Oct 23, 2017

Art and Pickles

Made some art for a friend, made some pickles for me (and roommates!)
Art is chalk pastel on scrap matboard.




This one is derived from some panels of Saga, an excellent (but super NSFW) comic series.

Pickles are various kinds of fermented or refrigerator pickles.
(avoiding botulism in canned pickles is too much effort to bother with)

I cut the radishes into flowers for one of the carrot-radish jars, but the small red radishes were a little too delicate and most of the slices lost their petals in the process.





Oct 22, 2017

Calipers Box round 2 (post 1/2)

Still working on box#2! (I borked my leg and picked up a dining-table project, so this is still ongoing)
This means I have the opportunity to take more photos of the steps I missed last time.

After milling out the calipers profile, etch-a-sketch style and seating the tool in the bottom half, I traced out the remainder on the top blank. This one needed a void for the round bezel-dial and not much else.

Here I'm also matching up the ends to mark where the hinges go.


Milling out the bezel void with a ball endmill for extra aesthetics. Hinge and magnet mortises happened afterwards.


I was lazy and decided to lasercut all my inlay pieces. They are 3/32" thick and would've been a pain.







Oct 11, 2017

Mario-style jumping spider experimental rig

Hey! A lot of things have happened in my life in the past two months, so I've been pretty delayed in posts.

A little bit about what's going on in my life currently - I'm now a researcher in the Shamble Lab! We study jumping spiders and try to figure out how they work. (The PI already has a bunch of cool discoveries about these critters, check them out)

We do a lot of serious experiments, but also some silly ones. This one is a play on a map from SuperMario3 (it's World 1-1)

We were creating repeatable horizontal/vertical obstacles to measure how far spiders were willing to jump, then realized we effectively were making a side-scrolling game. So we made one!


This version was made from lasercut acrylic, with tape on the tops and sides of the platforms to provide grippy surfaces. The entire setup was mounted on 2ft tall foamcore panels, to dis-incentivize  the spiders from simply leaving the stage.

However, the spiders behaved as you would expect a 3D character to behave in a 2D world - exited as soon as they got bored.

At least we got a video.

Aug 27, 2017

2015 EC Rollercoaster Construction


It's rollercoaster season again! If you're around MIT campus this week, the 2017 coaster will be up and running (should be completed tonight, actually!)

So with that, and me being laid up in bed with a busted knee... I should hurry up and do a construction post about the 2015 rollercoaster! (only two years delayed........)

Digging holes for a cinderblock/pier block foundation
East Campus has the best back-to-school traditions. Freshman at MIT get to pick which dorms they want to live in, and each dorm has its own character it wants to show off. At East Campus?

"We're cool people! Build cool things with us!"

So that's where the tradition of building large wooden structures comes from. The upperclassmen come back at the end of summer, set down luggages, and immediately get to work. From groundbreaking (above) to wiring lights and putting together a DJ set, all time and effort is devoted to making a student-led, student-made hell of a dance party (the coaster, the fort, all of it is ostensibly for this dance party)... all within around 10 days! Then everyone admires the fruits of their efforts for around another week, and then the following weekend it all gets deconstructed before classes start.

The story of an East Campus rollercoaster, of course, begins way before construction starts. In January, about 7months before, several enterprising groups of students propose projects they want to lead. Sometimes these include a rollercoaster (~2005-2008; 2014-2015; 2017), almost always they include a wooden fort, sometimes they're mildly different. The student chairs in charge of organizing REX (the back-to-school-event), select projects to fit a budget and get to work planning logistics and fundraising. The project leaders get to work too.



There's drawings and models and acceleration simulations to be made, and a whole bunch of math. Luckily most of the structural calculations can take advantage of years of tradition with previous wooden projects. The few novel calculations for the 2015 coaster were to figure out how to safely handle a 90deg drop (safety both with regards to acceptable G-forces on the rider and structural loading on the track)

Some examples of the structural calculations
By the end of spring, the designs are done and drawings are ready to send off. Rollercoasters have to get approval from MIT EHS, a structural engineering PE, an architect, Cambridge Fire Dept, and City of Cambridge (it's classified as a temporary building). The REX chairs do a fantastic job of organizing the meetings and getting everyone convinced; over the summer they and the project leads get to sit in a ton of meetings.

Anyway once the designs and BOMs are finalized, safety plans and assembly instructions (think lego manuals) get written up, and everyone gets hyped for rush to start!


The loading tower is the first structure to go up. Its placement determines the location and orientation of the rest of the rollercoaster, and it also requires the highest concentration of manpower. 

The 2014 and 2015 towers were attached to the fort. The 2017 coaster tower is separate.


Starting to install 3rd floor spandrels

First things to go up on every floor are temporary safety railings.
Once the third storey of the tower is done, then it's time to attach all of the frames. These frames were created while the project leads focused on the tower - they're good things to delegate and use to teach freshmen how to use power tools.

