Dynamic Spreader Spar for Variable Wind Conditions

edited April 2014 in Technique
Winds have been very strong and gusty here for the last few days since the start of WWKW. KAP conditions have not been ideal, however the higher winds have given me a chance to experiment with an idea I’ve been wanting to try for a while that will allow my Levitation to adapt to varying wind conditions.

The 9ft Levitation delta has become my primary kite for moderate to high winds from 15 MPH and above. (My Jones 7ft Rok is my primary kite for lighter winds up to ~15 MPH but that's another topic). For winds between 15 and 20 MPH, I’ll typically use the Levitation with a standard length spreader spar which is 46.75 inches long. For higher wind conditions I’ll use a shorter spreader (39.5 inches) which extends the wind range up to 30MPH+ and significantly reduces the amount of pull and provides a more stable, pleasant flight. Additional discussion of the dynamic spreader spar can be found here:


The short spar works well as long as the winds stay consistently above approx. 20 MPH, however, it doesn’t provide the desired lift in lighter winds.

I’ve been working on a dynamic spreader spar that adjusts its length based on the wind conditions. I’ve only had a few hours of flight time with this, but it looks favorable that it will allow the Levitation to be flown from ~5MPH to 30MPH+ with a single spreader spar configuration. The dynamic spreader is basically a spring loaded telescoping graphic tube that keeps a constant spreader tension on the kite wings. At low wind speeds the spreader is fully extended maximizing wing surface area and lift. As the wind increases, the spreader is compressed resulting in a swept wing profile reducing surface area and line pull.

Dynamic Spreader Spar

The dynamic spreader I’m using is an initial prototype and is constructed from graphic arrow shafts (roughly equivalent to P200 tubes). The uncompressed spreader length is 46.75 inches in light winds (length of stock spreader) and compresses down to approx. 42 inches in winds up to 30MPH. Its compression limit is set to 39.5 inches which it will reach in winds above 30MPH. The spring tension can be adjusted to provide different amounts of compression for a given wind speed.

With the current settings, I’ve been able to fly very comfortably in winds from approximately 15 MPH to 30 MPH with a line pull between approximately 2.5 and 12.5 lbs.

The dynamic spreader weighs 38 grams. The stock fiberglass spreader that comes with the Levitation weighs 67 grams.

Here are a few videos and photos showing more detail:

Basic operation video.

Launch and flight video showing the spreader compress as the wind increases.

Picture: No spreader compression (light wind flying, standard 46.75 inch length spreader)
Stock Spreader_5138

Picture: Medium compression (winds up to 30MPH, 3.5 inch compression)
DS Partial_5158

Picture: Full spreader compression (winds greater then 30MPH, 7.25 inch compression)
DS Full_4694

Additional pictures and videos showing construction details and operation can be found on my Flickr site.

I plan on developing this further over the summer. It would be interesting to see how this could be applied to other types of kites.


  • Mike, Love the design and there certainly is a need in this area. I use the shorten spreader with my Levitation Light in heavy wind, but has you point out, during a lull you need all the wing span you can get to keep the cameras in the air.

    Paul's Fishing Kites (PFK) has a similar concept in a small delta. The name of the kite is Flex Wing Delta. The PFK Flex Wing delta uses two sturdy rubber bands, one for each horizontal spreader pocket. I have flown this kite for over 5 years. The kite does well in high wind with this simple design. Uses a fix spreader length with the rubber bands pulling the wings out to full width in light breeze and allows the wings to sweep back in hard blow thus increasing the wind range. The kite is too small to lift a KAP rig on a regular basis. Recommend you take a look at the design.

    A different but similar concepts is to add an elastic trim line between the kite keel and the main kite line where in light wind the kite relative angle to the wind increases (more lift) and decreases in high wind (less lift). Thinking about combining these two ideas.

  • edited May 2012
    I had not seen the PFK Flex Wing, thanks for pointing it out Jim. It would be interesting to modify the Levitation spar pockets and try this approach.
  • No confirmation of Mike's dynamic spreader is necessary; nevertheless, I comment to record that I made a copy of his design for my Levitation Delta, and I can report that it works just as Mike has described. It is wonderful to have this flexibility in the air, and responding to conditions in the air! In addition to the new spreader I did replace the wing and center spars with lighter ShyShark spars. Essentially I made the standard Levitation Delta into a Levitation Delta light. This kite is a fabulous flyer, stable and reliable in flight, and a beautiful object in the sky.

    Thanks Mike for the design and for the personal call re my inquiry about updates to your design.

