it will not be stiffer in side to side flex where is actually matters. Now if one could make an oval stay fit when it is on its side we'd have something!

For this reason - with a metal tube - it's most often best to use as large a round tube as you can fit. The round tube will deal with torsional flex as well as an oval tube, it will be just as stiff side to side and it will be lighter. Round is really a good shape.

I've done much more testing of different shapes of c-stays that I care to admit. If I told you how many bottom bracket - chainstay assemblies I've made and done load/deflection tests on you would make fun of me..........and be justified! The tests show what one would expect to see when looking at the mechanical properties of the components involved. It becomes pretty clear when you watch stuff move on the bench.

--------------------------------------------

FWIW - if you make two bikes that each with everything the same but on one of them you bend the c-stays and then you do load/deflection tests on the two bikes you will most often see zero difference. I suppose one could argue that the bending work hardens the tube and thus makes it stiffer. But the area that is bent is almost always in the middle of the stay (front to back) where the stress is very low anyway and most of the load is transferred to the big end by the BB shell. It's very easy and consistent to test. The only reason to bend c-stays is to buy room (cassette/c-ring/heel...etc) and that can be a very good reason but stiffness isn't a benefit.

--------------------------------------------

Earlier someone said the stay size shouldn't matter much because the rear end is a triangle. I understand what you mean. The issue is that it's such a long skinny triangle that each stay pretty much acts like a single cantilevered beam and less as a unit. Combine this with the fact that one leg of the rear triangle made up by the two c-stays is the rear axle and the dropouts and you have bigger issues. Large plate style dropouts move A LOT (shocking to see when bench testing) and the rear axles in most hubs are far from being super stiff. To this add the flexi interface between the dropout and the axle and you end up with something that is far from a rigid triangle.

--------------------------------------------

Down tube size as it related to BB stiffness - - This is a large and complicated topic that I don't have time to get very deep into now but it was addressed pretty well in the old thread that the OP linked early in this thread. Suffice it to say that a big DT will make the BB feel stiffer (could be a good thing in some cases) but it will have little effect on drivetrain stiffness as that is largely determined by chainstay size and material. Tom Kellogg goes into this in that linked c-stay thread too as I recall. If you bench test two bikes that are the same except that one has a larger diameter DT and do a lateral load/deflection test the results will surprise you. The two bikes will in most cases deflect the same amount overall but they will flex in different places. The bike with the smaller DT will have a rounder more overall flex and the bike with the larger DT will flex less in the front end and more in the rear. I was confused when I first saw this but it makes sense when you watch it move. The big DT won't move so it transfers the load to tubes that will. It's fascinating to watch.

Now the larger DT may be a good thing for other reasons such as providing the frame with better torsional stiffness in the front end but BB stiffness is not one of the large benefits. This goes against what we see in all the marketing but I doubt this is a big shock to anyone that marketing might not reflect physical reality. The Slingshots as always are great at illustrating this point. I've ridden some that are super stiff in the drivetrain with a cable as a DT. Strange but true.

---------------------------------------------

Now all this stuff I'm talking about pertains mostly to metal bikes made from tubes. When you switch to carbon all bets are off as they can play with lay up of the material to make things stiff in one direction and less so in another.

Time for me to go lay some fillets - thanks for reading.

Dave

[93legendti]

FWIW slingshot is still alive and well, selling more stuff every day!

http://www.slingshotbikes.com/

http://www.slingshotbikes.com/images/img_farmboy1.jpg

I had one...fastest bike i owned. It climbed unbelievably well.

[FlashUNC]

FWIW, I'll pass along some perspective I got from Mr. Della Santa during a chat about my latest frame order.

He's doing a series of chainstays now he calls Ossobucco. Basically a series of reinforced holes on the chainstays that, in a vacuum, increase the stiffness of the chainstays from his standard offering.

But the reality, in his own words, "Once you attach them to a bottom bracket, and attach some seatstays to form a triangle, then fit a wheel in there which is clamped down. You don't really notice the difference."

