the Exo-Foot 3d printed biomechanical “Footware” platform (w/video build diary)

You can not imagine how long ive been wanting to write this post.  I have been contemplating a shoe/foot interface for at least 3 years.

I had ideas about what I wanted it to do.  things like wirelessly send positional and movement data.  but that was then; over 3 years ago-which might as well be another lifetime.  since then I became a sort of an open source designer and along the way, promised myself that I would stop buying things that I could make, even if I couldn’t make whatever at the time that I needed/wanted it.  in this spirit I promised myself that I would never buy another pair of shoes because surely I could design something better than shoes, right?Picture 367

better than shoes?

whats better than shoes?!  how does one even approach that question?  what IS a shoe?  it’s a strange question considering how much of a part of western society they are.  they seem to be “normal”, whatever that means.  they were something I have always had and never questioned what they were or why they were.  are they foot protection?  are they a status symbol? are they a cultural identifier? are they a magic spell meant to bind you within a a certain range of movements?  once I began asking questions about them I started getting strange answers back, in my head.

the foot is a very complex and strong appendage.  I never knew that the heel is a sort of ball-of-bone that protrudes down from the ankle area.

an equivalent concept would be having a golf-ball sized bone in the palm of your hand.  the forefoot and toes are very springy and when barefoot, one tends to eventually end up walking more on the balls of their forefoot than on their heels, scanning the ground for anything that might cause harm, before stepping down with their full weight.  and theres a whole range of pressure points that lead to all the organ in the body (reflexology). 

in a shoe the foot does not act in this manner.  shoes make the feet land on the heel first then roll toward the front of the foot.  ive come to realize that this is not good.  the entire weight of the body was not meant to land in the heel, looking at the way the foot is composed.  it was meant to be sprung between the forefoot and the heel.  so as I started to dream-design foot interfaces, I realized that it could not be a shoe.  I wanted the same freedom for my foot that I have for my hands.  I knew that I wanted something that would allow me to put my weight naturally on my forefoot and not my heel.


Picture 357this was the beginning of a 2  and a half year investigation because most shoes I would research were, well, shoes and I didn’t want a shoe.  I wanted something that allowed for full foot motion range while also being a platform for design iteration.  if I could design a relevant biomechanical form, I could iterate it over time.  and as a shoe, daily useage becomes the best laboratory for what is possible.

so at first the designs focused on being a semi soft shell for my feet. Picture 356Picture 360 the thinking was that it would be the barest possible covering , allowing my foot to act as it wanted, while having a slight tread for grip.  this immediately sucked…hard.  the toecoverings were painful as was the rest of the design.  it basically just hurt everywhere at once, so I put it away for a while and let the process keep marinating in my mind.20140620_00401420140620_001533

after a couple more iterations and some interesting insights from designing the new mask, the concept changed.  with the new exo-voice “cicada” mask, I found that interlocking small pieces of nylon allowed me to design a kind of softness from the relationships between rigid pieces, Picture 226so I designed a biomechanical “exo-foot”; a design of snap together hinged pieces that would be rigid when necessary and completely biomechanically compliant the rest of the time.

Picture 363Picture 364

the design was deceptively simple; a hinged heel that rotates freely, which was hinged to the mid-foot by a hinge underneath the ankle (because you don’t want scratchy nylon touching your ankle, trust me)., which then connected to a mid-foot hinge, which then connected to a mechanical forefoot which moved fluidly with the forefoot.

by this point, ive been designing daily for 3-4 years, so I don’t rush…I let the iteration happen mostly in my head, then once the formula works there, I draw it and once the formula works there, I CAD it, then start printing.  at any given time, ive got a dozen completely unrelated designs of various interests iterating in my mind. I decided to work on this one at this point for two reasons;

1. my shoes are fucked up and I needed something else to put on my feet and

2. the exo-voice iteration, especially since I am working to get a “proper” form out to my long-waiting contributors, burns me out after a certain level of pushing so from time to time its nice to change gears for a few days and come back to it fresh, later.

