Ski flex as a complement to your normal ski specification

[Woppe]

Ski brands don't normally show any form of information about the flex of skis like they do with ski boots.

I just thought some of you might be interested to know that we've measured mechanical properties of skis - both flex (bending stiffness) and twist (rotational stiffness) - and made this information available on a website. https://www.betterskis.com.

There have been attempts made at providing this sort of information before. A number of years ago, for instance, the Norwegian ski magazine Fri Flyt showed graphs with ski stiffness distribution along with their regular ski reviews. I'm a big fan of this idea but I believe it failed at the execution. This information needs to be easily digestible in order to help guide anyone to better understand what ski is a good match for them. Our attempt is focused around transforming such detailed information into one single Flex number just like with ski boots, where Flex 100 is "medium stiff" and anything higher is stiffer and anything lower is softer. Same goes for the Twist numbers as well.

Our idea is that this can be used to better understand the feel of the ski if you don't have the chance to ride it, and also for comparing different skis.

I just wanted to share this with you all and hope some of you find it interesting. I'm also happy to hear your thoughts on this concept altogether. Stay stoked!

[Core Shot]

Interesting.
Twist I get.

But one number for flex is hard to do.
Had some shitty skis that were stiff with floppy tips. No one likes floppy tips.

A flex profile would be more helpful

[adrenalated]

Agreed. A single flex number tells you more than nothing, but the profile along the length of the ski makes a huge difference in how the ski will actually feel.
Two skis with the same overall flex number, but one with a stiff tip/soft tail and the other with with a soft tip/stiff tail are going to feel very different on snow.

[toast2266]

It's cool that someone is doing tests of ski stiffness again. But agreed with others in here that one flex number for the whole ski isn't all that accurate. There should at least be a finer breakdown if you click into the page for the ski.

Also, the ski's full dimensions don't seem to be listed - I only saw waist width (which should probably be the first number listed, not buried at the bottom).

I tried the ski recommendation tool, and it gave me a kind of huge array of options. Seems like there needs to be a few more questions to give a more accurate recommendation. It also only recommended fairly short skis, but I post on TGR so obviously I need something longer.

[Woppe]

It's cool that someone is doing tests of ski stiffness again. But agreed with others in here that one flex number for the whole ski isn't all that accurate. There should at least be a finer breakdown if you click into the page for the ski.

Also, the ski's full dimensions don't seem to be listed - I only saw waist width (which should probably be the first number listed, not buried at the bottom).

I tried the ski recommendation tool, and it gave me a kind of huge array of options. Seems like there needs to be a few more questions to give a more accurate recommendation. It also only recommended fairly short skis, but I post on TGR so obviously I need something longer.

You guys are good! Yes agreed, flex distribution is very important.

The way we dealt with that is that we've let both the tip and tail contribute more towards the flex number than the central parts of ski. The rocker profile also plays a role, where the flex number is reflective of the fact that bigger rocker makes for a softer feel.

You'll find an indication of the distribution of both flex and twist of you go into the "Details" tab on ski, e.g. https://www.betterskis.com/ski/Head-...oy-163-75-2021

Also thanks for feedback on how to display the info, appreciated!

[stealurface831]

anyone know how praxis does their flex ratings?

[Marshal Olson]

You guys are good! Yes agreed, flex distribution is very important.

The way we dealt with that is that we've let both the tip and tail contribute more towards the flex number than the central parts of ski. The rocker profile also plays a role, where the flex number is reflective of the fact that bigger rocker makes for a softer feel.

You'll find an indication of the distribution of both flex and twist of you go into the "Details" tab on ski, e.g. https://www.betterskis.com/ski/Head-...oy-163-75-2021

Also thanks for feedback on how to display the info, appreciated!

It is easy to measure ski thickness at 1cm stations along effective edge of ski. Then, with the measured overall stiffness (which is the surface area under the core profile curve), you can then have meaningful flex comparison graphs.

essentially the stiffness (Deflection / length @ fixed weight) normalizes the thickness data (up/down since a thicker ski with lower modulus material vs thin ski with high modulus material ) to give helpful information. I am 95% sure this is what Friflyt used to do, and it is 100% what ski brands do.

it can also be really helpful this way, to understand how longer/shorter skis change vs the standard (18x) length ski.

