Thursday, December 31, 2015

Happy new year! Of goals, robots, rockets and motivation.

2015 has been...a year.

I did and lost the best job I ever had. Working with FIRST and the Jean-de-brébeuf students have been, not only the closest I've been to actual engineering, both the project management aspect and the design aspect, but also the most fun I've had in a job...no matter how I often felt like a headless chicken...on fire.

I also got to meet great and fun and motivating and nerdy people by joining GAUL. I've been wanting to join for a while, since it's inception actually. But back then I thought I was too green, not knowledgeable enough about...well, anything related to engineering. Now I know it actually kind of was the same for the founders!

I've had 3 goals this year;
I failed at doing the first and the last ones...

But I did do better academically! Only marginally, but I believe I know what my problem is. I've never been actually GOOD at math. I'm at most okay...compared to other engineering students. 

I've also managed to finish a few long projects: Cleaning my workshop and building a new table, Elucidator and it's scabbard, which I had been working on for years...and other stuff, i'm sure.

This brings us to this year's goals. I've been weirdly motivated to improve myself during this holiday, beside a short 4~5 days depression.


As such, I started watching MIT opencourseware classes on youtube, restarted practicing guitar and ocarina, and cleaning up my workshop. I've also decided to set a list of goals for 2016. Goals, not resolutions, which means my list has to have quantitative things in it, and not just "be better at math durrr hurr"

So withouth further ado, here is my list of Goals for 2016:

Engineering:
  • Be better at Math
    • Review Linear Algebra, do 10 problems a week from kreyszig and cegep book
    • Review Calculus, do 10 problems a week from the kreyszig
    • Study discretization and equation-setting in mechanism
  • 1h of programming practice a week
  • Finish Re-entry Simulator
  • Review Dynamics
  • Design a High powered rocket for VIPE 2016 (July ish?)
  • Design a combat robot and an Arena

Cosplay:
  • Finish Kirito
  • Monster hunter foam build (Full armor set + weapon)
  • Spacegun (Warframe/Borderlands)
  • Lightsaber with sound effects


Life:
  • 30 minutes of Rocksmith every 2 days
  • 30 minutes of Ocarina practice every 3 days, once a week without fingering chart
  • Build 1 plane from the book “Planes A complete history” every week
  • Read a novel every 2 months


I
With that, I've also created an Excel Spreadsheet to keep track of everything that has a X/week, so I can hit myself at the end of the year for not following it!

Finally, I have a post I should be finishing about Dark repulsor (Which i've been working on a bit) and my FEA class. If my cold continues, it should be done by next week!(the post, not my class and my prop) I could have done it first, but then, new year is today, not next week.

Happy new year everyone!


Sunday, October 11, 2015

This week on Big Fred's Cosplay and Rocket Shop

ok...not really week, but who keep counts, right?

Elucidator Scabbard

After having finished Elucidator, I decided to give a try at making my own scabbard. As with any project where you make something for the first time, research is key. Thanks to our overlords at google, I found this page describing two historical, and an easy method of making scabbards.

http://www.yeoldegaffers.com/project_scabbard.asp

So, because I'm a bit lazy, I decided to go with the easy method.

First I made the rough shape.


Then I used scrap pieces to build the details. I also thinned the piece near the opening because it looked even fatter than your mom Weird Al in "Fat"


And the final result.

Now, those make it look like everything went well, but it didn't. At first, because of the thickness of the felt I lined it with, it was way toot tight, to the point where paint on the sword itself would get scrapped off. The only way I found to fix it without opening the whole thing, or sanding the whole sword ~1mm thinner, was to get a rough piece of wood, a whiff thicker than the sword was, and repeatedly insert that...implement, in the scabbard. As a result, some fiber on the felt stuck on the wood, and after a while, removed just enough for the sword to have a tight fit, without destroying the paint.


Dark Repulsor

Just a quick glimpse of what I'm working on. I'll probably be making it during the October school break, in between researching on carbon-fiber layup and CFD on rocket bodies.


Speaking of rocket bodies...


ROCKETS!