Picture 346

technically, this is draft v2.5 but the first draft that is actually wearable. this is the visual and narrative storyline of the preceeding ones.

not only is it a shoe replacement, it is now an iterable expression I call “Footware” (trust me…nike et al will crib this word as soon as they see it).  I can evolve what my footware does as easily as any other aspect of my system now and just as multidimensionally (form, sensing, robotics, light, spatial position, etc).  for instance, currently its got a sort of sandal like forefoot area, but now tht it exists, I can iterate each toe, easily over time.  its much easier to adapt something that exists than to invent it in the first place.  I can make bike pedal clips and integrate controllable lights for biking, performance, computation…whatever.  but for right now, I just want to walk around with this one exo-foot for a while an see what breaks, then print the other foot with the upgrades.

This ends the basic overview and now it gets geeky so stop here if you don’t really care about any of that stuff

the build-printing

the peculiarities of designing for the foot evolved as my interest in it evolved. for something that interfaces with the ground and holds our entire body weight, they are profoundly sensitive.  possibly a result of wearing shoes, or maybe an adaptation for barefoot humans to better scan the ground for sharp things or. more esoterically, they do more thinking on their own than we give them credit for seeing as how they are  the furthest appendage from the brain yet simultaneously the closest body part to the earth. I will return to this question in the future. 

nylon, precisely, taulman bridge nylon, was and is the primary material I use for design work. my first thoughts were to create a semi flexible shell around the feet using printed nylon components.   I find it to have unmatched strength as well as flexibility determined by the thcikness of the piece being printed which allows me to design softness and rigidity into the same component.  in the areas where a more “true” or rubber-like flexibility was required, I used a different material called eco-flex pla (just because I had a roll of it from two years ago that never seems to end and which seem to not be manufactured anymore).  this is a rubber like pla based filament from Form futura.  between these two materials I could go the full spectrum from hard and rigid to soft and pliable.  I plan to upgrade my printer in the coming few weeks to allow these elements to be design together and printed together, but currently I design them to be snapped together manually.

I took into account that the foot interfaces our shifting body weight with the ground, so has lots of very powerful micro adjustments that must be accounted for and each of those micro adjustments is distributed throughout the entire foot, while, I must state again, carrying your entire body weight.  this means that slight roughness that wouldn’t irritate the hand, will wreak havoc on the sensitive skin of the feet.

with nylon, I use a 15mm retraction setting (this pulls the nylon back so that when you move the hotend across the print it doesn’t drag a string of melted nylon across its surface).  nylon drizzles more than abs or pla but at this setting it is perfectly fine..  because I didn’t want delamination over time, I printed the layers at a very hot setting for this material.  although suggested temp is 235-250c, I find that 285c ensures that the layers will stay welded.  this is the same temp I use for the eco-flex PLA, although at a greatly reduced speed since flexible material will easily get caught in the gears if pushed too fast.

I use scotch clear glue stick to hold all prints down to the bed during printing.  I don’t use a heated bed anymore. when the voltage regulator on my heated bed dies a couple of years ago, I started using glue stick and it, in addition to a 3-5mm brim on the prints has sufficed for all materials I throw at it as long as I print the first layer slowly and, in the case of nylon, it must be dry or any moisture will cause the corners of the print to draw up as the water is evaporated. even now that I could power the heated bed with the new RAMPS board I bought last fall, I prefer the simplicity of glue stick.  simply put a layer evenly on the bed and let it dry.  seems to work everytime.

I printed all the nylon pieces first and finalized their individual forms before I started printing the padding pieces.  I designed them to be both male and female. for instance, the heel innersole has a heel-plug that is inserted into the nylon heel shell and extended to the bottom. Picture 342Picture 355 the heel outersole had a hole that this plugged snuggly into while having 3 small plugs that inserted into the bottom of the heel shell and through into 3 holes in the inner sole.  all of the padding pieces have this type of locking mechanism.  the plugs are drafted outward so they are a little wider at the top of the plug than at the base so they fit snug and wedge in even tighter when stepped on.  no glue necessary. currently the top of the foot is held together with a Y shaped connector that will most likely be deprecated as the platform evolves.Picture 354

Biomechanical concerns

there were a number of biomechanical variables that needed to be accounted for.

  • the sensitivity of the skin of the feet
  • the ability to bend the foot properly across its full range of motion
  • Durability
  • modularity
  • upgradability

as I mentioned earlier, the feet are very sensitive and the design had to take this into account even while being printed with interlocking rigid pieces.  I found that the more flexible design was much more painful because it didn’t provide support where it was need and was too tight in other places to make up for this.  so a rigid, modular form made up of interlocking smaller pieces was a better choice.  it bends where I design it to bend and it is rigid unless I specifically design it to be not rigid.this left me play with the idea of a “mechanical shoe”; something made of interlocking mechanisms, which is a vastly easier task for a reprap.  it has the possibility of being evolved mechanically as well but more on that in a bit.

there are actually 3 main sections.

