Deck sweeping sails and effective aspect ratio

Discussion in 'Hydrodynamics and Aerodynamics' started by Will Fraser, Mar 31, 2015.

  1. daiquiri
    Joined: May 2004
    Posts: 5,373
    Likes: 247, Points: 73, Legacy Rep: 3380
    Location: Italy (Garda Lake) and Croatia (Istria)

    daiquiri Engineering and Design

    Not with a lifting-line spreadsheet.
    Either by performing a CFD analysis, or through measurements on a scaled or full-size boat.
     
  2. Remmlinger
    Joined: Jan 2011
    Posts: 243
    Likes: 20, Points: 18, Legacy Rep: 172
    Location: Germany

    Remmlinger engineer

  3. Will Fraser
    Joined: Feb 2014
    Posts: 168
    Likes: 17, Points: 18, Legacy Rep: 11
    Location: South Africa

    Will Fraser Senior Member

    I have some cfd results for the heeled monohull Ragtime versus its upright equivalent.

    All cases were simulated with an apparent wind of 37kts at an angle 28.4deg off the bow. This is equivalent to a sailing angle of 45deg to a true wind of 25kts with a boatspeed of 15kts and 5deg leeway.

    No wind gradient or twist was simulated at this time.

    The hull was submersed down to the DWL in the upright cases, and when heeled (35deg) it was trimmed to yield the same displacement and LCB. Subsequent sail loads indicated that it should have been submersed to yield 110% of the gross weight due to the big downward force of the inclined sails.

    The sealed gap cases had the sails (and spars) drop until they made contact with either the deck or coachroof. Additional endplates were added to seal aft parts of the sail feet extending beyond the deck or coachroof.

    Since the effect of a sealed deck gap would have a great effect on induced drag, and since induced drag is a function of CL^2, I tried to adjust the sail sheeting angles to yield the same values for CL. Sheeting angles were also eased until no extensive areas of separated flow existed (the mainsail just behind the jib head, for instance, always had a little residual separated flow). This resulted in a lower CL than might be expected in practice.

    To calculate CL, I used the total force component perpendicular to the wind, and as a reference area I used the projected sail area onto a plane normal to this force vector.

    The following images shows some of the differences in projections depending on which angle it is viewed from.

    View dead downwind (apparent):

    [​IMG]


    View directly across the wind, parallel to the water:

    [​IMG]


    Planar projections of sail areas for upright and heeled cases, showing wind orientation, maximum span and chord:

    [​IMG]


    The numbers, for what it is worth:

    [​IMG]

    The upright case indicates some benefit of sealing the gaps while the heeled case actually shows a lower sail L/D with a sealed gap.

    Just to explain the drive/vertical force ratio: the sealed gap upright case had less upward force (i.e. lifting the hull) than the stock version. The sealed gap heeled case had more downward force than the stock version.

    Despite almost identical CL values, both upright cases showed approx 17% more drive than their heeled counterparts.

    The stock version however had 3% more heeling moment when upright, while the sealed case had 7% less heeling moment than its heeled counterpart. Note that it would be inappropriate to compare heeling moments between stock and sealed gap in either of the two cases since the sealed gap's cases had their rigs lowered.

    Some more interesting screen-captures for comparison. Note that the sea pressure signature images had some graphics issues when rendering the hull.

    Flow over gunwhales and deck (stock left, sealed gap right):

    [​IMG]


    Pressure signature on sea for heeled cases (stock left, sealed gap right):

    [​IMG]


    Pressure signature on sea for upright cases (stock left, sealed gap right):

    [​IMG]
     
  4. Joakim
    Joined: Apr 2004
    Posts: 892
    Likes: 52, Points: 28, Legacy Rep: 422
    Location: Finland

    Joakim Senior Member

    What software and which models are you using?

    Why did you use an unrealistic boat speed of 15 knots? If this is the old Spencer 65, it won't do more than about 9 knots on a beat. Also it will need a lot smaller headsail in 25 knots TWS.

    Well the wind speed you chose doesn't really matter. The flow field will be (almost) identical at a more reasonable apparent wind. The only problems remaining are the too high AWA and wind shear, especially when you are interested in the foot sealing.

    According to a VPP Ragtime should have about 25 degree AWA at 20 knots TWS and that is referenced to line of travel, not to keel line. So with 5 degrees leeway that should be about 20 degrees. The AWS is 28 knots.
     
  5. Will Fraser
    Joined: Feb 2014
    Posts: 168
    Likes: 17, Points: 18, Legacy Rep: 11
    Location: South Africa

    Will Fraser Senior Member

    The model is rough one I created from a lines drawing. The cfd solver is Solidworks Flow Simulation. I have yet to run a few calibration models (know of any useful ones?) before I put any trust in the actual loads calculated. It is nevertheless useful when just testing comparisons.

    Like you mentioned, it is the flow field that is of interest, and that would not be expected to change much. It is also why I deliberately tested the upright cases as well. The aim was to see if heeling has a notable effect on the gap seal's effectiveness.

    The absolute values in the table is of no real use since wind-shear and twist could not be modelled accurately. What is of interest is how the values compare between upright and heeled cases for the stock and sealed gap configurations respectively.
     
