Locked Prop vs. Freewheel vs. Boat Speed

Ross · Jan 9, 2009

I have a PSS system almost no friction there just transmission drag.

Al - s/v Persephone · Jan 9, 2009

Some experimental results

At the risk of not fitting-in with the theorizing about the subject, I can offer two real data points from personal controlled experiments.

While the spring and fall offer good winds between Capes Ann and Elizabeth, we have frequent light airs during summer. The subject of getting the most speed out of them has been a bit of an obsession with me, and the prop drag minimization question led me to do some experimenting.

My last boat had a 2-blade prop and, asI had been told it would, exhibited the least speed reduction when fixed.

With no other changes to sails, course, etc., when doing 4 or 5kts in flat water, fixing the transmission in reverse consistently gained me 0.25kt which I lost when going back to free-wheeling.

(Each time I tried this, I repeated the changes several times, with 5min in each state, which allowed the speed to stabilize.)

That boat, and my present boat, are similar hull-forms (1999 Hunter 310 and 2005 Hunter 36).

I was very surprised, therefore, to find that my present boat's 3-blade cost me 0.5kt in the same conditions when I fixed it, but the reduction was quickly gained back when I free-wheeled.

Since I had taken the precaution of verifying that the cone-clutch transmissions on my Yanmars in both boats could free-wheel without lubrication issues, I have put over 6,000nm of shaft-spinning on my transmission with no apparent ill effects on it or my cutless bearing.

When I'm squeezing 3.5-4kts out of 5kts of true wind it makes a big difference. The extra 0.5kt has often made it possible to get somewhere sailing without overly trying my wife's patience, when others were motoring.

Fair winds,
Al
s/v Persephone

Brian M H23 · Jan 9, 2009

Here's what I don't understand. There are 2 main variables: speed of the water, and rotational speed of the prop. If driven at some rotational speed, the prop will be neutral with the water, and cause no drag. Faster it will push you, slower it will cause drag.

As you continue slowing the prop down, it should cause more drag, provided you speed is constant. But then at some point (slow, but not stopped), the drag starts to go down again. why?

RichH · Jan 9, 2009

Brian M H23 said:
Here's what I don't understand. There are 2 main variables: speed of the water, and rotational speed of the prop. If driven at some rotational speed, the prop will be neutral with the water, and cause no drag. Faster it will push you, slower it will cause drag.

As you continue slowing the prop down, it should cause more drag, provided you speed is constant. But then at some point (slow, but not stopped), the drag starts to go down again. why?

"will be neutral" ... NOT SO - water has viscosity (liquid friction) and to 'move' water is to do 'work'.
The properties of fluids involve 1 of 3 regimes (Reynolds number): Laminar flow - where the flow is all aligned at the streamlines: little friction; Turbulent flow - where the flow velocities cause extreme turbulence (relatively high friction values); transitional flow - somewhere between high and low friction.

There is ALWAYS 'shape' of cross sectional drag. The other component, Parasitic drag is dependent on speed (and roughness of the body) in relation to the fluid - the faster the speed the more the amount the parasitic drag. The 'faster' the prop spins, the more the parasitic drag.

"But then at some point (slow, but not stopped), the drag starts to go down again. why?" because the flow regime changes from turbulent (or transitional) to Laminar Flow .... Laminar flow involves the least amount of friction - less work.

tsmwebb · Jan 9, 2009

Maine Sail said:
Rich,

How is it that your rather scientific explanation differs so much from the study conducted by MIT that specifically measured sailboat props & drag? While your points certainly sound convincing MIT researchers disagree with you on fixed vs. locked..

Here's a direct quote from the study:

I still lock my prop to avoid undue wear and tear but how can you and MIT be so vastly opposite on this point?

I hate to link another (doubtless inferior) community but while googling to find the MIT study I came across this:

http://www.sailnet.com/forums/seamanship/654-gear-2.html

Which answers your question so directly that I don't want to paraphrase it.

I didn't find a link to the MIT study in question. Anybody have one?

