Sidecut profile & long axis flex profile - interaction and performance

[daveo]

12 minutes ago, SunSurfer said:

@daveoGreatest popularity with GS racers on their boards.

SL, too, it seems.

13 minutes ago, SunSurfer said:

@daveoNot necessarily all the secrets.

I have a 4 day trip booked in August. With my lack of skill and poor work ethic, that's enough time to reveal literally every single secret ever. Mine is the 'Super H' variant from Allflex. Seems more carving/SL oriented- pretty chill plate by the sounds of things.

Oxess in the mornings. Contra in the arvos. Gonna be interesting. Maybe my world will be turned upsidedown.

[Jack M]

1 hour ago, SunSurfer said:

@daveoNot necessarily all the secrets. Some of us are still trying to identify which of the various design features of the AllFlex plate design(s) are responsible for it becoming the GS racers current favourite.

Isolation / separation of board from rider

Long distance between board attachment points

Limiting the amount of board flex before the plate must also bend in order for further board flex to occur.

Splitting front and rear of plate, with spring controlled resistance, potentially allowing the racer to manage mid section torsion of the board.

Variation in plate flex characteristics due to different rib patterns on the underside.

Greatest popularity with GS racers on their boards.

Having ridden a bunch of Allflex designs this season, I can see why Allflex completely replaced the Apex style isolation plates in racing.  A few weeks ago I got back on my K168 with my Apex X-plate and suddenly didn't like it.  It felt very vague and disconnected.  The Allflex plates by contrast feel very direct and powerful.  This was a surprise as previously I really liked my Apex/K168 combo.

The traditional Allflex aluminum plate is a match made in heaven for my stock K185.  The Allflex aluminum spring plate was awesome on my SL board, but it is heavy.  Non-athletes like me need not apply.  That said I used it and my F2 Proto 163 to carve down a steep icy double black in December and it performed with zero mistakes.

So I'm a convert, but I still can't quite get my head around why it works.  An Allflex plate basically creates three separate zones of flex in your board: nose, mid, tail.  I disagree with that design in theory, but I like it in practice.

[SunSurfer]

3 hours ago, Jack M said:

So I'm a convert, but I still can't quite get my head around why it works.  An Allflex plate basically creates three separate zones of flex in your board: nose, mid, tail.  I disagree with that design in theory, but I like it in practice.

@Jack MAbsolutely, I can't get my head around why the basic design works, but also what the variants (springs/plate underside rib designs) contribute to the overall performance. Board flex limitation is a unique design feature. The total amount of board flex in the mid-section in use would seem to be a function of the plate flex pattern.
The long inter-hinge distance (70+cm) creates a very long midsection where torsional stiffness is presumably very high. Presumably the spring versions allow effective rider input to the board mid-section torsion/twist in race courses where slide/carve turns are a necessity, or it benefits their particular riding style.

 

[daveo]

4 hours ago, Jack M said:

An Allflex plate basically creates three separate zones of flex in your board: nose, mid, tail.  I disagree with that design in theory, but I like it in practice.

After speaking to Marcel during my last build, his boards are also designed with the same three separate flex zones... Whether or not that means something useful, I don't know.

[crackaddict]

 

On 4/16/2021 at 7:18 PM, Lurch said:

think it was @crackaddicts 25m moon rocket that raised the topic of "sidecut depth being better predictor of turn size than sidecut radius"

Ya, that's pretty much a direct quote, and it is true.

I came to that realization after measuring the depths on a bunch of boards in my quiver and finding that there was a surprising amount of consistency between boards of vastly different lengths.  If you don't believe me, go to your boardroom and lineup your boards in the order of the turn size you feel they make.  Then measure the sidecut depths and see if the order isn't almost identical.

 

13 hours ago, johnasmo said:

The sidecut depth can't be a better predictor of turn size than radius since it's the same predictor.  It describes the same thing about the geometry of the sidecut.

