Edison Lamp Station

IMG_0981

My initial idea for a lamp included a light switch, a rubber mat to rest your phone on, a metal post for sunglasses, and a pocketed area for change.

Once I started drafting this idea in Vectorworks, and became aware of the size limitations of the 4-Axis, I realized I needed to rework my idea. I knew I wanted something functional for my bedside table, and I wanted to incorporate my skill builder into my final. (For the skill builder I made a small charging dock for an iPhone.)

Vectorworks prototypes:

Screen Shot 2018-04-27 at 11.26.02

 

Screen Shot 2018-04-27 at 11.12.24

Screen Shot 2018-04-27 at 11.06.23

Screen Shot 2018-04-27 at 11.04.09

Screen Shot 2018-04-26 at 17.27.50

Screen Shot 2018-04-26 at 17.02.49

2018-04-27 17.40.02

Once the cherry wood arrived in the mail it became easier to see the dimensions of everything I wanted to mill.

2018-04-29 10.12.42

Because I wanted to maximize the amount of wood I could use I needed to add tabs in Vectorworks to allow the machine to take off the minimal amount on the front and back. This also prompted me to create a third roughing and finishing tool path to ensure the front and back would be milled.

2018-04-29 11.51.52

Unfortunately, the drill bit was not as long as I needed it to be and at one point the callet, collet, caullet however you spell that, ran into the chuck. Oops!

2018-04-29 13.32.00

After restarting the machine due to that error, I was pleased to find I could restart the job without losing all my axies! I edited the tool paths to not mill on either side of the piece.

2018-04-29 14.03.23

Thankfully, the bit was just long enough to be able to mill down to the center on each side.

2018-04-29 15.37.05

Once the 4-Axis started milling the back I saw that the bit wasn’t in fact long enough make the hole for the charger.

2018-04-29 15.42.04

I was a little baffled as to why the machine left such rough lines on the front near the back, but I thought the final tool path would correct that.

2018-04-29 16.12.39

It didn’t unfortunately.

2018-04-29 16.17.44

It left some pretty nasty grooves on the top.

2018-04-29 16.29.12

I thought that maybe if I ran the last finishing step one more time it would get rid of the grooves on the surface.

2018-04-29 16.29.23

So I deleted all the other paths and highlighted the top of the surface where I wanted the machine to mill.

2018-04-29 17.04.07

After milling air for half an hour it finally started to smooth out where I wanted it to.

2018-04-29 17.43.44

But to my dismay, it didn’t smooth it all that much.

2018-04-29 17.43.48

After taking it off the 4-Axis and cutting off the extra wood on the sides using the bandsaw, I was left with some sanding to do. But first I wanted to drill a hole where the bit didn’t go through.

2018-04-29 17.54.12

Hastily, and after 7 hours milling, I slightly damaged the wood around the hole which was frustrating.

2018-04-29 18.03.22

I tried the drill press but it didn’t do any good, so I started filing the area down to make the hole big enough to fit a charger cable.

2018-04-30 16.36.32

Then I got to sanding the sides using a miter guide which made it easy to create an even surface area.

2018-04-30 16.38.16

The finished result looked great!

2018-05-01 14.33.03

Then it was time to discuss how to fit the lamp parts into everything before I lathed the remaining cherry wood bits.

2018-05-01 14.36.00

Ben suggested making a small lamp post and affixing it on the lathe with the chuck and then using the drill to create a hole large enough to fit the guided nipple post — is that what it’s called?

2018-05-02 09.47.40

The cherry wood pieces.

2018-05-02 10.08.58

The failed drill press attempt.

2018-05-02 10.25.06

The cherry wood prepared for lathing.

2018-05-02 10.36.17

2018-05-02 10.43.20

2018-05-02 10.52.59

I needed to make a .7 diameter, 1/4″ deep part at the bottom of my lamp post in order to securely glue it into the 4-axis part.

2018-05-02 11.17.45

Once I had the correct size for that part I started shaping the rounded wood.

2018-05-02 11.25.01

Using three grits of sandpaper, I sanded it until it was really smooth.

2018-05-02 11.28.13

And used the drill to easily make a centered hole through the middle.

2018-05-02 11.36.44

2018-05-02 11.40.40

After band-sawing off the ends, it was ready to be glued!

2018-05-02 11.55.34

Tightly clamped.

2018-05-02 12.00.19

Now it’s time to wait 8 hours.

2018-05-02 15.31.35

Since I had all this time on my hands to wait, I wanted to utilize one more machine, the Othermill.

2018-05-02 15.50.16

Out of aluminum, I wanted to make a lightning bolt.

2018-05-02 15.53.40

It took me a while to remember that the illustrator file had to be filled in in order to show up in Bantam. While the Othermill was doing it’s job I realized a crucial mistake.

2018-05-02 16.28.13

My charger cable wouldn’t fit!