Since the FIRST robotics team I was managing doesn't exist anymore due to budgetary constraints, I decided to join a student project group once again. I was torn between Avion Cargo Laval, the  SAE Aero design group, which really need more people, Robocup, a project to make a team of robots that plays soccer, with one of my friend, or the GAUL, which I actually wanted to join since they started up again a few years ago, but thought I was too inexperienced to join.

Since it's my last year and a half of university, in theory, I decided to not help my old team, or my friend. Instead, I decided to do what I always wanted to do; what actually put me on the path to being an engineer: rocket science.

The first team activity we did was build rockets together. We had a choice to either work on a team rocket, or get our own kit. Since I like diving in the deep end on my first try, and I really like making stuff, I got my own kit. I'm cheap though, so I went with the cheaper, smaller kit, called the Screech, thinking it would be a one time thing and that it would either break, get lost, destroyed, or that i'd have my fill with that one and the competition's rocket...

The completed rocket, Thjalfi-1, pronounced Thyalfi, just like Mjollnir is pronounced Myollnir and eyjafjallajokull is pronounced eiyafyallayokul.
Boy was I wrong. We went to Ciel d'Octobre and the rocket launching community seemed just so nice compared to the RC plane community. We launched 5 rockets in total, mine and another small kit like mine, the communal DX-1, and 2 rockets made by the team's senior launchers. My launch didn't go so well: first the wind got pretty strong near the end of the day, so we angled the launch pad maybe a bit more than we should have. Then, my color choice wasn't so great. Chrome is actually hard to see against a cloud cover AND cut corn, so everyone lost sight of the rocket at it's apogee. Finaly, the parachute didn't deploy properly: the nose was probably a bit too tight. I was afraid that was going to happen, because it was really tight...but since it was my first rocket, I really didn't know how tight it had to be. In any case, that is only conjecture because we didn't find the rocket in the end; maybe the chute did deploy and we just didn't see it; maybe it deployed but was burned by the explosive charge maybe...

I didn't see any cows, that's proof right?
In any case, that was fun. It reminded me of my youth, launching model rockets in the sand quarry in my hometown. I'm not satisfied with my launch though. I'm going to build a bigger rocket for next year, and get the certifications needed to launch it.

Welp, add rocketry to my infinite list of hobbies...which really could mostly be summed up as:
My wallet ._.;

AND IN THE NEXT EPISODE! WILL ROCKET CFD BE AN ACTUAL THING? WILL FRED BE ABLE TO FINISH HIS COSPLAY ON TIME? OR WILL HE BE KILLED BY THE OVERWHELMING COMPUTER-AIDED DESIGN CLASS?? FIND OUT...SOMEDAY.



Sunday, August 23, 2015

Summer extravaganza

Things happened this summer! This was supposed to be a single post about the worktable project, but summer went by so fast, and projects so slowly, that I decided to add everything at once!

Here are the things that happened, and that we will be talking about!

Worktable Project
La Baniere and Bowmaking
A distinct lack of quadcopter and electric scooter
A new computer appears! Skylake i5-6600k
Elucidate!


Sunday, May 24, 2015

Mango Scooter

Ah, the boringness of working at minimum wage at a convenience store while, at the same time, wrapping up my job as a FIRST Robotics team manager! The nice part about it is all the money it brings!

Now, since I have absolutely no self control when it comes to buying making related stuff, especially when stuck with a boring job for a while, I ordered a bunch of stuff off Hobby King. After lurking Charles Guan's blog for years, I knew I wanted to make a silly electric powered vehicle. A kick-scooter looked like the safest (yeah, right) and cheapest way to make an exploratory project about electric vehicles, that is, a project where I start with the main part (in this case, the motor and speed controller) and make it run, instead of starting, as every engineering projects should: with a problem description and a requirement table.

For this purpose, I chose the following ICBM: Turnigy Aerodrive SK-3 6354; which I believe, if I understand right, is a close relative of Charles' Melon Motor. Since it's only a relative, and it's name clearly also has to start with M, I will now christen this motor; the Mango Motor, and as such, Mango Scooter.

Kind of looks like a Mango, right?