--Tom.

gettinthere · Jan 9, 2009

3 blade Max-prop, I shift to reverse to fold it.

tsmwebb · Jan 9, 2009

Maine Sail said:
Here's the study: Comparison of Ten Sailboat Propellers (LINK)

Thanks, that is interesting. I have just skimmed the paper. It might be worth noting that no tests were done wrt the difference between fixed and free wheeling drag. Technically, no tests were done on either fixed or free-wheeling props as the fixed drag numbers were derived from slow turning props. More importantly it appears that the conclusion about drag differences between fixed and folding is derived solely by doing two eyeball extrapolations followed by a bit of by guess and by golly fudging about friction with drag then derived by a simplified calculation for a single large area fixed three bladed prop. The received wisdom is that props that have large blade areas are the ones most likely show reduced drag when rotating. They only considered the prop most likely to benefit. So, it seem reasonable to read their conclusion to be that some fixed props will show a substantial reduction in drag when free-wheeling. That is, I think they are trying to suggest a potential rather than making a rule for the entire class. If I'm wrong then they have provided exactly no evidence or theory to back up the claim that all fixed props have less drag when free-wheeling in the paper you cite.

--Tom.

RichH · Jan 9, 2009

Maine Sail said:
Rich,

How is it that your rather scientific explanation differs so much from the study conducted by MIT that specifically measured sailboat props & drag? While your points certainly sound convincing MIT researchers disagree with you on fixed vs. locked..

Without seeing the exact parameters (other than presented in the 'laymans version', I cannot as a fluid/thermal scientist comment directly.

However as far as I can ascertain from the 'laymans version' certain parameters that would validate the 'test stand' are astonishingly absent:
a. isokinetic sampling of the velocity of fluid through the piping to assure constant and equal velocity through the cross section of the CONDUIT. Sailboat propellers do not operate in conduits, they operate in open flow or zero flow conditions.
b. that there are 4 corners or 90 degree angles to the test stand ... no verification nor mention of turbulence control or 'vaning' at the 90 degree to prevent undue turbulence at the elbow sections - thus easily skewing the results (for all samples).
c. the testing was done in a conduit (pipe) and is not equivalent to the flow regime of 'propellers' on boats: open flow or in a stream where the relative velocities of the fluid are NOT influenced by the pipe/conduit walls NOR the turbulence generated in 'elbows'. I think the rejection of other hydrodynamicists (sailnet discussion) is based on tank testing where shapes are pulled/dragged through the fluid which more closely relates to the 'actual' flow found in open water - not in the turbulent flow regimes of a PIPE. The diameter of test stand piping is not stated but at the stated max. velocities would indicate very turbulent flow.
d. if sailboat propellers operated in a 'ducted' or enclosed piping system this would be more valid. The test as presented would only (to me) to seem valid if the sailboats were normally navigating the Niagara or Colorado Rivers or caught in the penstock of a hydro-plants turbine intake .... in probably completely TURBULENT flow.

The sailnet discussion (I didnt read it all) seems allude objection from those hydrodynamicist who I assume would do or did 'tank testing' where the flow TO the propeller is at a laminar condition BEFORE impact with the propeller - a more valid & real world comparison.

My initial conclusion from the 'laymans version' (and without reading the 'real report') - ***HIGHLY DOUBTFUL and DUBIOUS*** ....
... due to apparent undefined control of turbulence INSIDE the teststand (isokinetic states are undefined) AND non-linear comparison of DUCTED flow vs. OPEN Stream flow regimes AND ESPECIALLY the usage of those 4-90 degree 'elbows' creating even more turbulence --- all of which can easily skew such results. Tank testing will result in less contrived results as it is closer to the 'real world' for propellers on sailboats.

The MIT (laymans version) paper seems VERY flawed to me with respect to especiallly the freewheeling props ... but I havent read the 'real' report.