    Robert R. Llewellyn
  • edited June 2013
    That's great to hear Robert! Thank you for the feedback, I wish you many enjoyable flights with it :)

    Any lessons learned and suggestions for improvement are welcomed.

  • I like the way that the Levitation Delta can be folded and fit into the standard case in its small size configuration. This size is great for travelling. I am pondering making a dynamic spreader with the two main pieces cut so that they can be reduced in length and fit into the standard case. Other than this I have no lessons learned or suggestions for next steps.

  • Good suggestion Bob. Let me know if you try it.

  • @ Mike: I wanted to let you know I built one of these for myself and put a kite up with it at a festival last weekend. It flew well unattended for hours in changing winds. Later I KAP'ed with it and believe it helps maintain a consistent line angle as well. Thanks for the contribution, it's appreciated.
  • edited July 2013
    Great to hear Ric!

    I was thinking about the aerodynamic effect that the dynamic spreader has on a kite, and recalled a point that Brooks made a while back. I think he hit the nail on the head when he pointed out that the kite aspect ratio changes when the wings sweep back in high winds (the kite wing span decreases and the effective length of the kite increases). This moves the center of pressure toward the back (trailing edge) of the kite which effectively moves the keel towing point forward, which is what you want for more reliable flight in higher winds. I think that this is the key reason that the dynamic spreader is so effective at increasing the Levitation's upper wind range.

    FYI, I've made some recent design improvements to the spreader that I'll try to post in the near future. One significant improvement is to apply 3 or 4 wraps of fiberglass cloth around the bushings and soak the cloth in fiberglass resin or epoxy glue for added strength prior to applying the heat shrink tubing. If you use the fiberglass cloth you can get rid of the zip ties. I also found a better way to attach the bushings to the spar that I will post later.

    I'm curious, how high of winds were you flying in, and what are the dimensions of the bushings and spars you used? I've found that the spreader will flex a small amount when under high load, so it's important to make sure that the inside diameter of the bushings are large enough so that the spars don't bind when it does flex. The spreader should compress smoothly even when experiencing a small amount of flexing. I noticed some binding when I made some additional spreaders using P300 and P400 spars.

    For my baseline dynamic spreader, I used P200 spars with a nominal outside diameter of 0.284 inches. The nylon bushings I used were approximately 1 inch long with an inside diameter of 0.325 inches. I found that if using a larger diameter spar (P300 or P400) you should either shorten the nylon bushing length to approx. 1/2 inch or increase the inside diameter of the bushing to keep the rods from binding when the spreader flexes under load.

    Thanks again for sharing your recent flying experience.

  • Mike, I was thinking about the same concern, binding when flexing under load conditions.

    Perhaps a third bushing in the middle may help to keep everything in alignment.

  • edited July 2013
    A third bushing in the middle would get in the way of the kite flex strap. You may be able to distribute four bushings across the spar as long as you leave the middle clear for the flex strap to re-center itself when the spar compresses.

    I have not noticed a problem with binding when using the P200's if I maintain appropriate diameter clearance within the bushings. I made a couple of spreaders with P300's and found if you make sure the movable spar is nice and smooth (may have to sand out any bumps) and if you increase the bushing diameter slightly, then binding isn't an issue. I played around with some P400's and found that I had to cut the bushings down to 1/2 inch length to make them work smoothly when flexed. Using a shorter bushing also means you need to beef up it's attachment to the spar (i.e. use fiberglass cloth or a lot of strapping tape, etc...). I'm not sure you would have to go much bigger than P200's since they are in parallel for most of the spreader length. I've not had a failure with P200's in gusts ~35MPH, but not sure of their absolute upper wind speed capability. If you make one, I would strongly suggest you get a decent amount of flight time without a camera attached if you plan to push it much beyond 30 MPH.

    There's a ton of experimenting that can be done to evolve this design, so go for it.

  • edited August 2013
    FYI, I made some updates to the dynamic spreader design and switched from P200's to stiffer spars (P300 and P400). My Levitation flies very well in high winds with the P200s and a tail but I noticed that it pulls significantly more to one side with the stiffer spar design. A few other KAPers have tried this and have noted their kites also tend fly hard to one side with the stiffer spars. The behavior is worse if you fly it without a tail. The winds have been calm here in the midwest lately so I haven't had much opportunity to investigate this behavior first hand but thankfully the other flyers are willing to engage in troubleshooting via email and telephone correspondance... a slow an painful process but will hopefully shed some light on the issue soon... stay tuned for updates.

