Friday, May 31, 2013

My first workable ariel self portrait.

I arrived by bike. and got to work so quickly I forgot to take off my bike helmet.
Posted by Picasa

Monday, May 28, 2012

The All-Round Picavet

Last Saturday I took the chance to do some kite aerial photography, but time was short and I could not build a picavet to steady my camera before I left my house.

So I built one on site.

After 40 minutes of snipping, cutting and taping on the side of a footpath close to the water in between the Canada Goose droppings I ended up with this:
A picavet constructed from plumber's strap.

The pivacet I constructed out of All-Round,bottom rear view
The camera is a Canon PowerShot 410 I previously bought for 20$ Canadian in a pawn shop, so I was not two worried about losing it into the drink or dropping it from a height or both. I have a 4GB SD card in it so it can take thousands of 3.2 megapixel images before running out of space.  I did not have timed photos working (I plan to use CHDK) so I simply launched with the camera in video mode.

All-Round is usually used for suspending plumbing or reinforcing electrical wire in construction. It cuts easily with aviation snips and forms easily with the hands. It acts a lot like Meccano beams but comes in cheap long rolls, one variation is covered with nylon!  I estimate that I used 1.25m of strapping and that cost me all of a Canadian dollar.

I made a bender out of  the same material to make angled bends easier, but most of the forming was done by eye with my hands. The frame is deliberately squashed to one side to balance out the camera below the suspension point.

The cross was more all round with a spacer made from strapping to allow the frame to twist on a 2 inch long number 8 nylon bolt with two nuts on top jammed together to prevent them twisting off.  The camera was bolted on with a shortened 1/4"-20 nylon bolt and stabilized with some electrical tape (I ran out of duck tape early this month).

Next was a cradle of dark fishing line, some carabiners on the flight line and a trial launch for balancing. I had another camera with me (a better one that I was not trusting to my first picavet) and here is what the result looks like in the air:
Kite, Picavet, Camera, and Moon.
If you look closely just below the line between the kite and camera the crescent moon just stands out from the high cloud cover. Whoever identifies the kite can has bragging rights for as long as they want them. :) Also, the safety line attached to the camera strap is visible swinging downwind.  I wanted to be able to try again if my construction was not up to snuff!

The wind was strong -- Beaufort 4 according to the video evidence -- and the kite was on the edge of being overpowered despite flying high and stable.  The only disturbances to the camera I saw were how puffs would pull the line tighter and pitch the camera along the axis of the line as it straightened and bowed.

I now think that I should have added some split rings or other low friction loops to run the cradle line through to counter this , but I'm still very happy with the results of a first try.

Look for stills from the 640x480 video later, the video itself is too bouncy in pitch  to compress nicely.

Oh, I used my KAPstan to pull in the line, and it made what would have been a chore a simple workout.  I would not have sent the kite up ~100m more in that wind if I didn't have a good way to pull it down and hand winding onto a halo at high tension (I had no way to measure it but there was no problem lifting that 150g camera with 2 alkaline AA's and a steel picavet!

This effort is not as clean and tidy as one of the refined units from Brooxes shop, but I sure had fun with it!

Sunday, April 22, 2012

Kite Mobile in Cube Land.

I must confess that I work in cubicle land to feed my hacking habits.

But my co-workers, neighbours and random visitors know that someone odd sits in my cubicle when the see this:

The hanging gnome is an in-joke from my wife based on a trip to Adelaide South Australia where we saw some Christmas elf decorations that had blown(?) off light stands on Rundle Mall and were hanging from their safety cables.  I'll find that surreal photo someday...

The fighting kites are from Lee Valley Tools [1] ten years ago, the tee-bar hook/adaptor is from Home Depot and the rest is just coat hanger wire and fishing line.

Lots of folks now give directions along the lines of 'two cubes toward the windows from the kites' with great effect.

I sometimes wonder what people really think about the guy under that mobile...

Are there any other mobiles out there in the cubicle wastelands?

[1] Warning! Lee Valley Tools is an enormous time and money pit for those of us who love beautiful functional tools.

Thursday, March 8, 2012

Easier letting out of line, proof of concept.

