Has anyone ever made a composite mast using glass fibre, or considered it? Seems like a good idea to me, I don't need or want CF but am experienced with vinly ester or epoxy resin and stiched glass fabrics. Have nearly completed a 48' cruising yacht, it's a classic yacht replica type sloop, long overhangs etc. With a composite mast I can easily taper it etc etc. Any help appreciated or shoot me down.
The Boat Design & Boatbuilding Site
Epoxy/glass mast for a cruising yacht?
- Thread starter Robjl
- Start date
Inquisitor
BIG ENGINES: Silos today... Barn Door tomorrow!
- Joined
- Nov 1, 2005
- Posts
- 276
- Reaction score
- 11
- Location
- North Carolina Mountains
If you want backyard mechanics... I ran across this site during a related research. It might be of interest...
http://www.geocities.com/Yosemite/3387/mast.htm
If you are of a more technical bent, here are some key things...
Bending stiffness is a function of EI. (Stiffness * Cross Sectional Property)
For a round cross section “I” is a function of r^4. (radius to the 4th power)
Thus a trade study can easily be set up using:
rg = rc*(Ec/Eg)^0.25 c & g are subscripts for carbon and glass.
Say someone is making a mast with carbon that is 3 times stiffer than glass. Using the same lay-up you could theoretically get the same stiffness by just upping the diameter by 32%!
The obligatory disclaimer: This highly simplifies the total problem. Among other things these need to be checked: Beam buckling (better for glass case), Local wall buckling (worse for glass case) and weight aloft (worse for glass case).
Although highly technical, the high points can be lifted. The main one in your case might be a 5 to 10% chord mast that has enough cross sectional chord and width to achieve the stiffness you require.
Hope this helps!
http://www.geocities.com/Yosemite/3387/mast.htm
If you are of a more technical bent, here are some key things...
- “Cheap" (relative term) Carbon Fiber is only twice as stiff as Glass. You can pay for a stiffness that is a multiple of 4+, but you'll pay dearly. I've designed fiber structure for satellites that have extremely high stiffness but costs in the neighborhood of $10K/lb. America’s Cup Syndicates probably do too!
Bending stiffness is a function of EI. (Stiffness * Cross Sectional Property)
For a round cross section “I” is a function of r^4. (radius to the 4th power)
Thus a trade study can easily be set up using:
rg = rc*(Ec/Eg)^0.25 c & g are subscripts for carbon and glass.
Say someone is making a mast with carbon that is 3 times stiffer than glass. Using the same lay-up you could theoretically get the same stiffness by just upping the diameter by 32%!
The obligatory disclaimer: This highly simplifies the total problem. Among other things these need to be checked: Beam buckling (better for glass case), Local wall buckling (worse for glass case) and weight aloft (worse for glass case).
- Another aspect, since you are building a “classic” design, a rotating mast may be out of character. However, if you can stand non-traditional mast and the complexity of a rotating mast, here is another good paper, I found…
Although highly technical, the high points can be lifted. The main one in your case might be a 5 to 10% chord mast that has enough cross sectional chord and width to achieve the stiffness you require.
Hope this helps!
Thanks for the reply Inquisitor,
Yes, I've read the sites you mention. I think I need to feel a little more confident about the design before building a backyard design. In my case a mast failure would have serious consequences for those aboard. My math is not up to the task though. I've built three 40' cruising yachts, (including the rigs) so far and have cruised each before starting another, I'm still married too. I think I have a good idea of the Aluminium mast size needed. About 240mm x 160mm oval shape 4-4.5mm wall thickness. This has worked in the past and I expect would work again on the new boat. But the untapered auminium stick just didn't do it for me. This section weighs about 7.5kg/metre in aluminium. Some slightly heavier sections are currently available here, up to 9Kg/metre. But in most cases I would need to buy it in two sections to transport it. I don't like the mid joint. The figures that I can find on the strengths of Alumin versus epoxy/glass lead me to believe that if I increase the cross section by 50% I could get the same result with slightly less weight. Even less overall weight if the thickness of the spar is "Thicknessed" to the compression loads in each section. The rig is a double (parallel) spreader non rotating mast(deck stepped) shroud base is 1.5metre on deck 6.4 metres to lower spreaders then 5.5 metres to top spreaders then 4.5 metres to cap shroud attachment then .6 metres to the masthead fitting: 17 metres total length. I have (with backyard maths) come up with a composite layup of 5000gms/squ-metre 60% in unidirectional and 40% Double Bias (45 deg) If I vacuum bag or infuse I think the section will be about 6mm thick. The cross section area about 3800sq-mm.
Any comments appreciated.
Yes, I've read the sites you mention. I think I need to feel a little more confident about the design before building a backyard design. In my case a mast failure would have serious consequences for those aboard. My math is not up to the task though. I've built three 40' cruising yachts, (including the rigs) so far and have cruised each before starting another, I'm still married too. I think I have a good idea of the Aluminium mast size needed. About 240mm x 160mm oval shape 4-4.5mm wall thickness. This has worked in the past and I expect would work again on the new boat. But the untapered auminium stick just didn't do it for me. This section weighs about 7.5kg/metre in aluminium. Some slightly heavier sections are currently available here, up to 9Kg/metre. But in most cases I would need to buy it in two sections to transport it. I don't like the mid joint. The figures that I can find on the strengths of Alumin versus epoxy/glass lead me to believe that if I increase the cross section by 50% I could get the same result with slightly less weight. Even less overall weight if the thickness of the spar is "Thicknessed" to the compression loads in each section. The rig is a double (parallel) spreader non rotating mast(deck stepped) shroud base is 1.5metre on deck 6.4 metres to lower spreaders then 5.5 metres to top spreaders then 4.5 metres to cap shroud attachment then .6 metres to the masthead fitting: 17 metres total length. I have (with backyard maths) come up with a composite layup of 5000gms/squ-metre 60% in unidirectional and 40% Double Bias (45 deg) If I vacuum bag or infuse I think the section will be about 6mm thick. The cross section area about 3800sq-mm.