2018-05-02 16.28.18

So I stared at it for a long time thinking of what I could do to remedy the situation.

2018-05-02 16.47.52

Every now and then glancing over and stopping the milling process to vacuum up the pieces!

2018-05-02 17.06.31

Thankfully Ben had a solution for me. And we carefully drilled a round hole in the bottom so I could thread the cable through.

2018-05-02 17.09.08

The lightning bolt came out nicely, and just required a little scraping off the ends to make it smooth.

2018-05-02 17.23.17

Ben was also kind enough to teach me how to properly wire the lamp so I wouldn’t die turning it on.

Screen Shot 2018-05-02 at 17.35.28

Although the two marked areas in red would be rounded, I think the bolt came out nicely.

Screen Shot 2018-05-02 at 12.46.27

The original Illustrator file without it being properly filled in.

Screen Shot 2018-04-29 at 10.00.30

Screen Shot 2018-04-29 at 09.34.34

Screen Shot 2018-04-28 at 15.38.30

Screen Shot 2018-04-28 at 15.36.52

Screen Shot 2018-04-28 at 15.26.09

I realized something wonky that took me a while to sort out when subtracting pieces from the main block in Vectorworks. I needed to use the arrows to have the part I wanted subtracted to be highlighted in yellow and the rest to be highlighted in red or else everything around it would be subtracted.

Screen Shot 2018-04-28 at 15.26.03

What it shouldn’t look like, but what it defaulted to.

Screen Shot 2018-04-28 at 15.05.10

The error I got when improperly trying to subtract the whole shape from the tiny part.

Screen Shot 2018-04-28 at 15.02.13

2018-05-03 00.03.50

After waxing on the lathe and by hand in the pocketed areas.

2018-05-03 00.03.53

2018-05-03 00.20.08

2018-05-03 11.10.25

Felt on the bottom.

2018-05-03 11.05.29

And it’s done!

A Bookshelf

Now that I’ve spent about 7 hours using the CNC machine I’m getting pretty comfortable with it. I’ve learned that the machine definitely is a little crooked because of the flooring, and that the wood I used was bent toward the center resulting in pieces that didn’t breakthrough. Although Mastercad is an ugly interface, I’m used to the work flow now, and I barely have to go back to my Vectorworks file to edit before milling. I definitely didn’t expect each piece to take about an hour to mill. And if I didn’t get sick over break, I would have had enough time to finish this before class. I drastically underestimated the time I would need on the CNC machine, and having everyone using it around the same time didn’t help.

PROCESS:

First, I sketched out each piece on paper to visualize how much wood I’ll need.

2018-03-21 14.38.18
2018-03-21 14.38.37

Then I created 3 Vectorworks files, one for the back, one for the two sides, and one for the top and bottom. As you can see below, I didn’t add any pockets on the top and bottom of the side parts, which I realized was a mistake later.

In order to get the dimensions right, I made a tiny model out of paper with the measurements to see how the parts connected.

Bookshelf
Bookshelf

Once I got all my measurements in order I went to Mastercad and started with the pockets for each file and then added contouring. Because I used a single rabbit joint, I set the pockets to be half of the width of the plywood which was .725, so my pockets were set to .363 (that’s with .5 added tolerance). I also double checked that the settings were on Absolute.

Bookshelf

My plywood was cut to about the size of the bed (4 x 4) and screwed into place with 5 screws.

Bookshelf

I realized there was a slight bend in the middle of the wood. Unfortunately, no screws could fix this issue.

Bookshelf

I loaded the “back” piece and after pressing start I realized that I didn’t zero the mill far enough to the left on the X axis. The mill went off the side of the wood, so I loaded the side pieces and re-purposed that piece of plywood for those two parts.

Bookshelf
Bookshelf

Once the side pieces finished I started milling the top and bottom parts. Something strange happened during this process. The piece of plywood was also bent in the middle and this caused the mill to miss the middle of the pocket during the first couple of runs.

Bookshelf
Bookshelf

This surprised me because half way through the process, it started milling the middle in what seemed like an even line all the way from one side to the other!

Bookshelf
Bookshelf

The finished top piece didn’t go all the way through on the other side, but looked pretty good from the front.

Bookshelf

The bottom piece milled the same way the top did, leaving out the middle until creating a heavy line midway through the process.

Bookshelf
Bookshelf

The last piece I milled was the back. I made sure to zero the machine all almost all the way to the left on the X axis. To my dismay, and because of the wood being bent in the middle, the CNC started milling the tab in the middle upper portion of the wood and scraped the middle of it on it’s way over there. I stopped the process and started again, zeroing the Z axis a little higher, but that wasn’t enough either and I ended up with two marks in the middle of my nice piece of wood.