While dreaming of getting my bike rolling in a couple of months I struck on a way to take up the slack thrown out by the take up reel in the designs I have show so far. I just modelled the idea in proof of concept form on my existing version 2 winder - in modelling clay.
Proof of concept model of a dual drum capstan (Operator's view).  Feeding line out causes the line to hop up on  the larger drum to prevent slack line from building up and getting tangled.
The basic idea is to use the first fair lead from capstan to the take up reel to cause the line to jump up onto a second capstan drum profile that has a diameter just larger than the top layers of string on a full take up reel. The jump happens only when there is enough slack to make it possible, and then the line stays on the bigger drum as long as the line is unwinding.

When the capstan is used to wind in, the line is guided back onto the smaller drum almost instantly, and only a small amount of slack builds up for the few turns it takes for the line on the large drum to be unwound again. That smallish bunch of slack could be handled with a tensioning spring arrangement - unwinding with the single size drum would produce slack as long as the reel unwound and no spring could control that.

 So, I tried it out with modelling clay tonight and it works!

Some notes:

  • Clay is ideal for the prototype, but I'll have to use something heftier for the real thing. The tension to pull line off the reel is driving the line into the clay.
  • The profile of the large drum should be level or slope toward the smaller drum so the string will travel toward the kite-side fair lead smoothly. As it is the line over-runs itself regularly and cuts out bits of clay.
  • The positioning of the take-up side fair leads is now more important. I just drilled another pair of holes for the eye screw, but it seems the best position is between those holes. C'est la vie.
  • The natural positions for the fair lead screws seems to be angled further toward the axle than they have been. I have not hidden the screw threads on them yet, and the line keeps getting caught down there.
  • I didn't notice before but the combination of a reel mount a bit skewed from the axle axis and the slipper clutch does some level winding!  I'll look into this more later.
Next Steps

Build another capstan out of a pair of disks of 3/4" plywood and try that.  I want a unit tailored to my 8" halos so I can unwind and wind with little trouble.

Experiment with different fair lead positions.

Add some way to slow the line as it goes out.  I can brake on the drum or support disk for now, but something more elegant might be needed for large line tensions.

Tuesday, February 21, 2012

A video introduction is worth how many k-words?

I have slapped together a pedestrian video introduction to how the winder works now and posted it on YouTube:

Also: there have been some excellent comments, experiences and challenges raised in the KAP discussion I started to get some feedback on my first article

Sunday, February 12, 2012

Introduction to the KAPstan kite winder

Please note:  The graphics in the early part of the document are the best I can do for now.  I hope they get the point across!

A Kite Line Capstan With Winder Adaptors.


In the summer of 2009 I decided I wanted to get into Kite Aerial Photography, and quickly became dissatisfied with the tedium of winding in kite string; especially with a lot of wind.
So, I went looking on the internet for kite winders and was unimpressed by what I found since I did not want to limit myself to just one line weight per bulky and heavy-weight winder. I really like flying from a halo spool; thank you very much.
But! One cannot wind lots of line under significant tension onto a lightweight halo. The halo tends to collapse or explode due to the accumulated stress. The heavy-weight winders are heavy because they are built to take that stress.
One comment on the page about a halo winder exploding got me thinking:
I would think that the right approach would be to have the cord wind around a shaft a few times before going onto the spool; the shaft would thus supply most of the tension for pulling in the kite; the spool could be left a little bit to avoid an overly-loose wind, but would not have the huge force pushing on it.
supercat, May 26 2004
That sounded like a capstan and take up reel to me!

