Strut Keel Design and Engineering Grounding Forces

Discussion in 'Boat Design' started by an2reir, Apr 21, 2020.

1. Joined: Sep 2001
Posts: 645
Likes: 163, Points: 53, Legacy Rep: 110
Location: Bordeaux France

an2reirfifty boat designs

Hello Everybody

I am working at present on a study of Composite Bilge Keels and Strut Keels design and engineering and in particular on designing bilge keels that would protect the propeller shaft line
Bilge keels and strut keels are required to withstand without damage the loads and forces generated by the grounding of the boat
I did design the composite bilge keel of fibrglass shell; high density PVC foam and ideally with a carbon prepreg internal framing structure fastened connected to the internal hull web framing with stainless steel bolts
I am doing a series of FEA simulations of the composite keel structure to determine wich is the best dimensioning of the keel bolts
I did read the very useful study so well writen by Eric Sponberg on the design and engineering of keels- (a good study for wich I here I thank Eric Sponberg)
Keels and Rudders: Engineering and Construction https://www.boatdesign.net/threads/keels-and-rudders-engineering-and-construction.8240/

The ABS ORY guide does indicate the grounding force to be interpolated between 1.5x Boat weight foar boats less than 10m and 3x boat weight for boats longer than 20m

The ISO 12215-9 defines the loads and specifies the scantlings of sailing craft appendages on monohull sailing craft with a length of hull of up to 24 m

I did as well read the study Mechanics of Ship Grounding - Bo Cerup Simonsen 1997

Could anyone indicate where to find the standard or formula to calculate the grounding forces for appendages of motor boats (with a length of hull of between 10m and 20m)?

Thank you

Andrei Rochian

2. Joined: Oct 2008
Posts: 7,808
Likes: 1,704, Points: 113, Legacy Rep: 2488
Location: Japan

Hi Andrei,

There is no real "standard" formula simply because as always - it depends!
All you need to decide is how much strength do you want to give it?

Since you can image a simple grounding at less than 5 knots... or taken up to, ground in shallow waters or a rock not shown on charts at full speed. But will yield very different scantlings.
Thus, how strong do you wish the skeg to be?...since there comes a cross over point of if the skeg is built like a tank.. great... but it may damage the major hull structure when a collision occurs.

Whereas, most appendages are designed to - break free - under high loads, merely to avoid this situation.
The prop shaft and/or prop is much cheaper to replace than the large sections of hull bottom structure....

Thus, how strong/stiff do you wish it to be?

3. Joined: Sep 2001
Posts: 645
Likes: 163, Points: 53, Legacy Rep: 110
Location: Bordeaux France

an2reirfifty boat designs

Thank you for your message I do appreciate. Now I am in France nearby Bordeaux . Greetings from confinement France!
My thinking now is that we have some new tools to do some development on previous design techniques.
We have the carbon prepreg so we can make the internal framings light. And we have the new generation of software to make simulations of structures being smashed by all sorts of loads.
I am basically thinking of bilge keels to take groundings on beaches of sand or pebbles ; sort of like the Shannon british search and rescue boats they do posess capability of beaching on a beach that is not rocks.
And also when the break not to damage the hull.
I did find equations for calculating the beaching forces in the very good paper on Beaching of Ships published by Bo Cerup Simonsen 1997
but they are a little too complex
The ORY guide very empiric formula of beaching is a function of just Displacement to me it seems a little too simple.
I think - If I could integrate not just displacement but displacement ; speed ; acceleration of beaching on a sand beach in mathematical equation to give me the force generate by beaching on the bilge keel lower face- that wold be good enough for me

Attached Files:

• Grounding conditions 1.jpg
File size:
120.3 KB
Views:
388
4. Joined: Oct 2008
Posts: 7,808
Likes: 1,704, Points: 113, Legacy Rep: 2488
Location: Japan

Andrei,

Think of it then like a fender.
The max typical load you should experience, other than a major collision is simply 1/4g load.

So apply a force of 0.25g x Displacement

5. Joined: Sep 2001
Posts: 645
Likes: 163, Points: 53, Legacy Rep: 110
Location: Bordeaux France

an2reirfifty boat designs

Hi Ad Hoc thank you for your message, ok I will do an FEA simulation of o.25g x Displ on my Strut Keel structure and fastemers ; and will post some informations and images of the loadcase . Her e is an image that shows the very genearal geometry of one strut keel version ; the FEA means nothing - I was just at the very incipient stage of this study.
I guess the ideal would be to just simulate a number of situations of groundings in a FEA software
What concerns me is the lateral grounding when the force would be acting in a transversal direction on the keel lowest surface I think that is when the load ion the fasteners is the most.
Most simplified grounding criteria I have read like the one in the ORY Guide only mention frontal axial loads

Attached Files:

• FEA strut keel.jpg
File size:
134.5 KB
Views:
522
6. Joined: Oct 2008
Posts: 7,808
Likes: 1,704, Points: 113, Legacy Rep: 2488
Location: Japan

I think that is an overkill, but it is your choice.
It is a relatively straight fwd calculation, FEA not necessary. I did these calc's by hand, I only use FEA if clients demand pretty pictures!
The FEM just gives an insight into load paths and sensitivity analysis... it cannot be treated as an absolute, especially with so many unknowns.

