Do your own homework and save fuel

[Leo Lazauskas]

Attached is an interesting paper from the Naval Post-Graduate School
in Monterey describing some simple optimisation techniques to save fuel.

Obviously, it isn't applicable to most members here, but it shows:
(1) Why doing your own homework is important :)
(2) How applying what you learn can save a lot of fuel.

I still find it a bit unusual that the methods described in the paper have not been used more widely.
Has anyone else seen similar methods applied to large vessels elsewhere?

Leo.

[DCockey]

The same principal applies to any boat with a humps in its fuel consumption vs speed curve. Planing boats are a common example. Attached is an example of a 17' boat with 115 HP outboard. I took the data from an outboard manufactuer's website. It's clear for this boat least fuel for distance traveled for any average speed below planing speed is acheived by staying still or idling, then running at planing speed. Worst of course is to be digging a hole at just below planing speed.

[daiquiri]

Leo, that was a very interesting reading, thanks! :)

I just have a problem with a part of the text where authors say (page 348) that:
"The result of the constrained optimization is “soft” guidance in the sense that the optimized variables provide the number of hours at each speed, but not the temporal arrangement of those hours within the transit. In the example in Figure 10, the planner can do 20 knots in the beginning and 40 knots at the end, or vice versa, or mix the two speeds up in some other way. Because fuel consumption is not affected, the temporal arrangement can depend on other tactical needs."
Strictly speaking, the bold-letter part is not correct, imho, because as the fuel is consumed the ship's displacement diminshes - leading to a smaller hydrodynamic resistance and smaller fuel consumption. So it should be advantageous to move at slow speed first (when the ship is heavier) and than at high speed (when it becomes less heavy).
Thanks again, and cheers!

P.S.
Re-reading the text again, I note that, at the page 346 they say "The incremental fuel usage is negligible." Perhaps they were referring to the above consideration?

[masalai]

Hi Leo, Thanks for your post... I was hoping to find you or Rick Willoughby to seek some guidance on re-proping my boat - link to (my little piece of peace), http://www.boatdesign.net/forums/boat-building/my-little-piece-peace-25962-new-post.html No mast or sails at the present so I am thinking of fixed-blade propellers for better mileage/speed performance... Bottom line is range at a reasonable pace and capacity to match tidal currents around Torres Strait ...

post 1291...
The Nanni saildrive uses a 17 spline shaft for propeller attachment. I have a Bob Oram 39C fitted with mini-keels and weighs about 5000kg, waterline beam length ratio is about 11.5:1 on a waterline length of 11.99metres... The boat is guite a slippery hard chine hull-form and does not seem to generate a 'bow wave' or much wake at all... The engines are Kubuto 21 hp and the gearbox 2.56 : 1, peak torque is at 2400 engine rpm and max hp/revs is at 3600 rpm... Engines seem to have a definite sweet spot at 3000rpm...

Anyone following Rick Willoughby's/Leo Lazerides formula for propeller design? - I could do with some additional views on propeller selection. I have a 300mm radius clearance for the propeller, so max propeller diameter would have to be less than 17inches but that may be too big for the available power... Present fitting is a "Redici" 2 blade folding of 15inch diameter x 10inch pitch...

post 1305...
"Engine linkage controls rectified and peak revs (3750rpm, which is 150 over-revving a bit, and still not assuredly full throttle, as the linkages may be adjusted to travel more?)... yields 7.9 to 8.2knots (two runs, with and against tidal flow averaged)... Using just one engine gave: - 7.2 (stbd) and 7.7 (port) and with both at 3000rpm gave 7.5 and 2400 returned 6.5 knots by the gps...

Existing propellers are Gori look-alike, - Redici - Italian made 15" x 10" pitch... Anyone with advice as to the fixed blade replacement pair? 17" x 11" three blade or what?

Would further speed trials/measurements help? (say every 500rpm running with the tide and similar in the opposite direction, doing 3 sets of runs - both, port & stbd engine?)

[Ad Hoc]

Leo

Interesting paper. But as alreday noted above..not ground breaking. Just a paper for the sake of it really.

Masali
Your prop. What is the engine kW and rpm at the prop and what speed do you envisage or know the boat gets? Prop selection is just about basics, speed of water in way of the prop, the power at the prop and the rpm.

[DCockey]

What does propeller selection have to do with thread? Why not start a new thread?

[masalai]

Apologies DCockey,
Leo does not post often so I will PM Leo directly, to avoid your 'snarley' comments and others may not be aware of the true breadth and wisdom that Leo brings to this forum...

DCockey, you have been awarded POSITIVE rep with a 'snarley' comment... (I do not believe "neg rep" has any useful value, and possibly, nor do you...)

[Leo Lazauskas]

Leo, that was a very interesting reading, thanks! :)

I just have a problem with a part of the text where authors say (page 348) that:
"The result of the constrained optimization is “soft” guidance in the sense that the optimized variables provide the number of hours at each speed, but not the temporal arrangement of those hours within the transit. In the example in Figure 10, the planner can do 20 knots in the beginning and 40 knots at the end, or vice versa, or mix the two speeds up in some other way. Because fuel consumption is not affected, the temporal arrangement can depend on other tactical needs."
Strictly speaking, the bold-letter part is not correct, imho, because as the fuel is consumed the ship's displacement diminshes - leading to a smaller hydrodynamic resistance and smaller fuel consumption. So it should be advantageous to move at slow speed first (when the ship is heavier) and than at high speed (when it becomes less heavy).
Thanks again, and cheers!