Any comments appreciated.
Inquisitor
BIG ENGINES: Silos today... Barn Door tomorrow!
- Joined
- Nov 1, 2005
- Posts
- 276
- Reaction score
- 11
- Location
- North Carolina Mountains
Are you sure you really want to know?
First off, I’ll tell you my background so you can weigh my input accordingly. I am not an old salt and even several nautical terms you used, I had to go look up. Although, I have only designed and am building a 20’ beach catamaran, my next project will be a ~40’ cruising catamaran. I’d like to pursue designing and building a composite mast for both. Although the rig I plan for it is radically non-conventional, many of the design criteria will be the same as you need to address.
I was in the aerospace field. I designed composite structures for extremely low weight and with no safety factors. If we keep this thread going someone with more direct experience may weigh-in and help us or at least keep us from overlooking something important.
There are three schools of thought here. (1) Radical (I had to work this way in A.S.) – try to nail down every possible loading condition, every environmental condition, life expectancy and so forth and so on till we drill ourselves down and come out in Australia or in your case Atlanta! (2) Conservative – we build so that it mimics the metal structure. Rarely does this reduce weight significantly (or at all), but it gives a baseline. Besides, you may not really be looking for that last pound of savings or last psi of stiffness. (3) Start with 2 and evolve it with more specific knowledge and taking advantage of a laminates ability to readily change over the mast’s length. This is the one, I think we should take.
Hopefully you have access to Microsoft XL so we don’t have to try and write equations here. Plus, it allows for dynamic changes… for us and anyone else that wants to participate. I’m going to use inches and pounds as I can’t get a feel for it in metric. If someone else is interested in adding to it, they could flesh-out the metric columns next to the English units. I’ll leave room. Who knows… someone might take pity on us and just supply us with a better spreadsheet.
Anyway, this first-cut is based on simple mechanics of materials approaches. I’m currently reading “Principles of Yacht Design” by Lars Larsson & Rolf E Eliasson and I hope it will shed some more real world experience into our design.
See attached Spreadsheet. It might be a little daunting depending on your background, but apparently, you were doing some calculations concerning a fiberglass mast. If you need some clarification, let me know.
For a reality check, I’ve plugged-in your aluminum mast’s geometry into the spreadsheet. Using the same geometry (with your 4.5 mm wall thickness) it requires the longitudinal plies of a glass/epoxy design to have a thickness of 5.9 mm. This makes sense, since the glass/epoxy is less stiff than the aluminum and the design is driven by Euler buckling.
Placing your proposed 50% increase in the spreadsheet allows the thickness to be decreased to 1.63 mm. (Note again: This is only the zero degree plies)
It’s beyond the capability of the spreadsheet to do Laminated Plate Theory (LPT) calculations. So based on the zero degree plies needing to account for 1.63 mm, I’ve created a laminate using 0.008” plies with a stacking sequence of [90/45/0/0/0/-45/0]s for a total thickness of 0.112” (2.8 mm). Using an LPT program I have access to, resulted in a laminate with the following properties:
Exx = 5.9 Msi
Eyy = 3.7 Msi
Gxy = 1.6 Msi
thick = 0.112 in
SxxC = 22.6 Ksi (First resin failure)
SxxC = 81.7 Ksi (First fiber failure)
Plugging these back into the spreadsheet to get margins, we get…
Things to be concerned about
Things could definitely be improved; both the reduction in weight and even safety factors if the actual loads on the mast could be determined. You may even be able to go to a single spreader (if desirable). If you have some ideas on stay and sail loads, we could tackle this problem also.
First off, I’ll tell you my background so you can weigh my input accordingly. I am not an old salt and even several nautical terms you used, I had to go look up. Although, I have only designed and am building a 20’ beach catamaran, my next project will be a ~40’ cruising catamaran. I’d like to pursue designing and building a composite mast for both. Although the rig I plan for it is radically non-conventional, many of the design criteria will be the same as you need to address.
I was in the aerospace field. I designed composite structures for extremely low weight and with no safety factors. If we keep this thread going someone with more direct experience may weigh-in and help us or at least keep us from overlooking something important.
There are three schools of thought here. (1) Radical (I had to work this way in A.S.) – try to nail down every possible loading condition, every environmental condition, life expectancy and so forth and so on till we drill ourselves down and come out in Australia or in your case Atlanta! (2) Conservative – we build so that it mimics the metal structure. Rarely does this reduce weight significantly (or at all), but it gives a baseline. Besides, you may not really be looking for that last pound of savings or last psi of stiffness. (3) Start with 2 and evolve it with more specific knowledge and taking advantage of a laminates ability to readily change over the mast’s length. This is the one, I think we should take.
Hopefully you have access to Microsoft XL so we don’t have to try and write equations here. Plus, it allows for dynamic changes… for us and anyone else that wants to participate. I’m going to use inches and pounds as I can’t get a feel for it in metric. If someone else is interested in adding to it, they could flesh-out the metric columns next to the English units. I’ll leave room. Who knows… someone might take pity on us and just supply us with a better spreadsheet.
Anyway, this first-cut is based on simple mechanics of materials approaches. I’m currently reading “Principles of Yacht Design” by Lars Larsson & Rolf E Eliasson and I hope it will shed some more real world experience into our design.
See attached Spreadsheet. It might be a little daunting depending on your background, but apparently, you were doing some calculations concerning a fiberglass mast. If you need some clarification, let me know.
For a reality check, I’ve plugged-in your aluminum mast’s geometry into the spreadsheet. Using the same geometry (with your 4.5 mm wall thickness) it requires the longitudinal plies of a glass/epoxy design to have a thickness of 5.9 mm. This makes sense, since the glass/epoxy is less stiff than the aluminum and the design is driven by Euler buckling.
