Analysis of pile groups with battered piles?? 3

[MrStohler]

Is anyone aware of a good reference for the analysis of pile groups that include battered piles to resist lateral loads.

A member of DFI has indicated that all that is required is to select the piles required to resist the vertical loads and then add the battered piles required to resist the lateral loads.

This seems too simplistic. It does not appear to account for any vertical load that might be resisted by the battered piles or other interaction effects.

Thanks for your asisitance.

 

[Focht3]

Your skepticism is justified - at least for large projects. It may not be worth the effort for smaller ones. It's hard to answer your question without considerably more information about the soil conditions, project location, structure, the intended use and expected loads, and the possible consequences of failure (other than economic loss), if any.

Have you evaluated the lateral capacity of the individual piles using a p-y curve approach, such as LPILE? Do you have any experience with this approach, or does your geotechnical consultant have any P-y analysis experience?

Give us more information - perhaps we can be of assistance.

 

[MrStohler]

Thanks for your response. To answer you question, I am not familiar with P-y analysis. Is there a good resource to describe this analysis. Many of the geotechnical consultants I speak with claim that design of the piles falls the the structural engineer (are there any industry standards to demark general limits of responsibilty?). The consultant generally would provide only an allowable skin friction (for pile design, other data is provided for footings) Unlike steel, concrete, timber, masonry etc., there seems to be a lack design guides or methods of analysis. Software is nice to use, but I do not wish to use it without understanding its basis of analysis. Any suggestions?

 

[Hoitsma]

Each battered pile should be associated with a vertical pile (like the joint of a truss where a vertical member intersects a diagonal member). The "truss joint" occurs therefore in the foundation. A batter pile subjected to a lateral force will react with two components of force (the batter pile is a two force member in statics). If you do not have a vertical pile adjacent to the batter pile, then the foundation must take the vertical reaction component. In a dynamic loading (say machinery) the vertical force component into the foundation mat may be far too noticeable, or worse, damaging.

I would agree with the simplistic approach: design vertical piles for vertical load and then design batter piles for horizontal loads.

Visit

http://www.ensoftinc.com

for information on LPILE software. Ensoft, Inc. is an excellent resource for information on piling design.

The piling design IMO should be a team effort between the civil engineer and the geotechnical engineer although the responsibility of each is to the adequacy and correctness of their respective sealed documents. Don't hesitate to ask questions and to cross check your work with your geotechnical consultant.

 

[Focht3]

MrStohler:

Ouch! Did your geotechnical consultant really say that? You need to find a better geotechnical consultant!

If you don't know anything about p-y curves, and your geotechnical consultant is clueless / of no help, then forget p-y curves. That is, unless the platforms are sizable...in which case you should do one of two things:

1. Hire a new geotechnical consultant; and/or
2. Contact Ensoft / Lymon C. Reese & Associates, Inc. in Austin, Texas and hire them as a consultant. Talk to Bill Eisenhour. He's very knowledgable (and a good friend.) You can reach Bill through Ensoft.

[Focht3 ascends his soap box.]

I was troubled by one part of your message - quoted as follows:

Many of the geotechnical consultants I speak with claim that design of the piles falls the the structural engineer (are there any industry standards to demark general limits of responsibilty?). The consultant generally would provide only an allowable skin friction (for pile design, other data is provided for footings)

In my opinion, the geotechnical engineers that you have spoken to either don't understand what you are inquiring about, or don't have relevant education and experience. I suspect the problem is not your inquiries. Shame on them for not confessing their ignorance and directing you to someone that can help you! Hoitsma is absolutely correct - pile design should be done jointly by the structural and geotechnical engineers. To do otherwise courts disaster.

[Focht3 now descends from his soap box.]

I was very pleased when you said,"Software is nice to use, but I do not wish to use it without understanding its basis of analysis." Good for you - that is the only attitude to have. And the misuse of p-y analysis programs is all too common. I would encourage you to learn about this method of analysis if your project time and budget will support the effort. Again, Bill Eisenhour is a great resource.

Let us know what happens, please -

 

[BigH]

Focht3's advice, again, is more than valuable and - - - he deserves his soap box time!!

I have found M.J. Tomlinson's book on Pile Design and Construction (ViewPoint Publications) to be the best book (with examples) that I have seen on the subject. Tomlinson is the Wimpey expert.