As for transverse...well, there is your problem. If you want strength, the skeg needs to be stiff/strong. That increases it thickness, = more drag = influences flow into the prop too.
Hence my original opening gambit... how strong/stiff do you want it to be?

I would suggest you're over thinking this...keep it simple.

bajansailor and jehardiman like this.
7. Joined: Sep 2001
Posts: 645
Likes: 163, Points: 53, Legacy Rep: 110
Location: Bordeaux France

an2reirfifty boat designs

I am however planning to do the FEA's in order to check and triple check my fasteners are going to be sized adequately in order to take the loads of multiple groundings without the strut keel failing
My boat should ideally withstand reasonably multiple groundings on sand or small pebbles liker the Shannon Search and Rescue boat does. In real life water these groundings are not really axial some forces are transversal. Like for example the grounding keels of the SAR Boat attached image

Attached Files:

• wire0002.jpgf767897d-565d-4b00-82f8-cb1466278176Zoom.jpg
File size:
201.9 KB
Views:
374
Last edited: Apr 21, 2020
8. Joined: Aug 2004
Posts: 3,795
Likes: 1,183, Points: 113, Legacy Rep: 2040
Location: Port Orchard, Washington, USA

jehardimanSenior Member

I agree with Ad Hoc here, you are over thinking this, or maybe not thinking about it the correct way.
You want a force, but grounding is an energy problem. Think about it...what is the energy available? 0.5*mass*velocity^2. I think if you look back through the COSTA CONCORDIA thread I calculated the energy available was equal to 3 heavyweight torpedoes or enough to punch a 1 m^2 hole through 900mm of steel. Additionally, think about how long the load you actually use is applied. Lets say you use 0.25g...that means that it takes a vessel traveling 9.8 m/s (19 knots) 4 seconds to stop and the distance it traveled was 19.6m. That means you have to apply that 0.25g force for 19.6m down the length of the vessel, not just to one spot.
Suffice it to say, that I was involved in looking at the whole ship to ground interface back in the early 1990's and the real issue is in picking the loads because not only the ship structure, but the ground must support the load. We burned through 3 soils PhD's who couldn't answer the basic issue of what the soil was going to do when impacted with that much energy. Eventually we came to the conclusion that the only way to get our arms around the problem was to select a stopping distance.
Finally, avoid trivializing your load case. When looking at the nose structure of a AUV for impact, the contractor assumed that the point of impact was always directly on centerline...I asked if he had ever heard of the RMS TITANIC.

9. Joined: Oct 2008
Posts: 7,808
Likes: 1,704, Points: 113, Legacy Rep: 2488
Location: Japan

Indeed.

It was the same back in the 90s. I was part of the structures committee looking at the new (as it was then) HSC code. The issue of collision damage and a "Gcoll" and how to calculate and apply this became problematic very quickly.
The Gcoll being the horizontal (de)acceleration value of a vessel when hitting a vertical rock that is up to 2m above the waterline. The assumption being a near dead stop... hence JEHs example in more focus!!

The eventual over simplified calculations and hence formula used was introduce in the 1994 code, but owing to endless irregular results, we had to revise or rather attempt to revise the formula to exclude those grossly irregular results. When i say "we" I refer to our IMO country committees along with others around the world in each country...as it was a consensus decision to the final method. But it spawned endless research (which still continues) and many many more than 3 PhDs Thus the HSC 2000 code came out with a revised formula. Suffice to say, it is still not ideal.

It just emphasises the difficulty in arriving at a satisfactory solution for a "simple" load case.

Thus, just use common sense and whilst using formulae like 0.25g or 1/2mv^2 may seem somewhat over simplistic.... it provides you with a path to a solution.
Is it enough... who knows... the acid test is satisfy yourself... then apply it. If the boat fails...you'll know...if there are know issues... you'll know.

Good luck ..

10. Joined: Aug 2004
Posts: 3,795
Likes: 1,183, Points: 113, Legacy Rep: 2040
Location: Port Orchard, Washington, USA

jehardimanSenior Member

Here is an interesting report on a modern grounding incident of a small boat.
https://www.volvooceanrace.com/stat...d-inquiry-report-released-on-9-march-2015.pdf
Leaving aside all the navigation issues, here is the actual grounding.
The grounding of DM Grömitz (M1064), USS Guardian (MCM-5) and USS Darter (SS-227) were similar in that they happened over several seconds and left the vessels well up on the reefs and at significantly less draft than normal though without too much initial damage.