Placing your proposed 50% increase in the spreadsheet allows the thickness to be decreased to 1.63 mm. (Note again: This is only the zero degree plies)
It’s beyond the capability of the spreadsheet to do Laminated Plate Theory (LPT) calculations. So based on the zero degree plies needing to account for 1.63 mm, I’ve created a laminate using 0.008” plies with a stacking sequence of [90/45/0/0/0/-45/0]s for a total thickness of 0.112” (2.8 mm). Using an LPT program I have access to, resulted in a laminate with the following properties:
Exx = 5.9 Msi
Eyy = 3.7 Msi
Gxy = 1.6 Msi
thick = 0.112 in
SxxC = 22.6 Ksi (First resin failure)
SxxC = 81.7 Ksi (First fiber failure)
Plugging these back into the spreadsheet to get margins, we get…
- You would hear matrix cracking at 60% of the yield strength of aluminum.
- Final failure (mast breaking away) at 210% of the yield strength of aluminum.
- Euler Buckling at 130% of the aluminum mast
- Bending stiffness 130% of the aluminum mast.
- Weight 69% of the aluminum mast.
Things to be concerned about
- Some of the material properties are questionable. They came from aerospace grade pre-preg material, vacuum bagged and autoclaved at 350F and 100 psi pressure. Doing that with room temperature resins will obviously be impossible. This should not affect the buckling and bending stiffness numbers much. But it would most likely reduce the strength numbers by some. It’s hard to tell, until actual properties are generated with the glass/epoxy chosen.
- Also, the numbers are based on achieving 78% fiber volume in the laminate. Again, this is very hard to achieve even with an autoclave and pre-pregs.
- Local buckling was not checked as doing that for an orthotropic laminate is rather difficult and I currently don’t have the equations available to run that. However, having the 90 plies on the outside reduce wall buckling considerably. And if really concerned using rigid foam for a core would totally eliminate any possibility of wall buckling.
- Getting the 0 degree plies absolutely strait. When I finally get around to trying it, I will try to make some end pieces so that I can stretch the plies while they’re curing to ensure they are absolutely strait.
Things could definitely be improved; both the reduction in weight and even safety factors if the actual loads on the mast could be determined. You may even be able to go to a single spreader (if desirable). If you have some ideas on stay and sail loads, we could tackle this problem also.
Thanks for all the work Inquisitor. But the math is my achilles heel, I'm not sure of the units you use for cross section properties eg x_g = 14.1732. What is 14.1732? After I re-read my #3 post I noticed another piece that I had omitted...the righting moment at 60-90 degrees (maximum) is 9.50 m-MT. I read somewhere.. enlightened I thought.. that when calculating loads on a mast a simple tack is to use the righting moment. Sounds logical to me as if this load is exceeded the yacht simply lays over. Add a safety factor of x5 ?
Will it work?? Oh, Princ of Yacht Design is a gem isn't it. I'll contact our local composite materials supplier next week to see if they can give me any strengths on available materials.
Regards.
Will it work?? Oh, Princ of Yacht Design is a gem isn't it. I'll contact our local composite materials supplier next week to see if they can give me any strengths on available materials.
Regards.
water addict
Naval Architect
- Joined
- Jun 30, 2004
- Posts
- 325
- Reaction score
- 6
- Location
- maryland
Why do you want to do it out of glass? You will likely have something that is heavier than aluminum, cost quite a bit anyway by the time you outfit it with hardware, take lots of time if you do it yourself, and likely be of lower structural reliability- because of the detailing required around load bearing fittings. There are any number of aluminum spar makers out there that will provide a quality spar that will be professionally done. Glass is a bad choice for mast material. Carbon has found some favor in the high price world because of weight savings. Glass won't give weight savings, and it will be just as hard to manufacture as carbon.
Last edited:
Water addict, I still don't get the aversion to glass, by my calcs an epoxy/glass mast won't be heavier than an aluminium stick, not a great weight saving I agree but that's not my prime motive, even when using CF the weight saving is not great when the overall weight of standing & running rigging, spreaders, masthead fittings etc and sails are considered. Won't the composite mast in epoxy/glass have better fatigue resistance, also it can easily be strengthened around shroud and spreader attachment points. It can also be tapered easily and will probably hold a paint finish better than I've ever been able to get on aluminium. Also superior corrosion resistance. As to the fitting out with hardware... won't that be the same regardless of the mast material. And my time to make it, well I just love the challenge, my time costs me nothing. I'm interested in the detailing you mention around load fittings, can anyone help me with any references or pics of examples? I'm still hopeful. Thanks all.
Inquisitor
BIG ENGINES: Silos today... Barn Door tomorrow!
- Joined
- Nov 1, 2005
- Posts
- 276
- Reaction score
- 11
- Location
- North Carolina Mountains
I didn't just fall off the turnip truck...
... bang my head, and dream myself a rocket scientist.
Dear water addict,
I have always believed that anyone can learn from another. I am open to it. I learned things last week from illegal Mexican labors! An opportunity for an experienced Naval Architect like yourself to see through a fresh pair of eyes of an aerospace engineer might not bare any fruit, but you’ll never know if you have don't have an open mind. This A.E. can certainly learn from and would soak up any wisdom an N.A. cares to offer. But to address your issues…
Carbon
Carbon has found favor mainly in the racing crowd. For the short life of a competitive race boat, it is the perfect material. Its stiffness and lightness can easily build a superior mast. However, It has several major strikes against it for any longevity and thus any practicality for us wanting a low maintenance, reliable mast. (1) The main one would be the ability to attach to it. If any aluminum touches the carbon, you have a battery that erodes the material till something fails. (2) The act of fastening. Drilling any holes is a bad thing. Continuous fiber structures DEPEND on their fibers being continuous for their strength and longevity. Yes, they can be reinforced successfully, however, bonding is far more effective. Oh and yes, aluminum is far superior in this regard! (3) Another strike against carbon is its extreme stiffness as compared to its resin. This, over time and cyclic loading will cause microcracks in the resin. It’s the equivalent of aluminum’s fatigue. In my opinion, an improperly designed (and most… properly designed) carbon masts have a defined life expectancy. I’d never purchase a boat with a carbon mast! Unless, I was Turner and wanting to go racing.