One thing regarding lateral loads on piles - it was mentioned earlier - is dynamic forces. I have seen recommendations that in such cases pile caps have been extensivelly damaged due to none uniform movements associated with minute variations of time of "impact" on batter piles. I think the 1964 Nagata Earthquake pointed this out - in Vancouver, unless things have changed, they design lateral loadings (earthquakes) to be taken by vertical piles. Pile groups would have a reasonable resistance to earthquakes compared to single piles (of course subject to proper verification of design parameters and the like).

 

[austim]

My comments mostly originate from experience with pile groups for marine structures and bridges, where it is common for the piles to extend a considerable distance above the adjacent soils.

I would question the statement that "each battered pile should be associated with a vertical pile". Pairs of battered piles work equally well.

I have always analysed a pile group containing raking or battered piles with all piles included for all loadings (even in the days before computer software simplified the artihmetic). With any significant projection of piles above the soil, I can see no reasonable argument for doing anything else.

The 'fancy' software that will now permit you to treat the pile group and surrounding soil as a continuum has always been beyond my budget. I also wonder just how reliable the inbuilt assumptions in such software can be, and whether the complexity of analysis involved really results in more 'accurate' results.

In any case, I hold the view that if a pile group can be shown to be safe and stable without the added lateral stability provided by the soil, then all will be well (so why bother with the extra analysis?). Of course, I have always allowed for the presence and nature of existing soil when considering the buckling capacity of individual piles.

On the topic of 'realistic' results, I am reminded of the learned geotechnics professor who did an extensive analysis of the many pile capacity formulae available and compared them with a raft of test values. His conclusion was that none of the formulae gave a lower coefficient of variation than the simple assumption that all piles have a capacity of 93 tons!.

Bowles "Foundation Analysis and Design" has a bit to say on the subject. His third edition has a Fortran program for the analysis of three-dimensional pile groups, which should not be unduly difficult to convert to Basic or whatever if Fortran is not convenient.

 

[MrStohler]

Thanks for your input, austim's comment about how "realistic" the various methods of analysis are seems to strike a point. Considering the variability of the material (soil) are the methods described above more reliable (and if so should a lower factor of safety be used)? Given that soil behaviour is not as predicable as steel or concrete, when is it appropriate to use the more detailed analysis?

Conversations with a local (Ohio, USA) senior engineer who has encountered similar problems suggested this approach:

1)Given that the allowable skin friction for the piles is usually provided (auger cast piles are typically specified in applications where piles cannot be driven) in the geotechnical report;

2)For any particular configuration; distribute the vertical loads to all piles in accordance with thier position in the group. Account for any variation in the loads distributed to each pile by usisng an analysis that considers Ixx , Iyy and Ixy of the pile group.

3) At battered piles, multiply the vertical load by 1/cos(batter angle) Example:a pile battered at 4v:1h would have vertical load from part 2 multiplyed by 1.0308

4) Distribute horizontal load evenly to battered piles in group and multiply by 1/sin(batter angle) Example: for a 10kip horizontal load at a 4v:1h battered pile, multiply load by 4.1231, pile load is 41.231kips

5) Add pile loads from parts 3 and 4 to obtain required resistance at battered piles.

6)Use group efficiency reduction as appropriate, allowable skin friction, and forces from above to calculate pile required pile length.

This analysis only coniders the skin friction component of the soil resistance and it only considers the pile to be loaded axially. Too conservative? Not conservative enough?
Any other comments.

 

[BigH]

MrStrohler:
1. Where in Ohio? I'm from just east of Cleveland myself. (Being in India, it does give someone a bit homesick to hear from a Buckeye!! at Christmas time.
2. Go OSU!
3. I doubt your pile lateral load is 41.231! Yes, when you use sinA you get this nice 5 significant figure but what happens if batter is not, say 4v:1h but 4.1v:1h?? or 3.9v:1h?? or . . . to much accuracy quoted!
4. Merry Christmas to all.

 

[Focht3]

BigH:
Thanks. And I agree with your discussion on battered piles in dynamic events. Battered piles are significantly stiffer, and can cause problems where vertical piles would not. Sound advice.

austim:
I generally agree, but a few of your comments do concern me.

1. The 'fancy' software that will now permit you to treat the pie group and surrounding soil as a continuum has always been beyond my budget. I also wonder just how reliable the inbuilt assumptions in such software can be, and whether the complexity of analysis involved really results in more 'accurate' results.