11. Joined: Sep 2001
Posts: 645
Likes: 163, Points: 53, Legacy Rep: 110
Location: Bordeaux France

an2reirfifty boat designs

Hi Jehardiman

yes I agree noted, of course you are right .

The mechanics of boat grounding involves a number of variables wich I think were very well expressed in terms of energy and force and mathematically described here in the attached paper Mechanics of Ship Grounding signed by Bo Cerup Simonsen from Denmark

I work to redesign effective " grounding keels" similar in some ways to those that were done on the RNLI search and rescue boats like the Severn , and Shannon classes ( attached image)
that would ideally allow the motor boat multiple groundings on a sand or pebbles beach in such way that the propellers and shafts are protected - with more modern materials and ( at the lowest possible weight ) .

For testing the proper dimensioning of the mechanical fasteners and the keel internal structure (wich is steel frame and ideally carbon prepreg frame ) I need to input in my FEA simulations the max load the keel is to bear- in the form of "force " because this is how the FEA software works.

1. One loadcase is the empiric formula from the ABS ORY Guide and consider the grounding case load as interpolation of 1.5xDispl and 3x Dispacement in axial direction on the lowest surface of the keel
2.Second loadcase will be as the one indicated by Ad Hoc above and do a loadcase with a force 0.25g xDispl

3.I think the calculation of Estimate Collision Impact Force would be good way to have yet another estimation -in full knowledge this is a simplification.
In a simplified case of collision - the collision impact force is function of Displacement, Speed , Collision distance , Collision time and we will assume the speed of the boat to be very low and the collision distance at the order of meters when beaching on sand or pebbles beach . The beaching time will be of an order of minutes.

I think the RNLI did even some real life scale models of their boats and tested them on beaches in Dorset

Nowadays a simulation of the grounding on a software like this one -
will also be a good validation - I think .

Thank you and have a great day

Attached Files:

File size:
174.5 KB
Views:
422
• grounding keels.jpg
File size:
225.4 KB
Views:
554
Last edited: Apr 22, 2020
12. Joined: Oct 2008
Posts: 7,808
Likes: 1,704, Points: 113, Legacy Rep: 2488
Location: Japan

Hi Andrei

How did you get on.. did you make any progress and/or come to a final solution?

13. Joined: Sep 2001
Posts: 645
Likes: 163, Points: 53, Legacy Rep: 110
Location: Bordeaux France

an2reirfifty boat designs

I am doing work on the design of "grounding keels " to take the forces of grounding on a sand beach or pebbles beach .
A type of grounding situation of a boat on a pebbles beach is this one

I did three designs: 1. Design of "grounding keel " made in fiberglass side thickness 10mm top and bottom of keel thickness 20mm with divinicell foam core no interior reinforcement; glued to hull with cyanoacrylate structural adhesive 2. Design of strut keel made in fiberglass thickness 5mm with divinycell foam core top and bottom of keel thickness 20mm with interior reinforcement web structure of vertical fibeglass beams and longitudinal beams bonded to hull with adhesive cyanoacrylate structural adhesive 3. Design of "grounding keel " made in fiberglass thickness 5mm with divinycell foam core top and bottom of keel thickness 20mm with interior reinforcement framing aluminium 7075 bolted on the hull bolts M30mm made of weaker metal like perhaps iron or stainless steel of lower Tensile Strenght.
All three are supported inside hull bottm by load distribution structure web type of transversal and longitudinal beams at the location of bolts.
In order to determine the grounding forces:
I found the paper writen by Eric Sponberg "Keels and Rudder Engineering";
and also I did consider the two methods to calculate grounding forces that were indicated by yourself above; having in mind as well the ABS Ocean Racing Yacht Guide and the ISO 1225-9
I did find the new software from Autodesk namely the Autodesk Fusion 260 that does offer the functions of FEA Static Analysis but as well the function of "Event Simulation" - Simulating collision
I did test the three versions of strut keel designs in a series of FEA loadcases with frontal forces , vertical forces but also lateral force
My design concept is to make the " grounding keel " be a "fusible" structure that will prtect the propeller in case of brounding but will break a a certain maximum charge in order not to damage the hull in the case of grounding
In order to achieve this I thought to establish a breaking or " fusible " part in my keel structure that are the bolts wich are to be made of a metal that is to be weaker than the keel reinforcement or the hull bottom reinforcement
I am working to determine the best solution by
1.determine the breaking point of my structures by FEA simulations ;
2. but as well I am thinking to make real life 1/2 scale models of the keel version attached by bolts and version attached by structural adhesive and subject them to real life test forces to see wich case the damage on the hull is the less.
3.As well I am working to simulate the grounding in the form of "Event Simulation" in Autodesk Fusion 360 like this:

