Sunday, December 30, 2012

Change to the Handcyle's Tail and Power Calculations

Aerodynamics and a Tail


One interesting aspect of aerodynamics shapes is that the tail shape can be as important as the frontal shape.  For instance, a square plate moving through air has less aerodynamic efficiency (Cd -- Coefficiet of drag) as compared to the same size square plate with a long tail.



I decided to add a more efficient tail to the handcycle.  Of course I am guessing that the shape I produce for the tail, is indeed more efficient as compared to not adding the shape.  I could be wrong!  For my bike, the tail is relatively short and therefore, the tail will not reduce drag very much as compared to a recumbent bike with a tailbox.  Here is an example of an efficient tailbox:



Since my ams are moving, I cannot build something as efficient as above and my interest is not in doing so.  

On the otherhand, a $150 TT (time trial) helmet could save me 4 seconds per 40K...


Or $10-$15 worth of carbon fiber may save me more... Here is the tail before trimming the trailing edge:

After - New Tail - Side View
After - Top/Rear tail View
I added the tail shape by crafting a few pieces of the pink foam board.  The foam took a number of hours to shape with the wire cutter, band saw, and 80 grit sandpaper.  The foam was fitted to the existing frame with 3M's 77 spray adhesive and then covered in carbon fiber.

Before Tail added

Design, Power and Speed

A friend and I rode together -- he on a very nice road bike and I rode on a 3-wheeled handcycle.  We maintained just about 20mph over a 20 mile ride.  I was pushing quite hard.  He had a power meter that kept the stats for the duration of the ride.  At the end of the ride, he found that he averaged just over 200 watts output for the duration.  Of course, I wondered what my power output averaged.  There is a site that does calculations based upon SRM Power Measuring Cranks.  There is no 3-wheeled recumbent bikes listed, but I presume that I had a very similar power output to a racing bike due to my 3-wheels.  

This raises a question -- what do I expect for speed with the new bike at a reduced wattage of output (I cannot hold 200 watts for 5 or 6 hours)?  The calculator's result for a lowracer:

If this bike can achieve 22.3 mph average (and 19.5 mph on a 0.5% grade) with my power output at 160 watts, I will indeed by happy with the results.

BTW, by choosing the "Lowracer with streaming tailbox" type of recumbent on the calculator, the speed increases to 23.3 mph.  Since my tail is much shorter, I would expect only a percentage of that increase in speed as compared to not having a tail. That still calculates out to about 20-40 seconds deduction over 40K.  Not bad if it holds true!

Finished Front Fork Views

The front fork is finished -- well almost.  I still have to cutout the dropouts on the added layers and re-drill the holes for the disc brakes. Otherwise...

The front fork has the two finish coats of West Systems epoxy over the many layers of carbon fiber. To this, I added four final coats of a polyurethane (with UV protection).  There was a lot of sanding in between all but the final coats of polyurethane.  While the end result is not perfect, I am relatively pleased with the look.  Daylight may change my mind! 

(There are black covers over the headsets in order that they did not received the urethane.)





I personally do not mind seeing the interweaving of various layers and direction of carbon fiber.  This is the way that the carbon fiber is laid up -- intermixing the direction of carbon fiber's grain through various layers.


Left Side of Fork

Right Side of Fork -- Displaying Chainline and Idler

The following picture -- from the rear of the fork -- is somewhat distorted due to the closeup of the camera. Notwithstanding, the picture does give an indication of the thickness of the fork down to the dropouts.


Rear of Fork 
Closeup of Idler in Fork




Tuesday, December 25, 2012

Somewhat Finished Fork

The fork is much closer to completion.  I have added a coat of epoxy over the last layer of carbon fiber.  This epoxy is still wet in the picture below.  BTW, I find that the epoxy brushes on much more easily with a small amount of acetone added to it.  This thins the epoxy for brushing -- but acetone evaporates very quickly.  So if one attempts to brush the epoxy too much, one will find it to thicken within just a few brush strokes.  From what I read on West Systems epoxy, this seems acceptable for a finish.