We should always worry about how reliable any analysis is - including all computer software. Having acknowledged the potential shortcomings of all analytical techniques, it is also important to understand that some of the newer engineering software that includes three dimensional visualization of the problem may provide the designer with new insights into soil / structure interaction that could not be easily foreseen with manual or text-only techniques. These approaches have not (yet) supplanted our "tried and true" approaches, and may never replace more conventional methods. But that does not mean that we should dismiss them out of hand, either. And the cost of three dimensional software has become much more reasonable - some programs are less than $10,000 USD. Not cheap, but about the same cost as an equivalent STRUDL run 20 years ago (if you could even get STRUDL to run the analysis...)

I concur with your skepticism about more "accurate" results. To quote a past Terzaghi lecturer, "It is better to be approximate and correct than precise and wrong." But sometimes the more complex analysis yields insights that simpler techniques do not. Maintain your skepticism - but make use of more complex analyses on more complex problems. And carefully review the results, of course.

2. In any case, I hold the view that if a pile group can be shown to be safe and stable without the added lateral stability provided by the soil, then all will be well (so why bother with the extra analysis?). Of course, I have always allowed for the presence and nature of existing soil when considering the buckling capacity of individual piles.

I'm a little puzzled by this comment; I assume that you mean that you prefer to pick up the lateral loads by frame action rather than by lateral load analyses. That's okay on smaller structures, but can be expensive for the owner on larger projects. And it ignores almost 50 years of successful offshore design experience using p-y curves to model soil behavior. I can't imagine designing fixed offshore platforms without using p-y curves, or some similar approach. Your initial statement, "My comments mostly originate from experience with pile groups for marine structures and bridges, where it is common for the piles to extend a considerable distance above the adjacent soils" suggests that your experience does not include much offshore design experience, where p-y curves are used on virtually every fixed platform project. How reliable are offshore foundation designs - most of which included lateral load analyses using p-y curves? To my knowledge, not a single platform has failed as a result of the failure of the foundations themselves. That is an enviable record.

3. On the topic of 'realistic' results, I am reminded of the learned geotechnics professor who did an extensive analysis of the many pile capacity formulae available and compared them with a raft of test values. His conclusion was that none of the formulae gave a lower coefficient of variation than the simple assumption that all piles have a capacity of 93 tons!.

You're probably talking about Ralph Peck's excellent - and brief - criticism of the use of pile driving formulas to predict static pile capacity. His paper compared the use of the Engineering News formula (not to be confused with the Engineering News Record) with drawing pile capacity numbers from a hat. It's a real classic in every sense of the word - and a must-read for any young engineer enamoured with predictive formulas. A copy of that paper is included in the excellent book, "Judgement in Geotechnical Engineering - the Professional Legacy of Ralph B. Peck", which was published in 1984.

But be careful here - Ralph did not mean to imply that the use of refined predictive abilities should not be used. He was simply cautioning the practicioner against using ill-considered "cookbook" formulae in place of sound engineering design - and sound engineering judgement. If you doubt my interpretation, call him. While he is advanced in years, he is still listed with directory assistance in Albuquerque, NM. He is a great gentleman and a real giant in the civil engineering world. I will miss him when it is his turn to travel to that land from which no man returns.

It is also possible that you may be referring to the paper by Roy Olson and Norm Dennis, or one published by Mike O'Neill. I have read them both; I do not recall the "93 tons" in either of them, although Olson and Dennis were reviewing pile driving formulae and could have referred to Ralph Peck's paper in theirs. (Olson was at Illinois with Peck for a period of time before he went to The University of Texas.) If I haven't identified the source you are quoting, please give me the citation - I'd like to read it.

4. Bowles "Foundation Analysis and Design" has a bit to say on the subject. His third edition has a Fortran program for the analysis of three-dimensional pile groups, which should not be unduly difficult to convert to Basic or whatever if Fortran is not convenient.

My first quote (above) from your posting includes the statement, I also wonder just how reliable the inbuilt assumptions in such software can be... I am very skeptical of much of Bowles' work because his approaches are often simplistic and overlook many relevant factors. That criticism applies to his 3-D pile analysis routine.

Frankly, I don't care much for Bowles' approaches to more complex geotechnical problems. That is not to say that I believe that complex analyses are always better; they are not. But he makes no effort to educate the reader about more sophisticated approaches, and leaves the reader with the impression that his (Bowles') methods are always sufficient. They are not.

I am reminded about a comment that J.E. Bowles made about engineering software in about 1984 or 1985. As I recall, he declared that serious engineering programs would never run on PCs. The words "serious" and "never" struck me: even I knew better than to make such an absolute comment. I believe this comment was made at an ASCE conference on computing in civil engineering; it both annoyed and amused me because I was already running COM623 on a PC with only a 5 MHz 8088, 256k of RAM and no hard drive! That was a very foolish prognostication by a well-known engineer. And I lost a lot of my respect for him at that point.

