Hey guys, a young beginner here with a question regarding sail design.

I've attached a graph of an RSX 520 mast bend under different loads, but this question applies to all (most?) windsurf sails.
As you can see there is a big difference between the mast-bend of a rigged sail and the IMCS defining ~30kg load @50% of mast length.

If I were to make a sail from scratch - only having the mast, how would I go about defining the bend of the mast under which ill design my sail ? If I define my draft amount via additional luff curve and my draft posit. via broadseam for the ~30kg load as displayed, am I missing out on some black magic that is achieved by designing the sail around a bigger bend, and if so, is there an optimal bend ? Thanks.

Wow !
Over to someone else ......

Thanks for the bump I guess, to clear things up - in the upper right corner where it says D24O20 etc. Those are different settings where the D stands for Downhaul and O for Outhaul amount expressed in cm.

Masts don't bend that much

That pictures wrong

Select to expand quote

Dishpet said..
Thanks for the bump I guess, to clear things up - in the upper right corner where it says D24O20 etc. Those are different settings where the D stands for Downhaul and O for Outhaul amount expressed in cm.

Dishpet, I am a Windsurfer, but you have completely lost me. Could you explain the graph. What are the units of each axis ?. Are you measuring the excursion at a single point on the mast from bent to at rest ?, or is the graph representing actual bend of the mast ?. RSX250 ?. Is this a Neil Pryde mast ?

BTW, nice graph. Are you using Matlab, or Excel ?

Select to expand quote

shoodbegood said..

Masts don't bend that much

That pictures wrong

I am surprised as well. What is the unit if the bottom scale? Where did the graph figures come from?

If it is a linear measurement oflength.....?

I can do some measurements to confirm, but I would be surprised if any the sails I have bend the mast approximately 200% more than the MCS test! (3 times the deflection!!!)

If the unit on the bottom scale is a load/force, perhaps it is correct.

With respect, I think you may be approaching this on the wrong way.

Try laying a decent sail out unrigged on a flat surface and trying to measure out the luff curve (in relation to a straight edge) for an approximation.

Select to expand quote

jn1 said..

Dishpet said..
Thanks for the bump I guess, to clear things up - in the upper right corner where it says D24O20 etc. Those are different settings where the D stands for Downhaul and O for Outhaul amount expressed in cm.

Dishpet, I am a Windsurfer, but you have completely lost me. Could you explain the graph. What are the units of each axis ?. Are you measuring the excursion at a single point on the mast from bent to at rest ?, or is the graph representing actual bend of the mast ?. RSX250 ?. Is this a Neil Pryde mast ?

BTW, nice graph. Are you using Matlab, or Excel ?

Hi jn1, I wish I could take credit for the graph but it's from a reaserch paper on rs:x aero and hydrodynamics.

The graph represents actual bend, on the left side it represents the height/length of the mast 0-100%/0-520cm, the bottom represents the bend amount as a percentage of the mast length - rigged it is 10ish percent bend @50% of length (260cm) - that translates to about 52cm deflection @50% height/length in comprasion the the mast not being bent/rigged. Hope that clears it up, English is not my first language.

Select to expand quote

Dishpet said..
Hey guys, a young beginner here with a question regarding sail design.

I've attached a graph of an RSX 520 mast bend under different loads, but this question applies to all (most?) windsurf sails.
As you can see there is a big difference between the mast-bend of a rigged sail and the IMCS defining ~30kg load @50% of mast length.

If I were to make a sail from scratch - only having the mast, how would I go about defining the bend of the mast under which ill design my sail ? If I define my draft amount via additional luff curve and my draft posit. via broadseam for the ~30kg load as displayed, am I missing out on some black magic that is achieved by designing the sail around a bigger bend, and if so, is there an optimal bend ? Thanks.

I think I understand your graph , the left side shows that top of the mast is softer than bottom, that applies to all windsurfing masts, the NP ones are a little softer at the top, in comparison to some other makes, were the difference between the top and bottom is less.

And the right shows the resultant luff curve under different down haul and out haul tensions? What is missing is the luff curve of the sail without the mast as comparison, but I am guessing its pretty close to its curve under tension.

so to an the simple answer to 'is there an optimal bend ?' I would say the answer is NO! All the designs are a little bit different I think, even before you get into the different styles of sail, etc. Some have more luff curve and less broad seeming and vice versa.