[Woppe]

It is easy to measure ski thickness at 1cm stations along effective edge of ski. Then, with the measured overall stiffness (which is the surface area under the core profile curve), you can then have meaningful flex comparison graphs.

essentially the stiffness (Deflection / length @ fixed weight) normalizes the thickness data (up/down since a thicker ski with lower modulus material vs thin ski with high modulus material ) to give helpful information. I am 95% sure this is what Friflyt used to do, and it is 100% what ski brands do.

it can also be really helpful this way, to understand how longer/shorter skis change vs the standard (18x) length ski.

It would be interesting to discuss this in further detail as I don't agree with you but I would love to be challenged.

1) "Overall stiffness is the [...] surface area under the core profile curve"
I see two problems with this.
a) It would work great if the skis were uniformly stiff along the length - but of course they are not. Imagine a ski with very soft at the very tip and tail, but stiff otherwise. It would deflect a lot, but feel stiff because most of the ski is indeed stiff. We had this idea as well to begin with, but ruled it out because we see this method is not being accurate enough.

b) Another prerequisite for this to be true is that skis of the same lengths are compared. Imagine a veeery long ski with the same thickness and material as a shorter ski. It would feel very flimsy. We have approached this differently to also take the length of the ski into account for setting a flex number.

2) Fri flyt might very well have measured the thickness at different segments rather than actual local stiffness, but I'm surprised if this is also what ski brands would do. I'm curious to understand how they would determine the effect of a metal sheet close to the top sheet, which will stiffen the ski significantly more than a thicker wooden core. Do you have anything further information on this?

I do agree that measuring the thickness of a model could indicate relative stiffness between different lengths of the same model - but I cannot see this working very well between different models. Like I said - I'm happy to be challenged!

[GoldMember]

Just a thought as I'm not an expert by any means on ski construction or design but for the purposes of establishing a flex test and scale for skis, I would think the following method could account for measuring stiffness at various points of a ski.

1 - Set up a jig with support points 25 cm apart
2 - Use a consistent pressure device, like a press, that can exert equal pressure at the center point of the support points
3 - Measure a ski at 25 cm intervals (tip, upper front of center, lower behind center, tail) and press the ski at those locations
4 - Determine deflection of ski at each point of measure, in mm or whatever unit you choose
5 - Based on deflection at each point, numerically define a stiffness
6 - As a final test, place supports at 170 cm as overall length representatives and test for overall stiffness

Alternatively, use a press to take deflection to a pre-determined value, say 5mm and determine how much pressure it took to get to that deflection measurement and delineate stiffness based on the pressure required to rate various skis.

It's just a thought but I think would provide a fairly accurate comparison of varying ski stiffness within the ski itself and relationally to competing brands. Again, I have no experience in this area but view it as a structural engineer. This also doesn't measure overall torsional stiffness but I would think there'd be methods with a similar approach that could provide some linear values.

[gregL]

Ski brands don't normally show any form of information about the flex of skis like they do with ski boots.

True, but ski boot flex index numbers are all over the map, and only a mild indicator of whether a boot might be suitable for you.

Outside of FIS race skis and comp mogul skis, pretty much everything these days is sort of "medium" flex (super stiff does not necessarily make a ski a better option for an "expert" skier) - do you think you could take a set of numbers (like Friflyte's, for instance) and predict which ski you'll like?

[Marshal Olson]

It would be interesting to discuss this in further detail as I don't agree with you but I would love to be challenged.

1) "Overall stiffness is the [...] surface area under the core profile curve"
I see two problems with this.
a) It would work great if the skis were uniformly stiff along the length - but of course they are not. Imagine a ski with very soft at the very tip and tail, but stiff otherwise. It would deflect a lot, but feel stiff because most of the ski is indeed stiff. We had this idea as well to begin with, but ruled it out because we see this method is not being accurate enough.

b) Another prerequisite for this to be true is that skis of the same lengths are compared. Imagine a veeery long ski with the same thickness and material as a shorter ski. It would feel very flimsy. We have approached this differently to also take the length of the ski into account for setting a flex number.