Sailing Winches, Capstans, and Take-up Reels

Now, my other summertime hobby is sailing, and there we deal with lines (ropes) under horrific tension and make them tighter with sailing winches, or capstans as they are properly called. The lines to be tensioned is wrapped clockwise around the barrel of the winch 2 or 3 times and pulled taught so the loops are tight against the barrel. Manually operated capstans have pawls inside (somewhat like a bicycle freewheel) that allow it to rotate clockwise only and snug winds of the lines have a lot of grip against the textured barrel; so the line will not slip back out again as long as a light tension is maintained on the tail ( that is the bit of line that is already 'past' the winch)
Usually it it possible to plug crank handles into the top of the capstan and use them to crank the barrel around to pull in more rope. The handles swing out 2 or 3 times the radius of the barrel so there is some mechanical advantage working for the cranker even with direct drive. For situations needing even more pull there are winches that use a lower gear ratio if you crank the handle counter-clockwise, and even fancier units that will gear down again if you crank clockwise again. Some sailing winches have an additional rope gripper that handles the tail automatically to avoid dedicating one person to tailing while another hefty soul grinds1 away on the winch handle.
Of course the hefty soul could be replaced with a motor. But motors are heavy and illegal in wind-power-only competition. And sailing is all about competition.
What happens to all that line when it has been cranked in? Generally it gets dumped on the deck because the lengths are manageable: meters only, not tens of meters. Its usually too long (and too thick) to wind onto the capstan of the winch, but O.K. to sit on or coil up a bit after the fact. Quick access may be more important than utter neatness all the time.
But if there is a lot of line, it should get wound up on a reel under just enough tension to keep it neat. That means the drum can be of lightweight construction (even cardboard!) because the windup effect is not going to happen once the winch/capstan has dealt with the tension..Managing the tension difference is a detailed subject I don't want to completely understand; I just want to preserve my halo kite string reels so I can take aerial photographs of racing!

My kite string winder wish list

  • Make it possible to use a halo as a take up reel
  • quick changes of take up reels so:
    • I can fly a a kite on the halo alone, and retrieve it on the winder.
    • several kite lines on compatible reels can fly and pulled down with the winder
    • I can chain several reels worth of line together.
  • handy enough to be used on a boat, especially a sailboat.
  • provide some mechanical advantage for high pull situations
  • have a speed handle for low pull situations
  • lightweight, but robust enough to go in a backpack, be used on a boat, and by a child.
  • home build-able from cheap material

The Basic Ideas

I started with the basic capstan to reel idea ( the reel is on a slipper clutch to control the winding tension):

and put the winch and reel slipper clutch on the same axle to ease construction and save space. This required a pair of fair leads to direct the line from the capstan to the reel

This is a schematic of the top view:

One Real World KAPstan2 Winch

Here is an operators view of one real world implementation:
  • The spine of the winch is made from 3/4 inch dowel with a simple hole drilled crosswise for the 1/4 inch bolt axle.
  • The top and bottom cross pieces are also dowel, glued and screwed onto the spine where it was cut with a 3/4 inch drill for a flush fit. The top cross piece forms a handle for the left hand, and the bottom rests against the gut or hips to support the whole for the right hand to operate the crank.
  • The white capstan is made from the plastic hub of a lawnmower wheel, the ribbed texture is a neat bonus I only discovered after cutting off the tire!
  • The green 50 pound test fishing line is being wound on the reel it was sold on with the spider cut out to make it small halo.
  • The 3 fair leads are eye-bolts with their eyes twisted a bit out of true so the line can be slipped in from the side at treading time.
  • The pink stuff just visible to the right of the reel is 'rubber' packing foam cut into a disk to support the reel from the inside with a hole in the center for the axle to go through.
  • The wing nut just visible to the left is on the end of the 1/4” x 4” bolt that forms the axle. The axle passes though the slipper clutch/reel holder, spine, and capstan to a double ended crank on the far right.
  • The line is threaded in from the bottom of the reel to the kite as follows:
    • it feeds up from the bottom of the reel
    • to the reel-side fair lead
    • to the capstan takeoff fair lead,
    • around the capstan 4 times in the same direction as the takeoff reel,
    • then through the capstan fair lead
    • to the kite.