Bookshelf

Eventually, I had to bring the Z axis much farther up so the bit could pass the middle bent section without hitting it. That resulted in my entire piece not breaking through to the other side.

Bookshelf
Bookshelf

I had to do something about the fact that most of my pieces didn’t breakthrough so I grabbed a hand tool to take off the edges of some of my pieces.

Bookshelf

That left me with quite a bit of sanding to do. I made sure to not sand down the pockets like I did when I practiced making my joints.

Bookshelf

The sanded pieces looked pretty good after a while, aside from the back piece which had the two track marks in it.

Bookshelf
Bookshelf

Then it was time to glue the pieces together. I started with the top and bottom and glued it to the back with 4 90 degree clamps.

Bookshelf

Once I had both sides clamped there wasn’t much more I could do because the CNC was booked until the next day when I had 3 more hours in the evening to finish the shelves and two doors. That was around the time I got a text from NYU saying they would be closed the next day due to a snowstorm. I left the top, bottom, and back pieces for the snowday clamped together for a strong hold.

Jointery – 2nd Try

While thinking about what I want to ultimately do for my midterm, I realized that it would be a good idea to try and perfect the joint I’ll be using in order to save time. A double rabbet joint should suffice for the credenza I’ll be adapting from this work plan.

Screen Shot 2018-03-08 at 10.46.17 AM

I ran into a few problems when attempting this type of joint. First, I created two identically measured pieces in Vectorworks and brought it into Mastercam and then to the CNC machine. I laid down the wood only to realize once I had everything set and I was about to press “start” that I had aligned the wood improperly on the bed.

2018-03-06 16.55.09
2018-03-06 16.59.17
I needed to set it up so that the blank area of the wood was along the X axis not the Y. So I re-positioned it and then I was all set to go.
2018-03-06 17.04.18
2018-03-06 17.05.55

After about 12 minutes here’s what I ended up with:

2018-03-06 17.29.58

Not a great fit, I know, but it allowed me to see where I needed to subtract from each piece.
2018-03-06 17.49.45

I went back to Vectorworks and Mastercam and made my adjustments.
2018-03-06 16.37.21

By the end of that run cycle I ended up with two pieces that fit together almost perfectly!

2018-03-06 18.28.44

When I looked at the two pieces fitting together very closely I realized that in Vectorworks I didn’t round the corners on the inside of the pockets which could be why they weren’t sitting entirely flush.

2018-03-06 18.38.11

Nevertheless, I was very satisfied with the progress I had made and what I learned from taking the time to really get this joint down. I’d like to know how I can figure out the math behind this without having to make two identical parts first.

2018-03-06 18.28.52

Jig & Coin Mill

My goal for this assignment was to engrave an “S” into a coin. At first I thought I might want to carve my own coin from aluminum but once I realized that I couldn’t get the 90 degree angel’s I wanted from an 1/8th inch bit I thought it may be cool to carve into a Euro instead. I chose an “S” for my wife’s name Sadie.

To begin, I fit my 1/8″ bit to the Othermill and then had to remember to download the Delrin custom tool library for brass because Euro’s are made of a combination of brass and nickel in the middle of the coin where I’ll be engraving.

Once I put together the file for the Delrin rig in Vectorworks I couldn’t figure out how to export it as an .svg file in order to import it to Bantam. I asked around and realized that I needed to bring the file into Illustrator in order to re-save it….and then it took me a good 15 minutes to figure out what to export it as. Turns out you need to export it to an .eps file in order for Illustrator to read it.

Once I did that it wasn’t entirely obvious whether I should keep the art board that automatically came with the .eps file once I imported it. After importing the .svg file from Illustrator I did not get what I expected in Bantam. After deleting the square I made outside the circle and adding a bit of a stroke to the circle constraining the art board to the size of the circle, I was able to get things looking the way they should.

I set the advanced properties on Bantam Tools to cut in the inside because I feared that leaving the settings as they were on default would cut the circle larger than I wanted.

I’m having a lot of trouble getting the “S” to show up in the Bantam program. It seems like it’s only taking the art board because I’m getting an outline of the board but not the “S”. The problem was that it was saving as a text file which wasn’t being read properly by Bantam so I had to rasterize the S to make it an image in Illustrator. I resized the image and set the engraving so it would engrave the middle.

Of course I took the jig off the bed which entirely ruined the point of making a jig so I quickly put it back and made another circle on the other side of the Delrin so I could make a precision “S” in the coin. I’m pretty happy with the results although I would have liked it better if it didn’t make a crazy noise and I had to stop it 3/4 of the way through the process once I stopped it to clean the debris.

I tried making it again with another 2 Euro but it immediately sounded bad once the bit hit the material so I halted production.

I really like using¬† the Othermill and could’ve spent many more hours experimenting with it if I didn’t have to share the machine!

 

CNC Milling a 2 Euro Coin from Marco Wylie on Vimeo.