 

This is (almost) correct.  @johnasmo knows what he means but he didn't say it quite clearly; the two metrics do of course describe the same thing so long as the boards are the same length (well, same effective edge).  The depth metric takes into account the length of the board and is therefore the better metric and more accurate predictor of turn shape if you only look at one or the other.  If the length is constant then of course they tell you the same thing.  But radius tells you very little if you don't know the ee, and if you look at both, then you're also looking at the sidecut depth, even if you haven't done the math.

 

Of course, there are many many factors that contribute to turn shape.  My thesis is only that sidecut depth is a better predictor than radius, not that either of them tells the whole story or predicts perfectly every time.

 

Let's take an example with two boards of the same depth.  The depth  for this example will be 2.25cm.  The first board is 165cm long (150cm ee) and the second is 185cm long (170-cm ee).  Do the math and find that the radius of the 165 is 12.5m and the radius of the 185 is 16m.  Now it gets a little subjective but assuming that the boards were made for the same rider and well designed, the turn shapes wouldn't be that different.  A little bigger maybe for the 185, but not 28% bigger as the pure radius number would suggest if you only looked at that one.  So in this case, neither number fully expresses the expected turn shape, but the 0% difference in depth is closer, I maintain, than the 28% difference in radius, and therefore sidecut depth is a better predictor than sidecut radius.  Q.E.D...

 

On 4/16/2021 at 10:08 PM, SunSurfer said:

I can't agree with the text in italics. Turn geometry doesn't work like that. 
So to reduce that to the absurd, that would mean a 6 metre long board and a 1.5 metre long board, both with a 3cm deep sidecut would turn roughly the same.

 

No point getting absurd @SunSurfer, I can't compare the turn shape of a 6m long board (which would be 149m radius by the way with your 3cm depth) with the turn shape of a 150cm board at 7.6m radius any more than I can compare the radii...  149m radius is meaningless in our world, we don't ride anything close to that.  But let's look at a more extreme example if you like, but something in a more familiar range.

 

Keep your depth of 3cm, which is a bit high but not absurd.  Let's take a 1.5m board as you suggest but a 1.5m ee and so around a 165 total length (about as short as grown men ride for carving), and let's compare it a 2m long board with a 185cm ee (about as long as I've ever seen).    The radii would be 9.4m and 14.3m respectively, so that's a 165 with a 9.4m scr and a 200 with a 14.3 scr.  The 200 will probably turn bigger sure but not 52% bigger as the pure radius number would suggest.   A 14.3m scr is tiny for such a long board, and it would turn very tightly indeed.

 

So the depth is still a better predictor, even in the extreme, though the thesis is more valid among comparable boards.  It can give one an idea of how much to increase the radius of a new longer board to achieve the same turn shape as a favorite current board, for example.

 

One more extreme example: two boards with the same radius but different lengths.  Say... the 165 and the 200cm boards above, both with an scr of 25m this time.  Now which one turns tighter?  The 200 right?  The one with the bigger sidecut depth.  See what I'm getting at?

 

So again, the sidecut depth is not a perfect predictor, but it is better than the radius if you're only going to look at one number.   Which of course you're not; you're going to look at least two of the numbers (out of the three: ee, scr, and sidecut depth), and any two define the third uniquely anyway so the point is moot.  But I'm glad someone noticed my idea, even if it took a year.  It was meant to provoke discussion and analysis, which it finally has, even if it's not really the topic of this thread.

Edited April 20, 2021 by crackaddict

[daveo]

I think sidecut depth was being referred to when effective edge is being held constant...

[SunSurfer]

@crackaddict We'll agree to disagree, but the disagreement will be friendly and discussed over beers if this awful epidemic ever ends, and I manage to find my way back to the USA for the Montucky Clear Cut.

[johnasmo]

6 hours ago, crackaddict said:

So again, the sidecut depth is not a perfect predictor, but it is better than the radius if you're only going to look at one number.

