Circoflex Hints and Tips

For a fast track to the bridle table use this link


Silver aluminised mylar makes for a very fancy distintive `UFO' like circoflex. It is also cheap (when purchased in a bulk roll) and strong. It however can only be `taped' together, preferably with double sided tape.

Unfortunately while very very strong the material has an annoying habit of ripping right accross the kite in a split second, from a small puncture or edge nick. Mylar is also used for `potato crisp' packets for the same reason, it is tough, but easy to rip open, from appropraite `nicks' on the edge.

Where to buy mylar...
Simple. Silver mylar is used by flower sellers to make a bunch of flowers more interesting. You can buy either pre-cut sheets or 250 meter or even 500 meter rolls of the stuff from florist supply warehouses. You can also try a local florist too. One roll would be enough to supply all your circoflex and other silver kiting (tetrahedral?) needs for a very long time and works out to less than 10 cents a meter!
Ripstop is also a good material and provided the means for creating a pocket for a removable spar. It is also great for fancy designs and applique. It is heavier and more expensive (100 times more) than mylar but much more duriable and open to different construction techniques.
Tyvek is in my thinking rather heavy for a circoflex, however I have had an email from Patrick Mann <patrick.mann@bigfoot.com> who has built a 7.5 meter circoflex with tyvek. It is also quite cheap.
Know your material
Be sure you understand how to work with the material, be it ripstop or mylar, or something else. The more experience you have with working it the better the kite you can build. Practice on scraps before tring something different on a new kite design.
Bridle Point and Line Lengths...
The calculation of bridle lines is fairly straight forward. Basically it is the distance between the appropriate point on the circoflex ring to a single point in front (above) the ring.

This point is always (in all plans so far) directly above (in front of) the ring halfway between the center and the 6 o'clock position. The distance of the bridle point from the plane of the ring however varies from plan to plan and the size of the circoflex. Typically it looks like this distance is about 1/4 of the rings radius. Though it does not seem to be very critical to the kites success.

Bridle Point for Mini-Circoflexi (less than 5 meters)
While building mini-circoflexi, I tryed use use the same bridle point position as for the larger circoflexi. However positioning the bridle point half way between the center and the lower edge, just seems to cause no end of problems. The main ones being

The solution was to instead of moving the bridle point inward, move it downward instead. this ment the weight ballance was still fine, and the bridle point was not `too close' to the spar ring plane (See the flying problems below, and bridle line calculator).

Bridle Line Table...
Size(m) 7 + 7.5 * 7.5 * 9 10 * 12 12
Bridle(cm) ** 25 20 30 30 30 35 45
Radius(cm) 111.4 119.4 119.4 143.2 159.2 191.0 191.0
Bridle Line Lengths (by clock face)
12 169.0 172 181.5 216.9 240.6 288.6 290.0
11 , 1 164.0 167 176.2 210.5 233.5 280.0 281.4
10 , 2 150.0 153 160.7 191.8 212.7 255.1 256.6
9 , 3 127.0 133 136.8 162.9 180.5 216.4 218.2
8 , 4 99.7 108 107.6 127.6 141.1 169.1 171.4
7 , 5 73.4 75 79.8 93.7 103.1 123.4 126.6
6 61.1 58 66.8 77.6 85.0 101.7 105.6
+ My prototype circoflex was this size, using 2 x 2mm fiberglass rods
* Measurments from magazine articals
** Distance of bridle point in front of (above) ring leading edge.

In the original plan the upper 7 bridle lines and the lower 5 lines are usally attached to two seperate rings. These rings are then seperated by a short length of line (10-20cm) to allow for ``tow point adjustment''. I have however not found any need for this seperation or adjustment. See below for the two bridle ring methods I use.