Aluminum
Great material. That built-in track is a blessing! But it’s not perfect either. I’ve seen fasteners ripped out. It’s near-on impossible to add reinforcement in any effective way. The only answer is to distribute the load more. In some cases you are carrying a wall thickness over the entire mast just to attach a fastener to it in one location! I’ve seen if fatigue and fail. It does corrode. I haven’t seen any variable geometry… tapered or cross section (CS) changes in an aluminum mast. So you have a 10” oval mast with a ¼” wall where you need it, but you also have a 10” oval mast with a ¼” wall where a 1” mast would do just fine! With all that, I still think it’s the mast of choice for just about any boat and most people! In my personal case, it’s trying to get it. Even here in Atlanta, I can’t just go anywhere and pick up a 50’, 60’, 70’+ aluminum mast! And if I could, it would be rather expensive. Like Robjl, I don’t like the idea of shipping pieces and welding/bolting/or even bonding them together.
Fiberglass
My whole career was spent using carbon, boron, Kevlar and more exotic fibers and matrix. My total use of fiberglass was as a barrier to eliminate the dielectric problem described above. I think fiberglass has a legacy burden of all the chopped fiber boat builders to overcome. Fiberglass is far more forgiving than carbon. The fiber is as strong as most and stronger than some carbon fibers. It fatigue strength is better than aluminum and as good as carbon. In a laminate, its properties being closer to epoxy provides a more forgiving material. However, its main advantages are: (1) the main one – You can change the cross-section. (2) There are an infinite number of possible designs to tune to the exact problem instead of the limited number of sections from our friendly Alcoa. (3) You can bury reinforcements at specific locations to make it cosmetically and structurally aesthetic. (4) It can be made lighter than Aluminum. (5) It can be made stiffer than Aluminum. (6) It can be made stronger than Aluminum. (7) It can be made to last longer than Aluminum. (8) With a proper design, anyone who is proficient laying-up fiberglass can build this in their basement! (9) The price difference between an extruded 10” oval, ¼” wall thickness that is 60’ long delivered to my house versus some 200 lbs +/- worth of fiberglass and epoxy delivered to my house! (-1) THE ONLY ONE I’ll give you is it takes more of my time!
I hope the purpose of this thread is to spread and foster ideas. Advances often come from throwing out old concepts and using a fresh sheet of paper. Advances CAN come from someone who doesn’t have a clue about your industry and blends some pieces in from his own! With a mentor, with the expertise in the industry, advances are a certainty. Now, don’t get me wrong about old concepts... I’ve been on and love the experience of an all-WOOD boat. Nothing fiberglass (or aluminum for that matter) can give you that feeling!
I hope the purpose of this thread will be to blend ideas. With the support of experienced people such as yourself to keep us from bouncing off the walls, this could have some far reaching goals. By using a XL spreadsheet, I hope the end result will be an analysis tool that is both flexible and useful. If not, it might at least provide an analysis methodology. At the very least, it’s an interesting problem to solve.
As our first trivial cut illustrates, we’ve already save 30% of the weight of the aluminum. It does not take into account track hardware yet, but it also doesn’t have variable cross section yet either! It is already stiffer and stronger than the aluminum one. When all is said and done, I am quite confident; we can easily exceed an aluminum mast on ALLpoints. With your help, I’m sure we can!
Looking at your credentials, I’m sure you understand:
1. Mast design is driven by bending stiffness…
2. Bending Stiffness = function of (EI)
3. I = function of (r^4)
Therefore: Bending Stiffness = function of (Er^4)
Therefore: Its FAR, FAR easier to get stiffness with r than E.
Therefore: So glass/epoxy’s weaker E can easily be made up for with r.
Aluminum can NOT take advantage of this principle ONLY because you can’t change its cross section!!!! You are stuck with the same section at the top as you have to have at the bottom! Thus you are carrying a tremendous burden just to use Aluminum. Glass/Epoxy does not have that burden.
Looking forward to your mentoring,
Inquisitor
P.S. There are many homebuilt/production aircraft out there (some that can exceed 300 knots) that use fiberglass exclusively! I can assure you (because it is my field) that their design is mainly driven by stiffness. They must have things fastened to them. They must not fall out of the sky. They must have a useful life expectancy. I can assure you the fiberglass that you are thinking about is not in this mast design!
... bang my head, and dream myself a rocket scientist.
Dear water addict,
I have always believed that anyone can learn from another. I am open to it. I learned things last week from illegal Mexican labors! An opportunity for an experienced Naval Architect like yourself to see through a fresh pair of eyes of an aerospace engineer might not bare any fruit, but you’ll never know if you have don't have an open mind. This A.E. can certainly learn from and would soak up any wisdom an N.A. cares to offer. But to address your issues…
Carbon
Carbon has found favor mainly in the racing crowd. For the short life of a competitive race boat, it is the perfect material. Its stiffness and lightness can easily build a superior mast. However, It has several major strikes against it for any longevity and thus any practicality for us wanting a low maintenance, reliable mast. (1) The main one would be the ability to attach to it. If any aluminum touches the carbon, you have a battery that erodes the material till something fails. (2) The act of fastening. Drilling any holes is a bad thing. Continuous fiber structures DEPEND on their fibers being continuous for their strength and longevity. Yes, they can be reinforced successfully, however, bonding is far more effective. Oh and yes, aluminum is far superior in this regard! (3) Another strike against carbon is its extreme stiffness as compared to its resin. This, over time and cyclic loading will cause microcracks in the resin. It’s the equivalent of aluminum’s fatigue. In my opinion, an improperly designed (and most… properly designed) carbon masts have a defined life expectancy. I’d never purchase a boat with a carbon mast! Unless, I was Turner and wanting to go racing.