Here is a capstan-side view with an Premier Kites 8 inch winder mounted. Only the foam insert was changed to fit the larger reel. Note the path the line is threaded through the fair leads and around the capstan and reel.
The crank bar is made from hardware aluminum3 1.25” by 1/16 bar stock, the knobs are plastic drawer pulls. The frame is made from 3/4” dowel. I cut the spinal dowel with a 3/4” drill to make clean T-joints at the head and tail ends.
The head of the axle bolt is visible here, clamped to the crank by crimping it around the bar with a bench vise after starting the bend with a hammer.
The black backing disk on the other side of the spine is made of 1/8” expanded PVC board sometimes sold under the name 'Sintra' in Canada and the United States. This stuff is nice to work with: I cut that circle out with aircraft snips after drilling a center hole and scribing a big circle as a guide.
Here is the whole winder from the other side with the 8 inch halo mounted and tow other reels with their inserts to show the range of what sizes work. The holes were punched though the foam with thin walled 1/4” brass tubing because drills just don't work on this foam. I cut the outlines with scissors and I did not try to be precise at all. There is a lot of room for eccentricity and run out here since the rotation speeds are so low.

The black retaining disk for the foam is also made from Sintra. This was cut with the biggest hole saw I had and it fits nicely inside the smallest reel while working well with this big one.
Here is another view from the reel side with the fishing line mounted to show the big backing disk along with the foam insert, insert retaining disk, and the axle hardware (a jammed pair of nuts) that keeps the business from spinning off while slipping.
The foam insert allows the reel to be aligned to the fair lead that feeds it. Level winding can be approximated by nudging the line to the appropriate sides with the thumb of your hand on the top crossbar while cranking with the other.


Slipper Clutch and Reel Holder

Here is the winder, upside down, with the reel support and slipper clutch removed from the crank axle.
The big backer disk would go on the axle first, then the foam reel adaptor, then the small disk, the washer, the hex nut and lastly the wing nut.
The hex nut is finger tightened enough to squeeze the backer plate against the hairpin clip just visible in this photograph. The tighter this pressure, the tighter the line will wind on the reel since this is the slipper clutch that controls that tension.
The wing nut is necessary to hold the hex nut at the position selected since normal operation will either tighten the nut (right handers) or loosen the nut (left handers) as the clutch slips. Just tighten the wing nut enough to jam the two nuts together. Remember you will have to un-jam them with your fingers at some point!

Crank Axle Through Spine and Retaining hardware

Here the crank axle has been removed from the spine and the retaining hardware laid out in order of assembly. The two washers on each side reduce friction and slop between the capstan, spring clip, and the spine between them.
The spring clip is threaded through a hole cross drilled in the axle far enough away from the capstan to leave room for washers.
The spine is cross drilled with a hole (barely visible here) big enough to allow the crank axle to turn freely without slop. Start small and work up in the smallest increments you can until things run freely. If needed lubricate with candle wax later, but will swell the wood and make things tighter.

Attaching the Capstan to the Axle

I used a piece of plastic hose to re-size the bore of the lawnmower wheel hub to fit tightly on the bolt. This is where most of the kite line strain and leverage from the crank racks the capstan so it must be strong and solid. Fortunately that is not hard to accomplish.
The hose was picked to be a tight fit for the wheel bore, inserted and cut to length, then drilled out for a tight fit on the 1/4 inch crank axle.
With the axle pressed in (and the crank installed, see below) I drilled through the face of the capstan, through the hub, hose, and axle to install a cross pin that fixes the wheel rigidly onto the axle. I use coat hanger wire for the pin; I will use a rolled spring pin for a cleaner look next time.

The Crank

The crank is a bit of aluminum bar stock with drawer pull knobs for handles. The bolt of the crank axle is threaded through the a hole at about the 2/3 vs 1/3 point and the bar bent up to pinch the hexagonal head of the bolt. Note: the gaps between the crank handles and the capstan are excellent traps from stray kite line. So eliminate them in your editions!

Obvious Possible Improvements

These have been left as an exercise for the reader.
  • The hex nut jammed with the wing nut lacks class but that is the best I have for now.
  • The wire pin through the capstan and axle should be a rolled spring pin.
  • The length of the spine should be adjustable to the body of the user.
  • The sloppy nature of my crank handle leaves space for the line to get wound between it and the capstan when threading or winding out (see below). This is just a teething pain until someone comes up with a better construction technique.
  • Clean up all the places the line gets caught while threading.
  • Add a holding strap for the crank.