2) Fri flyt might very well have measured the thickness at different segments rather than actual local stiffness, but I'm surprised if this is also what ski brands would do. I'm curious to understand how they would determine the effect of a metal sheet close to the top sheet, which will stiffen the ski significantly more than a thicker wooden core. Do you have anything further information on this?

I do agree that measuring the thickness of a model could indicate relative stiffness between different lengths of the same model - but I cannot see this working very well between different models. Like I said - I'm happy to be challenged!

Hey man, all good. To share a little background:

Regarding #1 - This is high school calculus (Derivative). Deflection at a fixed mass divided by the length of the effective edge = Surface area under the curve of the thickness profile within the same effective edge length. You need both to make any meaningful comparison (as you highlight in your post).

Regarding #2 - Based on 23 years working for ski brands, with 15 as the head of sales and product at them, this is exactly how engineers at ski brands calculate it, and how they properly scale lengths shorter/longer within a model, or to give similar feelings within a collection for different models.

From my experience, rocker profile is easier to understand visually than with tables/metrics.

Hope that helps.

[Dee Hubbs]

Hey man, all good. To share a little background:

Regarding #1 - This is just high school calculus (Derivative). Deflection at a fixed mass divided by the length of the effective edge = Surface area under the curve of the core thickness profile within the effective edge.

Regarding #2 - Based on 23 years working for ski brands, and 15 as the head of sales and product at them, this is exactly how ski brands calculate it, and how they properly scale lengths shorter/longer within a model, or to give similar feelings to similar skis at different waist widths.

Hope that helps.

But Woppe has 9 posts here! Of course he does not agree with you, or you lifetime of experience.

[Marshal Olson]

Oh, and to add:

To account for effective edge length variance, anchor the curve profiles at Chord Center for each ski

[Woppe]

Just a thought as I'm not an expert by any means on ski construction or design but for the purposes of establishing a flex test and scale for skis, I would think the following method could account for measuring stiffness at various points of a ski.

1 - Set up a jig with support points 25 cm apart
2 - Use a consistent pressure device, like a press, that can exert equal pressure at the center point of the support points
3 - Measure a ski at 25 cm intervals (tip, upper front of center, lower behind center, tail) and press the ski at those locations
4 - Determine deflection of ski at each point of measure, in mm or whatever unit you choose
5 - Based on deflection at each point, numerically define a stiffness
6 - As a final test, place supports at 170 cm as overall length representatives and test for overall stiffness

Alternatively, use a press to take deflection to a pre-determined value, say 5mm and determine how much pressure it took to get to that deflection measurement and delineate stiffness based on the pressure required to rate various skis.

It's just a thought but I think would provide a fairly accurate comparison of varying ski stiffness within the ski itself and relationally to competing brands. Again, I have no experience in this area but view it as a structural engineer. This also doesn't measure overall torsional stiffness but I would think there'd be methods with a similar approach that could provide some linear values.

Yes you are exactly right, this is one feasible way of determining piecewise stiffness. The tricky bit is translating this into something that is easily digestible.

True, but ski boot flex index numbers are all over the map, and only a mild indicator of whether a boot might be suitable for you.

Outside of FIS race skis and comp mogul skis, pretty much everything these days is sort of "medium" flex (super stiff does not necessarily make a ski a better option for an "expert" skier) - do you think you could take a set of numbers (like Friflyte's, for instance) and predict which ski you'll like?

Yes agreed. This is an effect of that boot flex is measured by the manufacturers themselves instead of measured with a consistent method across different brands - or even just set to a number they see fit their marketing needs I imagine?

While skis generally exhibit quite "medium" stiffness, there are still indeed variations - some are more playful and some are more big mountain charger type of skis. Could I take a set of numbers from Fri Flyt and make out which ski would suit me? No. I think they fell short in translating the stiffness curve into something they conveys the feel of the ski. I couldn't make that out despite the fact that I have a good number of years of experience in the field.

