I sketched up a split rig design for my little Pelorus motorsailer, which I was (am) considering – and made the same mistake as Bert. I under-estimated the balance by 3% (as now explained by Slieve).

Fortunately in my case it is only a paper exercise at this stage, and the numbers I am considering are conservative anyway so an extra 3% is probably tolerable.

However, a mistake like this is potentially expensive.

In my case, when considering balance, I carefully consulted PJR for a definition of balance, and also consulted the JRA website glossary, which lead to the mistake.

Newcomers can be easily confused and as I am a newcomer, here is my tuppence worth.

I would suggest that the JRA Glossary definition of “balance” ought to be revised.

Firstly: terminology. To remove the confusion with that other question of balance (the balanced helm, or balanced hull, and the relationships between the centres of underwater hull profile and sail profile) – it would be better to follow PJR and refer to the balance we are talking about here, distribution of sail area about the mast, as “aerodynamic balance.” (It is notable that in this thread, which is essentially about aerodynamic balance, some of the discussion seems to have wandered into the domain of helm or hull balance.)

Secondly, and more importantly, the definition of aerodynamic balance needs to be revised, to be made a little more precise and to take into account the various split rigs which have a slot in the sail plan.

The JRA Glossary definition is currently: “The area of a junk sail (or lug sail) extending forward of the mast line, often expressed as a percentage of the sail’s total area”. Evidently, taken literally, this is not always appropriate in the case of a slotted sail, and could yield a slight under-estimate of the amount of aerodynamic balance.

In this thread, Chris has not directly given his definition of aerodynamic balance, but by implication it is the position of a chord line which coincides with the centre line of the (vertical ?) mast, expressed in percentage terms. This, too, might not be appropriate for some quite different planforms, especially considering the the shape of the upper panels, and probably leads to a slight over-estimate.

PJR (page 10 of my edition) has a quid each way. First, it defines aerodynamic balance in 2-dimensional terms as percentage of sail forward of the mast – almost (but not quite) the same as the definition used in the JRA Glossary. Then in the next sentence PJR gives an alternative definition, in linear terms, as the percentage of batten length forward of the mast – somewhat in line with the approach taken by Chris.

(“Forward of the mast” is ambiguous too – is it supposed to mean the mast centre line?)

In terms of a conventional rig, both definitions more or less give the same result though not with any great precision. But junk sail plan forms seem to have moved forward since the days of PJR, and in addition some rig designers are now pushing boundaries where just a few percentage points could be significant.

Since the axis of (weathercock) rotation is the mast centre line, perhaps the definition should be given as: “percentage of sail plan, including slot if any, which is forward of the mast centreline.” 

Just a suggestion.

By the way, under this definition, the linear approximation used by Chris (based on a chord line of the lower panels) then becomes a slightly conservative rule of thumb as it tends to over-estimate the amount of balance by a tiny amount, due to the shape of the top panels. This is a practical approach which saves a lot of time calculating areas - provided it is made clear as Chris now does. And it is in line with the similar linear rule of thumb evidently used by both Chris and Arne in estimating centre of area.

Bert appears to have followed the JRA sail area approach, literally, in his calculation of aerodynamic balance, so his 33% was a slight under-estimate of what Chris was slightly over-estimating when he gave his report of 33% for Poppy. Not only that, but Bert's full 33% is forward a little, by the width of the slot. Near the limit of what is possible, it is easy to see now that the few percentage points between the two methods of calculation could go a long way towards explaining the problem which Bert found. Hence it is important to be clear about definitions, even for a theoretical measure, like this one.

And it is, too, just a theoretical measure. Arne raises an interesting point: if the fore part of the rig is behaving differently from the aft part, then a simple geometric measure might not be an accurate general predictor of weathercocking tendency. 

So working near to the limit of balance might perhaps be left to those, such as Slieve, who have developed a good understanding of the characteristics of their rig, by practical experience - and for the rest of us, to start by having a more clear, agreed definition of what is meant by aerodynamic balance.

Uah, indeed I overbalanced my sail because I did not include the gap in my balancing!!!

Slieve is nearly right: The balance is 37%. Now, with the shortened jiblets, balance is reduced to 29% (calculation with gap).