P.S.
Re-reading the text again, I note that, at the page 346 they say "The incremental fuel usage is negligible." Perhaps they were referring to the above consideration?

I thought of that too.
In my work I use the so-called Breguet Range which does take the fuel loss into account in a fairly crude way. I attached the equation as a gif because I still haven't figured out how to enter equations on BDN.
(I'm fairly sure aero engineers are familiar with this formula).

smax is the range (in kms), W is the total weight (in Newtons), RT is total resistance, Wf is fuel weight, eta is the overall propulsive coefficient, and sfc is the specific fuel consumption (in g/kW-hr).

DCockey: Thanks for another perspective!

Ad Hoc: I agree it's nothing earth-shattering. (In fact I wonder if it is even a refereed publication).
The most interesting part for me was that it came about as a result of a class exercise. That, and the fact that US Navy personnel are trained to bring turbines on-line in a practised and orderly way.

And it has maths as the hero of the story :P

Enjoy your collective weekends, chums!
Leo.

[rxcomposite]

Leo,

Glad this came up. I found the breguet formula in one of your papers. I tried to feed the formula into my excell spreadsheet a long time ago but am not sure if I have done it correctly. Excell does not have a log e function so I have to insert a base number of 2.718 for the log. I also have to multiply the constant by a neg 1 because I am getting a negative value for the answer.

Any hope I can use Breguet's formula for the English System? What constant do I use? I have to write both in English and metric.

Thanks,
Rx

[rxcomposite]

Interesting paper. Nothing new except the linear programming approach. Maybe I can use that.

You wrote a paper about ship's efficiency, comparing monohull, cats, tri, and other hull forms. I think that is where the breguet formula is. I use that as an inspiration to write a program in excell so I can predict my range by choosing a hull form. In this I have included the equipment number to forecast the probable weight to come up with estimated ship's weight. Adding crew, provisions, FW, and FO was easy as it depends on the ships SOR.

[daiquiri]

In my work I use the so-called Breguet Range which does take the fuel loss into account in a fairly crude way. I attached the equation as a gif because I still haven't figured out how to enter equations on BDN.
(I'm fairly sure aero engineers are familiar with this formula).

Yes they are, though not in that form (of course).

[rxcomposite]

imho, because as the fuel is consumed the ship's displacement diminshes - leading to a smaller hydrodynamic resistance and smaller fuel consumption.

I have read somewhere in one of my books, that by reducing the coefficient of drag (Cd), the equivalent of hydrodynamic resistance, an airplane is able to go faster. By going composites, all this bolt and rivets sticking out (contributing to the drag) was eliminated. Hence, this small reduction in drag resulted to higher speed/efficiency (at higher froude number). There was a formula posted as proof.

[daiquiri]

I have read somewhere in one of my books, that by reducing the coefficient of drag (Cd), the equivalent of hydrodynamic resistance, an airplane is able to go faster. By going composites, all this bolt and rivets sticking out (contributing to the drag) was eliminated. Hence, this small reduction in drag resulted to higher speed/efficiency (at higher froude number). There was a formula posted as proof.
That's correct and also quite intuitive, isn't it? :)
By the way, I've received your mail, will try to play with your spreadsheet this weekend...
Cheers!

[rxcomposite]

Thanks Daiquiri. Really appreciate that.

[Frosty]

I have read somewhere in one of my books, that by reducing the coefficient of drag (Cd), the equivalent of hydrodynamic resistance, an airplane is able to go faster. By going composites, all this bolt and rivets sticking out (contributing to the drag) was eliminated. Hence, this small reduction in drag resulted to higher speed/efficiency (at higher froude number). There was a formula posted as proof.

Countersunk rivets are used in high speed wings.

[DCockey]

The Breguet range formula which Leo posted assumes constant resistance to drag ratio, constant specific fuel consumption and propeller efficiency, and constant velocity. How reasonable those assumptions are for a given marine vessel would need to be determined for any application of the formula. Presumably the fuel weight would need to be a not too large fraction of empty weight.

In the aircraft world a different version of the formula is used for jet aircraft, which leads to higher speeds for maximum range than for propeller driven aircraft due to the difference is efficiency characteristics of jets vs propellers. I wonder if something similar would hold for vessels with waterjet propulsion?

[Leo Lazauskas]

Leo,

Glad this came up. I found the breguet formula in one of your papers. I tried to feed the formula into my excell spreadsheet a long time ago but am not sure if I have done it correctly. Excell does not have a log e function so I have to insert a base number of 2.718 for the log. I also have to multiply the constant by a neg 1 because I am getting a negative value for the answer.

Any hope I can use Breguet's formula for the English System? What constant do I use? I have to write both in English and metric.