He said locals who have bikes with both say they think the stays a bit stiffer climbing, but you can't tell the difference elsewhere.

He freely admits the does them because "they look cool" rather than any real performance gains.

[wallymann]

He's doing a series of chainstays now he calls Ossobucco. Basically a series of reinforced holes on the chainstays that, in a vacuum, increase the stiffness of the chainstays from his standard offering.


are della-santa's reinforced-hole chainstays something like those on a select few confente/masi frames?

http://farm8.staticflickr.com/7141/6818748567_9259a30208_b.jpg
http://farm5.staticflickr.com/4122/4882588006_7a947cda65_b.jpg

[FlashUNC]

are della-santa's reinforced-hole chainstays something like those on a select few confente/masi frames?

http://farm8.staticflickr.com/7141/6818748567_9259a30208_b.jpg
http://farm5.staticflickr.com/4122/4882588006_7a947cda65_b.jpg

Yes. In fact, Roland said he was inspired by those old Masi's and Confente's for the design.

[wallymann]

Yes. In fact, Roland said he was inspired by those old Masi's and Confente's for the design.

even ernesto seems to have taken inspiration from confente when creating his HP chainstay design in CF:

http://farm4.staticflickr.com/3084/3217774941_7970a13301_z.jpg?zz=1

[cnighbor1]

In an article by Jan Heine has proved a less stiff frame is superior. with a less stiff frame the energy you put into the frame comes back out to push you down the road
He test an OS tubing frame verus non OS tubing steel lugged frame. the non OS tubing was faster

[LouDeeter]

In an article by Jan Heine has proved a less stiff frame is superior. with a less stiff frame the energy you put into the frame comes back out to push you down the road
He test an OS tubing frame verus non OS tubing steel lugged frame. the non OS tubing was faster

Da plane! Da Plane!!

[LouDeeter]

There's more than one way to skin a cat. I'm guessing that this bb from Dave Porter in Albuquerque is plenty stiff!

[Joachim]

In an article by Jan Heine has proved a less stiff frame is superior. with a less stiff frame the energy you put into the frame comes back out to push you down the road
He test an OS tubing frame verus non OS tubing steel lugged frame. the non OS tubing was faster

Here we go...

[firerescuefin]

Here we go...

Did someone put Jan and "proved" in the same sentence together:rolleyes:

[JLP]

Terrain! Terrain! Pull up!

But seriously, I love that DK, who has actually looked at the engineering and measured results, can join us for the conversation. Thanks so much. Joe

[carpediemracing]

I see what you are saying but in my experience it doesn't work this way.

Thanks. I try to speak from experience but I don't have much with actual frame testing so thanks for explaining what you've experienced.

[Chance]

..........
The one place where room is truly limited is between the tire and the chainrings and since these two things fully overlap there is no way curving anything will allow for a much larger horizontal cross section to fit in there.

I say 'horizontal' cross section because one can make the stay oval in cross section with the major axis being vertical. This way you can get more tube into that small spot. Unfortunately an oval tube isn't much stiffer than a road tube when flexed against its minor axis. In other words the tall oval stay will flex as much side to side as a round one will as long as the minor axis of the oval stay is about the same as the diameter of the round stay. And since we can't make more room between the tire and the rings that means that an oval stay with the same wall and minor diameter as a round stay will not be any stiffer is side to side flex. It will be stiffer vertically (not needed) and it will be heavier but it will not be stiffer in side to side flex where is actually matters. Now if one could make an oval stay fit when it is on its side we'd have something!

For this reason - with a metal tube - it's most often best to use as large a round tube as you can fit. The round tube will deal with torsional flex as well as an oval tube, it will be just as stiff side to side and it will be lighter. Round is really a good shape.