This motor has the following specs:


  • kv: 260 RIP'EM (RPM) per volt
  • Max Voltage: 37 Volts
  • Max Power: 2360 W (About 3.16 HP) I doubt it can actually output that much mechanical power, I don't doubt it can eat that electrical power though
  • Max Loading: 70 Amp
  • Internal Resistance: Supposedly 0.023 ohms, I kind of doubt it.

To go with it, I got a "100A" Red brick ESC, why 100A while I only got a 70 amp motor you say? Because:
  1. Safety Factor
  2. It's Cheap Chinese Stuff so it's probably gonna explode as soon as I put current through it.
  3. It's entirely made of MOSFETs (Whatever those are) And (almost) no heat dissipation.
Literally a bick of MOSFETs

...Thinking about it, I should probably have went for 150 or something.

Let's see how fast that could go...suppose a 10cm diameter wheel and direct drive. 10 cm is nice and round.

At 35 Volt, because a 37 volt battery would be at 37 volt only for the first few minutes anyway, the motor spins at 9100 RPM. Thanks to the 10 cm diameter wheel, that's 285 740 cm/min or 171.44 km/hour.

Hey, that's nice! I don't need to go that fast though!...and the motor can't even start if I leave it like that anyway. The red brick controller is a sensor-less controller, as such, it can't know where the motor poles actually are, and which poles needs to be switched. What it means is that the motor just won't work if the inertia it sees is too big, or if the motor is going really slowly. In both cases, we need reduction, so that high speed is actually good.

Here's some interesting reduction cases:

  • a 9x reduction in 2 3x stages reduction gives a nice 19km/h
  • a 12x reduction with a 3x and a 4x stage gives 14.25 km/h
  • 16x reduction with 2 4x stages is 11 km/h
Now, the question is, which is the lowest reduction I can get that will still let the motor start. I will need to test in order to answer this...which will require me to get a bigger iron, because my little 50 watt sparkfun hakko-copy iron just doesn't have the thermal weight to solder those beefy connectors.

Now, obviously I didn't just buy a mango off Hobby king. I always wanted to make a quadcopter, and try my hand at writing the control code for it myself, so, I bought 5 itty bitty motors too! (along with their ESC, obviously)

Here's one of them next to their big brother Mango:

Thinkpad for scale?


So that will be fun. Next week, I will be introducing one more project i've been working on.

Sunday, May 17, 2015

Re-entry simulator - Part 1

Ages ago, I've had to make a launch simulator for a multi-staged rocket. The goal was to find the efficiency of multi-stageg rocket compared to single staged ones. That simulator was pretty simplistic; 1D (altitude only), no easy way to change initial conditions, lift was neglected, and, as it was only 2D, there was no way to change orientation or go into orbit.

Now, for my space sciences class (Phy-2100), I have a homework that requires me to calculate the re-entry window of an Apollo capsule so that astronauts don't feel more than 6,5G. Now, I had 2 ways to do it: either calculate a few values by hand and easilly find the maxima and minimum, or do it the Kerbal way; Trying every possibilities and see which sticks. So I took my old Simulink program, and used that as an excuse to fix the problems with it.

First was making it simpler to change initial conditions. That was pretty straightforward; I simply changed some values to be inputs from the workspace. The initial conditions I added are:
  • Initial speed (Absolute value and angle)
  • Initial weight of propelant
  • Initial altitude

Next was turning my 1D sim in 2D...now that was an headache. I'm not great at setting up motion equations, so the following took a few textbooks; along with wikipedia, to set up and make sure I understood. Movements, especially relative motion, and in polar coordinates to boot, are a pain.