Thanks,
Rx

The constant at the front of the equation I gave is
367097.8 = 60 X 60 X 1000 / 9.80665.

gravitational acceleration is 9.80665
60 seconds per minute
60 minutes per hour
1000 to get grams

In Excel you can use the LN function for logarithms to base e.

To convert from lbs/hp-hr to g/kW-hr multiply by 608.277

Leo.

[Leo Lazauskas]

The Breguet range formula which Leo posted assumes constant resistance to drag ratio, constant specific fuel consumption and propeller efficiency, and constant velocity. How reasonable those assumptions are for a given marine vessel would need to be determined for any application of the formula. Presumably the fuel weight would need to be a not too large fraction of empty weight.

In the aircraft world a different version of the formula is used for jet aircraft, which leads to higher speeds for maximum range than for propeller driven aircraft due to the difference is efficiency characteristics of jets vs propellers. I wonder if something similar would hold for vessels with waterjet propulsion?

Yes, it is a little crude, but it is not as pessimistic as other formulas that are routinely used by naval architects and that do not include the "log factor".
See, for example, the discussion by L.J. Doctors following the "High Speed Sealift Technology Workshop, 1997".
The discussion is in the report:
"Hull Form and Propulsor Technology for High Speed Sealift", ed. Chris B. McKesson, 13 Feb. 1998", John J. McMullen and Assoc.
I think it is still around somewhere on the net.

A better way is to calculate the drag separately for, say, the full fuel condition, the half-fuel state, and the "burned-out" state. Each calculation should be made for the appropriate displacement weight after reducing by the weight of fuel loss.

Whether OPC and sfc remain constant is, as you say, open to question.

Leo.

[daiquiri]

Just found the answer to my doubts in the previous page, regarding the variable ship displacement: http://www.mckesson.us/mckwiki/index.php?title=NAME_4177_-_The_Practical_Design_of_Advanced_Marine_Vehicles_-_Chapter_21&redirect=no

Citation:
"Finally, note that this effect (Variable Displacement) is only realistic if the owner uses it: if he doesn’t refuel, and doesn’t ballast. The military practice, for example, of never allowing the ship to get below ½ or ¾ “tank” will obviate the benefits of this calculation: In effect the owner is running his ship in a Constant Displacement mode, and it behoves the Naval Architect to perform the calculations accordingly."

[rxcomposite]

Thanks Leo. I used LN and the numbers came out the same. I did right except for the method. I feel more confident now.

However, If I use
g= 32 feet/sec
60 sec/min
60 min/hour
608.277 to get lbs

In the English system, I get 68,112 for the constant. The number 74,445 seems to work best. Anyway, I used;

Np= 0.6
sfc= lbs/hr
Rt - lbs
W = Long tons
Wf= Long tons
Range= Nm

I guess all this convertion added to the error or complexity of the equation.

[Adler]

Dear All,

I would like to attach on this thread some information.
Last two years we participate on a similar design/project that will be applied on FAC
(High Powered Semi - Displacement Boats).

[Adler]

Dear All,

On the attached file there are some data regarding the relation between Fuel Load and the corresponding measured Speed that were extracted due to Sea Trials on a semi-displacement vessel.
Should be interesting to graph the Derivative Curve of the presented Curve there.
Following this method you can define the range that the variance of fuel load could be have less influence to the fuel consumption rate.
Important when is examined the possible adding advantage of a Stern Flap or any other Hull's factor regarding to reduce the drag at a given speed.

[Leo Lazauskas]

Thanks Leo. I used LN and the numbers came out the same. I did right except for the method. I feel more confident now.

However, If I use
g= 32 feet/sec
60 sec/min
60 min/hour
608.277 to get lbs

In the English system, I get 68,112 for the constant. The number 74,445 seems to work best. Anyway, I used;

Np= 0.6
sfc= lbs/hr
Rt - lbs
W = Long tons
Wf= Long tons
Range= Nm

I guess all this convertion added to the error or complexity of the equation.

I was brought up with feet,tons, versts, pounds, shillings and pence etc but I am hopeless with them now. You should give up that Masonic nonsense and go metric :P

[rxcomposite]

I did went metric a long time ago but the USAnians in this forum have a hard time deciphering what I write so I have to do both.:):)

Anyway, thanks a lot Leo. I know my program is correct.

[Frosty]

I went metric Oh donkeys years ago. I went the whole 9 yards.

But I often do both --like 14 inches x 140mm.

In for a penny in for a pound I say .

Its the beer that confuses me --is 3x 330 Ml bottles 1 pint. or half a Quart.

Maybe its neither maybe its just 3x 330 Ml bottles.

On the hole I prefer mm for measuring drills though. 3 mm or 13mm conjures an immediate size in my head rather than a ridiculous 11/64ths. mind you I still keep good 1/8 for a pop rivet or two.

[rxcomposite]

I grew up using the English system but have to convert to metric due to standardization. Funny thing is my subconcious still pray tricks on me.

I am working on a drawing on ACAD. Metric setting of course. This afternoon, I went down to measure the additional part, I started measuring in inches. Only when I started drawing the part in ACAD did I realize my mistake. My mind is playing tricks on me.

It is like language. You learn to speak it but then, you start learning a second language.