Aluminum
Great material. That built-in track is a blessing! But it’s not perfect either. I’ve seen fasteners ripped out. It’s near-on impossible to add reinforcement in any effective way. The only answer is to distribute the load more. In some cases you are carrying a wall thickness over the entire mast just to attach a fastener to it in one location! I’ve seen if fatigue and fail. It does corrode. I haven’t seen any variable geometry… tapered or cross section (CS) changes in an aluminum mast. So you have a 10” oval mast with a ¼” wall where you need it, but you also have a 10” oval mast with a ¼” wall where a 1” mast would do just fine! With all that, I still think it’s the mast of choice for just about any boat and most people! In my personal case, it’s trying to get it. Even here in Atlanta, I can’t just go anywhere and pick up a 50’, 60’, 70’+ aluminum mast! And if I could, it would be rather expensive. Like Robjl, I don’t like the idea of shipping pieces and welding/bolting/or even bonding them together.
Fiberglass
My whole career was spent using carbon, boron, Kevlar and more exotic fibers and matrix. My total use of fiberglass was as a barrier to eliminate the dielectric problem described above. I think fiberglass has a legacy burden of all the chopped fiber boat builders to overcome. Fiberglass is far more forgiving than carbon. The fiber is as strong as most and stronger than some carbon fibers. It fatigue strength is better than aluminum and as good as carbon. In a laminate, its properties being closer to epoxy provides a more forgiving material. However, its main advantages are: (1) the main one – You can change the cross-section. (2) There are an infinite number of possible designs to tune to the exact problem instead of the limited number of sections from our friendly Alcoa. (3) You can bury reinforcements at specific locations to make it cosmetically and structurally aesthetic. (4) It can be made lighter than Aluminum. (5) It can be made stiffer than Aluminum. (6) It can be made stronger than Aluminum. (7) It can be made to last longer than Aluminum. (8) With a proper design, anyone who is proficient laying-up fiberglass can build this in their basement! (9) The price difference between an extruded 10” oval, ¼” wall thickness that is 60’ long delivered to my house versus some 200 lbs +/- worth of fiberglass and epoxy delivered to my house! (-1) THE ONLY ONE I’ll give you is it takes more of my time!
I hope the purpose of this thread is to spread and foster ideas. Advances often come from throwing out old concepts and using a fresh sheet of paper. Advances CAN come from someone who doesn’t have a clue about your industry and blends some pieces in from his own! With a mentor, with the expertise in the industry, advances are a certainty. Now, don’t get me wrong about old concepts... I’ve been on and love the experience of an all-WOOD boat. Nothing fiberglass (or aluminum for that matter) can give you that feeling!
I hope the purpose of this thread will be to blend ideas. With the support of experienced people such as yourself to keep us from bouncing off the walls, this could have some far reaching goals. By using a XL spreadsheet, I hope the end result will be an analysis tool that is both flexible and useful. If not, it might at least provide an analysis methodology. At the very least, it’s an interesting problem to solve.
As our first trivial cut illustrates, we’ve already save 30% of the weight of the aluminum. It does not take into account track hardware yet, but it also doesn’t have variable cross section yet either! It is already stiffer and stronger than the aluminum one. When all is said and done, I am quite confident; we can easily exceed an aluminum mast on ALLpoints. With your help, I’m sure we can!
Looking at your credentials, I’m sure you understand:
1. Mast design is driven by bending stiffness…
2. Bending Stiffness = function of (EI)
3. I = function of (r^4)
Therefore: Bending Stiffness = function of (Er^4)
Therefore: Its FAR, FAR easier to get stiffness with r than E.
Therefore: So glass/epoxy’s weaker E can easily be made up for with r.
Aluminum can NOT take advantage of this principle ONLY because you can’t change its cross section!!!! You are stuck with the same section at the top as you have to have at the bottom! Thus you are carrying a tremendous burden just to use Aluminum. Glass/Epoxy does not have that burden.
Looking forward to your mentoring,
Inquisitor
P.S. There are many homebuilt/production aircraft out there (some that can exceed 300 knots) that use fiberglass exclusively! I can assure you (because it is my field) that their design is mainly driven by stiffness. They must have things fastened to them. They must not fall out of the sky. They must have a useful life expectancy. I can assure you the fiberglass that you are thinking about is not in this mast design!
water addict
Naval Architect
- Joined
- Jun 30, 2004
- Posts
- 325
- Reaction score
- 6
- Location
- maryland
Inquisitor said:
Yes you can gain stiffness by changing the cross section and get it back with larger r. You can also do that with aluminum. But you will have a dogmeat slow, dragging mast that creates a lot of turbulence. Also you can get tapered aluminum spars- its been done for decades.
I work in composites, steel, and aluminum structures. we do testing of all. The biggest problem with composites is quality control. You can take test specimens from a prof. manufacturer that are supposed to be the same, and get structural properties that look like a shotgun blast. So in order to overcome this, most safety factors for composites are double that of metals. By the time you do this with glass, you get a part that doesn't save you any weight. Carbon spar manuf. claims are savings of 30-50% in weight over Al. I'm sure they try every trick to maximize the weight savings, since that is the principal selling point over aluminum. So how you can get a comparably shaped glass composite section (ie not twice the diameter of the AL spar) that saves you any significant weight is a good trick. Post your calcs when you are done, I'd love to see them- perhaps I can learn a thing or two.
Of course if you want to build a spar of glass/epoxy cause it will be fun for you, by all means go for it! I'm in your corner on that!
Inquisitor
BIG ENGINES: Silos today... Barn Door tomorrow!
- Joined
- Nov 1, 2005
- Posts
- 276
- Reaction score
- 11
- Location
- North Carolina Mountains
Ooops...