Though the amount of luff curve versus the mast bend is probably representative of a big slalom sail, if that help?

Or if there is an optimal bend then it's a secret...

Select to expand quote

racerX said..

Dishpet said..
Hey guys, a young beginner here with a question regarding sail design.

I've attached a graph of an RSX 520 mast bend under different loads, but this question applies to all (most?) windsurf sails.
As you can see there is a big difference between the mast-bend of a rigged sail and the IMCS defining ~30kg load @50% of mast length.

If I were to make a sail from scratch - only having the mast, how would I go about defining the bend of the mast under which ill design my sail ? If I define my draft amount via additional luff curve and my draft posit. via broadseam for the ~30kg load as displayed, am I missing out on some black magic that is achieved by designing the sail around a bigger bend, and if so, is there an optimal bend ? Thanks.

I think I understand your graph , the left side shows that top of the mast is softer than bottom, that applies to all windsurfing masts, the NP ones are a little softer at the top, in comparison to some other makes, were the difference between the top and bottom is less.

And the right shows the resultant luff curve under different down haul and out haul tensions? What is missing is the luff curve of the sail without the mast as comparison.

so to an the simple answer to 'is there an optimal bend ?' I would say the answer is NO! All the designs are a little bit different I think, even before you get into the different styles of sail, etc. Some have more luff curve and less broad seeming and vice versa.

Though the amount of luff curve versus the mast bend is probably representative of a big slalom sail, if that help?

Or if there is an optimal bend then it's a secret...

Thanks for the reply, the only thing that comes to mind regarding "optimal bend" is that perhaps the more bent the mast and the luff curve of the sail is - the more the sail/mast form as a whole is rigid, but im pulling this one out of my a**.

This graph needs to have the axis' labelled, I'm assuming that the 'Y' axis is percentage length (a 'per unit' value) and that the 'X' axis is deflection measured as a 'pu' value of the percentage length. ie for the 31.2kg point load (assuming it is placed at the halfway length), the mast will deflect just under 4% of the mast length for the mast showing this graphed characteristic. How this data then translates into the sail design is past my knowledge as I have no clue about sail design.

Remember though, that this graph shows a bend created by a load at 90deg to the mast axis. The bend created by a sail being rigged is created by the mast being compressed with downhaul, although you would have to clarify what those other numbers/curves are actually showing (the ones that aren't the 31.2kg point load)

Select to expand quote

Paddles B'mere said..
This graph needs to have the axis' labelled, I'm assuming that the 'Y' axis is percentage length (a 'per unit' value) and that the 'X' axis is deflection measured as a 'pu' value of the percentage length. ie for the 31.2kg point load (assuming it is placed at the halfway length), the mast will deflect just under 4% of the mast length for the mast showing this graphed characteristic. How this data then translates into the sail design is past my knowledge as I have no clue about sail design.

Remember though, that this graph shows a bend created by a load at 90deg to the mast axis. The bend created by a sail being rigged is created by the mast being compressed with downhaul, although you would have to clarify what those other numbers/curves are actually showing (the ones that aren't the 31.2kg point load)

You got it right regarding the representation of the two axis.

Regarding the other curves - those show the real life mast bend of the rig (sail,mast,boom) under different outhaul/downhaul tensions. It's not the difference in those settings that confuse me, what confuses me is if you were to design a sail around a mast, how much of a bend would you use to design the luff curve of the sail around?

Select to expand quote

Dishpet said..

Paddles B'mere said..
This graph needs to have the axis' labelled, I'm assuming that the 'Y' axis is percentage length (a 'per unit' value) and that the 'X' axis is deflection measured as a 'pu' value of the percentage length. ie for the 31.2kg point load (assuming it is placed at the halfway length), the mast will deflect just under 4% of the mast length for the mast showing this graphed characteristic. How this data then translates into the sail design is past my knowledge as I have no clue about sail design.

Remember though, that this graph shows a bend created by a load at 90deg to the mast axis. The bend created by a sail being rigged is created by the mast being compressed with downhaul, although you would have to clarify what those other numbers/curves are actually showing (the ones that aren't the 31.2kg point load)

You got it right regarding the representation of the two axis.