  • the heel and ankle bracing system
  • the mid foot and forefoot mechanics
  • the toes area

Picture 350in this version the toes are static, meaning I didn’t do them yet for the reason cited above. the toe area is very sensitive and any experimentation done in that area must be done carefully, so that is a longer term investigation.  at present, I gave them a zone give them freedom while being wide enough to allow the toes to spread nicely when walking on the balls of my feet, which I expect to begin doing more and more often with these.

Picture 336just behind the toes is the forefoot/midfoot mechanism.there is a brace that rotates when foot goes to balance on the forefoot. this provides complete freedom for walking or dancing on the balls of the feet.  I am currently not comfortable with attempting to run in these although my walking tests have all gone very well.

the heel is the most crucial area as this is where the whole construct locks into place on the feet.  there is an ankle brace that goes across the top of the foot at a 45 degree angle with the bottom of the foot and the Achilles tendon.  the Achilles brace is necessarily soft as it sandwiches the Achilles tendon providing a stop for the foot backing up out of the construct, no matter what position the foot is in.  it also plugs into the ankle junction where the heel, ankle brace and Achilles brace converge into a 3way structure that doesn’t slip backward or forward while allowing full range foot movement by way of a rotating junction where they all interconnect.  Picture 348

this thing  was a pain in the ass to design.  early on I discovered that if you put any non-cloth-like material against the ankle, it will not be enjoyable.  anything, hard or flexible, against the ankle is a no go.  too tight across the top of the ankle area, equally fucked.  things just seemed to saw their way into the skin very easily in that region because it has such a huge range of motion.  what might be ok while simply standing, would be horrific if you attempt to walk or climb a stair, let alone dance.  by designing everything from the heel structure, forward, it gave the whole design a strong foundation. the ankle brace rotates so that the heel doesn’t have to move at all and the Achilles brace cradles the tendon assuring that the ankle brace doesn’t pop off because the design is tools free/snap together

all together, the exo-foot feels like it was designed for foot freedom. when I am barefoot, I like to walk on my tip-toes just because I find it fascinating that it is possible to easily balance so much human on such small areas…and because its fun.  in shoes it is possible but more of a chore because the shoe wants to be flat. in the exo-foot the sensation is strange.  its like being barefoot but its also obvious that you're not barefoot.  there is a little tightness in the forefoot area and the padding can get slippy, although not dangerously so.  speaking of which, the eco-flex black based soles grip like regular shoe soles and didn’t seem to erode too badly on my first walk out of the house. I walked to the store and back after printing. and the walk was extremely favorable.  not great but considering the previous version never made it to the curb, a mile round trip stroll speaks pretty highly of this first draft. Picture 355

what this means

The exo foot is currently a shoe replacement.  and as a shoe replacement it is a construct for investigating what that means. the exo-foot is biomechanically sensitive meaning that its primary function is to support the function of the foot as it is currently expressed.  after that function is to extend the range of functions that the foot is capable of.  the feet are obviously as dexterous as the hands so iterating an interface designed around their “voice” is a whole constellation of questions I don’t even know how to ask yet.

of course initially the first things that come to mind are sensors for input into my exo-voice system.  things like gyroscopes, and pressure sensors.  next might be decidedly non-musical yet still sort of cyborg functions like reflexology points and things like clipless pedal clips and programmable lights for biking at night that either don’t have to be taken off or can be taken off and snapped onto the exo-foot system easily.  but then if its programmable, it can be integrated into the sonic system .  there is no distinction between functions when self iterated. beyond that, I could see the toes becoming more expressive and robotic enhancement is a fuzzy possibility off slightly in the distance.

the exo-foot is a a Footware platform.  these words will create their own meaning over time and will evolve exponentially as the tools to iterate them evolve (2nd extruder and ultra-conductive filament ordered and on its way).  so rather than peer too far into futures, it is most useful to enjoy this plateau and concentrate on make it comfortable so I can recycle these old ass Nikes.