Robjl,
... I got a little carried away in the math!
I’ll take it as an action item to clean that up and make it more human friendly.
I think this can be a great thread that will have many far-reaching advantages. Hail, that’s why I want to make it general, so everyone (including me) can use it. With more people offering constructive criticism, I know we might see our work in production some day.
I’m comfortable with the math and engineering techniques required. But I also know that all the fancy math in the world is worthless without good input. And I know that is where I’m not connected!… at least in this industry. None of us is going to have access to a 70’ autoclave so we can use pre-preg material in a clean room.
This is to everyone who might be willing to contribute,
The two things we really need that I don’t have access to:
1) Loads – I think what you have supplied is a great start. I’ll work at it backwards from the point that any load greater than X will just push the boat over. It’s a self limiting system. This covers most things, but shock loading will be the hardest to get our hands around. And from what I understand, a catamaran (what I want) has higher shock loads.
2) Material Properties – We need to get some properties so that I can plug them into the spreadsheet and Laminated Plate Theory (LPT) programs.
A little LPT 101 lesson for those interested:
LPT uses mathematics. It’s not real complex. No irrational math. No calculus. No differential equations. It is mostly simple matrix manipulations… adding, multiplying, inverting matrices. It predicts the behavior (stiffness and strength (S&S)) of a generalized stack of plies that can be any material, in any orientation. It is considered to be fairly accurate. Meaning… it’s fine for safety factors of 3 or more. Building satellites, we designed with the attitude; “we ain’t got no stinking safety factors!” Couldn’t afford them. Every ounce we had extra in structure meant one less ounce that couldn’t go to optics, electronics and such. We always tested for laminate (the exact stacking sequence) cut up into coupons and pulled.
As input, it needs, the S&S properties of a lamina (not laminate. One ply’s S&S properties) of the materials used. There are two way of getting lamina properties…
(1) Micromechanics – this is a theoretical way of calculating those properties. Basically (although there are finer points) it’s just a weighted averaging of the fiber properties and resin properties. In AE, we felt comfortable doing trade studies for a proposal, but would do coupon testing of the actual material combination before committing a design to production.
(2) Making coupons out of nothing but unidirectional material with all plies oriented in the zero direction. Some would be cut out along the “grain” and some across the grain. These are then pulled, pushed, sheared to get the properties.
Unfortunately, to do it reliably, statistics requires many samples. A-basis is basically some “fudge factor” based on the number of samples. The more samples, the tighter the estimate. The fewer samples, the “gray” area is wider.
Hopefully, someone out there can help us with these properties. If so, any help would be appreciated.
In LPT parlance, we need:
Subscripts:
1 – this is the fiber direction of the lamina (one ply).
2 – this is transverse to the fiber direction. Often called the matrix direction.
T – In Tension
C – In Compression (Most times properties are different in tension/compression)
E11T – Stiffness in the fiber direction, in tension.
E11C – Hopefully, you get the idea.
E22T
E22C
G12 – Shear stiffness in the 1-2 plane
v12 – Strain in the 2 direction based on straining it in the 1 direction. You know when things stretch they get thinner. That’s what this number describes. It usually uses the Greek symbol Nu.
S11T – Axial strength in the fiber direction/tension
S11C - Hopefully, you get the idea.
S22T
S22C
T12 – Shear strength of the lamina in the 1-2 plane.
Any values would be great, however, since the stays of a mast system tend to compress the mast, I believe, the compressive values are by far the most important!
Thanks to all willing to help.
Robjl,
... I got a little carried away in the math!
I’ll take it as an action item to clean that up and make it more human friendly.
I think this can be a great thread that will have many far-reaching advantages. Hail, that’s why I want to make it general, so everyone (including me) can use it. With more people offering constructive criticism, I know we might see our work in production some day.
I’m comfortable with the math and engineering techniques required. But I also know that all the fancy math in the world is worthless without good input. And I know that is where I’m not connected!… at least in this industry. None of us is going to have access to a 70’ autoclave so we can use pre-preg material in a clean room.
This is to everyone who might be willing to contribute,
The two things we really need that I don’t have access to:
1) Loads – I think what you have supplied is a great start. I’ll work at it backwards from the point that any load greater than X will just push the boat over. It’s a self limiting system. This covers most things, but shock loading will be the hardest to get our hands around. And from what I understand, a catamaran (what I want) has higher shock loads.
2) Material Properties – We need to get some properties so that I can plug them into the spreadsheet and Laminated Plate Theory (LPT) programs.
A little LPT 101 lesson for those interested:
LPT uses mathematics. It’s not real complex. No irrational math. No calculus. No differential equations. It is mostly simple matrix manipulations… adding, multiplying, inverting matrices. It predicts the behavior (stiffness and strength (S&S)) of a generalized stack of plies that can be any material, in any orientation. It is considered to be fairly accurate. Meaning… it’s fine for safety factors of 3 or more. Building satellites, we designed with the attitude; “we ain’t got no stinking safety factors!” Couldn’t afford them. Every ounce we had extra in structure meant one less ounce that couldn’t go to optics, electronics and such. We always tested for laminate (the exact stacking sequence) cut up into coupons and pulled.
As input, it needs, the S&S properties of a lamina (not laminate. One ply’s S&S properties) of the materials used. There are two way of getting lamina properties…
(1) Micromechanics – this is a theoretical way of calculating those properties. Basically (although there are finer points) it’s just a weighted averaging of the fiber properties and resin properties. In AE, we felt comfortable doing trade studies for a proposal, but would do coupon testing of the actual material combination before committing a design to production.
(2) Making coupons out of nothing but unidirectional material with all plies oriented in the zero direction. Some would be cut out along the “grain” and some across the grain. These are then pulled, pushed, sheared to get the properties.