I managed to accomplish most of my wish list and have hopes for the rest. It is light and easy to transport, the parts are cheap to buy, alter, and assemble. It is fairly handy to use, and gets out of the way easily when it is not wanted.
The mechanical advantage of the long throw handle is amazing to use, especially when the line is being reeled in so nicely. The small size of the capstan wheel helps a lot.
The high speed, short throw handle works well too and its amazing how fast the line comes in when all one has to do is crank; instead of swinging the arms all over to dump line on the ground for later low-tension pickup or for the slow one wind at a time spooling directly onto a halo.
A small capstan can wind onto many (larger) sizes of reels. The example 8 inch halo seems to be about the workable limit. Larger capstans would match better with larger reels for ease of operation, and a longer crank handles would preserve the mechanical advantage.
The cranking speeds are low enough that sloppy construction will still work. The critical fits are all along the axle, but most are self aligning or decreed by the drill bit for the axle hole. I don't think I measured any linear dimension on the prototype; it was all done 'by eye' and 'to fit'.
This thing should scale up to handle strong pulling kites easily. I suspect a heavier spine, a back strap, bearings on a thicker axle, and a heavier crank will be most of the changes. The wheel hub should be able to handle a lot of pull as it is.

Disadvantages (with some suggested fixes)

When the reel is bigger than the capstan, then unwinding line while threaded is a tricky business. The problem is that the reel lets out more line per revolution than the capstan will wind out; and the extra line usually bunches up between the reel and the last fair lead before jamming between the backing plate and the spinal column. Retarding the reel with a thumb is usually enough to prevent this from happening, but it is a pain. I hope to add a one way retarder between the spine and the back plate to do this automatically. For large line releases it is usually nicer to un-thread the line and use the reel free-hand anyway.
This winch does nothing for twisted line. It leaves the twist as it finds it as long as there is some tension. If there is a lot of twist and the line goes slack, then the line will twist on itself and probably tangle somewhere inconvenient4. The free-hand reel technique used to let out line will affect how much twist added or removed from the line so pay attention and alternate sides when letting the line rip off the reel.
Threading and un-threading the line under tension is tedious but not horrible. I worry more about teaching others how to do it than the personal time and attention needed this ritual.
The capstan has to be smaller than the winding diameter of the reel for the slipper clutch arrangement to work. If the capstan is bigger than the winding diameter, a gearing up mechanism is needed and that seems too complicated to be worthwhile.

Further Work

Addressing the obvious improvements is a priority, but I see other riffs on the basic concept.

Cordless Drill Power

I have made an electrically powered version. In fact that was first! I learned just enough to tease you with.
It turns out that a cordless drill, the rubber core from a sanding drum, some O-rings, the top of a soup can, plus some coat hanger wire can develop enough pull to snap 50 pound test fishing line while leaving the winds on the reel loose enough to fall over. Fine trigger control is necessary with that monster.
I hope to refine this beast for wider consumption later. It seems to be the perfect use for a cordless drill whose batteries have died. Replacing those expensive batteries with a heavy sealed lead acid battery means that the unit can be left on the ground with a few traction spikes to haul down kites with no cranking!
However, that is a story for another day. The existing kite fishing winches are safe for now.

Bicycle Powered

I usually am on my bike when I am kite flying so using my legs via gearing would be nice. Even better would be to use my 3 year old's boundless energy.

Wrap-up and Challenge

Kite fliers tend to be inventors and hackers. My goal with this missive was to throw a challenge out so others will improve on the basic concept and construction insights that struck me this summer.
If someone wants to construct and sell winders based on this, go ahead! I hope you have competition! I want a sample, and some form of credit, but I don't want to block real innovation in a pastime I enjoy.

Contact Info

Albert den Haan
Ottawa, Ontario, Canada.

A Kite Line Capstan With Winder Adaptors. by Albert den Haan is licensed under a Creative Commons Attribution-ShareAlike 3.0 Unported License.
1With the handles installed these winches do look like manual coffee grinders. The slang term for the tedious winching task becomes too obvious.
2Yes, I had to use the Kite Aerial Photography acronym.
3'Aluminium' to those outside North America
4This is not unique to the KAPstan, it just has more inconvenient places for the tangles to go than some other reels.