Bridle Line Calculator...
The following `pseudo program' can be used to calculate the bridle lengths needed. Just code in your favorate computer language. Curtsey of Bruno Diviacco <Bruno.Diviacco@elettra.trieste.it>.
 # Kite size -- figures for a specific 3 meter mini-circoflex 
 c=306           # circumference (all measures in cm) 
 h=25            # height of the towing point in-front of ring. 
 f=3/4           # displacement of bridle point (3/4 from center) 
 ndiv=8          # number of bridle lines 

 # For a normal circoflex (10 meters such as table above) use 
 #  c=1000   h=30  f=1/2  ndiv=12 

 PI=3.1415       # PI 
 r=c/2/PI        # radius 
 nlen=ndiv/2+1   # number of bridle measurements needed 
 for i = 0 to nlen-1
   a=PI*(1-2*i/ndiv)                           # angle (in radians) 
   len=sqr( r*r + d*d - 2*r*d*cos(a)) + h*h )  # Length 
   print i,len 
 next i
Example output (for Mini-Circoflex_306 the above was set for)
 0 88.8           # 12 O'clock 
 1 82.7 
 2 65.8           # 3/9 O'clock 
 3 42.6 
 4 27.8           # 6 O'clock 
I give no assurances of correctness of the above. Use at your own risk.
Here is a Perl Version which seems to work well and outputs the information nicely formated.
Three Ring Bridle Arrangement...
To keep the bridle lines neat, I larks head the 7 upper bridle line to a small solid (no gaps) brass ring, and the 5 lower lines to second brass ring. I don't bother with a 10-20cm line between the two rings (as given in the original plan), but instead use a split ring to link the two small brass rings directly together (like a chain).

This split ring is in a vertical arrangement, which makes it very nice to clip my flying line to directly (using a fishing swival clip or snap). This arrangment ensures than the fan-out of bridle lines is neat-n-tidy when flying, and does not tangle easilly in storage.

Single Ring Bridle Arrangement...
If you only wish to use one solid (no gaps) brass ring for all the bridle lines, (as I use with my mini-circoflexi), I attach the lines around the ring, alturnating from side to side. That is if I started at the top with a 12 line circo, the lines are attached around the ring in the following order...
12 o'clock, 1, 11, 2, 10, 3, 9, 4, 8, 5, 7, and lastly 6 o'clock.

This wierd side to side ordering means that the ring will naturally want to sit veritically. Thus making it simple to attach a flying line to the ring between the top (12) and the bottom (6) bridle lines. All the bridle line will also still fan out neatly from the ring.

Pigtail Bridle Attachment...
To allow a bridle line to be removable (for replacement and adjustments) I recommend the larks head and `pigtail' approach that is now used in ultra light sports kites. See kiteflyers knot gallery.


First I add `pigtails' (short lengths of line with a `stopper knot' on the end) to each of the bridle points on the main (only) spar. The length of the pigtail does not matter as it will be measured when measuring the full length of the bridle line.

Then tying a loop in the end of the line comming from the roll, I larks head that to the pigtail.

I measure out the line from the main spar (including the pigtail) of the appropriate length + a small amount to take care of the line used up in a final knot and larks head (+1 cm for the line I use).

I fold the line at the measured point, and tie another over hand loop to create a loop at that end. I now cut the bridle line (melt cut with a lighter) from the rest of the roll.

I take the line off the pigtail (undoing the larks head) and thread it through the appropriate brass bridle ring. I then thread one end through the loop of the other end of the line and pull it tight to larks head it to the brass ring. The loose end is then re-larks headed back on to the same pigtail.

Repeat for all the other bridle lines.


This arrangement means that by reversing the process the bridle line is very easilly removed for replacment. Or in one case swap two bridle lines I attached to the wrong places on the clock face. In fact this arrangement means you could very easily replace ALL the bridle lines to try out different bridle point heights from the leading edge plain, or other positions. Though it is tedious, but easier than untieing it all.

Spar Pockets...
Be generious with mylar spar pocket!

I have had a lot of mylar circoflexi fail simply because I made the spar pocket fold along the leading edge too small. If it is too small the spar and mylar in time starts to stick to each other. Also during the folding up for transport (see below) a small spar pocket may allow the spar to contact the double sided tape closing the mylar pocket.

As such be generious! If the spar is 4mm thick ensure at least 1.5cm of material is folded over the spar to keep it free.