Front Fork -- with a finish coat of epoxy.


As you might remember, I had said that I would probably finish the bike with paint and not take the time to get a nice finish on the carbon fiber.  But, carbon fiber has an attraction -- and maybe this is the lipstick on a pig.  


I will shoot some more pictures that display the idler on the drive side as well as the fork thickness at a later point  -- the epoxy is curing!

Tuesday, December 18, 2012

New Handcycle Design

The winter is here and I needed a winter project. This is especially true since I overworked tendons in both elbows with the training over the last year and thus require a rest.  I had established a goal to set the course record for the Air Force Marathon in September (I did accomplish that).  But the heavy training -- 200 to 300 miles per week -- starting last March took a toll. The tendons were getting increasingly sore as the training continued, but I felt that if I took extensive time off from training, I could not meet that goal.  Now I am taking the time off to heal (and it is a VERY slow process) and filling the time with the design and building of another 2-wheeled handcycle.

My design objectives have not changed much from my original handcycle design:

  • Build a handcycle that can be balanced.  My first handcycle design had a trail that fell well out of the normal range for today’s 2-wheeled bikes.  That was due to the very low angle of the front wheel’s headtube.  This bike has a Trail that is about 60mm and an headtube angle that is consistent with many of today's bikes.
  • Keep the base/sitting area low enough to the ground that I can use my hands to balance the bike at stops.
  •  2-wheeled.  If each wheel of a 3-wheeled handcycle is essentially equivalent to a “front” wheel of a regular bike (no protection), and if a bicycle’s front wheel requires 40 watts of power at 20mph (whereas a protected rear wheel may only require 5-15 watts), then the handcyclist must overcome a real disadvantage compared to a 2-wheeled bike. Rough math would predict that the 3-wheeled handcyclist must generate about 65 or more watts for wheel rotation at 20mph as compared to a 2-wheeler.  That is a lot.
  •   I have to lie down and cannot use a kneeler-type handcyle. So the design reflects this.
  • Paint it.  I am not going through the extra work for beautifying the carbon-fiber.  Of course, I can smooth out the carbon fiber with light weight body filler and then either paint it or put on one more layer of large pieces of carbon fiber.  But I will probably opt for the paint.

Newest desgin
·        Sweet-looking.  Hey, if I get to design it and build it, I may as well build something for which I like the looks!  Of course beauty is in the eye of the beholder.  One only a parent could love...
  • Transmission hub (I-Motion9) for gearing.  This somewhat simplifies the design and potentially gets rid chainline problems.

This design (above) is therefore quite different from the first handcyle (below).  

First Handcycle design I built...sans wheelcovers...


The leg-holders will be added after I position/fit myself to the bike.  The dummy in the design is fairly accurate to my size -- but by no means perfect. ...maybe better looking though. 

As you can see, the front wheel is a small one in order to accommodate the much-more-normal headtube angle.  Again, this headtube angle is much closer to that of a road bike as compared to my previous design/attempt.  A large 650 or 700 wheel would cause the steering and crank to completely obliterate my line-of-sight unless I lift the base.  By lifting the base, I would no longer have easy reach of the ground. Therefore, the small wheel...


The front wheel is a 451 (20” x 1.125”) and home-built since I am putting in a transmission hub.  The hub has 36 spoke holes.  

A number of features have percolated from this design. One is that my feet will be closer together and therefore -- hopefully -- I will be a bit more streamlined.  The small wheels of this design keeps the wheelbase shorter and therefore the weight of the frame lower.

Another design consideration -- Since I am maintaining front-wheel drive (rear wheel drive is a whole other topic) on the bike, I have a chainline that must be redirected on the return side in order to allow for the chain to go through the fork in a placement that still allows for the integrity of the fork.


For the transition from design/paper to a foam plug, I built a foam hot-wire cutter that can cut through a large block of foam (very large!).  The foam cutter is essentially built from a three-legged shelf for which the fourth leg is the hot wire.


The battery charger was from my last (and much smaller!) foam cutter design.