@crackaddictis correct.  I retract my statement that they are the same indicator geometrically because depth gives you a single number that is affected by both length and radius rather than just radius.  It contains more information than just radius alone.  But if you do have values for both radius and length, discard depth altogether, it does not add more information.  If you know only radius and depth, but not length, use depth as the better predictor when comparing two boards.

We usually know both length and radius, so this is somewhat academic.  Even knowing length and radius you can't accurately predict turn size without knowing something about the stiffness of the board, but we lack a standard quantitative measure for that.  As tilt increases, the influence of stiffness on turn shape surpasses that of sidecut radius.  IMO, radius predicts how turn initiation feels, how quickly the board begins to steer underneath to keep you from falling into a turn as your roll it on edge, but flex quickly takes over the main role if the board can penetrate the surface.  Even straight skies can turn in soft snow.

GS-like stiffness produces a GS-like turn size, regardless of sidecut radius/depth.  One of the Coiler prototypes we just tried at Big Sky bears this out.  It paired a long effective with a short radius and a go-fast flex.  It's was super fun, but you'd swear it was 3 meters straighter than it is.  A previous iteration (same radius) that was kept softer rides like a slalom board.

[Jack M]

@crackaddict isn't wrong.  An example was the pair of c.2000 Donek Freecarves, 171cm long and 179cm long.  Both had a sidecut radius of 11.2m (they were actually parabolic, but that doesn't matter).  The 171 was very popular.  I had one, it was one of those boards you demo and then you have to open your wallet and buy one without even asking your wife.  I heard a number of complaints about the 179.  I'll bet that the 179 carved a shorter turn due to the deeper sidecut depth, and was probably more finicky.  On another note, a board with less sidecut depth is easier to skid and feather.

The calculation of turning radius = sidecut radius * cosine(edge angle) is an approximation.  It kind of gives you the projection (shadow) of the sidecut onto the snow when the board is up on edge.  In real life the ends of the board will pinch in further, because the length of the board is fixed while the length of the projection is not.  I think deeper sidecut = more pinch.

Personally I just don't think of sidecut in this way, and I don't know of any board makers who will discuss turning radius with you in this way.  I guess unless you want to.  You certainly can't talk about a multi-radius sidecut this way.  I ordered my Kessler 180 with an avg 15m sidecut because I wanted it slightly longer than my Coiler Stubby 170 at 14m, and similar to my old Donek 186 at 15m.  It rides as predicted.  Hansjuerg did discourage me from making the board 185cm long due to the "short" sidecut.  He didn't say it but I guess he was concerned about sidecut depth.

[SunSurfer]

@johnasmo
Thanks, that relationship between sidecut/board inclination angle/flex all determining turn shape, with the inclination angle changing the share the other 2 factors contribute, is the concept where I began this thread.

I'm starting to get a clearer picture of the forces that make a board flex, and the desirable flex patterns in the front, midsection (between bindings or outermost plate attachment points). I still don't have a sense of what is desirable in rear section flex.

@johnasmo & @dredman
watching the Whitefish videos, John tends to begin his turns with a little sideways kick of the tail, and to my eye seems (correct me if I'm wrong) to ride the tail more in his turns. David seems to ride more balanced over the centre and begins his turns just by engaging the edge and rolling onto it.

Does this difference in style translate into different perceptions of the tail flex goals with the boards you ride together, and then compare notes on the chairlift?

Edited April 20, 2021 by SunSurfer
I'm gonna give up posting on my phone, my keyboard keeps doing really odd things about where the characters go!

[daveo]

3 hours ago, Jack M said:

Hansjuerg did discourage me from making the board 185cm long due to the "short" sidecut.

Marcel encouraged me to get a sidecut of 11-14m on a 182cm long board... Just sayin'

[SunSurfer]

@Jack MDid you discuss with Hansjuerg Kessler how the board flex might be adjusted, or was the discussion purely around sidecut?

The whole point of this thread, and the point that I made at the beginning, and John has made just above, is that SCR is the major determinant of turn shape only at shallow board inclination angles, and flex rapidly increases the contribution it makes so that it forms the majority of turn shape geometry at angles above 45 degrees.