The problem with a mylar kite is that once the spar is sealed inside a spar pocket you can't really unseal it easily without destorying the kite, even when you are being very very carful. The smallest mishap and a rip will zip across the kite. It has happened to me twice and is not a enjoyable experience.

This is of course NOT a problem with ripstop circoflex. For one way of adding a ripstop spar pocket to a mylar circoflex see my page on Mylar Circoflex Construction Notes.

Wieght placement...
The English plan says...
`` The weight should be spread evenly over 30-35cm inside the bottom of the kite in the middle of the hours 5 to 7, some 8 to 10cm before the end seam. ''

This is a bit confusing, so I will try to explain.

Basically the weights should be spread along a line starting at the 5 o'clock mark, past the 6 o'clock point (bottom of kite) to the 7 o'clock mark. This is NOT critical, I myself have often don't apply weights over the full length of this line, just along it.

The line of weights is 8-10 cm in from the trailing edge. That is it is in the `hollow' of the kite material formed by the `leech line' in the trailing edge. In this position the weights are NOT visible from the ground except posibably as a shadow through light colored ripstop, or plastic. Not a problem with silver mylar or dark (eg: black) ripstop.

One thing that I do do is that while I spread out the weight along this 5 to 7 o'clock line, I put more weight round the 6 o'clock mark and thin out the weight toward the ends of the line. This seems to make the kite roll slightly less and you don't get a sudden boundary between the weighted sail and un-weighted sail.

Leech Line...
The leech line basically creates drag which produces the internal pressure on the sail. Only a little pressure is needed to `puff out' the sail.

The leech line I have found must only be minimally shorter than the real circumference of the circoflex. Just enough to `fluff' out the ring (stopping any wind ripples) and no more. Any more just produces drag and stops the kite flying as high as it should.

The smaller the circoflex the more this is the case. In some of the mini-circo's I have built I only use the leech line to `wrinkle' and `pucker' the sail, after that the leech line was left loose, with the ends sticking out.

The only case I have for more leech line is for when the wind dies suddenly and the circoflex is too high. The weight of ithe flying line in this case may pull your kite in a ``nose'' dive. More drag to stop the kite getting too high might be the trick in this case. Though it is probably more of a case of just not being a good day to fly circo's.

Making Leech line gather evenly spread
To make the gather even when building, I do the following BEFORE inserting the main spar. I tie the two ends of the leech line together (20cm or so larger than the circoflex circumference) temporaly so they can't be lost. then pull as much leach line (both lines) out of the kite, gathering the whole trailing edge on as small a section of leech line as posible!

This puts lots of little folds and creases into the trailing edge, making the gather the same over the whole trailing edge. In fact in my smaller mini-circos this creasing is often enough to do the job leaving the leech line just loose in the pocket.

After that I tie off the leach line about 2 cm from the line marks (EG: 4-6 cm shorter than the kites circumference) and just spread the material out again, drawing the whole leech line back into the leech line pocket.

The main spar can now be inserted (and if desired, sealed) into the leading edge of the ring.

Folding into three for transport..
A circoflex will fold without twists into three. Experiment with a rubberband and you will see what I mean very quickly. Of course somehow removing the spar all together (ferules?) is even better!

WARNING: Be careful that you fold it the right way or you could really twist the spar, which in fibreglass can cause it to fracture and splinter.

Flying problems
Too floppy...
The kite just collapises in a heap when a wind gust hits it or gets a bit stronger, then the spar is probably not rigid enough for the close in bridle point OR the wind is too strong, or turbulent.

Remember a circoflex is a light wind sea breese kite. Medium and strong winds tend to be too much for them :-( My mini-circoflexi however while smaller can handle a medium (but steady) breese, unlike my full sized mylar circo.

One side always seems to collapse...
The trailing edge may be uneven. This basically improves with time and more evening of the trailing edge leech line. It maybe also be that the spar is just not springy enough. this was my problem with my first few circoflexi. With a springier spar I have not seen this problem since.
Rolling to one side only...
If the kite likes to suddenly roll to the same side when a gust of wind hits it, or continually leans in a strong wind but straightens up when the line is released a bit, or wind drops slightly, then the leach line is uneven from one side to the other. This is pushing the kite over, due to increased drag on one side.

