Savitsky Power Prediction

High aspect ratios

I have compared probably the simplest Savitsky equation to the high aspect ratio calculation in Fluid-Dynamic Drag by S. F. Hoener. I found that somewhere before an aspect ratio of 5 the Savitsky calculation gave a higher lift coefficient than airfoil theory would give for even an infinite aspect ratio.

With a stepped hull the pitch angle is controlled by the geometry of the step and you typically can have higher aspect ratios, at least at higher speeds. The higher aspect ratios have better lift to drag too. I'd just suggest being a little careful and confirm with some reasonable reference that you are in a regime were Savitsky is valid.

Savitsky is I think also simplified compared to many real boat hulls. Ocean racing deep-V hulls all have strakes that are spaced and optimized, including ones that control trim angles by not going all the way to the transom. Maybe for this case Savitski would over estimate drag.

Many high speed boats also use air pressure to produce lift and isn't this outside Savitsky too?

 

XXL,

Please give me your e-mail address?

 

XXL,

Will you provide us all with the enhancement to support steps?

Dingo,

Can you provide a version without macros? The visual basic macros don't work on Macs.

 

u4ea32,

Sorry, but without macros it would be a tedious process to do. I would suggest running the spreadsheet on Windows, print out the macro code, and rewrite the code in whatever language you desire.

XXL, I agree with u4ea32; will you provide us all with the step mods?

 

Go to this web page to get the code I wrote in 1981 for the old programmable HP-41C. The search technique I describe on page 3 to solve for those terms may be helpful to you. Frankly I don't know if Excel can search for solutions (within a given tolerance) but it may be worth a look.

http://www.boatdesign.net/forums/design-software/planing-boat-power-prediction-22579.html#post203228

Good luck!

 

I use the "Goal Seek" tool in Excel to find solutions in my Savitsky short form spreadsheet. It's easier than writing macros for me.

 

1st of all Dingo, congrats and thanks for providing such a wonderful resource... don't suppose you have one for scantling calcs too do you!?!

I was wondering if you could tell me what some of the correct entries would be if using the spreadsheet for outboard or sterndrive installations...ie I presume the inputs for all the rudder shaft and strut entries ought to be set to zero..?

 

Mat-C,
A outboard or sterndrive can still have a angle of thrust, and added resistance due to the leg. In my opinion (note, this is my opinion, so by definition I can't be wrong!) I would add a factor for the extra resistance (drag) to the final results. This would be the best practically; a CFD model will give better results.
Note that this theory (and especially the spreadsheet) is only a basic estimate of the resistance; model testing will give better results.

 

Mat,
Dingo's right - there's absolutely no question that there is drag created by the 'leg' of an o/b or sterndrive. In practice, however, I've found that the results are very close with the inputs that you mentioned set to zero even though this is technically not correct. This is probably down to the choice of factor one chooses in converting from effective horsepower (EHP) to the actual required shaft horsepower.
Depending on the type, I've found good correlation using a factor of between 0.6 and 0.65

There's at least one scantling spreadsheet (using Gerr's rules) in the spreadsheet library....If you come across any others, be sure and let me know!

 

I was also planning to do this but if you beat me to it, good. Please send me a copy.

A planing boat is like an airplane that is cruising. Similarly, a kite (flat plate) is in a perpetual stall condition with the speed of the wind as velocity and the pull (tug) on the string as thrust. A program was written in Java about this, Search Aerodynamics/NASA about kite design.

I plan to write the code in excel but I have to figure out a way to present the integration of finding the Center of Pressure at varying angle of attack. The projected area would be similar to a series of triangles, one on top of the other. Any ideas?

If I can present varying areas at varying angle of attack, maybe I can integrate it into Dingo's work to balance it with the other forces (thrust, drag, vector quantities, lift, ect). I intend to work only on the projected areas of the hull.

Rx

 

Mat, For my outboard, I used the dimensions of the lower unit below the keel for "Rudder", and set "Shaft" and "Strut" to zero. I/O stern drive should work the same. I'm getting very close to observed results (depending on how you evaluate the efficiency of the drive train).

dingo, I really admire your spreadsheet work.
just a couple things I'm unsure about --
Angle of thrust line -- I'm entering as a negative value as my propshaft is pointing down (as opposed to the shaft in your illustration), does that sound right?
Lwl -- not sure -- is the waterline at rest, or the waterline on plane?

thanks!!

 

Negative Thrust Angle

Waterline is static waterline.

It seems that using a negative value for the angle of thrust line is a no-no. I looked at the formulas and using a negative angle would give a negative drag (!) and a positive "lift" force, whereas in fact it should be positive drag (acting aftwards) and negative "lift" (since, if a shaft, a normal force acting downwards would be experienced).

I will have to tweak the spreadsheet and re-submit it.

Dingo

 

I'm getting barely 1% difference between -5º and +5º, with +5º being favorable (reduces hp requirement).

"positive lift force" seemed reasonable since the downward force on the stern tends to lift the bow. It also seems to reduce drag, as RPMs rise as you trim out. I haven't got a clue how to model it, though. On a boat with power trim you adjust the angle on the fly. Never gave it much thought, but I wonder if direction of thrust aligns with direction of travel....

any rate, in my attempt to validate my understanding of your model by applying it to a boat with known parameters, I'm getting about 3% variance (est 33 knots vs. 32 actual). That's the equivalent of about a 3-inch error in my LCG estimate.

The implication here is that I could increase top speed by 5% just by moving one deepcycle battery about 10 feet aft. ...hmmmmmm...

It's a better prediction than I expected, and especially considering hull & engine both over 30 years old.

Most difficult parameter to measure might well be "actual horsepower". I know I'm losing right at 10% at the prop, and assumed a 15% crankshaft/propshaft difference in rated horsepower. Most would assume some power loss due to age of the engine; RPMs tell a different story though.

 

Try software at AeroMarine Research for tunnel hull design/performance and for vee hull design/performance. I think the algorithms used are modified Savitsky formulae, but seems very accurate. They've got alot of good graphic presentation of results. Their website also has several free papers written on powerboat performance.

 

dear mr.dingo

This spreadsheet is very helpful in doing my final project. please send me the english version of this spreadsheet. it would really help me in working on final project.



best regards