  • Maybe you've thought of this but perhaps it's the weight imbalance. I think if one side is fixed and the other slides, then it can't be balanced for both uncompressed and compressed. Maybe the stiffer spars are heavier, so the effect is stronger.
  • Good point. The spars, however, should center themselves on the flex strap when the DS compresses, so the total spar length on each side of the kite shoule remain balanced (assuming the flex stop slides to the middle). So, I don't think there is a weight imbalance due to the larger spars. There is, however a small weight imbalance when the DS compresses due to the nylon bushings which are attached to only one spar. The overall center of balance will shift slightly off center when the DS compresses because of this, but I didn't think it is significant (I may be wrong).

    There are two other factors that I think play into this (it's a bit complicated to describe but I'll give it a try...):
    1. Non-symetric spar flex when the DS compresses. This is due to the fact that the nylon bushings are only attached to only one spar. When the DS compresses, the stiffness is no longer constant along the length of the spreader because one side of the spreader has a longer length of parallel spars than the other. This causes the spreader to flex asymetrically on either side of the flex stop, which could result in one wing bending back more than the other in high winds. Adding a 3rd bushing attached to the "movable" spar should help correct this.

    (I'll add a picture here to illustrate this better)

    2. The fact that the two spars are mounted next to each other (offset) instead of directly in line with each other results in a small offset between the two wings. Modifying the flex strap attachment point should help correct this.

    (I'll add another picture here to illustrate better)

    I think that items 1 and 2 above along with the additioon of stiffer spars results in non-straight flight. I need to try some of these changes out to see how they work, but won't have much opportunity to until we get some higher winds here...

  • edited August 2013

    First thanks for letting me experiment with one of your Dynamic Spreaders.

    Still a neat idea...that must move forward.

    As previously communicated (via e-mail / pictures / video). My 9' Levitation delta kite did pull to the right under stress from gusts. I flipped the Dynamic Spreader (DS) end for end and the kite few near normal balance. The DS that I am experimenting with is set up with the SkyShark P400 tubes (my preferred strength) and I fly with no tails (just not my style due to tight urban KAPing sessions).

    I agree with your line of thinking above. Possible root causes for the kite flying to one side include:
    - Asymmetric wing bending due to a possible number of factors:
    - Non symmetric spar flex during compression
    - Flex stop loop on back of the Levitation delta that the spreader (DS) spar fits through hangs up or impedes the compression or relaxation of the DS
    - Slight shift in the center of gravity due to two parallel spars now occupying the same space where just one spar was present (I am also thinking of forward / aft center of balance - not just right to left).

    As noted, several options exist to confirm the root cause hypothesis above and identify solutions.

    As previously posted and pictured below, one or more small light weight plastic springs internal to the existing spar may offer a more robust design with less sources of variation that could impact stable kite flight.

    The discussion above points out the obvious...kite set up tolerances are very tight and that we sometimes overlook the impact of very small changes. I still wonder at all this stuff stays up in the air as much as it does.

    On the experimentation front....the wind speeds have been very light across my KAPing locations in North American and Europe...my Dopero is getting just about all the flying time....perhaps the fall will bring stiffer wind and some time to experiment....I have reserved some free time at a beach I know at the end of September....time to experiment!


    Dynamic Internal Spring Spreader Spar for delta kites.

  • edited March 2014
    Here's a quick progress update on the Levitation Dyanmic Spreader (DS) design. I was able to isolate the problem that caused pulling to the right or left in high winds (see earlier threads) and found that the problem was due to the spreader not staying centered on the kite Flex Stop strap as the spar compresses. I developed a "Tracking Bridle" that addresses this issue and results in consistently stable flight in high winds without the need for a tail. Here's a photo of the DS with a tracking bridle.

    DS - Tracking Bridle

    I have built several DS's with this new bridle and they have all performed consistently well in winds of 30MPH+ with no tail. For more information and videos of how to build one yourself you can go to my Flickr "Dynamic Spreader (DS)" Set. I can also send a .pdf file with additional construction details if you contact me via Flickr email.


  • Mike,

    Love these Dynamic Spreaders (DS). Ben using them for several months on both Levitation Light and standard Levitation. The tracking bridle helps. The DS gives the Levitation delta a much wider wind range (I have parked my PFK Nighthawk)! I sometimes use two elastic cords on windy days to keep the kite a bit stiffer.

    Keep up the ingenuity and thinking out side the wing spars!