Regarding the other curves - those show the real life mast bend of the rig (sail,mast,boom) under different outhaul/downhaul tensions. It's not the difference in those settings that confuse me, what confuses me is if you were to design a sail around a mast, how much of a bend would you use to design the luff curve of the sail around?

This diagram of the curves may help www.sailrepair.co.uk/makingsail.php , but keep in mind this RSX sail design is frozen in time, it's a pretty big sail with a lot of tension, designed with cams.

What missing from that diagram is the actual luff curve without any tension.

Interesting figure (from cmst.curtin.edu.au/wp-content/uploads/sites/4/2016/05/gourlay_2011_02_aero_hydrodynamics_of_an_RSX_Olympic_racing_sailboard.pdf in case someone is interested). It illustrates why we use rig winches .

Select to expand quote

Dishpet said..
... am I missing out on some black magic that is achieved by designing the sail around a bigger bend, and if so, is there an optimal bend ? Thanks.

The "black magic" that's achieved by higher rigging forces, which lead to a larger mast bend, is increased stability in stronger winds. The 9.5 m RS:X sails are still used in > 20 knots wind, when many recreational windsurfers would use a sail about half that size, so stability is very important. The same is true for other race sails. Cambers and a large number of battens also increase stability; you'll usually find all three used together.

What an "optimal" bend is depends a lot on what the sail is used for. For recreational sailors, ease of rigging (and therefore less tension and bend) may be more important than high-end stability. For wave and freestyle sailing, being able to "turn off" a sail in turns and tricks is very important, so you'll generally find sails with less mast bend, fewer battens, and no cambers. For just cruising around, something in the middle is often preferred - intermediate rig tension, 6 battens (compared to 3-5 on wave sails and 8+ on race sails), and no or 2-3 cambers (compared to typically 4 on a full race sail).

The luff curve has to take in the side ways bend in the mast in this equation. The sail will be designed for its optimal "flying" shape in its wind range with a optimal load produced by the sailor,down haul etc: (nominal 30kgs from memory). More down haul and the sail will lay off sideways more when sailing to depower decreasing angle of attack at the top of the rig and flattening at the same time.
"Flying shape' is the optimal depth, distribution of depth (draft fore and aft and top to bottom) and angle of attack whilst sailing. Windsurfing sails look very twisted at the top (angle of attack) on the beach but work when on the water when rig is loaded and sideways forces come into play.
I coach RSX and can say that they didn't get the 9.5m rig right from day one. Close, but could be better. The 8.5m rig is a good deal better in design scope.
The RSX rigs use an adjustable down haul that is changed on the go whilst racing and is critical to keeping an optimal flying shape. The difference in down haul is up to 6 cm whilst racing.
You have to consider and factor in the sideways component not just fore and aft.
Try pushing down on the tip of the mast (with weight on the base) on the beach when rigged and you see the sail start to get closer to the flying shape on the water.
I probably made it more confusing for you now.

Select to expand quote

Fez said..
The luff curve has to take in the side ways bend in the mast in this equation. The sail will be designed for its optimal "flying" shape in its wind range with a optimal load produced by the sailor,down haul etc: (nominal 30kgs from memory). More down haul and the sail will lay off sideways more when sailing to depower decreasing angle of attack at the top of the rig and flattening at the same time.
"Flying shape' is the optimal depth, distribution of depth (draft fore and aft and top to bottom) and angle of attack whilst sailing. Windsurfing sails look very twisted at the top (angle of attack) on the beach but work when on the water when rig is loaded and sideways forces come into play.
I coach RSX and can say that they didn't get the 9.5m rig right from day one. Close, but could be better. The 8.5m rig is a good deal better in design scope.
The RSX rigs use an adjustable down haul that is changed on the go whilst racing and is critical to keeping an optimal flying shape. The difference in down haul is up to 6 cm whilst racing.
You have to consider and factor in the sideways component not just fore and aft.
Try pushing down on the tip of the mast (with weight on the base) on the beach when rigged and you see the sail start to get closer to the flying shape on the water.
I probably made it more confusing for you now.

Good insights. What's wrong with the 9.5m rig?