4. I did some researach finding informations on motor boats fitted with strut keels that did suffer grounding of beaches and rocks and I did find some videos and some photos of such boats

The grounding keels may be positioned typically on the bottom of the boat offset from the proplleer shaft . I here attach a photo of a" grounding keel"of a british boat
One serious aspect I think is good to consider is the situation of the boat hitting submerged rock with keel, the keel is to protect
the propeller and the manoeuvering capability
Is there a possibility that such grounding keel if and when breaking due to contact with submerged rock - to be perhaps driven toward the center ax of the boat and stop the propeller from turning?
Is there the possibility that if the beel is bonded to the hull by adhesive only when keel may beak off the adhesife braking may as well do damage to the hull laminate?
Is the best solution to bond such a keel to the hull by adhesive or bolt this kind of "grounding keel " to the hull by bolts?

My boat displacement is 24000 Kg - I was considering grounding condition loads frontal and vertical forces on the lower surface of the keel of of 60700 kg that is 595263 N

Attached Files:

• 20140814_113837.jpg
File size:
185 KB
Views:
383
Last edited: May 16, 2020
14. Joined: Oct 2008
Posts: 7,808
Likes: 1,704, Points: 113, Legacy Rep: 2488
Location: Japan

Hi Andrei

Nice to hear how the work is going, and I trust you are well and safe too.

Whilst I understand the "appeal" for simplified FEA software, I would hesitate to use it. They are overly simplified to the point you are not controlling the element types or the method of solution and in some cases you can't even control the mesh size how you want it to be. One of my DO staff decided (in his own time) to use the FEA side of Inventor - did lots of lovely looking models and pretty colours and animations. But upon interrogating the FEM, little was known or able to be controlled by the user. Suffice to say, the input was wrong because there was far too much reliance on the software doing all the work - making it look easy! FEA requires a lot of understanding to get even half decent results and even then still treated with caution unless you can validate it. So unless you're familiar with FEA, it may be better to get someone with a dedicated FEA software to do it.

I would suggest you should perhaps consider approaching your problem in the same manner i which I have designed the foundation for T-Foil system on our catamarans.
But it is of course your choice. I define the load, in my case the foil hitting a rock at 45knots.

It was a bolted on structure, so the bolts were designed to be the 'weak' link. The surrounding and supporting structure designed to absorb the load from the collision with a good FoS. This this ensured the weak link would fail as designed.
The failure load selected was with no FoS i.e at yield... so the bolts would fail first. So in the event of a collision the foil would simply break free and not damage the hull structure.

It was confirmed several times, as this is indeed what occurred. At one time the foil hit an ISO container that fell off a container ship and was barely floating, unseen by the Capt. The foil broke free, relatively undamaged too, and the hull was sound, no damage at all. It also occurred again several years later, hitting what many believe was a large shark or whale. In both cases the foil had a small transponder inside so it was remotely located and then salvaged by a diver.

So if you take the assumption that the keel is an "add-on" - design it in that way. Make the keel/skeg be the sacrificial appendage.

This helps you to focus your mind on not making the appendage too stiff/strong that it does damage the hull upon a collision.
Otherwise you may well over engineer it being too cautious.

an2reir likes this.

15. Joined: Sep 2001
Posts: 645
Likes: 163, Points: 53, Legacy Rep: 110
Location: Bordeaux France

an2reirfifty boat designs

Thank you for your kind feed back. I do hope you do well and keep healthy out of this coronavirus mess.
Yes that is precisely my thinking as well , the bonding to be via bolts and the bolts to be the weak link. I think this is the general principle of designing the keels , with the keel body wich may have been iron or lead in sailing yachts wich was attached to a strong distribution system in the hull via the "weak links" wich are the keel bolts.
Yes of course despite the fact that I am already quite experienced myself with the FEA simulations I do not relying on the FEA
What I do is I set up a system of cross calculations and cross checking with making calculations as far as I can ; but as well doing FEA simulations with two different systems .
As well I plan to do real life sample laminate structure testing.
However, the triple checking of a collision situation by this new Event Simulation capability of the Fusion 360 looks pretty cool to me
Here is a photo of broken keel in grounding . Has this broken keel jammed the propeller and stopped it from turning?

Attached Files:

• 7073086-10827115.jpg
File size:
78.3 KB
Views:
325
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 potentially dangerous or financial decisions, always employ and consult appropriate professionals. Your circumstances or experience may be different.