A representative frame's construction sheet

Locations of each frame

Frame Assembly
While frames are going up, track pieces start getting added to the coaster skeleton. Each track section is a simple 2x4 unit assembled with a jig - another good thing to use for teaching freshmen - and slide against the previous section just like a puzzlepiece. These sections interpolate the desired track curvature onto the rollercoaster frame.

Putting track sections on the vertical drop.

Instruction sheet for a track section
The track sections are first attached to the previous section with a single screw (allows it to pivot.) Then a vertical support is attached to determine angle before the rest of the screws are added. This ensures that the rough curvature of the track follows the original design.

Once the rough curvature of the rollercoaster is established, multiple layers of plywood are tacked on to form the track. Other notable features of the rollercoaster include the work platforms that are not only useful during construction but also make safely dragging the cart back possible.


Adding vertical supports to the track units. 





And that's the coaster! There's a good construction timelapse video of the fort and rollercoaster below.

(video credit: Banti Gheneti)

You might be wondering about the cart. The cart is an ridiculously-heavy wood and aluminum-plate contraption bolted to a bucket seat (5-point harness!)

I didn't have any pictures of it from REX, but I did take some afterwards when it was in storage.


The 2015 cart has a total of 8 wheels rolling on top of the track, along with 4 wheels beneath (to prevent the cart from flying off) and 4 on the outer edge (preventing skidding). The side wheels received extra abrasion protection via heat-shrunk soda bottles (a MITERS tradition).


The cart clears the steep track curvature by having a bent frame, ~10deg total.


So what made this rollercoaster different from the previous year's? Not much, except for the vertical drop. How vertical is vertical?
Art credit: Emily Tencate
http://www.ultimaterollercoaster.com/coasters/records/wood-roller-coasters.php
We set out to beat this record and make a true vertical drop. And we did!



Technically beat a world record! (Steepest Drop, Wooden Rollercoaster)
We reached out to Guiness world records, but they didn't make it over in time for deconstruction. Oh well.


Jul 2, 2017

Calipers Box

Finished pictures first. I completed the first of my set of calipers boxes (this one was the guinea pig version) and learned a lot about RealPersonWoodworking! The wood is domestic walnut, with cedar inserts on the inside. This box is for me, to replace the incredibly mediocre stock case my calipers came with.

I got incredibly lucky with the selection of half-inch walnut boards at the store; found one with really nice figuring on one end. Of course the top side gets the prettier half. 

Finish was two coats of satin polyurethane rubbed on with paper towels, which did a surprisingly good job of preserving the grain texture. Would use again.

 Calipers! I couldn't find any reasonably-sturdy brass latches for this box thickness, so I inset magnet closures (two sets of neodymium magnets). I plan to use this in machine-shop settings with lots of steel filings around, so I covered the magnets with cedar-endgrain inlay. Filings can now be easily brushed off without getting stuck.

Here's a closeup of the cedar. I chopsawed a lump of scrap and lasercut the resulting endgrain slice (~1/16" thick). It was glued into the pocket and then made flush with finger planes and lots of sandpaper. I also have a rounded half-slot (don't know what those are actually called) for easier box opening. That was made with a router.

I missed some spots of glue (dark spots at the edges) when I was cleaning everything before putting the finish on, which makes me mildly sad. Lessons for next time.

All the pockets were freehand milled (I posted a fun video of me playing etch-a-sketch here) then cleaned up with chisels.

Here's what the rough blank looks like coming off the mill. Originally I was going to felt the bottoms and therefore didn't care about surface finish in the pocket, but I liked the plain wood more... so more chiseling/sanding than expected.

Hardware is the good stuff - brass 90deg Brusso stop hinges.

Here I'm checking fit; shortly after this I had to alter my magnet setup to use two magnets instead of one.
Finished box is smaller than the old carrying case in all dimensions, so it's much more portable. More importantly, my calipers don't rattle around in this new one.

Mitutoyo is a quality calipers company, I don't know why their cases are so inefficient.

Made a box! One down, one to go.


May 29, 2017

free-hand milling calipers pocket

In a shocking turn of events, I have a post that isn't about 2.70.

I'm in the process of making a set of calipers boxes, where I'm roughing out the pockets on the mill and then cleaning up with chisels.

Mill pocketing is done free-handed, because it's more fun. It's like playing with an etch-a-sketch, but better!


video 2x speed

May 22, 2017

[2.70] Final Testing & Documentation

Last 2.70 post woooo!
After assembling the vertical axis, all that was left was bolting on the wooden desktop and testing.

pretty computer on pretty desk
For the purposes of this class, I kept power & software fairly rudimentary. The motor is powered from a 12V supply (using alligator cables) and driven with an H-bridge/Arduino/USB setup hooked directly to my laptop. I'm just running the generic stepper library and not trying to use microstepping, so the motor is ridiculously loud and inefficient compared to its potential.