MrStohler:
While my discussion (above) was not specifically directed at your question of "how "realistic" the various methods of analysis" are, I hope they have shed some additional light on the subject. Soils are clearly more variable than steel or concrete; but the appropriate factor of safety depends a great deal on the design and method of analysis. And excessive factors of safety can actually cause failures in some circumstances. The approach you outlined in your December 22, 2002 posting seem okay for smaller structures where you do not need to use more sophisticated design approaches.

If you are in Ohio or northern Kentucky, you might look for a geotechnical engineer named Ron Ebelhar. I don't remember where he works, but he is a very competent and experienced geotechnical engineer. And he understands many different design methods, including p-y curves, the lambda method for axial capacity, and others. Look him up if you need help.

 

[austim]

Focht3,

You are right in assuming that my experience has rarely ventured more than 2 miles from the coast. And of course there is commonly a major difference between pile groups used for deep water wharfs, dolphins etc and offshore platforms. The latter are frequently (if not invariably ?) designed as braced jackets with triangulated bracing over the full height from bed level to the deck. That may be suited to the use of mainly vertical piles, with transfer of lateral loads via lateral soil pressures on the upper regions of the piles. (p-y analysis and all that).

In contrast, few major wharves are designed in that way; most use pile groups with nothing but piles between seabed and the deck.

Granted, if you are thinking of a jacketed platform, then the 'free standing length' of the piles is negligible, and the pile/soil transfer mechanism for horizontal loads is structurally scarcely any different from everyday land-based buildings on piles (except for the radically different soil conditions, of course).

However, those structures where there is a significant gap between 'ground level' and the loaded superstructure have to be designed so that all loads can be tranferred through the piles alone, (since there is nothing else available to assist). Under these conditions, a prudent designer will arrange the geometry of the pile group in such a way as to minimise bending in the piles (some bending is inevitable, from loads applied directly to the piles, such as wave, current and boat impacts). With due care, the group can usually be designed so that all loads applied to the superstructure are tranferred to the foundation soils by direct axial loads (compression or tension) in the piles.

That results in the pile-soil load transfer being largely by longitudinal skin-friction, and 'lateral' effects (p-y etc) only coming into play for resistance to end fixing moments from wave loads etc, and for limiting the buckling lengths of the piles.

Going back to my original comments, I would still be happy to treat most pile groups (other than groups of purely vertical piles) on the following broad basis:

1. Assume a position of 'effective fixity' of the piles at 8 - 15 diameters or so below ground level, depending on soil conditions. (Also refer back to ASCE papers by Francis, Stevens and Trollope on the analysis of pile groups and slender piles).

2. Analyse the entire structure (including both vertical and battered piles) assuming piles fully restrained at their lower ends. This can be done by any one of many general 3D structural analysis packages, or more simply by one of the more specific programs. (We were using some very simple programs that worked on this basis long before I had ever heard of Bowles).

With proper attention to pile group geometry, design by such a procedure will result in a pile group that is not critically reliant on the lateral stiffness/strength of the upper regions of the in-situ soils (since the longitudinal skin friction required my be developed well below the surface) and is therefore more able to cope with unforeseen soil conditions. Also, since pile bending stresses will be minimised, economy of pile material should be ensured.

I'm sorry, but I can't give you a precise citation for the 93 tonnes. That came verbally from a past colleague, who was quoting his Soil Mechanics professor at Monash University (from the late 1960s). The actual figure may have been anything from 90 to 95 tonnes (my memory is losing some cohesion with age ). It was only the basic principal that I was trying to bring forward, not the particular number.

 

[BigH]

What a wonderful thread - to see Engineers at "work" not only in theory but also on a philosophical plain - whether that of design, or outlook or life, in general.

One of the central debates that comes to the fore is that of the theorists vs the practictioners - the professors so to speak vs the doers. What a wonderful time it must have been (catching the coat-tails, I was) when professors and practitioners were more commonly tied than what appears today. 5th order equations vs 1st order input!!! The other day I saw a paper using genetic logic in engineering design of pavement. Interesting concept????