SCR is easy to design, and relatively easy to change. Flex, as we've discovered, is a much more difficult beast to tame. You can really only find out the flex pattern once you have the fully asembled board. And yet it would seem that getting a sympathetic relationship between the two properties is the key to getting a well performing board

[crackaddict]

3 hours ago, johnasmo said:

Even knowing length and radius you can't accurately predict turn size without knowing something about the stiffness of the board, but we lack a standard quantitative measure for that. 

 

Indeed we do lack a standard metric for board stiffness, but @BLOODTYPEZX10Rdeserves credit for trying.  Mark's "Flex Index" is about the closest thing we have.  Maybe it's time to take this brilliant concept and perfect it.

Just as a longer board should have a larger radius for a given turn shape, the longer board also needs to be stiffer.  When you apply pressure to a board (i.e. when you turn it) that pressure is applied largely at the ends of the effective edge.  So with a longer board you're applying that pressure further out from the centre, and therefore with more leverage.  

If you have two boards with the same sidecut depth but one is longer, that one will turn tighter unless it's also stiffer.  It's difficult to compare board stiffness if the effective edge is not the same.  Hand flexing doesn't work: a short board will necessarily feel way softer than a long board even though both might be perfectly suited for the same rider.  Enter the flex index...

To determine flex index, we place 2.25" wood blocks under the board so that the distance between their inner edges are lined up with the ends of the effective edge, and the distance between them is the same as the ee.  Then put a bathroom scale between the bindings and step on it until the centre of the board just touches the ground.  Record the weight showing on the scale (in lbs) and divide it by the effective edge (in mm I think) to get the flex index for that board, expressed as a decimal.

This is an excellent metric that can be used to compare boards of different lengths.  It measures the stiffness per unit of length.  It doesn't tell you anything about the flex pattern or the torsional rigidity of course, but it's a very useful number.  However, something always bothered me about it.  I wonder if dividing by ee isn't somewhat arbitrary.  For example, maybe we would get a more accurate metric if we divided by the square of the ee, or the square root of the ee (or 2x or 4x the ee...).

I believe Mark invented this metric so that he could make boards that are suitable for riders he hasn't met or riders who haven't demo'd Thirst boards.  He'll ask you to measure the index of a board you like and then use that as a benchmark for your new Thirst.  This works great among boards of similar lengths, but may or may not work as well as the difference in effective edge gets bigger.

Other metrics don't work at all.  We talk about stiffness sometimes as a target rider weight, but this is somewhat  meaningless because it's very subjective.  The rider's skill, aggressiveness, style and preferences all need to be taken into account, not just their weight.   I learned this season, for example, that the words "super stiff" will be interpreted very very differently by a builder like Jasey-Jay Anderson than it will by a Sean Martin.

So how do we perfect the flex index metric?  I think it's an empirical question.  I've never done it, but I've thought about it a lot and discussed it at length with Mark (who seems to be satisfied with his flex index as it is).  I propose that we each take a few boards of different lengths but which are all in a comfortable stiffness range, measure the readings on the bathroom scale and then see which formula best expresses their stiffness "feel" while riding.

This would be very useful when buying used boards sight unseen, or when trying to describe your preferred stiffness for a first board with a new builder.  Even on your second board with the new builder, terms like "10% stiffer" are somewhat subjective and meaningless without a flex index or similar metric.  (Does 10% stiffer translate to 10% heavier target rider weight?  Not really.)  So let's play with this formula a bit and see what works and what best expresses the subjective stiffness feel while riding.

Or not.  The flex index is very close and very useful as it is.  It would be extremely useful if more builders used it and if we all used it when we're selling our used boards.  

Edited April 20, 2021 by crackaddict

[JRAZZ]

When you say "sidecut depth" do you mean the value measured between a line that runs between contact points to the waist or do you mean the theoretical value calculated as a sagitta using the average sidecut radius and running length? 