Hey man, all good. To share a little background:

Regarding #1 - This is high school calculus (Derivative). Deflection at a fixed mass divided by the length of the effective edge = Surface area under the curve of the thickness profile within the same effective edge length. You need both to make any meaningful comparison (as you highlight in your post).

Regarding #2 - Based on 23 years working for ski brands, with 15 as the head of sales and product at them, this is exactly how engineers at ski brands calculate it, and how they properly scale lengths shorter/longer within a model, or to give similar feelings within a collection for different models.

From my experience, rocker profile is easier to understand visually than with tables/metrics.

Hope that helps.

#1 - I absolutely agree that that is one way measuring overall stiffness of any generic structure. We have chosen a different approach to determine flex of skis however, which we have found to better represent the feel of the ski. Our reasoning is that one single measure of stiffness (P/d, i.e. force per displacement) can underestimate the flex for skis with bigger variations in local stiffness.

#2 - I'm very happy to gain this information, appreciate it! I understand that this serves as a good proxy for stiffness distribution as stiffness is proportional to thickness^3 given that the material is homogeneous throughout the length of the ski. This of course would create a larger error when introducing various materials along different parts of the skis, such as metal plates or "air tips" (e.g. Rossi soul series), but again - good proxy. Interesting to hear this is the method that is commonly used. We've gone above and beyond this approach for determining flex on betterskis.com.

But Woppe has 9 posts here! Of course he does not agree with you, or you lifetime of experience.

Haha yup - good thing I've picked up a few things here and there without posting much. Try me

[GoldMember]

Yes you are exactly right, this is one feasible way of determining piecewise stiffness. The tricky bit is translating this into something that is easily digestible.

I would think you could determine a gradient that would correlate with the stiffness variables. Blister does a pretty good job of stating their interpretation of stiffness at various points of the ski on a 1-10 scale with 1 being toilet paper and 10 being a 2 x 4. I would think a similar gradient scale in your current range of 50-150 would be doable?? I guess my kind of one last comment is, don't overthink it. Keep it simple and relative from one ski to the next.

[Core Shot]

Hey man, all good. To share a little background:

Regarding #1 - This is high school calculus (Derivative). Deflection at a fixed mass divided by the length of the effective edge = Surface area under the curve of the thickness profile within the same effective edge length. You need both to make any meaningful comparison (as you highlight in your post).

Regarding #2 - Based on 23 years working for ski brands, with 15 as the head of sales and product at them, this is exactly how engineers at ski brands calculate it, and how they properly scale lengths shorter/longer within a model, or to give similar feelings within a collection for different models.

From my experience, rocker profile is easier to understand visually than with tables/metrics.

Hope that helps.

Pwned

Snap.

But seriously, the op website flex numbers are useless.

Blister gear reviews flex numbers are far more useful.
With their numbers and a pic of rocker profile and sidecut dimensions, I have a pretty good idea what I’m in for.

The op website is too dumbed downed. Good for rec skiers, but useless if you understand physics.

[glademaster]

it can also be really helpful this way, to understand how longer/shorter skis change vs the standard (18x) length ski.

I think somebody messed this up with the Koalas. My 179's are more of a handful than the 184's I demo'd. They're planky as fuck.

[Vitamin I]

1.) I applaud these kinds of attempts. I encourage you to keep developing this idea, and we'll see if users can ever actually harness any value from it. I don't even know if I like high twist or low twist or both, and I'd love to use your measurements to learn more about my own personal tastes (after you measure a lot more skis).

2.) Why does your website display the same flex value for all lengths of a given ski model (e.g. same value displayed for both a 172cm K2 Mindbender 108 Ti 2021 and a 193cm K2 Mindbender 108 Ti 2021)? Did you actually measure every length? (I doubt it.) Better to just display a value for the actual ski length you measured, and then display "not measured" for all the lengths that you did not actually measure. Otherwise, the data you report seems imaginary, without credibility.

.