I've done much more testing of different shapes of c-stays that I care to admit. If I told you how many bottom bracket - chainstay assemblies I've made and done load/deflection tests on you would make fun of me..........and be justified! The tests show what one would expect to see when looking at the mechanical properties of the components involved. It becomes pretty clear when you watch stuff move on the bench.
............


Technical data to the contrary. This is just not correct regarding stiffness. Anyone with the proper technical skills to know what they are looking at can look it up for themselves.

An oval as you’ve described with equal minor axis to a circle and of equal wall thickness will be stiffer in side to side also. Heavier yes, but also stiffer. The difference in stiffness won’t be as pronounced as in the direction of the major axis but there is a difference nonetheless. No doubt that’s why so many chainstays over the years have been made oval.

Don’t know what mechanical properties you look at, but it’s interesting that your test confirm what is impossible given your assumptions. Either the results of the tests are wrong (confirmation bias???) or you are making assumptions about tube dimensions and or properties that are not accurate.

[David Kirk]

Technical data to the contrary. This is just not correct regarding stiffness. Anyone with the proper technical skills to know what they are looking at can look it up for themselves.

An oval as you’ve described with equal minor axis to a circle and of equal wall thickness will be stiffer in side to side also. Heavier yes, but also stiffer. The difference in stiffness won’t be as pronounced as in the direction of the major axis but there is a difference nonetheless. No doubt that’s why so many chainstays over the years have been made oval.

Don’t know what mechanical properties you look at, but it’s interesting that your test confirm what is impossible given your assumptions. Either the results of the tests are wrong (confirmation bias???) or you are making assumptions about tube dimensions and or properties that are not accurate.



Hey Chance -

Thanks for the input.

I make very few assumptions as frankly I'm not very good at them. What I have done instead (during both my years at Serotta and my time on my own) is do a good bit of controlled and repeatable testing.

One of the areas of testing has been frame stiffness and what contributes to stiffness and what type/direction of stiffness results.

I've done much testing of lateral stiffness in 3 different ways. One is just an isolated c-stay with a load/deflection test.............another is a pair of c-stays attached to a BB and constrained by a hub and the last is a complete frame. In all of these tests the chainstays were either round (22.2mm OD) with a given wall thickness of an oval stay with a minor diameter of 22.2 and the same wall as the round stay.

The tests were set up in a simple and pragmatic way and ended up being very repeatable with the same results time after time.

The results I saw with this consistently showed that the oval stay did not provide any real world lateral increase in stiffness as compared to the round stay. One would see a very narrow scattering of results (1 or 2%) but in the real world the stiffness was what most folks would think of as the same. There are of course small differences in individual stays (manufacturers tolerance) but in the end any differences were on the order of 1-2% and could fall either way - from round to oval being stiffer.

I did the bulk of this particular testing while I was working at Serotta and frankly I thought the oval stay was going to be the way to go and I was a proponent of moving Serotta to using oval stays on most everything. To help make the decision the tests were devised so we could have a better idea of how they would change the ride and performance. In other words I expected the oval to be better and maybe even wanted them to be but the tests didn't bear this out.

Now I suppose there could be any number of things that could have skewed the test........but frankly it was a brutally simple test and it worked the same time after time, across materials (steel and Ti) and with either a single stay, a pair of stays hooked to a BB or a complete frame. So while I can imagine that the tests were not perfect every effort was made to learn as much as we could and to design the bikes based on our findings.

After the bench testing was done we made bikes where the only variable was the stays to help give a real world feedback to the deal. A few of us road them and we, not surprisingly, couldn't tell the difference between the two shapes of c-stays. Pretty obvious stuff that if we couldn't measure a difference that we wouldn't feel one but it was a good check on the bench tests for if we felt an obvious difference then we'd know that the bench tests were flawed.

That has been my experience. I understand it might fly in the face of what you would expect to see and I respect that and would be open to new and different testing. But I can only share what I have done and seen myself over time.

Thanks again for reading and I look forward to your thoughts and feedback.