Focht 3 is right in that the complex analyses may/will permit insights that we could only dream about in our more simplistic approaches - but we all need to gain the experience and/or have the advantage of engineers like the ones contributing heavily in these forums to show us the light of complex insight vs comlex fallacy. We need touchstones to keep our feet on the ground (or in the ground for us dirtmongers) so we stay true. I still suggest to all young engineers to get the books/papers by the pioneers - Terzaghi, of course; Tchebeterioff; Krynine and Judd; Tomlinson; Chellis; White; Bjjerum; Skempton - these guys pioneered the field of geotechniques and wrote in a most logical fashion to show us how to think; not just apply numbers and codes. The road to proper visualization and application of our field is what appears, to me, to be mostly missing in today's literature.

Keep up the great discussions in 2003; these help me keep my sanity and "base" while in the sub-continent.

Best new years cheer to you all and families, most.

 

[MrStohler]

Thanks for all of the feedback. It appears as though I opened a bigger can of worms than expected.

More background about my design problem: Altough many of your responses address situations where portions of the piles are exposed, that is not the arrangement I face. Do to underground utilities and existing foundations, the pile groups for several foundations at an industrial facility addition take on less than conventional arrangements (column not at centroid of group, batter piles to resist lateral load may be only on one side of group, or these two conditions combined). All piles are 12" diameter auger cast.

As I understand your comments: The method I described in the Dec 22 post is reasonable for smaller (what determines small?) structures. There is not a "usual" method for performing analysis of pile groups with battered piles. LPILE is one software package that can perform this analysis, the software user should be aquainted with P-y analysis. The ASCE papers by Francis, Stevens and Trollope provide valuable insight (are these available from the ASCE website?)

Please correct me if I am mistaken.

Is anyone aware if an organization such as DFI is attempting to bring together the reaserch and publish manuals or design guides, as ACI, AISC, AISE and many other organizations have done for thier respective areas of expertise? Am I alone in thinking that this could help aoid many problems, even if the guide directs the structural engineer to seek out a specialized design proffessional?

Thanks

BigH, Greetings from Wadsworth, GO OSU & Browns

 

[BigH]

Yeah, the Browns - what a phrase - THE BROWNS!! great feeling on the cold shores of Lake Erie tonight!!

 

[Focht3]

To All - I hope that your holidays were filled with joy.

MrStohler:
Don't worry about the can of worms - I love to fish.

Your proposed 12 inch auger cast piles can't take much moment, so you will have to use frame action to pick up the lateral forces. Or use a different foundation type.

Other resources on group action: EPRI (Electric Power Research Institute) has supported research on single and group piers, piles, etc. Their studies are not public information, and belong to the member electric utility companies. Contact any friends you have at your local power company's engineering department and see if they will let you look at the materials. (I doubt they will give you copies.) You'll find some of Fred Kulhawy's work there -

BigH:
A great post, as usual. We need to be grounded in the works of the "great masters" if we are to truly understand the principles they have left for us. It is also important to understand that most of their works were not intended to be "the" final answer on a given subject, but rather the best answer that the researcher could conceive given the available resources and knowledge of the time. I have had the priviledge of meeting quite a few of the well-known "names" in geotechnical engineering, and not one ever told me to stop thinking and researching a topic because he/she had found "the end of the path"! (Well, Fred "the Troll" Kulhawy did once, but that's another story...)

austim:
I've done more near shore design than offshore, so I am very familiar with your problem types. Most of my marine work has been along the Texas Gulf coast (Ports of Houston, Galveston, Corpus Christi and Brownsville) with some consulting in Singapore and Pensacola. The sites have typically involved mudlines deeper than -40 feet MSL.

I agree that most designs are done without the benefit of p-y analysis, but would argue that this is primarily due to a lack of familiarity with the technique, and not because it is of no benefit. Most docks see relatively small lateral loads under typical service conditions; hurricane wind loads on moored ships an obvious exception to the "typical" case.

But protection, breasting and mooring dolphins can all benefit from their use. By designing these elements as relatively flexible single members rather than as fairly rigid battered pile groups, cost savings can be realized since they are less likely to be damaged by the hull of ships (and barges) and can incorporate smaller fenders to absorb the force of impacts. And economy is obtained by the use of only one member rather than four or more. Installation time for a few larger members can be significantly less than for four to ten times as many smaller piles.

 

[JB69]

Some non-technical partners and I have a floating fishing cabin on the Texas Gulf Coast near Matagorda Island. We have had much trouble with the minimal jetted pilings we've installed getting worked/pulled out by Norther storms during the spring months and the cabin ending up high and dry on the island adjacent to it. Obviously, the single pilings we installed have been inadequate for the lateral loads imposed by a) winds pulling on the steel cable draped around the pole or b) dynamic wave loads that cause the cable to yank on the piling with the full force of the 9000# cabin. The cable is about 100' long and is attached to the cabin on two corners of the 16'x24' cabin.