Unfortunately, to do it reliably, statistics requires many samples. A-basis is basically some “fudge factor” based on the number of samples. The more samples, the tighter the estimate. The fewer samples, the “gray” area is wider.
Hopefully, someone out there can help us with these properties. If so, any help would be appreciated.
In LPT parlance, we need:
Subscripts:
1 – this is the fiber direction of the lamina (one ply).
2 – this is transverse to the fiber direction. Often called the matrix direction.
T – In Tension
C – In Compression (Most times properties are different in tension/compression)
E11T – Stiffness in the fiber direction, in tension.
E11C – Hopefully, you get the idea.
E22T
E22C
G12 – Shear stiffness in the 1-2 plane
v12 – Strain in the 2 direction based on straining it in the 1 direction. You know when things stretch they get thinner. That’s what this number describes. It usually uses the Greek symbol Nu.
S11T – Axial strength in the fiber direction/tension
S11C - Hopefully, you get the idea.
S22T
S22C
T12 – Shear strength of the lamina in the 1-2 plane.
Any values would be great, however, since the stays of a mast system tend to compress the mast, I believe, the compressive values are by far the most important!
Thanks to all willing to help.
Inquisitor
BIG ENGINES: Silos today... Barn Door tomorrow!
- Joined
- Nov 1, 2005
- Posts
- 276
- Reaction score
- 11
- Location
- North Carolina Mountains
Thank you water addict,
I had hoped you will help keep us on the strait and narrow. And I totally agree, that qc (or actually... variability) is by far the biggest issue. I may be deluding myself that anyone (including myself) can be fastidious enough to get anything near theory. Aerospace takes great pains (and money) to achieve close to theoretical values with low safety factors. That’s why I plan on doing it on my 20’ cat first before I attempt it on a 40’ cruiser. I can always paddle back.
Also, the calculations are already there in an attachment (XL spreadsheet) above for your perusal. Be aware it was only an 8 hour puking… with very little double checking. By all means point out any errors in math or theory. I take constructive criticism pretty well.
For that matter it’s on there for people to add on to also. I hope this to be like an open-source software project.
Regards,
Inquisitor
I had hoped you will help keep us on the strait and narrow. And I totally agree, that qc (or actually... variability) is by far the biggest issue. I may be deluding myself that anyone (including myself) can be fastidious enough to get anything near theory. Aerospace takes great pains (and money) to achieve close to theoretical values with low safety factors. That’s why I plan on doing it on my 20’ cat first before I attempt it on a 40’ cruiser. I can always paddle back.
Also, the calculations are already there in an attachment (XL spreadsheet) above for your perusal. Be aware it was only an 8 hour puking… with very little double checking. By all means point out any errors in math or theory. I take constructive criticism pretty well.
For that matter it’s on there for people to add on to also. I hope this to be like an open-source software project.Regards,
Inquisitor
I would actually prefer to work with glass than with carbon as it appears to be more visible if the fiber has taken on resin or not.
Otherwise, I was looking into the same resources of Georgia Tech. when I hopefully start building my dream-boat with free standing masts.
In general, if I can’t make the math myself, I’ve to trust in the math + safety margin. Assuming that the safety factor may be a bit higher from an A.E. than a N.E. considering that a broken wing will likely end in a fatal accident, I’ve no problem with this picture.
Where I get a bit puzzled is the missing torsion load apparently carbon masts builders anticipate by wrapping the fibers in an angle as well as unidirectional.
I either wait until my son advances a few semesters to proof otherwise. Until then, just by guts feeling I would place 2 layers of fibers in a 45 degree angle.
Otherwise, I was looking into the same resources of Georgia Tech. when I hopefully start building my dream-boat with free standing masts.
In general, if I can’t make the math myself, I’ve to trust in the math + safety margin. Assuming that the safety factor may be a bit higher from an A.E. than a N.E. considering that a broken wing will likely end in a fatal accident, I’ve no problem with this picture.
Where I get a bit puzzled is the missing torsion load apparently carbon masts builders anticipate by wrapping the fibers in an angle as well as unidirectional.
I either wait until my son advances a few semesters to proof otherwise. Until then, just by guts feeling I would place 2 layers of fibers in a 45 degree angle.
Inquisitor
BIG ENGINES: Silos today... Barn Door tomorrow!
- Joined
- Nov 1, 2005
- Posts
- 276
- Reaction score
- 11
- Location
- North Carolina Mountains
45/-45 plies
stewi,
That sounds like another good reason (resin taken)!
Only designed satellites… no planes. Yes, planes have safety factors. Satellites do also… at least NASA ones. Most black ones didn’t have S.F. We had 0.1 margin on loads, so, I guess we had a S.F. of sorts.
If I understand your quandary correctly…
Most carbon fibers have a negative coefficient of thermal expansion (CTE) and resin has a high positive CTE… often greater than even aluminum. I’ve cured many 0/90 laminates at 350F for my college thesis and sometimes you could hear them pop and crack (the resin) just by cooling down to room temperature. Rather depressing!
They can also start micro cracking just from heating and cooling cycles.
Also interlaminar shear stresses (stresses trying to shear the plies apart) are very high for 0/90 laminates under load.
Once the resin is shot… it’s like pushing a string… millions of little black ones… Not very conducive to taking loads of any sort.
Even though this is a glass tube design and glass has a high positive CTE, old habits die hard. If you’ll note, I made this preliminary laminate with a 45 degree ply between the 90 and the zeros. It kind of shields the 0s from the 90s and vice versa.
stewi,
That sounds like another good reason (resin taken)!
Only designed satellites… no planes. Yes, planes have safety factors. Satellites do also… at least NASA ones. Most black ones didn’t have S.F. We had 0.1 margin on loads, so, I guess we had a S.F. of sorts.
If I understand your quandary correctly…
Most carbon fibers have a negative coefficient of thermal expansion (CTE) and resin has a high positive CTE… often greater than even aluminum. I’ve cured many 0/90 laminates at 350F for my college thesis and sometimes you could hear them pop and crack (the resin) just by cooling down to room temperature. Rather depressing!