Cause these aren't the same and give surprisingly different answers.

 

I always liked using sidecut/running length. I think that's easier to understand and is somewhat more descriptive of how a board turns.

[workshop7]

1 hour ago, crackaddict said:

If you have two boards with the same sidecut depth but one is longer, that one will turn tighter unless it's also stiffer.  

Given two boards with the same side cut depth, the longer one will make a bigger turn as this board will have a larger radius.

[johnasmo]

2 hours ago, SunSurfer said:

watching the Whitefish videos, John tends to begin his turns with a little sideways kick of the tail, and to my eye seems (correct me if I'm wrong) to ride the tail more in his turns. David seems to ride more balanced over the centre and begins his turns just by engaging the edge and rolling onto it.

Just my bad form.  Hesitation to throw my body downhill early when not carrying enough speed off the last turn.  Side effect of completing turns too far, taking too long in transition.

[Chouinard]

 

I think I'll quote myself.

On 4/14/2021 at 9:44 PM, Chouinard said:

A singular “flex” value is a coarse measurement. There’s magic in how the “flex  coefficient” changes as the board shape changes.

If you've got a lot of time on your hands due to government lockdowns maybe you could suspend a board at the ends of it's effective length using rotating sleeves to relieve friction, load the board on the two binding locations with 5, 15 and 25 weights [each location] and use a dial indicator to measure board deflections at 1" increments see what the data tells you about flex under different loads.

 

 

 

[Jack M]

6 hours ago, SunSurfer said:

@Jack MDid you discuss with Hansjuerg Kessler how the board flex might be adjusted, or was the discussion purely around sidecut?

I told him about my weight and dadbod if that's what you're getting at.

[SunSurfer]

@Jack M

I'll chuckle and take that as a "No".
My thinking is that makers have core profiles and composite layups that produce a flex profile for that length that is in sympathy with a known range of sidecuts.  Ask for a sidecut outside that range and you're heading into unpredictable territory.
Watched a YouTube video of a computer guided ski core sander.
 A modified core would be a whole new code task and unknown performance. Hence the caution

 

 

 

 

[johnasmo]

On 4/19/2021 at 9:13 PM, Jack M said:

Having ridden a bunch of Allflex designs this season, I can see why Allflex completely replaced the Apex style isolation plates in racing.  A few weeks ago I got back on my K168 with my Apex X-plate and suddenly didn't like it.  It felt very vague and disconnected.  The Allflex plates by contrast feel very direct and powerful.  This was a surprise as previously I really liked my Apex/K168 combo.

The traditional Allflex aluminum plate is a match made in heaven for my stock K185.  The Allflex aluminum spring plate was awesome on my SL board, but it is heavy.  Non-athletes like me need not apply.  That said I used it and my F2 Proto 163 to carve down a steep icy double black in December and it performed with zero mistakes.

So I'm a convert, but I still can't quite get my head around why it works.  An Allflex plate basically creates three separate zones of flex in your board: nose, mid, tail.  I disagree with that design in theory, but I like it in practice.

Traction where it's needed, without unnecessary rider isolation.  Never rode one, but that's my guess.

I kicked the bees nest with my first post on page one of this thread, where I summarized my design thinking as:

"As I see it, sidecut shape, along with snow compaction or lack thereof, plays a role in influencing the shape it is being flexed into.   The board's resistance to being flexed into this shape affects the pressure distribution along the base.  How well the pressure distribution matches available friction along the edge affects edge hold.  Available friction along the edge is a function of downforce and tilt. "

We got sidetracked onto sidecut size issues, but let me take this opportunity to return to the issue of sidecut shapes.

A board that turns most aggressively near the tip requires traction there to match.  Tilted to 90 degrees, the lateral stiffness of the board would distribute the rider's down force (vertical pressure) all the way out there without loss.  But at lesser tilt, say under 45 degrees, it's being distributed through a platform that has longitudinal flex.  The tip can be flexed (and twisted) away, leaving the rider's vertical pressure closer to its input locations on the board -- where the rider's legs are attached.