On the side to which it rolls, stretch out the leach line gather a bit, moving it round to the other side. Test, then repeat until it flys straight.

Other less likely causes of sideways flying is: uneven streach or seams, or even uneven bend in the leading edge spar. The better and more even the kite construction is the less likely sideways flying becomes a problem, particularly in moderate winds.

A roll to one side can also be caused by the wind being less strong on that side due to a wind shadow from tree or tent. In fact getting a circoflex up out of a strong wind shadow is very difficult.

Goes up fast, then comes down doing loops...
Generally this is cause by either very large one-side drag problems (see above), a very bad bridle point (such as a large circo bridle point on a mini-circoflex), or the wind is just too strong or too turbulent (wind is more than the circoflex can handle). The latter being the most common problem.

Remember a circoflex is a LIGHT wind kite. Unless you are very lucky they will not stand up to moderate to strong winds. Large circoflex wind limits is determined by the spar colapsing from wind pressure into multiple loops. But small mini-circos (less than 7 meters circumference) usally has a really springy spar due to its size.

As such high wind limits is this `looping' behaviour.

Rolls back and forth to either side...
Simply not enough ballist weight has been added between the 5 to 7 o'clock marks, just inside the trailing edge. Add more.

In a lighter winds you may like to remove some of this `extra ballast' to lighten the kite, it will roll more but should fly.

`Nose' dives forward when wind drops...
When the wind dies the elastic streach of the flying line and its weight pulls the kite forward and downward. The top edge then over flys and the whole kite `nose' (not that a ring has a nose!) dives.

Use a lighter, less streachy flying line (try braided dacron, instead of springy nylon). In strong turbulant winds more ballast will also stop the top edge overflying, but it will also make the kite heavier.

Once a circoflex is in a dive the weight of the leading edge spar will keep it in a dive straight down. The result is sort of a slow motion ringed parachute. Rather preety, especially if it isn't your kite!

If you are very fast, you can pull the kite line HARD, to bring the lower edge back under the top edge. Then as the kite rises, you can again let out that line again. This is one case where pulling the line on a diving kite is a good thing to do. Most other kites require the oppisite behaviour to stop the dive.

It is for this reason that it is not recemended you fly the kite out over water. If it `nose' dives, the kite will probably end up on the bottom of the lake (or whatever), and if the bridle tangles with bottom junk, or a current is present, you might as well say goodbye to it.

Top colapases backward...
This is the oppisite of the last on (nose dives). Now you probably have too much ballist weight for the wind conditions. As such weight is pulling the kite to far backward. Also may mean the kite spar is not springy enough, but this is not a good indicator for that problem.

Note that it is normal for the kite to lean backward slightly when flying is flying correctly.

Flys with a low flying line angle (less than 30 degrees)
The bridle point it too far out in front of the leading edge. This means you need to re-calculate and re-bridle all twelve (or however many) bridle lines.

OR more likely...
You have tightened the leech line far too much, so it is producing a lot of drag. Very common for someone's first circoflex kite. It is a lot easier to over do the leech line than not enough.

Leans so far back it colapses...
That is when you let the ring go to fly up in a good wind, the kite leans too far back in the wind. Note this is different to ``Top colapases backward'' point above, The Top is not colapsing, the whole kite just leans too far back.

The problem is that the bridle point is to close to the leading edge ring plain, so that the kite even in a good wind just leans too far back, to fly properly.

The solution is the same as the last, re-calculate and re-bridle all the bridle lines, this time moving the bridle point away from leading edge ring plain.

NOTE: Once the bridle point is roughly correct it does NOT require adjustment for different winds, thank the powers that be. The ballist however may need some change for ultra light to medium wind changes.


Created: 4 May 1998 (copied from top level page)
Updated: 27 June 1999
Author: Anthony Thyssen, <anthony@cit.gu.edu.au>