They can also start micro cracking just from heating and cooling cycles.
Also interlaminar shear stresses (stresses trying to shear the plies apart) are very high for 0/90 laminates under load.
Once the resin is shot… it’s like pushing a string… millions of little black ones… Not very conducive to taking loads of any sort.
Even though this is a glass tube design and glass has a high positive CTE, old habits die hard. If you’ll note, I made this preliminary laminate with a 45 degree ply between the 90 and the zeros. It kind of shields the 0s from the 90s and vice versa.
Apologies for being in a different time zone, but all this makes for a great breakfast for me. Inquisitor, I live in the land of sheep, we've got millions of them, we call them a "mob"; everone here has seen a sheepdog work the mob. The sheep all stick together, there is no individual thought.. they all move together usually away from the dog.
Two year olds start asking why? at fourteen it becomes what if? Some of us never stop asking, it's got to be in the genes I think, but also a learned thing from that moment when you discover that the mob or conventional wisdom is not the only answer. Does anyone remember the 12metre that broke in half, the CF rudder stocks that fractured and CFmasts that went over the side. All of them engineered by experts. All promoted on the fantastic strength of CF. It seems when we learned to use CF we put glass on the bench.
But to respond the the comments;
Why glass/epoxy instead of Al (in no particular order);
1. For me it will be cheaper, and don't think I don't value my time, at the end of the day it's all I've got. The cost benefit favours Glass/epoxy.
2. Weight aloft, not in the bottom panel... up top where the leverage is greater, that is the significant saving that will be noticeable.
3. Windage, again in the upper sections ....
4. The ability to tailor the shape and cross section to the load.
5. Longevity and corrosion resistance.
6. And not least a shape that is a joy to behold.
The tapered Al sections that I'm aware of here are produced by cutting three wedges out of the top secion of an alloy stick with a circular saw, a few "G" clamps to squash then in, an abrasive disc to "V" it out and TIG welding them up. If you are talking about consitency of material properties what happens when mast grade Aluminium is heated significantly and then often with incomplete penetration...
Then we have the issue of composites and quality control, surely that also applies to CF, the CF mast manufacurers are on top of it or aren't they?
So why not unstayed? the hull is built.. unfortuately or not it's for a stayed rig.
And why not a hollow wood, well designed and varnished mast. Am I the only one who ever had to sand the mast from the Bo'sun's chair on the way up to the masthead and varnish it on the way down...annually? But there's more, at 5 years max we would unstep the rig, dismantle it, sand it back to bare wood and start again, I've done this on a magnificent Lion Class sloop several times, hollow Sitka spruce mast, no expense ever spared. It was the bane of our maintenance.
I love building yachts and I love the cruising, but varnished timber especially where I can't easily get at it is a pain in the stern.
Inquisitor, I really liked your observation about the history of chopped strand and the baggage/reputation that hangs with it, I think you are right on the mark there.
It also occurs to me that the CF industry is a fairly elitist mob with a product to promote and a design arm that by it's nature wants to stay on the leading edge, why would they do the hard yards on a product they have left behind.
Thanks to all, great comments, obsevations, this is great.
Two year olds start asking why? at fourteen it becomes what if? Some of us never stop asking, it's got to be in the genes I think, but also a learned thing from that moment when you discover that the mob or conventional wisdom is not the only answer. Does anyone remember the 12metre that broke in half, the CF rudder stocks that fractured and CFmasts that went over the side. All of them engineered by experts. All promoted on the fantastic strength of CF. It seems when we learned to use CF we put glass on the bench.
But to respond the the comments;
Why glass/epoxy instead of Al (in no particular order);
1. For me it will be cheaper, and don't think I don't value my time, at the end of the day it's all I've got. The cost benefit favours Glass/epoxy.
2. Weight aloft, not in the bottom panel... up top where the leverage is greater, that is the significant saving that will be noticeable.
3. Windage, again in the upper sections ....
4. The ability to tailor the shape and cross section to the load.
5. Longevity and corrosion resistance.
6. And not least a shape that is a joy to behold.
The tapered Al sections that I'm aware of here are produced by cutting three wedges out of the top secion of an alloy stick with a circular saw, a few "G" clamps to squash then in, an abrasive disc to "V" it out and TIG welding them up. If you are talking about consitency of material properties what happens when mast grade Aluminium is heated significantly and then often with incomplete penetration...
Then we have the issue of composites and quality control, surely that also applies to CF, the CF mast manufacurers are on top of it or aren't they?
So why not unstayed? the hull is built.. unfortuately or not it's for a stayed rig.
And why not a hollow wood, well designed and varnished mast. Am I the only one who ever had to sand the mast from the Bo'sun's chair on the way up to the masthead and varnish it on the way down...annually? But there's more, at 5 years max we would unstep the rig, dismantle it, sand it back to bare wood and start again, I've done this on a magnificent Lion Class sloop several times, hollow Sitka spruce mast, no expense ever spared. It was the bane of our maintenance.
I love building yachts and I love the cruising, but varnished timber especially where I can't easily get at it is a pain in the stern.
Inquisitor, I really liked your observation about the history of chopped strand and the baggage/reputation that hangs with it, I think you are right on the mark there.
It also occurs to me that the CF industry is a fairly elitist mob with a product to promote and a design arm that by it's nature wants to stay on the leading edge, why would they do the hard yards on a product they have left behind.
Thanks to all, great comments, obsevations, this is great.
Forum posts represent the experience, opinion, and view of individual users. Boat Design Net does not necessarily endorse nor share the view of each individual post.
When making any potentially dangerous or financial decision, always employ and consult appropriate professionals. Your circumstances or experience may be different.
When making any potentially dangerous or financial decision, always employ and consult appropriate professionals. Your circumstances or experience may be different.
Advertisement:
