A question of planform - Replacement Trimaran Centreboard

[nzl51]

Hi Guys, Long time stalker. First time post.
Can i just start by saying that I appreciate some of the excellent topics and intelligent discussion that i have come across on this page.

Now - We have a Farrier F25A Trimaran with the pivot up centreboard arrangement that has extensive warping due to water getting into the cedar core. Too hard to fix so I am currently designing a new arrangement which we will construct using a CNC shaped foam core.

I have been doing a lot of reading and research. Many informative websites and forums and of coarse both the Bethwaite High Performance sailing books. Generally I want the new board to be the best compromise I can make given the limitations of the existing CB case and 25 years of thought on foil design.

Below is how i have approached this. Im not a hydrodynamicist. My thoughts are at the bottom but i thought it a good idea to seek other opinions before i pull the trigger on the CNC. Please let me know if you think i'm up the wrong garden path.

The original board (below) as per the 1995 designs took a teardrop planform as shown below. The best we can figure the section is a NACA0009.

From my observations and research there has been a definite push towards higher aspect ratio foils due to the improved lift to drag ratio that can be achieved. This is also reflected in Ian Farrier's newer design such as the F22. But i am also aware the planform area is the biggest determinate of the actual lift the foil provides.

Option 1 - A very similar Farrier F82 design with increased length (+700mm) and sail area (+18% on mainsail) was constructed around 2011 with a modified CB planform and CB slot closer. The Original F82 board design basically matches the F25A. I would like to think that they had discussed the modified planform with Mr Farrier but have no way to know.
I have created a higher aspect ratio planform shape which fits well into our existing CB case. You will note that this board has reduced planform area from (75% of) the original, a straight but not parallel LE and TE with majority of the board tapering the chord to about 87% near the bottom. Also note the slot closer in green which pivots on the back of the board which in the up position fits above the TE in the case.

Option 2 - I am aware that the research says aim for a elliptical pressure distribution at the the tip can be got close to by tapering to a tip chord of about 40-50%. This seems to be what the F22does so I also drew a planform similar to the F22. Which has a further area reduction (65% of original) and increase in aspect ratio. It tapers to about 50% of the root chord at the tip.

Discussion.
The above options have allowed me to run a sensitivity check on the relative performance between the lift produce by the differing aspect ratios at different speeds/Re numbers and AoA using Xfoil.
For the option 1 board Re numbers i calculated at 1knot is about 200,000, at 20 knots 4,000,000 ish.
Board thickness will be maximised in the existing slot to ensure sufficient strength at the root. The reduced chords will result in a 12% chord - thickness section.
Higher aspect ratio improves L/D ratio. But does not improve the actual lift generated as they have smaller planform area.
The newer planforms allow space for the CB case closer which will create a more effective hull end plate than the original arrangement which has an open case.

A very rough comparison follows using the original foil as a baseline where i compared lift generated by the different planforms with NACA64A012 sections at a range of AoA. Bear in mind i could have cocked something up here.

My thoughts:
Adopt Option 1
The original board dimensions were also employed on a larger boat with more sail area.
The option 1 type and size board were more recently installed on the larger type boat which suggests upwards of 20% loss of lift is ok. If the larger boat can have a smaller area. The smaller boat (ours) should be OK also.
OR the benefit of closing the CB slot is worth losing some area for.
The option 1 board is not heavily tapered to maximise aspect ratio efficiencies which suggests they were still trying to maximise the area.
Loss of area will increase leeway and hence AOA so pick a section with a wide drag bucket.
Slow speed manoeuvring in a cross wind and down speed tacks may become trickier.

Questions.
Can anyone suggest further refinement to the Option 1 shape?
Is there anything to do to mitigate the vortex at the tip that will be created by not having a elliptical distribution?
Sections:
I started out looking at NACA 64A012. Now also looking at Eppler E168 and 836. Anyone have experience with these or could recommend another that works well across a range of Re numbers with the above considerations?

Thanks for reading my essay!
Dave.

 

[tspeer]

Instead of thinking of aspect ratio, you should instead consider span and area separately. Unfortunately, you are constrained in span by the board trunk, and that rules out the most effective way to reduce the induced drag. So almost anything you do will result in less area than the original 1995 design. Whether that is a good thing or not depends on how you want to sail the boat.

If you reduce the area, the board will be more susceptible to stalling out at low speeds and in light winds. It will have a little less drag at high speed. If you were sailing, say, on San Francisco Bay where strong winds are the norm, this might be worthwhile. However, if you were sailing on Puget Sound where the winds are light in the summer, it may not be the way to go.