  6. Will Fraser
    Joined: Feb 2014
    Posts: 168
    Likes: 17, Points: 18, Legacy Rep: 11
    Location: South Africa

    Will Fraser Senior Member

    As just shown, sealing the gap on a heeled boat might actually harm the sails' L/D ratio. So why bother sealing even just the headsail as seen on modern racers?

    The image below shows that the windward gunwale vortex seems to have some additional effects to leeward. Notice the prominent vortex attached to the tack of the sealed jib (right) compared to the stock sail on the left. One cannot help but notice the similarity with flow over the wings of an F18 Hornet at high AoA.

    The phenomenon is readily exploited in aircraft to delay flow separation, allowing a higher maximum angle of attack.

    (The colour gradient used for pressure in the cfd images differ between that on the sail surface and that used for the streamlines - just for display purposes)

    [​IMG]

    [​IMG]
     
  7. johnhazel
    Joined: Jun 2008
    Posts: 250
    Likes: 5, Points: 18, Legacy Rep: 60
    Location: Michigan

    johnhazel Senior Member

    Maybe the answer to that might come with a calculation of flow for the sealed headsail with stock main.

    Also you might be try bringing the foot of the headsail all the way to the bow (or moving the bow back to the headsail.)
     
  8. johnhazel
    Joined: Jun 2008
    Posts: 250
    Likes: 5, Points: 18, Legacy Rep: 60
    Location: Michigan

    johnhazel Senior Member

    Fun stuff.

    If positive is lifting force then sealed 4.15 is more than stock 2.35?
     
  9. Will Fraser
    Joined: Feb 2014
    Posts: 168
    Likes: 17, Points: 18, Legacy Rep: 11
    Location: South Africa

    Will Fraser Senior Member

    Fun indeed. You never know what new or unexpected you might learn or discover.

    The lifting force is the denominator in the ratio, hence 2.35 represents a stronger lifting force for a given drive.
     
  10. brian eiland
    Joined: Jun 2002
    Posts: 4,955
    Likes: 181, Points: 73, Legacy Rep: 1903
    Location: St Augustine Fl, Thailand

    brian eiland Senior Member

    Sorry for not being so computer analysis knowlegable as most of the posters in this subject thread.

    But I do have a question here. Are you saying that the rig with untwisted sails is going to point higher and produce more lift?....lift being both heeling and driving force??
     
  11. brian eiland
    Joined: Jun 2002
    Posts: 4,955
    Likes: 181, Points: 73, Legacy Rep: 1903
    Location: St Augustine Fl, Thailand

    brian eiland Senior Member

    I sure see a lot of disturbed air flow off the 'hounds area' of this fractional rig (above).

    But not so much, if any, off of the flow diagram below??


     
  12. brian eiland
    Joined: Jun 2002
    Posts: 4,955
    Likes: 181, Points: 73, Legacy Rep: 1903
    Location: St Augustine Fl, Thailand

    brian eiland Senior Member

    BTW,
    Will, it would help if you edited the size of that fighter plane down some so that one does not have to scan across the age to read the postings.
     
  13. Will Fraser
    Joined: Feb 2014
    Posts: 168
    Likes: 17, Points: 18, Legacy Rep: 11
    Location: South Africa

    Will Fraser Senior Member

    In this theoretical, hypothetical case, yes.

    There are however a number of factors that would prevent one from making the same assumption about sails in general, namely:
    - the wind has no gradient or twist in the way it was modelled
    - the sheeting angle is automatically limited as soon as a single point along the span reaches the user-selected maximum Cl value.
    - the sheeting angle used is more than the maximum suggested for accurate use of the theoretical model
    - the values quoted are for the sealed-gap versions compared to the stock versions. I do not recall how the stock versions compared to each other.
     
  14. Will Fraser
    Joined: Feb 2014
    Posts: 168
    Likes: 17, Points: 18, Legacy Rep: 11
    Location: South Africa

    Will Fraser Senior Member

    The program that I use is not ideal for modelling transient separated flow - it can be done but takes an awful lot longer.
    The disturbed flow is still there though, and if I increase the amount of streamlines in the region it becomes more apparent.

    [​IMG]
     

  15. Ben G
    Joined: Jun 2013
    Posts: 49
    Likes: 0, Points: 6, Legacy Rep: 10
    Location: Brisbane

    Ben G Junior Member

    A friend of mine and I have looked in to this for the skiff.
    I don't have the details at hand, but in summary we decided that practically there wasn't much to be had.
    - jib is pretty much sealed to the foredeck, and roughly aligns with the leeward surface of the hull, so there is little to be gained here
    - as the jib overlaps the main, there is little benefit in bringing the leading edge of the main to the foredeck.

    An idea I had, which did give a benefit, was to have a 'winglet' that protruded downwards from the aft end of the boom. this way you can still cross sides of the boat. At the aft end of the boom, the vortex curling under the boom was at an appreciable angle of attack. So a vane that would straighten this flow would generate useful driving force.

    I have some ideas on how this could be implemented.

    Give it a crack, now you know the answer to your question :cool:
     
Loading...
Forum posts represent the experience, opinion, and view of individual users. Boat Design Net does not necessarily endorse nor share the view of each individual post.
When making potentially dangerous or financial decisions, always employ and consult appropriate professionals. Your circumstances or experience may be different.