  • edited April 2014
    I have been able to replicate the effect Mike has achieved with what you might call a design 'hack' I have flown it and found it works well and, to my surprise found it to be self centering after a few flights has 'slackened' the tape.


    It is somethng of a 'there I fixed it!' bodge:


    but it works!
  • Very clever - and functional!
  • edited April 2014
    Nice job on your variation of the dynamic spreader. There are definitely many ways to "skin the cat". I also really like your blog. There's a lot of great information on it and it's very well organized, thank you for making it available.

    It looks like flexing at the center of your coupling piece helps keep the flex strap centered. I like how you're using a long, continuous coupling, and think that this can be a nice improvement to the overall design approach. Have you had a chance to fly it in higher winds? One thing to watch for is if the spars begin to flex too much under higher load and then bind in the coupling pieces. If both spars don't compress equally, then the kite will fly to one side when the wind picks up.

    FYI, I've made some modifications to my construction technique and now use epoxy putty and fiberglass mesh to eliminate the need for pinning the spacer. This results in a very compact and robust attachment.

    DS Nylon Spacer Directions-page-005

    DS Nylon Spacer Directions-page-006

    I've uploaded a series of 6 photos (converted to jpg from pdf) that show the construction technique and measurements of the spacers as well as a series of 6 photos that show construction details of the tracking bridle to the end of my Dynamic Spreader (DS) Flickr Set.

  • Thanks for the pictures and captions. It is very clear. Now for some construction work of my own. It appears by the way that you have experimented with cutting the two ends so that the DS will fit the kite case.

    Bob Llewellyn
  • edited April 2014

    So glad the pins have gone!

    About the 'tube version' (DSTV sounds a bit odd maybe 'bodge version' DSBV is better)

    I have had it out in about 25mph and it held up OK. It began to sharpen the arrow shape of the kite as it should. I'm a bit nervous about it as I imagine the tape twisting, failing and the consequential nose dive. On getting the kite back I found all was well: duct tape is strong (if heavy) stuff! But I chickened out and flew the PFK!

    I have 2 issues with DSBV:

    first is the over-sail of the nock ends, I haven't cut the 6mm stuff at all so there is enough spar to push into the sail when under full compression- I'm watching this closely to see if it's a potential puncture site. It's not a show stopper as the contact is with the sail when it's loose at full compression but I worry.

    second there is a reduction in the power of the kite. I have 2 Levitations one DSBV and one 8p sparred. I have found I can lift 2 rigs with the 8p sparred version (if downwind space permits) but try as I might I can't do the same with the DSBV!

    The self centring is a nice 'side effect' of the 'x' of the tubes. As it's carbon to carbon the friction is minimal, there is just enough play for a loose fit but not so loose as to bind. It's pure luck the bits I had worked so well. It can stick if the assembly fouls the rubber band, so far this hasn't happened in flight but it's possible. The angle between the tubes is very slight (less than 2 deg) but enough to 'lock' them in the centre giving a rise from end to end of about 2mm at the apex.

    On the kitchen scales it weighs about 26g.

    In the days of the bell-bottom flares I was a keen Peter Powell flier so I know the magic of dynamic geometry well!


  • Update on the DSBV: crashed it into a tree, duct tape stretched and whatever magic that made it work has gone: the self centre effect is lost. I'll see about a new method of bonding and report back...heigh ho.
  • edited April 2014
    Duct tape is amazing stuff! I don't know if you've ever watched episodes of "Red Green", but he would be proud of this use for the "handy man's secret weapon"! It's great for making proof of concept prototypes and looks like it did it's job again here. I was trying to think of some other ways to bond the two pieces of 8P carbon together.... you could possibly try some nylon zip ties around the two pieces along with the tape. Another method may be to use a couple layers of heat shrink tubing. It's somewhat flexible and tougher than duct tape. Another possible method would be to use fiberglass cloth and epoxy. Keep hammering on this, as I think you're on the path to a good solution to the problem.

    I really like your idea of using a floating sleeve with the two pieces of 8P Carbon. I think this has some promising possibilities of simplifying the design. I've thought of a potential way to incorporate this approach with the tracking bridle that will allow the sleeve to be rigid with no "x" flexing but will still keep the spar centered on the spine of the kite (see diagram below). The single floating center sleeve would have to be rigid and long enough (TBD length) to allow the spars to freely slide in and out when under high load forces and to prevent binding of the spars when they flex.

