After much boiling of Junk ingredients, in the cooking pot of my mind, the reduced content has two concepts -- a luff spar with a leading element before a soft wing(with composite battens), being split about  the mast, and then a more familiar split Junk sail with roped luffs to both elements, stretched between battens.The first mentioned luff spar option, requires carbon composites components, so has limitations to preclude immediate further discussion. Then the second soft luff option is based on the idea of carrying the split element configuration a step further, by using flexibility and hinged junctions to create variable camber with the two elements, by means of sheet control.

Perhaps this sort ofthing has been tried before, but I cannot find any info on it, so a wordy explanation might help to bring it out, from wherever it may be hiding. Instead of attempting to use batten curvature at the forward part of a single batten, the idea has an alloy tube wishbone lodged about the mast, adjoined to a bendy part that extends to a straight trailing length of stiff tube or other light rod, that prevents any hooking at the leach.Camber is created in the sail fabric, and is kept to max depth about 40% after the luff( of the combined element configuration), while being kept flat by the battens for the last 20% of chord -- as per DavidTylers recommendation.

Camber and set of the leading element produces a slot where the elements are split about the mast.This is achieved by means of 'boomlerts' at the foot of each jiblet'.

Variability in AoA of the leading elements luff ( by sheetlet adjustment) is/should take care of the problem being discussed here, where permanent batten curvature does not deal with variable wind velocity in their shaping of the sails leading area.

Shortly after the notes on the SJR first appeared on the JRA website I was in a private correspondence with a member about drawing camber curves, and as an 'off the cuff' method I suggested drawing an arc centred on a line drawn perpendicular to the mean chord line below the max camber point, and drawn through the luff, max camber point and back to the mean chord line, before adding a tangent from the arc to the leech. This was a possible way to get a consistent curve which would give an adequate and acceptable curve. This was further developed to draw the tangent from the arc to a point at about 95% chord and rounding of the aft end with a slight reflexed curve to the leech to make sure there would be no drag producing cupping in this area.

I remember trying to calculate the radius of the arc for differing camber depths to simplify the process though didn't complete it. David T's three point arc is exactly the same as an arc with max camber point at 40% will strike the chord line at 2 x 40 =80% chord.

This method does produce a workable result, though I personally prefer a slightly tighter curve near the luff.

Unfortunately this arc/tangent idea was not included in Appendix 5 of the SJR notes (which is rather waffley and needs to be re-written), although it does mention the slight reflex towards the leech.

Appendix 5 does have a small comment on the use of spline and the potential problem of not supporting them in the correct places.

The photo shows a spline being used to draw a curve during the making of the Amiina Mk 2 rig.  NB. The straight 'chord line' seen in the photo has been off-set to include a provision for a batten pocket and hem allowance. The actual mean chord line can just be seen by the second tin from the right.

The paper is a roll of lining wall paper, the spline is a 25 x 5 mm cross section wooden batten and the weights are full tins of paint. The one at the luff, on the right in the photo is positioned well forward of the actual luff to allow the curve to run to the luff. The second from the right is at the max camber position and the third one is placed to make sure that the batten bends round the max position correctly, and doesn't have a deeper camber elsewhere. The one on the left is positioned well beyond the leech and bends the batten with a very slight reflex to make sure the sail is flat towards the leech. If I remember correctly the depth of camber was taken from the Round & Broadseam spreadsheet and was used for the main panels.

Unfortunately such a spline does not give enough camber right at the luff of the angled shelf foot jib panels. Camber curves in the jibs are under constant review and the dust has not settled on the latest thinking. There are a couple of ideas floating around but no one has been persuaded to gamble (yet).

Perhaps this will cloud the issue a little more??

Cheers, Slieve.

I use this 'arc + straight line' approach, too, but with different proportions. I draw an arc through three points: at the luff, at the maximum depth of camber 40% of chord aft, and on the luff-to-leech straight line 80% aft; then I draw a straight line tangent to this arc from the leech. The intersection between arc and straight line is at 55% aft. I think this would put the curvature further aft than that which Paul describes, and closer to that which Chris Scanes uses.