Here's a simple challenge for you.
In still and quiet water, take a prop and fasten it to a float, the prop submerged and fixed to a shaft and held in the proper position while being towed. Get a simple force gage and pull the prop behind your boat but off to the side of the boat's slipstream (isokinetic regime) ... fix your speed carefully and measure the force on the 'tow line'.
Repeat with the prop spinning on the shaft with bearings, etc. ... compare the results.

gmeyerga · Jan 9, 2009

locked prop

This is my zero order approximation, derived from intuition rather than with a research grant:
When a prop srews itself forward at exactly the boat
speed I would expect to see no drag. When the prop is braked by shaft friction it sees drag. When it is locked the drag is larger than
when it spins. When the prop spins in reverse the argument is over.
Of course, the boat does not move through molasses, but I would
think that the flow across the prop at 5 knots or so is not all that
turbulent so that the energy argument should hold on average.

RichH · Jan 9, 2009

I scanned the MacKenzie paper and although it contains a lot of hypothesis it is clear with respect to the supporting empirical results.

Still and with respect to the physical laws that control hydrodynamics, until one can arrive at the 'reason' and mathematical substantiation for this apparent contradiction ... the jury on my part will remain 'out' ... - as others have reported just the opposite conclusions. Such acceptance is never like interpretive pursuits ... will have to wait until a detailed peer review and mathematical explanation/proof is completed and the 'hypothesis' portions are removed. Until then its still a lot of hypothesis even that the offered data supports the hypothesis.

Im 1800 miles away from my library and wont return until mid-summer ... so I'll have to think deeply on this until then.

kendall · Jan 10, 2009

gmeyerga said:
This is very confusing. Why would there be less drag on a fixed prop than on a
spinning prop? A system left alone wants to be in a state of minimum energy.
If it takes less force to drag a fixed prop then it would not want to spin.

Nonetheless, I usually put the tranny in reverse because I was TOLD it provides
better lubrication. Now I'm confused again. What difference does it make whether
it's in forward or reverse? Either way nothing turns inside the transmission (at least for my Yanmar 1GM).
Keeping it in neutral, on the other hand, makes the prop turn things inside at rpms vastly different from driving it with the engine. I can imagine that this might
compromise proper lubrication.

Agree 100%
If it took LESS energy to pull a fixed prop than to rotate the prop, then it would follow that an unlocked prop would be stationary. The fact that an unlocked prop rotates should show that it uses LESS energy.

Energy does not transfer to a higher load, (would be nice, you'd always have more power when you needed it) if it was easier for a prop to remain still it would, it doesn't. therefore less energy is used in turning the prop.

Simple experiment, grab a small prop or pinwheel and hold it under a running faucet, allow it to rotate and then again while you hold it stationary.
See which one produces more drag. (allowing the water to pass between the blades is a cop out and would only be used by those who own marked cards)
I await your responses.

An experiment I conducted a long while ago used pinwheels held out the car window. when free to rotate the stick bent back a small amount, (which meant some energy was being absorbed by the wheel) when it was held stationary the stick broke.
The ONLY possible interpretation is that there was more drag when the pinwheel was held still than when it was allowed to rotate. (which means that more energy than the same stick on the same pinwheel was able to handle being absorbed from the water)

My view is that the ONLY thing that should be considered is whether the transmission is able to handle freewheeling. If nothing will be harmed, let it spin.

Ken

Roger Long · Jan 10, 2009

Maine Sail said:
Perhaps you or Rich can share with us where this study is flawed (this was a tank test). Sailboat Propeller Drag - P.M. MacKenzie, M.A. Forrester - Journal of Ocean Engineering 2007 (LINK)

Well, if you saw the second History Channel Titanic show, you saw that that great scene at the end where I'm saying, "Uh, gee, I was wrong." Now SBO gets it's very own Roger Long mea culpa moment.

This is a much better paper than the one I read long ago and I think the experimental set up was sound. Experimental results with actual installations would probably yield slightly different results but I would bet now on freewheeling producing less drag in most modern prop installations. I'm less sure about props in deadwood apertures and the fact that this configuration was far more common back when I fixed my mind on the subject could be signficant.

The reverse is clearly true in aviation but I realized a major difference. The prop is on the front so the disturbance created by it's rotation effects the whole fusalage of the airplane (talking about singles here because twins almost always have feathering props).