If there's not enough vertical pressure where the turning loads are greatest, the horizontal component of the centripetal forces of turning can overcome the available traction.  Adding a structure between the rider and the board changes where the vertical pressure is applied to the board.  By the looks of the Allflex designs, going from two contact points to three and widening the rider's platform by about 300mm.  So it's a structure helping distribute the input of the riders weight closer to the tip and tail, which should be useful to increase traction at the tip and tail if that's there the board needs it.

I believe most plates have a similar effect, but the brilliance of Allflex is that it doesn't pursue trying to isolate the board from the rider's input.  The whole "letting the board flex naturally" idea.  The rider is the pilot; their inputs should be useful to control the board, including its flex.  A good plate should be giving them more control, not less.

The Contra designs go a different way, a contrarian way.  Instead of playing tricks (camber, plates, torsion) to redistribute traction to where your tires are, why not move your tires closer to where the traction naturally is?  I.e. Use sidecut shape to affect where load is being borne against the snow to keep it closer to your legs.

Plates also allow for tuning aspects of the board's stiffness and torsion to suit the rider, but at twice the cost of the boards I buy, there is an alternative -- find someone that can make you custom boards that cost less than these plates.  It's quicker and easier to hop on a different board than to re-tune a plate throughout the day.

[lowrider]

I have always said the purpose of a plate on a board is to allow a board to do what is was designed to do without influencing it . Whether or not the rider can match those parameters is always the elusive element. It's not realistic to think one board does it all but with each passing season it's evident that some are getting closer. Best example for me to date has been the effectiveness of a plate on a board in icy conditions. ( you have to be an east coaster to experience this )

[Chouinard]

On 4/14/2021 at 9:46 PM, dredman said:

So many factors....and how do they change performance? I am still trying to get my head wrapped around how all of these variables interact.  How do you quantify these things to test hypotheses?  Because we do not have the luxury of building and testing boards that only have one change in design parameter to help make these determinations it leads to lots of speculation  and even more questions.

If you want to understand the interaction then you may consider doing something like this if you have access to a large diverse quiver of boards.

Rider perception - Start with a screening DOE:

1)      Make a list of the critical input/output characteristics such as: Dampness, edge hold, ease of turn initiation, flex index, side cut, chamber, etc.  Some metrics will be rider perceived, some will be geometric as-built and some will have to be measured using a repeatable test method. Use at least 10 measurement bins.

2)      Record the geometric as-builts dimensions and measurables.  

3)      Make a survey form and take a diverse quiver of boards out and using the same rider/setup/condition/technique make several runs and rate the characteristics from 1 to 10.  Now change the rider setup to the riders optimized settings for that board based on past experience and rate the characteristics again. Repeat the trials for different surface conditions representing everything from powder to glare ice.

1)      Bin all the metrics into at most three levels.

2)      Use Minitab or other and generate a DOE for each separate output characteristic using all the input characteristics.

3)      Crunch the data. The analysis with statistically shake out any significant interactions and the primary inputs for each output separately. You will also generate a prediction equation that will allow you to identify the optimize input metric values.

4)      Now take the separate results and select the board that is a blend of all the optimal inputs [closely matches the individual prediction equations] and verify the results.

Board Design - all purpose board [for a condition specific board ignore data collected under non-specific conditions]:

Identify the design elements that control the significant inputs identified in the screening DOE and use a DOE again but this time look at the input levels of the design elements both geometrical and mechanically.

Hand off the results to the board builder for a prototype.

Ride the board and validate the design.

Start with a rider perception, translate it to numbers, optimize the numbers, translate it back to a rider perception with a prototype.

[SunSurfer]

@Chouinard
Not familiar with the DOE acronym. What does it stand for?

[JRAZZ]

We deep into engineer speak. I believe the intent is "Design of Experiment"

 

The use of Minitab warms my heart.

[lowrider]

DOE $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$