You already have a slot closer, and I think that is the most important thing for reducing the drag of the centerboard.

And you do know that the lift generated by the board is independent of the board design, right? Reducing the area increases the leeway angle - the lift remains the same.

The differences in drag between a NACA 6-series section, a NACA 4-digit section (like the NACA 0009), and the Eppler sections depends on how much laminar flow you can achieve on the board. The amount of laminar flow that is possible will depend on how well maintained the board surface is, and the conditions in which you sail. There is anecdotal evidence that transition occurs earlier in water than at the same Reynolds number in air, but I haven't seen anything that quantifies this. I suspect the NACA 0009 will actually have less drag than the NACA 64A012, because it is thinner and because you will be operating outside the drag bucket for the NACA 64A012. The Eppler sections are also thicker, so they may have more drag as well, especially if you can't achieve the laminar flow assumed by the Eppler code.

 

[nzl51]

And you do know that the lift generated by the board is independent of the board design, right? Reducing the area increases the leeway angle - the lift remains the same.

Thanks for your response Tom. Am I correct that what you're referring to here is that the centreboard only ever generates enough lift to counter that lateral force of the sail plan and hull windage? Yes I perhaps lost sight of that when I went down my rabbit hole in Xfoil and excel.

I suppose it comes back to what I can do with the CB Case space available.
In the simplest terms, and you have said it above, it comes down to - is it worth reducing a bit of plan area in order to have space in the case for the slot closer. Less drag (maybe significant) but at the expense of more leeway.
I have compared our local avg winds here in NZ and we are quite similar to SF. Generally we get a sea breeze of varying strength but very rarely would we be going out in wind less than 5 knots.

I guess I'm leaning towards the benefit of the slot closer being worthwhile but keeping the area large as i can.

My current plans above have the same ~50mm root thickness as the original board. By changing the chord to fit the slot closer into the case, the thickness /chord ratio changes from 9% on the current to 12% on the proposed.
I will maintain the 50mm thickness as it allows me to use the spar and laminate layup from the original foam board design drawings that we have.

Regarding the Section, you have got me thinking. The Bethwaite books do mention needing to keep the laminar sections surfaces very highly polished. The boat is dry sailed so shouldn't have issues with growth but the board will be sealed into the slot could get scratched while pivoting with no access to polish.
Perhaps its more appropriate to not push for a laminar section and use a more standard NACA0012?
I'm not sure of the time to benefit of learning how to make and assess a custom section but i do have a rudimentary knowledge of using XFoil.

Thanks again.
Dave

 

[tlouth7]

You can get more span for your trunk length by:
- rotating through a larger angle to bring the deployed board closer to vertical. This will of course change the balance so you might need to adjust your rig.
- Optimising the tip of the board. The large radius on the front and rear of the tip on the original board seems a bit daft. At the extreme your tip could be an arc centred on the pivot, giving an almost full-length leading edge.

Remember you don't need to taper the planform to taper the lift distribution.

 

[nzl51]

thanks Tlouth,
- I've so far been trying to maintain the position of the COR near to the original and the new to avoid having to muck around with the rig and the balance any more than necessary. Going more vertical is likely to move it more substantially.
- As you suggested I've had a look to see if i can get some more area. Without rotating the larger angle i get a weird looking tip arrangement. I don't know whether this arrangement would be more or less draggy than the previous? (the leading edge radius avoids a fragile point if it hits the ground and kicks up as its supposed to do.)

By your final comment you are meaning to change the cross section towards the tip to a thinner section (smaller lift coefficient)?

Option 1 - 75% area as before.
Option 2 - discounted as i don't want to try 65% of original area.
Option 3 - 86% Area. Parallel LE TE area at tip more maximised. Section would narrow near the tip.
Option 4 - 81% area. A further development of option 1. Parallel TE and LE. Section would narrow near the tip.

 

[Steve Clark]

An important input is your impression of how the boat handles with the current board. Does the boat make excessive leeway? Is it hard to hook up after a tack or other down speed maneuver?
Doe it feel like you can pinch all you want and never stall? Does the helm go light, or even become negative ( Lee Helm) at high speed?
It's entirely possible that Farrier got it right.
The force a centerboard has to generate is dependent on the stability of the boat. If the area is too small, it will have to operate at higher lift, which may push you out of the low drag bucket.
This can be maddeningly slow.
A board that isn't warped is going to be much better than one that is. So you are going to be better off even if you just reproduce the original design.
SHC