Select to expand quote

Fez said..
The luff curve has to take in the side ways bend in the mast in this equation. The sail will be designed for its optimal "flying" shape in its wind range with a optimal load produced by the sailor,down haul etc: (nominal 30kgs from memory). More down haul and the sail will lay off sideways more when sailing to depower decreasing angle of attack at the top of the rig and flattening at the same time.
"Flying shape' is the optimal depth, distribution of depth (draft fore and aft and top to bottom) and angle of attack whilst sailing. Windsurfing sails look very twisted at the top (angle of attack) on the beach but work when on the water when rig is loaded and sideways forces come into play.
I coach RSX and can say that they didn't get the 9.5m rig right from day one. Close, but could be better. The 8.5m rig is a good deal better in design scope.
The RSX rigs use an adjustable down haul that is changed on the go whilst racing and is critical to keeping an optimal flying shape. The difference in down haul is up to 6 cm whilst racing.
You have to consider and factor in the sideways component not just fore and aft.
Try pushing down on the tip of the mast (with weight on the base) on the beach when rigged and you see the sail start to get closer to the flying shape on the water.
I probably made it more confusing for you now.

Awesome reply Fez, thanks.

Correct me if I'm wrong here - The sideways bend component of the mast/sail is taken care of at the top via additional downhaul. This reduces the depth at the top of the sail / reduces the Angle of Attack resulting in the de-powering at the top of the sail? Is this what you meant by saying that the luff curve has to take the side-ways bend of the mast into the equation ?

Select to expand quote

racerX said..

the left side shows that top of the mast is softer than bottom, .

Okee Doeck.. if we apply the bend curve formula:

1/4 = 2.4% of length
mid = 2.8% of length
3/4 = 2.3% of length

(2.4% x 520cm x 100) / (2.8% x 520cm) = (2.4 x 100) / 2.8 = 86%

(2.3% x 520cm x 100) / (2.8% x 520cm) = 2.3 x 100 / 2.8 = 82%

86% - 82% = 4 which is a... hard top ? shouldn't this be a flex top (> 16) ? (where am I going wrong ?)

Dishpet: Thanks for explaination. Your English is very good :)

Select to expand quote

jn1 said..

racerX said..

the left side shows that top of the mast is softer than bottom, .

Okee Doeck.. if we apply the bend curve formula:

1/4 = 2.4% of length
mid = 2.8% of length
3/4 = 2.3% of length

(2.4% x 520cm x 100) / (2.8% x 520cm) = (2.4 x 100) / 2.8 = 86%

(2.3% x 520cm x 100) / (2.8% x 520cm) = 2.3 x 100 / 2.8 = 82%

86% - 82% = 4 which is a... hard top ? shouldn't this be a flex top (> 16) ? (where am I going wrong ?)

Dishpet: Thanks for explaination. Your English is very good :)

I think you calcs are correct, but even if it is hard top, its softer at the top than at the bottom... The RSX is special mast, so maybe its different or the graph is wrong, or they had a bad sample, or its correct.

I think its more important, to concentrate on the basic principle that the mast and luff curve do not match on purpose. The top of windsurf mast has more curve in the top, and luff has more curve in the bottom. This gives the bottom of the sail depth, and top twists off flat. Broad seaming also does this as well. You could of course still achieve this intentional mismatch of the luff and mast curve, with a perfectly symmetrical mast curve, its the variation between the curves that gives the desired shape.

IMHO there is a lot a magic here, and magicians aren't going share this info, other than perhaps the principles...

Select to expand quote

jn1 said..

racerX said..

the left side shows that top of the mast is softer than bottom, .

Okee Doeck.. if we apply the bend curve formula:

1/4 = 2.4% of length
mid = 2.8% of length
3/4 = 2.3% of length

(2.4% x 520cm x 100) / (2.8% x 520cm) = (2.4 x 100) / 2.8 = 86%

(2.3% x 520cm x 100) / (2.8% x 520cm) = 2.3 x 100 / 2.8 = 82%

86% - 82% = 4 which is a... hard top ? shouldn't this be a flex top (> 16) ? (where am I going wrong ?)

Dishpet: Thanks for explaination. Your English is very good :)

If you're reading the ~31kg load curve I think the math is as follows:

1/4 = 2.5% = 130mm = 67% of mid deflection
mid = 3.7% = 192mm = 100% (You somehow read 2,8% from the mid point deflection on the chart)
3/4 = 2.8% = 146mm = 76% of mid deflection

So the bend curve number I got is 9 which is still within hardtop/cons.curve territory, but the chart is not accurate enough for these calculations since we're approximating the .x decimal.