Until something better comes along, which could happen in a subject this complex, I'll go with this paper. However, my transmission is one of those that says to put it in reverse for sailing and I not going to ask it to change its habits at its ripe old age

BTW, in the Titanic show, I'm admitting to being wrong just seconds after the computer analysis being run in real time has conclusively proved my primary thesis. The secondary thesis was a lot sexier though so the producers focused on the dramatic TV moment.

Here's a short trailer for the first show:

http://video.aol.com/video-detail/titanics-final-moments-roger-long/592058749/?icid=VIDURVENT03

Grizz · Jan 10, 2009

Hail Martec!

Wow. 76 posts (and climbing) w/ all sorts of thoughts, opinions, papers dredged up by a wide range of members w/ a wide range of experiences. Now I appreciate my 2-bladed, folding, twice refurbished Martec soooo much more, 'cuz it avoids the issue entirely.

Since I'm off to shovel the 6+ inches of white stuff that fell last night...is a curved handled shovel better ergonomically than a straight handled?

tsmwebb · Jan 10, 2009

Maine Sail said:
Golly fudging... Interesting. I don't usually hear MIT and "golly fudging" in the same sentence but..

Well, that's exactly what they did

Maine Sail said:
Perhaps you or Rich can share with us where this study is flawed (this was a tank test).

Sailboat Propeller Drag - P.M. MacKenzie, M.A. Forrester - Journal of Ocean Engineering 2007 (LINK)

I'd take my thoughts on this with a grain of salt if I were you.

I haven't had a chance to read the paper. I did skim it really quickly. If I'm not mistaken they are using a single prop for testing and then using a polynomial interpolater to extrapolate to a broad class. Is that right? Did they ever get published? Was there any feedback? Anyway, it is a very interesting article and I will try to read it carefully when I have some more time. Thanks for pointing it out! It's more fun than most of the trash that we have in the book exchange here and I'm even father from my library than Rich is... And, pondering this is way more fun than dealing with Mexican customs who are currently holding my solar panels -- blast them! :cussing:

The place where this has been tested in real conditions was racing under the old IOR rule. As the rule evolved credit for fixed props was reduced to the point where all serious racers went to folders. However, there was a period when even serious racers considered fixed props. In racing conditions it was clear that small area two bladed props created less drag fixed than spinning. I think the strut and shaft even on boats with "modern" under-bodies creates a significant disturbance and it may be that hiding the top blade in their wake reduced the drag they would have seen in a free flow. This reasoning is purely speculative on my part, but when lab results conflict with real world observations the lab tests are flawed.

I have seen a theoretical paper on this that had a convincing, to me, argument for the assertion that the drag wrt fixed or spinning will depend (maybe one of the old AIAA symposium papers) but I'm darned if I can find it on the net. :cry: At any rate, with some folks reporting reduced drag and with the knowledge that in some instances (autogyros, auto-rotating heliocopters, maple seeds, etc) rotating props are "dragier" than fixed I'd strongly suggest that anyone concerned about this should do their own testing.

--Tom.

Roger Long · Jan 10, 2009

tsmwebb said:
The place where this has been tested in real conditions was racing under the old IOR rule. In racing conditions it was clear that small area two bladed props created less drag fixed than spinning.

That could well be true, in fact, almost certainly is. Racing boats almost all had at least a paint mark on the coupling so the two blade prop could be lined up in the keel wake. Many also used narrow blade props especially for racing that were pretty inefficient under power.

I would also like to see testing done on props AND shafts at the large shaft angles found on many boats. A spinning cylinder creates quite a bit of lift thus drag. The effect of the wake thrown off a spinning shaft intersecting prop blades that are constantly changing their angle of attack due to shaft angle could alter the results.

Rick D · Jan 10, 2009

Funny You Should Mention Shaft Angle, Roger...

...once I was considering a Kiwi feathering prop. In talking to them they said an asset of the blades individual feathering is that there was less drag than a other feathering props in which the blades did not feather independently, because of the effect of shaft angle. BTW, I opted to pass on it since my HP rating was at the top of their allowable range.