We've installed two pilings to date: 1) a 8-10" treated wood piling about 20' long, jetted about 10' into the saturated silt and fine sand bottom and 2) a combination piling consisting of an 8" 20' galvenized pipe (tinhorn) with a 3" ODx30' drill pipe in the center and concrete in the annular space, jetted to about 15' below bottom. Water depth is about 3.5-4.0 feet. The wooden pole was wallowed out by a storm in 2001 and we haven't seen it since. The combo piling is listing at 20-30 degrees to vertical, but is still intact. It will have to be jetted out to restore it to vertical.

We are limited to manual methods that can be accomplished by 4-5 weakling professionals and a couple of 22' fishing boats for piling installation, as pros will charge as much as $10K to install a dolphin.

My current plan is to install a 3-piling dolphin, including the current listing one, using a 3" trash pump and PVC stinger jet pipes to install these. After installation, sand will be pumped back into the conical holes from a distance away to backfill the holes to grade. That's a trick I didn't learn until recently.

Any suggestions on a) technique or b) an easy way to determine adequate piling number, type, grade, or depth? Jetting is definitely the cheapest and most accessible method of installation we've found, and my choices of pilings now include 1) treated wood pilings (20' long) or 2) helical screw anchors. We can't use a hydraulic power unit and impact driver with the latter, because the power unit is over 1200# and too heavy for an ordinary fishing boat. It should be possible to jet the screw anchor in though, I believe. Other threads have suggested that the helical anchor won't support as much lateral load as a larger diameter piling. I'd like any thoughts on that issue.

The design storm would be about 75 mph and the area of the exposed end of the house is about 200 sq.ft. Again, the house is about 9000#.

Thanks for any help on this you can provide.

 

[Focht3]

JB69:


Matagorda Bay - design wind of 75 mph? That would have been blown away by Hurricance Alicia - a minimally Category 3 hurricane. (My parents owned a beach house at Pirate's Beach on the west end of Galveston Island at the time. It survived quite well.) The appropriate "design" wind speed should be at least 105 mph to 130 mph if you want it to have a prayer of a chance of the house surviving even a near miss of a hurricane and not ending up in someone's yard in Bloomington or La Ward.

The current storm predictions suggest that as many as 8 hurricanes may form in the Atlantic / Gulf of Mexico this year. To top it all off, we have averaged a hurricane landfall along the Texas Gulf coast every 6 years; and we haven't had one for 20 years. We're long overdue. I found the following reference in the Handbook of Texas Online, In a short time, Indianola achieved the rank of the second port of Texas, a position it held until the catastrophic hurricane of September 16, 1875, devastated the low-lying city and caused great loss of life. (For those of you that don't know Texas geography, Indianola is on Matagorda Bay.)

You will need to jet your piling group to a minimum depth of 25 to 35 feet, since you need to account for the effects of scour of at least 8 to 10 feet (resulting water depth of 11 to 14 feet.) and let's not forget the effects of a storm surge of 4 to 15 feet, depending on the cabin's location and the hurricane's path. That will give you a total possible depth to mudline of 15 to 29 feet. Maybe you should make your cables longer...


Trying to protect your floating paradise in this way may be a fool's errand. Is it really worth the time, effort and expense?

What about using a deadman anchor made of scrap concrete beams or columns?

 

[JB69]

Focht3, thanks for the info. I hadn't intended to add to this thread, but instead to create a new one. Thus the repeat in the new thread 256-58237.

Glad you suggested use of concrete deadmen. Had considered that, but had dismissed earlier due to the weight considerations. These could be installed by jetting a trench and backfilling, I suppose, and assisted by driving long T-posts along the length of each beam.

Whatever we can do to keep the floater where it is in a hurricane is the desired endpoint. Perhaps the design wind should be 200 mph?!

 

[Focht3]

Nah - if the winds hit 200 mph, your paradise won't be worth keeping! I'd run the numbers at 105 and 130 mph, then make a judgement call. Over a few cold brews with your fishing buddies, of course.

Deadmen anchors have another advantage: they will sink over time, increasing the soil's passive resistance to pullout. The cables will pull through the mud under high loads, leaving the deadmen to be pulled through. That can give you a heck of a lot of resistance! Just be sure to add cable from time to time...