They tested 3 of the same masts in the research paper and found differences in the bend curves - picture attached.

Select to expand quote

Dishpet said..

jn1 said..

racerX said..

the left side shows that top of the mast is softer than bottom, .

Okee Doeck.. if we apply the bend curve formula:

1/4 = 2.4% of length
mid = 2.8% of length
3/4 = 2.3% of length

(2.4% x 520cm x 100) / (2.8% x 520cm) = (2.4 x 100) / 2.8 = 86%

(2.3% x 520cm x 100) / (2.8% x 520cm) = 2.3 x 100 / 2.8 = 82%

86% - 82% = 4 which is a... hard top ? shouldn't this be a flex top (> 16) ? (where am I going wrong ?)

Dishpet: Thanks for explaination. Your English is very good :)

If you're reading the ~31kg load curve I think the math is as follows:

1/4 = 2.5% = 130mm = 67% of mid deflection
mid = 3.7% = 192mm = 100% (You somehow read 2,8% from the mid point deflection on the chart)
3/4 = 2.8% = 146mm = 76% of mid deflection

So the bend curve number I got is 9 which is still within hardtop/cons.curve territory, but the chart is not accurate enough for these calculations since we're approximating the .x decimal.

They tested 3 of the same masts in the research paper and foune differences in the bend curves - picture attached.

Actually in the report is says that they used mast with serial # 907334 the 'hard top' one from the table for the rest to the analysis and I assume the graph.

'Therefore all mast and sail analysis was subsequently done using a single mast, which is the first mast shown in Table 3'

Pardon my question, but after reading thru this thread I feel compelled to ask something.
You said "Hey guys, a young beginner here with a question regarding sail design.", followed by "If I were to make a sail from scratch".............
Do you really want to make a sail? There are many great used ones on the market that can save you a bunch of time, money, and research. Or are you a DIY guy who loves physics, research, and have the extra time and money, maybe even more than the act of windsurfing itself?
There really is no wrong answer, I just find some approaches to windsurfing by some people very curious.

Select to expand quote

Mastbender said..
Pardon my question, but after reading thru this thread I feel compelled to ask something.
You said "Hey guys, a young beginner here with a question regarding sail design.", followed by "If I were to make a sail from scratch".............
Do you really want to make a sail? There are many great used ones on the market that can save you a bunch of time, money, and research. Or are you a DIY guy who loves physics, research, and have the extra time and money, maybe even more than the act of windsurfing itself?
There really is no wrong answer, I just find some approaches to windsurfing by some people very curious.

Perfectly reasonable question Mastbender.

I'm an apprentice sailmaker for yachts/dinghies who has taken up windsurfing last year and got hooked. I've repaired numerous windsurfing sails and have always admired how complex yet simple they are. Windsurfing sails are like scaled-down versions of high-end regatta sails.

What I mean by that - they have full length tapered battens with tensioners perfectly shaped to amount and position of draft, different materials are used on a single sail according to the load-bearing points, the sails are rigged onto carbon masts and booms etc.

A regatta yacht mainsail with the mentioned "options" (minus the mast obviously) can easily cost you above 10,000$ , but in my humble opinion a windsurf sail is more interesting. And yes, I am aware of the way the windsurf sailmaking industry functions and how closed-circle it is, I'm not going into this to make a buck (yet ).

Forget all the mathematical mumbo jumbo. Since you have the facilities and skills, you will learn a lot more of practical value by:

- Decide what type and size of sail you are interested in making.
- find a very cheap or damaged one of that type.
- pick it apart and measure up or copy the panels
- Pay particular attention to any broad seam shaping (mm's definitely matter!), luff panel curve and Luff panel shaping!
- Make experimental changes and try your creation.

Simples!

Select to expand quote

jDishpet said..

Good pick up Dishpet :)

Select to expand quote

Dishpet said..

Mastbender said..
Pardon my question, but after reading thru this thread I feel compelled to ask something.
You said "Hey guys, a young beginner here with a question regarding sail design.", followed by "If I were to make a sail from scratch".............
Do you really want to make a sail? There are many great used ones on the market that can save you a bunch of time, money, and research. Or are you a DIY guy who loves physics, research, and have the extra time and money, maybe even more than the act of windsurfing itself?
There really is no wrong answer, I just find some approaches to windsurfing by some people very curious.