Between assembly and testing the HobbyShop staff started storing stuff on my desk...

HobbyShop staff trust my desk as a shelf.
Powersupply below. Foam cup only holds screws (no liquid don't worry)
 It goes up and down! Tuning the Arduino code to behave with a not-actually-a-stepper-driver at a reasonable speed while not causing ridiculous vibrations took some time.



I conducted some repeatability abbe-error tests with a laser pointer. The laser fixture was clamped to the desk and pointed towards the wall, 1m away (there was a brick column in the way!).

Laserpointer assembly clamped to the center of the desk.

The test shown below has the desk commanded to move between two set points 14cm apart. The laser-projected locations of the lower setpoint were marked on a piece of paper, from which positioning error can be determined from average deviation. This positioning error (unweighted) ended up being an average of 2.39mm over the 1m distance, so 0.33mm error over the 14cm travel.

This desk therefore has an unweighted positioning error of 2.4 microns per mm-travel.


An unweighted desk is an unused desk. The desk-requirements allow for an unweighted desk when moving, but it still needs to hold stuff.

I grabbed 2.5lb (1.134kg) and 10lb (4.536kg) weights and observed their effects when placed on different areas of the desk relative to the rail-ballscrew shafts. 
2.5lb weight at (-430, 420)

Experiment with 20lbs at (0, 250)

Coordinate system for desk tests
After adding HDPE skids (see vertical axis build post), I experimentally determined max yaw displacement by pushing on the corner of the desk until it hit the hardstops. These projected errors were +19.83mm and -22.24mm for a vaguely 10lbf push.

Using the same Abbe error equations as the previous linear axis testing,

$\alpha_{pitch} = \frac{\delta_y}{L}$
and
$err_{pitch} = \frac{\delta_y y}{L+y}$

$\alpha_{roll} = \frac{\delta_x}{L}$
and
$err_{roll} = \frac{\delta_x x}{L+x}$

where I'm making the approximation that vertical displacement only comes from pitch and horizontal displacement from roll since I have so few samples and since the measurements are reasonably close within sets.

Using these error calculations, I calculated pitch, roll, and yaw stiffnesses of the desk, where roll stiffness > pitch > yaw by approximately an order of magnitude each.

I also did some qualitative testing, and discovered that my system is too low-friction and too backdrivable for my motor to support loads as predicted by the error spreadsheet. That's kinda expected, given that my system uses a high-pitch ballscrew.

However, an unfortunate consequence of this is that a load of ~7lbs (laptop + 2.5lb weight) is the most this actuator can take while traveling up and down at speed, and it sounds terrible doing it. I could have tuned the system to run at a lower speed, but this is somewhat difficult to do with my software setup. Soooo... meh.


Desk happilly traveling with 20N loads, then getting upset at 30N load.

This means I can type on my laptop and put my elbows on this desk, but putting my legs on the desk backdrives the motor (no power). If power were turned on, it would likely hold more; however putting 20lb on the desk draws 3A current, at which point I start to worry about thermal dissipation in my H-bridge driver setup.


Even small dynamic loads lower the desk.

The desk holds laptop and legs in static-loading just fine
Adding a gearbox between my motorshaft and the ballscrew would help make my desk sittable, but in its current setup I doubt it would hold close to bodyweight before either my motordriver overheats or my rails break. So, no-go on the "Real Desk" functional requirement.

The desk does hold at least static-108N (laptop + 20lb weights) without failing, so it does meet the class calculation-standards with 100N loads (albeit barely; it probably wouldn't meet the expected 2x safety factor). 

Circling back to the original error predictions for a 100N load, I had expected to get 0.23mm displacement from the theoretical desk. Instead, I got 2.87 - an order of magnitude higher.
What went wrong?
Searching through the error spreadsheet, I found a problem with my model.
Linear stiffness of the carriage in the model equals bearing stiffness
I had included flexures in my carriage to account for parallelism-errors with the rails, and had discovered that the rail-shafts bend before the flexures do when I was assembling the vertical axis. And that makes sense - the rails are only 8mm in diameter and have an unsupported length of ~350mm, whereas the ball-bushings are set in a thick block of aluminum.

I had considered shaft stiffness before (in that post, actually), but at the time I was only concerned with whether deflections approached yield stress. Returning back to those calculations, and changing rotation and linear stiffnesses in the spreadsheet to be an average of ballscrew and rail shaft stiffnesses, I get some more reasonable results.

Shaft compliance calculations

new results. Note the F = kX displacement.
Reality only matches models when the models are accurate - shaft stiffness is where my order of magnitude discrepancy came from.

That's it for the 2.70 desk!

Desk being a desk.