Dave

[nicrump]

the oval stays are almost always thinner than the round ones.

in my mind the newer(well in '92) tall ovals were about reducing wall/weight and increasing clearances while maintaining the mechanics of the previously heavier(thicker wall) rounder ones.

i think Mr Kirks testing demonstrates this.

but for my money i am looking for hub to hub torsional stiffness across the bike.

what i am looking for in my chainstay is one that takes the repeated compression/bend and keeps on ticking.

i have no earthly idea what bottom bracket stiffness means. it has not been properly defined as near as i can tell.

[11.4]

Hey Chance -

Thanks for the input.

I make very few assumptions as frankly I'm not very good at them. What I have done instead (during both my years at Serotta and my time on my own) is do a good bit of controlled and repeatable testing.

One of the areas of testing has been frame stiffness and what contributes to stiffness and what type/direction of stiffness results.

I've done much testing of lateral stiffness in 3 different ways. One is just an isolated c-stay with a load/deflection test.............another is a pair of c-stays attached to a BB and constrained by a hub and the last is a complete frame. In all of these tests the chainstays were either round (22.2mm OD) with a given wall thickness of an oval stay with a minor diameter of 22.2 and the same wall as the round stay.

The tests were set up in a simple and pragmatic way and ended up being very repeatable with the same results time after time.

The results I saw with this consistently showed that the oval stay did not provide any real world lateral increase in stiffness as compared to the round stay. One would see a very narrow scattering of results (1 or 2%) but in the real world the stiffness was what most folks would think of as the same. There are of course small differences in individual stays (manufacturers tolerance) but in the end any differences were on the order of 1-2% and could fall either way - from round to oval being stiffer.

I did the bulk of this particular testing while I was working at Serotta and frankly I thought the oval stay was going to be the way to go and I was a proponent of moving Serotta to using oval stays on most everything. To help make the decision the tests were devised so we could have a better idea of how they would change the ride and performance. In other words I expected the oval to be better and maybe even wanted them to be but the tests didn't bear this out.

Now I suppose there could be any number of things that could have skewed the test........but frankly it was a brutally simple test and it worked the same time after time, across materials (steel and Ti) and with either a single stay, a pair of stays hooked to a BB or a complete frame. So while I can imagine that the tests were not perfect every effort was made to learn as much as we could and to design the bikes based on our findings.

After the bench testing was done we made bikes where the only variable was the stays to help give a real world feedback to the deal. A few of us road them and we, not surprisingly, couldn't tell the difference between the two shapes of c-stays. Pretty obvious stuff that if we couldn't measure a difference that we wouldn't feel one but it was a good check on the bench tests for if we felt an obvious difference then we'd know that the bench tests were flawed.

That has been my experience. I understand it might fly in the face of what you would expect to see and I respect that and would be open to new and different testing. But I can only share what I have done and seen myself over time.

Thanks again for reading and I look forward to your thoughts and feedback.

Dave

What Dave is saying (if I'm not misquoting him horribly) is that there are more variables at work than even an experienced framebuilder was anticipating. Chance's doubts stem not from contradictory results in equivalent testing, but probably from having to digest Dave's having identified the existence of additional variables that contribute to the stiffness equation.

I've found in many frames over many years that frame stiffness and ride quality are both highly multivariate determinants and I've never seen anyone, anywhere, who actually captured the variables in a manner that was reproducible. Individual framebuilders each have their own take on how they influence certain variables and different framebuilders can create superb bikes via different paths simply because they have more variables to work with and choose the ones they can modify and respect those that they cannot. What becomes interesting is when classic constraints such as chainstay tire clearance can be modified either by carbon fiber or other engineered solutions or by different bottom bracket designs. Same for other dimensions on a bike. I suspect that while we bitch about eleven speed cassettes and having to replace older equipment, we have both front and rear wheel spacing, fork/head tube design, bottom bracket design, front and rear brake mounting and technology, and who knows what other variables that will all change in efforts to create greater degrees of freedom in frame design.