Perfectly reasonable question Mastbender.

I'm an apprentice sailmaker for yachts/dinghies who has taken up windsurfing last year and got hooked. I've repaired numerous windsurfing sails and have always admired how complex yet simple they are. Windsurfing sails are like scaled-down versions of high-end regatta sails.

What I mean by that - they have full length tapered battens with tensioners perfectly shaped to amount and position of draft, different materials are used on a single sail according to the load-bearing points, the sails are rigged onto carbon masts and booms etc.

A regatta yacht mainsail with the mentioned "options" (minus the mast obviously) can easily cost you above 10,000$ , but in my humble opinion a windsurf sail is more interesting. And yes, I am aware of the way the windsurf sailmaking industry functions and how closed-circle it is, I'm not going into this to make a buck (yet ).

At the end of the day that's what all the others do

Select to expand quote

sailquik said..
Forget all the mathematical mumbo jumbo. Since you have the facilities and skills, you will learn a lot more of practical value by:

- Decide what type and size of sail you are interested in making.
- find a very cheap or damaged one of that type.
- pick it apart and measure up or copy the panels
- Pay particular attention to any broad seam shaping (mm's definitely matter!), luff panel curve and Luff panel shaping!
- Make experimental changes and try your creation.

Simples!

At the end of the day that's what all the others do

woops, wrong quote first up

Good on you for exploring Dishpet. You are running over discovered territory. As RacerX eluded to, this is an esoteric discipline and it's one I don't understand, but to get really really good at something, you have to go through that process.

When I studied at college (TAFE as it's called here), I designed an instruction fetcher. This is a subsystem of a micro-processor and a subsystem that intrigued me at the time. It was quite a large and complicated circuit, but a necessary one. I did my research and as RacerX mentioned, you learn the "principles" only and so you end up having to fill in the gaps. I designed it to complete my understanding of machine concepts. I could've just read about it in a book and stopped there like everybody else does. Instead, I created one. So I totally get what you are trying to do. Keep up the good work !

Select to expand quote

jn1 said..
Good on you for exploring Dishpet. You are running over discovered territory. As RacerX eluded to, this is an esoteric discipline and it's one I don't understand, but to get really really good at something, you have to go through that process.

When I studied at college (TAFE as it's called here), I designed an instruction fetcher. This is a subsystem of a micro-processor and a subsystem that intrigued me at the time. It was quite a large and complicated circuit, but a necessary one. I did my research and as RacerX mentioned, you learn the "principles" only and so you end up having to fill in the gaps. I designed it to complete my understanding of machine concepts. I could've just read about it in a book and stopped there like everybody else does. Instead, I created one. So I totally get what you are trying to do. Keep up the good work !

You both make a good point, thanks for the kind words. Guess I'll start stalking the local used gear listing

I think windsurf sails and regatta sails are similar in some ways but wildly different in others.

Over the years I have sewn lots of things, some of which perform very well compared to "off the shelf products"- however I would never ever contemplate sewing a windsurf sail.

The design process is a fascinating one, in that each product has a very particular layout/design/construction/geometry that will reach an optimal shape over many iterations....whether this be a 29er mountainbike, a backpack, snow skis or windsurf sail. How a particular item will look in a highly refined iteration is very hard to predict, even with a good knowledge of the principles involved (I wonder, for example, how the "final" foil iterations will look in 5-7 years time).

The "static" mast load from a graph only gives a small piece of the puzzle- especially when compared the flex characteristics of the product in use. The fore/aft and lateral deflection and response of a boat sail is massively different to that of a windsurf mast, and even two windsurf masts that have the same static load pattern could perform very differently when rigged up.

The mast in a windsurf sail is probably doing a lot of what a vang would do as far as affecting upper leech twist and hence draft position in the body of the sail.

Part of the interest with windsurfing sails is this incredible complexity in design. Having seen the evolution of windsurfing sails since the 80s, and heard of various failures, trials and "offshoots" in design- it is indeed an amazing story. The "bird wing" geometry of the ART sails in the early 90s (as a method of twisting off a loose leech) and how this in some ways evolved into the general "loose leech" designs of the current day is one such story.

I admire your approach and interest, but suspect that looking at existing designs and using them is probably a far better place to start in this process.

Clarence