But Camber is done as required, jibs are made as required (angled shelffoot), only the mainsail is not constructed with broadseems but has the desired camber too by adding material in the middle of each panel which has the advantage that the edges of the panels are parallel to the battens (not rounded -> less wrinkles).

Hopefully to steer the boat where I want to now!

Edit: How did the mistake occur? The decision to give Runa a splitrig was while I was building the interior. Because of the new position of the mast I gave her a new layout according to the specs of the splitrig. One number which burnt into my mind was 33%. When I finally designed the sail (about 3 years later) I obviously lost sight of including the gap into the calculation...

------------------

Thanks to you all taking your time to solve my problem.

@ Ueli: Sitting in the lee position heeling the boat helped a little bit.

Hi Guys, 

This thread has raised an number of important issues and highlighted a surprising amount of misunderstanding which should be cleared up.

It started with two good questions by Scott in post #4793670. The first one asked, “Why not balance the sail as much as possible?”

I suppose the answer is that the classic 'pretty' fan topped Chinese rig with the high peaked yard pushed the sail aft and was limited to minimum balance for close-hauled sailing, and the Hasler/ McLeod westernised development and commercial Sunbird copies followed along the same lines. Once camber was incorporated in the rigs by whatever means it became desirable to include luff hauling parrels and fore and aft balance controls, and resulted in rigs that were rather loosely attached to the masts and required anti-fan up controls. The net result was a rig with its many adjustments that one sailor claimed cost more than the actual boat.

However some of the river junks used a more square topped low yard angle rig with greater balance, and the nearest westernised version of this rig was that drawn up by Derek van Loan. Following my analysis of the various designs available I realised the many advantages of this simple rig.

At university we were told that engineers should kiss. Our all male group agreed whole heartily, and where I really cannot remember much about some of the lectures I have spent the rest of my life applying the advice to everything in my life. The KISS principle brings the complicated fan-topped junk rig back to reality. The split rig is based on this KISS principal. It includes as many good aerodynamic practices as I could include in a simple cloth and stick, low tech, low cost rig and easily sailed rig that still returns impressive performance.

In answer to Scott's question, it does incorporate as much balance as would appear to be practical, with all the advantages that go with it. It also answers Scott's second question about moving the rig back and forward on different points of sail. With the sail fully forward on all points of sail there is no need for such controls and the rig can be securely attached to the mast all the time. KISS.

The problem appears to arise in David T's response in #4793778 where he states, “but as much as possible is in fact with the mast at about 25% of chord, and if you go for more than this, there is a danger of the sail over-balancing -”. This is not the case, but I can fully understand why David has got this impression which I also held when I was initially looking at rig designs.

It is possible that David has designed and built more westernised junk and wing sail rigs than possibly anyone else in the world, and not only reported on their good points but had the strength of character to criticise his own work where he has found problems. I would love to read an ordered list of his many rigs and their main features, and the boats they were tested on, as I believe we would learn from the information. He has, however, possibly only a couple of hours experience of the split rig on Amiina, and if I remember correctly that was at a time when the rig had not been installed as per the design specification. I understand that despite his wide ranging experience he has no experience of sailing on a boat properly rigged with balance around 33%.

It is easy, with hind sight, to understand the 25% mental block, which I also struggled with initially. Open any aero/hydrodynamic book on symmetrical foils and you will read, taking an example, that the NACA 00 series foils have a neutral/ balance point of 24% chord, and with imperfections in manufacture this tended to be 25%, an easy figure to remember. So in theory a rig like the split rig which is the same on both tacks is theoretically symmetrical and should follow this rule. This is not the case as the rig changes camber from one tack to the other, and is asymmetric when in use. The NACA 00 series starts producing lift even at 1° on either tack and stalls at around 15°, but with a split rig the tight main leach will be trying to weather-cock the rig and the jib luff will be luffing until the jib luff fills at around the entry angle defined by sheeting angle plus camber which may be about 15 to 20° (from the top of my head).

It was only when I bumped into Roger Stollery of model yachting fame that I discovered that he had run a full set of experiments with balanced model rigs and discovered that a rig with up to 33% area balance was perfectly safe to use, and there are thousands of model boats throughout the world using these figures successfully. The Aero Rig by CarboSpars was a full size copy of his rigs which did not make full use of the rigs abilities.