With this design, I attached a spring to the end of the nichrome wire.  This worked much better than I expected.  The nichrome wire never broke in all the cuts. Additionally, I used a router to sink a movable square into the table top in order to insure straight cuts.

To assist in cutting the foam according to design, I attached a paper outline (printed 1:1) of the base (or fork) to 2 sides of a large block of foam.  The foam was built up from 2" thick foam board -- the pink stuff from Home Depot. I employed 3M's 77 adhesive for holding the foam together.  The adhesive works quite well -- but be liberal with it.  The benefit of a spray adhesive is that during sanding through the adhesive, there is no problem.



The cutter worked very well in building the rough shape of the base and fork.  I then used a drywall sanding screen to add rounded edges.  Within a day, I had the foam plugs of the base and front fork pretty well completed.  When cutting the foam, I made a number of changes to the design as you can see below:

View from the rear of the bike


Fitting the foam together

Testing the headrest/base for fit


When building the foam plug, I decided to keep the whole base deeper than the design drawings.  Originally, the design drawings had a center tube.  I will use the additional depth/space for hydration bags when I remove the foam with acetone.

I attempted to concoct a filler for the dings and such of the foam plug. Others have used the foam crumbs from sanding in combination with white glue as a filler.  If you do this, you must water down the glue substantially since the glue is not very sandable otherwise. 

Homemade foam filler -- Add water to glue!


From lessons learned, I covered as much of the foam plug with a layer of carbon fiber rather quickly. The last handcycle's foam plug I built got dented and even broken after falling on it).  Addding the carbon fiber helped in stabilizing the foam.  

Wood form to hold handcylcle square and plumb
In order to align the foam plug -- before any carbon fiber is added -- I built a wood base/holder with 135mm front fork offset and 100mm offset for the rear wheel in the base. The wood base is absolutely mandatory in order to square and plumb the plug. If I did not construct this, I could be assured of having a bike that would not track properly.  Even with this wood base, I used a "laser" level that shot a light the length and height of the holder. 

Into the foam, I mounted front and rear dropouts cut from carbon fiber plate.  Two of the dropouts contained the disc brake holders.  With the dropouts in place, and with the wood base built for holding the plug square and plumb, I then applied a carbon fiber layer to much of the foam and drop outs to hold everything in place.

Adding dropouts into foam plug
Drawing Displaying Disc-Brake Bracket

This picture below displays the bottom bracket (with crank) as well as the idler in place.  The idler is taken from an old derailleur's return cage. The idler changes the chain’s return direction.

Bottom bracket

The next phase of work required that the headsets/steerer tube be built into the frame and fork.  I used AHeadset ZS (Zero Stack series) for a 2” headtube and 1.125” steerer tube. The total height of the headset configuration took more space than I initially estimated.  Therefore the design changed slightly in the front fork to accommodate the height.  The change in height of the foam ply was fairly easy to achieve by merely gluing on more foam and shaping it. 




In order to ensure that the headset would align properly, I install the headset parts in carbon fiber tube.  Each carbon fiber tube was sized to the component first (lower fork, base, upper fork). Then the tube for the base case epoxied in with ar carbon fiber layer to the base.  After that cured, I epoxied in the fork's upper and lower headset tubes -- with the steerer tube installed through all headsets. I used the laser level to plumb and level each part and the bike.  I then applied a layer of carbon fiber throughout the headset to hold all parts accurately in place.  After that cured and after a thorough check to ensure that the parts were properly positioned, I added many layers of carbon fiber.

With the headsets in place, I was then able to check out the chainline, disc brakes and steering:

Well, maybe not beautiful...

The disc brakes and headsets aligned very well – phewwww.

During the downtime as the fork's epoxy is curing, I add light-weight body filler to the base. I think I sand away 98% of what I apply... and then I do it again and again. I guess when I paint it, I will learn whether I performed a good finish job.

Lightweight body filler for smoothing out carbon fiber

The front fork requires quite of bit of work in order build out the thickness of the fork where it carries the majority of the stresses.  This will be the basis of a later update.

Front fork before the buildup of carbon fiber