PDA

View Full Version : Freewheeling ?


Ian McColgin
12-13-2004, 09:42 AM
Below I've C&P'd a puzzler from Brion Toss's web site (www.briontoss.com) just in case there are WBF members so benighted that they don't turn to his Spar Talk and Puzzler sections already. Below the C&P I remark on some inadequacies to both explanations. I know we've folk out here with extensive air and water experience and hope there are formulae that indicate when it's better to freewheel and when not.

November 2004 Puzzler Winner
Hello all,
Last month's Puzzler, which came to us to us courtesy of Bertram Balch, was:
"What creates more resistance: a freewheeling propeller, or a fixed one?"
Bertram has seen conflicting data on the subject, and indeed many of you weighed in emphatically on both sides of the question. And indeed it is a bit of a trick question, because, as Ian McColgin noted, "... Like the answer to the question, "What's the best lead for a jib sheet?", the answer here is, "It depends."
First, a non-drag consideration. Some transmissions are harmed by allowing them to freewheel while the boat is moving. Not all transmission manufacturers' claims accord with boat owners' experiences on this matter, which simply tells us that many manufacturers' reps and engineers never went to sea.
The forces that contribute to prop drag when the boat is moving and the engine is off include: Resistance from the presenting surface area of the prop; turbulence behind the prop; and hydrodynamic (or aerodynamic)resistance of the prop. The first does not change much whether the prop freewheels or not. To the extent that it might diminish a little if freewheeling, the net's about the same, as there's some transmission resistance to counter. Turbulence is probably reduced somewhat if the prop is allowed to freewheel.
Dynamic resistance is nearly non-existent when the prop is held still. However, as anyone who's survived a helicopter crash can tell you, the power of a prop's autogyration is considerable. If a chopper loses power, the pilot will steepen the prop pitch a bit to get them spinning nice and fast and will then flatten the pitch to the point where they gain useful lift and are still spinning fast enough. The chopper can then be "glided" straight down.
On a sailboat with a fixed prop, the typical pitch is enough that the freewheeling prop is actually generating lots of reverse lift to resist the boat's motion. Most sailboats will experience drag from a free- wheeling prop equal to drag from a towed bucket a bit larger in diameter than the prop. The drag from a fixed prop is usually about equal to the drag from a bucket about 3/4 prop diameter if a three bladed prop and considerably less than that if two bladed.
Ian McColgin
S.V. Granuaile
A fine analysis, Ian, and it is seconded by this month's winner, Commander David E. Davis, USN. He writes:
Sorry but...it depends. According to the U.S. Navy Towing Manual, Appendix H, when you determine the strain for a towed vessel the propeller resistance is a function of the projected waterplane area of the prop, times a constant (3.737), times the velocity of the tow squared.
So the faster you tow, the faster the prop resistance increases. For example: for a typical destroyer the towing resistance attributed to the prop(s) at 6 knots is 34k pounds while at 8 knots the same prop(s) yields 61k pounds. This is based on a fixed controllable reversible pitch wheel. A controllable reversible pitch prop would be set max ahead to minimize resistance. But then some ships save fuel when steaming independently at higher speeds by simply trailing a freewheeling prop. This discounts any concerns re shaft seals, machinery lubrication, etc..
The bottom line is that there is likely a crossover velocity where the resistance of a fixed prop (braking effect) becomes greater than the force required to impart rotation to the entire gear train (in a freewheeling prop), but studies on the topic are scarce. Given the small waterplane area of most sailing vessel's props, the relatively slow speed through the water, and the overriding desire to minimize wear on the shaft seal and transmission, fixed is the way I'd go (at max ahead or better, feathered, if the pitch is variable).
CDR D.E. Davis, USN
Commanding Officer, Naval Diving and Salvage Training Center
Congratulations, Commander, Sandi will be in touch with you shortly to arrange for your Fabulous Prize.
******

OK - Seems to me that CDR Davis's remarks deserve the win due to their greater specificity for the fixed prop, but they are still about as speculative as mine for a turning prop.

I take it that the numbers must be an approximate average for the increasing resistance of a fixed prop as surely the dynamics would be more severe the deeper the prop is, assuming full immersion and no added complexities of air/water boundary effects. Additionally, a steeply pitched prop can freewheel more easily than a more normal pitched prop. The physics are the reverse of the phenomenon that it takes more engine power to turn a steeply pitched prop and, of course, is why the spinner on the patent log has such steeply pitched fins.

When a prop is powered, the venturi effect will be a pull normal to the forward face of each blade. When towed and freewheeling, the venturi is normal to the aft face.

Given that a helicopter can autorotate to safety if the freewheeling wings are pitched correctly and that every boat I've sailed would speed up just barely noticeably if the freewheeling prop was then locked, I believe that the drag from venturi effect on a free spinning prop increases more rapidly than the drag from turbulence behind a fixed prop.