    Dynamic Spreader - Floating Sleeve 2.vsd

    Regarding the reduction in lift with the DS, I've found that using a stiffer elastic cord helps quite a bit. I'm presently using a 13 inch piece 3/16 inch diameter shock cord and it does a good job of maintaining sufficient lift for my 450g rig in winds as low as 12-15MPH. Using stiffer spars (P300/P400) also help maintain good lift because they allow "linear compression" for reduced wing span, and minimize the distance that the wings flex back away from the spine in higher winds.

  • edited April 2014
    I had another thought regarding simplifying the design, but realized that it wasn't such a good idea and removed it from this thread....

    Something to watch for when using the 8P carbon for sleeves is any abrasive effects from carbon-on-carbon interfaces. The outside edges of the 8P sleeves could possibly cause wear and abrasions to the inner spars (P300/P400) as they slide in and out under load over an extended period of time. One advantage of a nylon spacer is that it is a softer material than the carbon and should reduce wear over long periods of use.
  • edited April 2014
    Mike, I think I cracked it!

    See vid here:


    To quote Uncle Red: "It's not smart, or correct, but it's one of the things which makes us what we are...."
  • Mike,
    Many thanks for all the work you’ve done on the Dynamic Spreader; it’s been a fascinating journey to follow. I built my version a few days ago and made some initial tests yesterday with encouraging results. (I may opt for 2 elastic cords next time to better manage the 40 km/h sea breeze!)

    The only real departure from your September 2013 design was in the spacer fixing arrangement. Rather than epoxy the spacer directly to the spreader, I opted for a two spacer version using aluminium external ferrules as the spacer material.

    I used a short length of 3/4 inch internal diameter PVC irrigation pipe which I put into a bench vice and compressed into an oval shape. Two P400 aluminium ferrules were then slipped into the tube before releasing the vice allowing the PVC to resume its original shape clamping the ferules tightly together. I then forced epoxy putty into the remaining gap locking the two ferrules firmly in place.

    In my version I cut the ferrules into four sections, each ½ inch long. These spacer sections were then slipped onto the P400 spreader tubes and epoxied in place as per the original September 2013 design. (Heat shrink was applied to each spacer section to cover any sharp edges which could damage the elastic cord.)

    This is a simple way of fixing the spacers to the spreader but it seems to work well and should be suited to a range of spacer materials including 8P carbon as used by Bill in the DSVB experiments currently under way.
  • edited April 2014
    Nice video and information that you posted on your most recent blog entry. I was glad to see that the zip ties are doing their job. They are a nice little building block to have around.

    Regarding your video using two independent rubber bands attached to the center point of the sleeve... as you pointed out, for this approach it's very important to have well matched rubber bands. It's also important to have well matched friction between the two spars and the 8P sleeves. I found that even a small amount of elastic or friction mismatch between spars can result in asymmetric wing compression and the kite flying to one side. I think that this approach will work well in lighter winds, but I would be cautious in higher winds until you can get some extended flight time on it to prove it out.

    When working on the DS, I found that it's really (really) important to make sure the wings compress evenly, especially in high winds when even a small amount of asymmetry will cause the kite to fly to one side. I wrestled with this for quite a while and ultimately came to the conclusion that I needed to mechanically couple the two spars together so that the compression of one (relative to the center line of the kite) tracked the compression of the other precisely. The "Tracking Bridle" is one solution to the problem, and while it does look somewhat complicated, it really isn't any more difficult to rig than a standard Picavet. One nice thing about the tracking bridle is that it doesn't rely on matching elastic / friction forces between spars to keep things symmetric, the rigging of the bridle forces each spar to compress equally.

    As I wrote earlier, there are many ways to skin the cat, so am looking forward to reading about your continued work on this!

    Create on!!!

  • Al,
    Clever approach to use PVC pipe and aluminum ferrules, I like it! Do you have any pictures that you can share of this?

  • Mike,

    Here is a link to a photo of a spacer section and as you can see there it's a good fit without being sloppy. It remains to be seen if the aluminium causes wear on the spar but it moves quite freely at this stage. https://flic.kr/p/mTJDAA

  • edited April 2014
    More fun & games here:


    even without the tracking bridle it's getting close when the flex stop is shortened. I'm not sure but I think as Mike has suggested I have asymmetric friction issues.

    On test with a peak gust recorded at 25mph the kite flew steadily but would heel for long periods.

    While I wait for the post to deliver the nocks I need for the tracking bridle I'm attempting an improved linkage ring, the simple keyring might be improved with a 'v' profile to help the flex stop centre itself.


Sign In or Register to comment.

Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

In this Discussion