One point which has emerged from this thread is that the definition of the 33% must be clarified. It is clear in my mind, but others have possibly found a different understanding. In the model rigs there is virtually no gap between the leech of the jibs and the mast, and on my 'Footy' model the jib leech actually slightly overlaps the mast at one point, the mast being only 3mm diameter! On Poppy, and subsequent rigs there has been a slot drawn between the mast centre line/ pivot point and the leech of the jibs and where I have included this in the 33% balance calculations some would appear to have added this afterwards, moving the 33% jib further forward and possibly causing stability problems.

When I draw up a rig, I take the total rig outline, including the slot and guestimate the centre of area at 50% of the total chord, jib luff to main leech, and position it on the centre of area line of the original Bermudan rig. This is not strictly accurate, as the aft part of the sail is taller than the fore part, but experience has shown that it is not that critical, and this is the KISS way to do it. I have then drawn the 33% chord line for the mast and drawn a minimum slot which I suppose is subtracted from the 33% area. All these simplifications move towards a more stable set up. I have drawn rigs up to 35% balance with these simplifications and they have been perfectly stable in use.

When on the dead run the split rig is not necessarily full stalled as the cambered luff of both the jibs and main seem to encourage the wind to flow round the lee side of the sail and appear to give additional drive. Whether this is a major point or not, I seen no reason to add complication in moving the rig from the 33% balance, particularly as this is the best ratio for bending moments of the battens.

I try to keep the gap between jib leech and main luff to a minimum but wide enough to let the air flow from the windward side of the jib get past the mast, thus a fatter mast gets a wider slot.

Regarding mast rake, I'm a great believer in a vertical mast well installed and left alone. KISS.

Reading Bert's posting, #4840027, there are a few concerning points, and Scott's comments in posting #4840083 show a clear approach to the problem in that there are two separate balance problems, and it would appear that in Bert'c case both are outside practical limits. We frequently talk of C of E when we are referring to C of A, and this causes no end of confusion. The Centre of Area (C of A) is the one we can calculate and draw, and this is generally used in rig design. The Centre of Effort (C of E) is a dynamic point which moves when either wind speed and/or direction vary, and will also depend on camber shape and depth. As expected, experience has shown that the C of E of the cambered split junk rig is a lot further forward than on a flat Hasler/ Mcleod rig, and if Bert's split rig is in the flat rig fore and aft position then lee helm is to be expected.

Add to this, there is the overbalance and instability problem with the actual rig. Looking at the photo in Bert's photo album it is difficult to see how much camber has been built into the rig, but by taking simple horizontal measurements from the photo it would appear that in the split panels the jibs have 33% of the total area but have been moved forward by the width of the wide slot so that the mast/ pivot appears to be at over 38% of the total jib luff to main leech chord. It is not surprising that this is unstable. It would be interesting to see a copy of the rig and hull drawings to see exactly what is happening. If the camber is as flat as it seems in the photo that would exasperate the problem.

All previous reports I have seen on split rigs have shown no problems in these areas, so this is a particularly interesting situation.

I hope this makes sense, and that we can get a good solution to Bert's problems.

Cheers, Slieve.

PS. Yes, Manchester is a worry to everyone anywhere in the world.

hi bert

Bert Qui wrote:

Hi Slieve,

…the boat pointed well to weather, but the rudder must have an angle to weather of 10 to 15 degrees! Therefore it is quite impossible to tack, because the rudder stalls...

10 to 15 degrees is a bit much for sailing straight – did you try to sit on the lee side in low wind conditions? – i do this often to battle lee helm.

i have no advice for the rest, as i'm still under bermuda sails.

ueli

Oh dear!

I cannot believe what I am reading. I was pointed to this thread today and a quick read through shows a lot of miss-information flying around which will do more harm than good. Unfortunately it will be a couple of days before I will have time to digest this, but please do not make any major decisions until this miss-information has been cleared up.

I do not fully understand Bert's set up, but there is obviously a major problem with it which will not be resolved by simply criticizing either the rig or the boat design. Before jumping to any conclusions we need to analyse the details to find where the problem lies. I would be happy if Bert contacted me directly by e-mail or give me his land based 'phone number as I have free calls to land lines in Germany and we could have a chat. If language is a problem my wife speaks German and could be persuaded to help.

In haste,

Cheers,  Slieve.