But no doubt not always. Besides blade pitch, I should think diameter must play a huge role. And compared to a boat's prop, helicopter wings have an enormous blade area to tonnage ratio.

I'd bet that there are some props that start with higher resistance fixed but at some speed maybe they gain lower resistance and visa versa and there are likely even some props that have two crossovers. Anyone got any numbers?

It's like Bob Dole's famous answer to whether he wears boxers or briefs.

"Depenz."

(And Michael, please make your answer in English!)

mmd
12-13-2004, 10:18 AM
Whazzamadda? Don' you like my Bluenoser idiom? ;)

Dave Hadfield
12-13-2004, 10:33 AM
Ian, what about propellor-driven multi-engine airplanes that suffer an engine failure? They all, without exception, feather the prop to reduce drag. The reason given is that if the propellor is allowed to free-wheel in the airflow, the drag it produces is equivalent to the entire disc of its rotation. In other words, if the prop is 10ft in diameter, the drag is as if you had a 10ft disc facing the airstream. Gigantic drag. Down you go.

I always just assumed the nautical case was the same. That the drag from the 3 stopped blades was less than the whole disc of their rotation if allowed to free-wheel.

Does Cmdr Davis' info come from a test tank, I wonder?

Ian McColgin
12-13-2004, 10:39 AM
I just don't want:

{[(K)(sa)/rpm)]/[(t)(sd)]}{V squared}{NASA Cord #}

mmd
12-13-2004, 10:48 AM
Awww, c'mon! Sometimes with some subjects it is easier to speak Math than English. :rolleyes:

mmd
12-13-2004, 01:01 PM
Ok, Ok; I'll answer in english...

A very simplistic way to look at the problem: I'll use the body sections of the lovely Herreshoff "Bounty" fitted with an on-centreline 20" diameter prop (yes, this is an appropriate size prop) and assume a hull velocity of 8.5 knots (theoretical hullspeed for this boat is 9.5 kts and both boats that I have been involved with easily make this speed, so 8.5 kts is reasonable). Assume that the prop is shielded somewhat by the skeg ahead of it.

If the prop is fixed so that it "hides" as much of itself as it can behind the skeg, as shown in Fig. 1 (below), the exposed surface area of the prop creates a drag of:

P= 3.28 x A x V^2 (Barnaby)

where
P = force in lbs
A = appendage area in square feet
V = velocity in knots

P = 3.28 x 0.528 x 8.5^2 = 125 lbs

http://www.imagestation.com/picture/sraid150/p94f2fb5fdf69169d86b42811b5fdbb3d/f5ebde37.jpg Fig. 1

If, however, the prop is allowed to freewheel, the average exposed blade area is greater (see Fig 2 below), resulting in greater drag.

P = 3.28 x 0.590 x 8.5^2 = 140 lbs

http://www.imagestation.com/picture/sraid150/pd300d067da17fe64e01966d5935c1e3e/f5ebde35.jpg Fig. 2

As can be seen, this results in a better than ten percent increase in appendage drag when the prop is allowed to freewheel.

Of course, the actual problem is much more complex than my dinky little presentation here, but you get the drift, I hope.

Tom Lathrop
12-13-2004, 01:12 PM
No science, no math, no conjectures, just observation by a 10 year old fishing with a casting rod. It was clearly obvious when the jitterbug plug I often used got fouled by a bit of weed. The drag decreased dramatically and I'd just reel it in fast to clear the propeller on the nose of the plug before casting again.

The reduced drag of a jammed prop compared to a free wheeling prop was apparent at all retrieval speeds. This has always been my simple answer to this puzzle.

mmd
12-13-2004, 01:32 PM
I turned up this interesting post in the SailNet.com website whilst Googling for freewheeling prop info:

Message Board: Seamanship
Topic: In Gear

Date: Feb. 05 2001 7:15 AM
Author: Jeff_H

I just spent two days last week attending the fifteenth Chesapeake Sailing Yacht Symposium (CSYS). There was a room full of some of the world's best yacht designers and researchers. There was a lot of information flowing. I tried to get some answers to some of the common issues that have been discussed on various BB's either from the lectures or in discussions with presenters and yacht design professionals. One of the papers dealt with a simplified VPP (velocity prediction program) written on Excel and designed as tool for quick analysis of varying designs. In the follow up discussion members of the audience questioned the use of the particular coefficients of drag for propellers as presented in recent testing results. Practical Sailor had published one such study performed at MIT that suggested that a free wheeling propeller had less drag than a locked propeller. These results were roundly questioned by design professionals who had worked in the field of propeller design. After this discussion I had the opportunity to discuss this issue with a Professor of Mechanical Engineering teaching Naval Architecture and whose students had actually performed the same type of tests as the MIT students. His results are very different than those presented in the MIT study. Here's what I came away with in this conversation. (I also had the chance to discuss this with a number of other people in the field this includes inf from the general trends of this discussion.) 4 or more bladed, steeply pitched, propellers such as used on big ships, typically have less drag when permitted to free wheel than when they are locked up. This occurs because of the small incident angle of the water flow on the blades (therefore they are not stalled out) and the comparatively small amount of rotational friction of the propeller shaft and bearing when compared to the large amount of drive generated by the blades. In the case of Sailboat propellers, the pitch is quite flat and so the blades are generally in a partially stalled condition when they are allowed to freewheel. This is even more pronounced when friction is applied to the shaft and the blades are turning slower than the flow of water passing over them. In this circumstance the propeller produces a ball of turbulent water. This ball of water generally has a greater drag than a locked prop. This professor explained the MIT results by saying that the MIT study actually used a propeller that was powered at equal speed as the water to replicate a freewheeling state and so had a very low drag. In his study they allowed the propeller to actually free wheel and then applied increased friction to the shaft measuring the resultant drag. His study concluded that there was more drag in the freewheeling prop than the fixed one. He went on to add that the differences between locked and freewheeling were much smaller in a three-bladed prop than a two-bladed one. I also had confirmed in later conversation that prop position was found to have a real effect. A two-bladed prop locked in the vertical position and a three-bladed locked with one blade vertical in the down position had substantially less drag than the same props in other points of rotation. We discussed the idea of using the freewheeling prop to generate electricity. He indicted that in terms of drag this was the worst condition because the partially constrained shaft would produce the greatest drag. Apparently, the propellers designed for water driven generators are specifically designed to have a minimal drag using blades with a very large pitch. Conclusion: You give up a fair amount of speed if you permit your prop to freewheel. You loose more speed free spinning a two-bladed prop than a three bladed prop but loose the most speed with a partially constrained three-bladed prop.

Tom Lathrop
12-13-2004, 05:25 PM
Originally posted by mmd:
I turned up this interesting post in the SailNet.com website whilst Googling for freewheeling prop info:

Message Board: Seamanship
Topic: In Gear

This occurs because of the small incident angle of the water flow on the blades (therefore they are not stalled out) and the comparatively small amount of rotational friction of the propeller shaft and bearing when compared to the large amount of drive generated by the blades. In the case of Sailboat propellers, the pitch is quite flat and so the blades are generally in a partially stalled condition when they are allowed to freewheel. -----------------------

--- This professor explained the MIT results by saying that the MIT study actually used a propeller that was powered at equal speed as the water to replicate a freewheeling state and so had a very low drag. -------------------------

-------- Apparently, the propellers designed for water driven generators are specifically designed to have a minimal drag using blades with a very large pitch.
1. Strange, are they saying that a stalled or partially stalled blade has less drag than a completely stalled blade? That would be news to anyone who has experienced a stalled sail, keel or centerboard in a sailboat.

2. Even more strange. What did they hope to prove from this kind of experiment?

3. This also seems wrong. For best power transfer, the blade pitch would be set to extract max power at the preferred speed. Minimal drag is not going to give max power to the generator, just the opposite.

[ 12-13-2004, 05:26 PM: Message edited by: Tom Lathrop ]

PVanderwaart
12-13-2004, 05:32 PM
Playing with a kid's pinwheel, I noted that as I waved it, the resistance increased noticiably when the wheel began to turn.

Hwyl
12-13-2004, 07:14 PM
I pondered this back and forth for years. I generally came to Ian's conclusion. But well before that, I broke a propshaft while sailing (only). It was a Taiwanese built Vagabond 47. The shaft had a keyway, a set screw dimple and a through bolt all in the same cross section. I could never really line up the motor and shaft bearing (I suspected hull flexing).So the shebang would vibrate if I lazily let it freewheel. One day (on charter) it just broke, luckily there was a coupling which made repair simple and also prevented the shaft from backing out (with possibly wet consequences). Ever since then, I figured that the assembly suffered less wear and tear with the prop fixed.

Luckily the charter guests were world class sailors (one son had just finished the Whitbread)and we finished the weeklong charter, sailing in and out of anchorages in the BVI (much to the consternation of some bareboaters)

Ken Buck
12-13-2004, 11:00 PM
What's a propeller? Which end of the oars does it attach to?

brian.cunningham
12-14-2004, 12:09 AM
Isn't this what folding props were invented for?

JimConlin
12-14-2004, 12:36 AM
On Magic, I had a two-blade prop and marked the shaft so I could align the prop behind the deadwood when racing or doing a long sail.

Stiletto
12-14-2004, 01:38 AM
Re the Airplane example: when a propellor is feathered on a plane, is it feathered to the forward motion of the plane or to the rotational motion of the prop?

Ross M
12-14-2004, 02:15 AM
Aircraft, on engine failure, coursen the pitch of the prop on the failed engine as much as possible. With luck, a "neutral" position is achieved.

Interestingly, high powered Drag racing boats frequently have a "whirl-away" feature that releases the propeller shaft from the engine (rotationally) in the event of catastrophic engine failure.

This feature purportedly reduces the severity of the resulting high speed crash.

Ross