First was setting up my hypotheses:
  • Coordinates are geocentric.
  • The earth surface is a round shell at mean earth radius inside of which relative speed is forced to 0. That shell rotates at 1 turn per 23 hours and 56 minutes.
The equations of movement for polar coordinates are the following (From Wikipedia, because it's easier to copy paste than trying to do rdotdot in blogger):

 \mathbf{r} = (x, \ y ) = r (\cos \varphi ,\ \sin \varphi) = r \hat{\mathbf{r}}\ ,
 \dot {\mathbf r} = (\dot x, \ \dot y ) = \dot r (\cos \varphi ,\ \sin \varphi) + r \dot \varphi (-\sin \varphi ,\ \cos \varphi) = \dot r \hat {\mathbf r} + r \dot \varphi \hat {\boldsymbol{\varphi}} \ ,
 \ddot {\mathbf r } = (\ddot x, \ \ddot y ) = \ddot r (\cos \varphi ,\ \sin \varphi) + 2\dot r \dot \varphi (-\sin \varphi ,\ \cos \varphi) +  r\ddot \varphi (-\sin \varphi ,\ \cos \varphi) - r {\dot \varphi }^2 (\cos \varphi ,\ \sin \varphi)\ =   \left( \ddot r - r\dot\varphi^2 \right) \hat{\mathbf r} + \left( r\ddot\varphi + 2\dot r \dot\varphi \right) \hat{\boldsymbol{\varphi}} \  = (\ddot r - r\dot\varphi^2)\hat{\mathbf{r}} + \frac{1}{r}\; \frac{d}{dt} \left(r^2\dot\varphi\right) \hat{\boldsymbol{\varphi}}

And as such, our forces equilibrium is the following (Again, from wikipedia):

F_r = m \ddot r -mr \dot {\varphi}^2 \
F_{\varphi} = mr \ddot \varphi +2m \dot r \dot {\varphi} \ .

Now, in order to make my things simpler, I need to isolate the acceleration terms.

Now comes the question: what are my forces? and what information do I need to calculate them?
The forces I've chosen to model are the following:
  • Drag (Dependent on speed, atmospheric conditions, and geometry)
  • Lift (Dependent on the same things as drag)
  • Thrust (Depends on ISP, Chamber pressure, angle and atmospheric conditions)
  • Gravity (Depends on altitude)
Just for the sake of completeness, here are the things I neglected/assumed:
  • The pilot is not Jebediah Kerman
  • For this particular case, Drag and lift are hardcoded, and are only dependent on altitude. They are based on this paper:
  • Trans-sonic effects (Additionnal drag between mach 0.8 and 1.2) are neglected
  • Capsule dynamics (Roll, pitch, yaw and such) are neglected. Capsule is assumed stable at a 25 degrees flight angle.
  • Flow is supposed incompressible below mach 1. Past mach 1, the drag coefficient depends solely on altitude.
  • No control whatsoever (although this could easilly be added later, along with the capsule/rocket yaw pitch and roll)
  • The atmosphere is the standard atmosphere, but it also has a speed, which depends on the altitude. It is 0 at 100km, and the same speed as the earth's shell at sea level. The distribution is assumed linear for now. (Hint: it's not, but that's easier to estimate)
Allright, now that everything is set-up, time to code my life away!

For now I got the orbit part mostly functionning, except for a weird "bug" where my orbit precesses. It shouldn't. I'm guessing it might have to do with rounding errors...time to debug!

EDIT on may 19th: Since it's been less than a day from my previous post, I thought I'd just edit it. I found my bug. Gravity was acting weird, because I had re-used my gravity calculator from my first launch simulator, I had added an "arbitrary" Earth Radius constant. As such, my gravity force was getting lower and lower and lower, due to it adding an earth radius every time! Which explains the precession.

Here's how it looks so far

It works, it's a circle tracer!

























Monday, April 20, 2015

FIRST!

Allright, after procrastinating and reading every blogs made by people from MITERS (The MIT student maker...thing), and especially Charles Guan's etotheiplusone, I've decided to make my own blog, with blackjack and hookers!

Ok, maybe not blackjack and hooker...maybe just silly builds I do from time to time along with design projects and...stuff.

I'm hoping to, like Charle's blog did for myself, spread the fun that is making silly things! Or just things, they don't really have to be silly do they? I'm also hoping to use this as a sort of portfolio for future employers.

Anyway, here's a list of projects i'm hoping to do in the near (or not so near) future:

6x6x6 LED cube
Use techniques from engineering in order to maximise matertial removal for bow making.
Modify a MOMUS CNC design to better fit my needs and then build it.
Make a combat bot for DragonCon (not this year)
Cosplay Stuff
Silly electric vehicle club