So I needed a part made for an experiment I was working on for my PhD. The experiment involved tracking the outflow of a dye from a tube. Since it was a coloured dye, the easiest way to track the concentrations as they flow through a tube was through optical measurements.

We decided to use a fiber optic spectrometer and to use an LED as a stable light source, and some how attach these to a cylindrical tube. The tricky part is getting the LED and optical fiber lined up and rigid so they don't move and change the intensity of the light and give a faulty reading.

I decided to come up with an idea to make a holder that would line everything up and could be glued into place to keep everything rigid. 3D printing was the perfect solution to this. A good, well thought out design saves a lot of trouble and would have to be done anyway, so the only additional step was to actually come up with .STL file from a 3D CAD program, and then save a lot on the fabrication process by having it automated using a 3D printer.

Materials Needed:
-Access to 3D CAD software
-Access to a 3D printer

The access to a 3D Printer is actually getting easier as they are decreasing in price while also increasing in quality. Thankfully, there are also Hacker Spaces filled with tinkerers and hobbyists who play with things like these and have a great amount of experience with them. I was lucky enough to have met the people at the Think|haus Hacker Space in Hamilton, Ontario (http://www.thinkhaus.org), who were an immense help.

Submitted by think|haus for the Instructables Sponsorship Program

Step 1: Designing the Holder

The first step in the design process was to figure out what the key features were:
- I needed a 0.25 inch hole through the holder so that the outlet of the flow tube could go through the optical path of the sensor and light.
- I needed a 630 um slit through one side in order to slide the fiber optic through the holder.
-> The fiber had to go through the acrylic tube to be in contact with the liquid in order to get a good signal, and we were thinking
of poxying it into the tube first. This way we can minimize the amount of surfaces the fiber could potentially scratch up against
and mess up the light coupling into the fiber. So if we put the fiber in first, and have a slit in, we can move the holder into place
and still have a good alignment.
- We needed a hollowed out center in order to fit the LED in without bumping into the acrylic tube. I am using a through hole mount
for the LED, so I needed two holes for the leads for the LED.
- I added a cut out on one side also in an afterthought to make it easier to put the LED in to holder.

Step 2: Drawing the Base Parts in Solid Edge

The first step was to make the general shape of the holder. In retrospect, I could have gone with a square, or really any shape for the outer shape, but I went with a circle both for smaller area/less cost, and also it would look better, which is really the only reason that matters.

I made the protrusion of the circle to make a 3D cylinder. I made it 20 mm in diameter so that there were no thin walls with all the cut outs so I wouldn't have to worry about the structural integrity of the part with handling. I then extruded it by 8 mm to make it a 3D shape.

After the base part is made, I then started adding the important features: First the hole for the acrylic tube. This was done by drawing a circle at 0.255 inches, slightly larger than the 0.25 inches that was needed. This was done to cover any slight enlarging done by the extruder and also to make it fit over the tube snugly, but not need to be forced on too hard.

Step 3: Adding the Features

The next step was to gouge out a circle in the center of the holder for the LED and optical fiber to have room. The easiest way to do this was to hollow out the whole circle - this saves on material, and again as this part isn't load bearing, and the walls were made thick enough.

I did this by making a plane parallel to the front face and moving it to the centre of the part. I made a circular cut in this plane and do a symmetric cut in both directions from this plane. I chose the circle diameter to be (0.255 (inner circle diameter) + 0.189 (height of the LED) +0.05 (wiggle room))x25.4 (mm/inch)

Step 4: Aligning the LED and Fiber

To make sure that the light source and detector were lined up I made two planes tangent to the outer part of the cylinder, and ensured they were 180 degrees apart, on opposite sides of the part.

On one of the tangent planes, I made a slit all the way through so it could abut up next to the acrylic tube that is going to go through the center hole. The slit was made by making a rectangle of the width of the optical fiber. I then made the cut out to extend all the way to the edge of the part, which stopped at the interior circle.

Step 5: Adding in the LED Placement

Rotating to the second plane that is aligned on the opposite side from the fiber, we are now working with the LED placement spot. To make things easier to see, I toggle the display settings to only display the edges, and not the faces. This lets me see the red line which indicates the centre line of the slit for the optical fiber so I can align the LED.

**Not pictured here, but in reality, I moved the centre of the LED to be lined up with the center of the optical fiber by moving it a distance of the radius of the optical fiber away from dead centre. I forgot to do that when I remade the part to get the pictures for this instructable.**

I measured the spacing between the LED electrodes to be 0.7 mm, so I added some lines of 0.3 mm on either side of the centre point so that the LED could slide all the way down to the resin and be set up perfectly. The square cut outs for the electrodes were made to be 0.6 mm squares, more than enough room for the leads to get in. Then I removed the lines for that I used to line up the holes and proceeded to cut out the rest of the way through the part.

Step 6: Adding Practicality

The final step I added was more to deal with the practicality of putting the LED in. Giving myself more than 0.25 inch to add in the LED but adding a cut out so it is easier to slide the LED in. I cut down just enough so that I didn't disturb the original cut out.

It made the final piece to look like a key hole, which generally means that everything is perfect because it resembles something else we all know familiarly and it was purely coincidental.

In conclusion, I know this may not be the most useful part in the world for everyone else as fiber optics aren't easily accessible, but I hope this lets everyone know how easy it is to get fairly precise pieces made with some foresight and 3D CAD skills, and access to a 3D printer. 3D printers are becoming more and more accessible as time passes due to places like ThinkHaus and other Hacker spaces. Be sure to find one in your location and see what fun things you can play around with.

I needed about 6 of these pieces, and making them by hand would easily have taken 3-4 hours each, given how well the alignment needed to be for my application, but with the resolution of the Reprap being about 0.2 mm, I was able to print out these 6 parts in 80 minutes.

Step 7: The Benefits of 3D Printing

After getting these pieces all spec'ed out and made, I had them printed at the Thinkhaus Hacker Space in Hamilton, Ontario (http://www.thinkhaus.org).It took about 60 minutes to design, 80 minutes for them to print out these pieces for me, and then the following day another 5 minutes to realize I had screwed up.

I incorrectly added the LED height to the diameter of the piece, when I should have been adding it to the radius. This made the slot for the LED half as big as it should have been with no way to get it in.

Thankfully, it was a 3D printed piece, and very small and didn't use up too much plastic and fixing it took no time at all. As I had saved the part file, all I had to do was go in and double the distance for the LED and it was fine. It took another 5 minutes to make the changes and then again get them printed out.

Also, after talking to the people at the ThinkHaus I found out I had wasted a fair bit of time hollowing out the inner ring all the way around to "save on plastic". 3D Printing software, specifically slic3r in this case, has the option to set the fill density of the part. As this wasn't a structurally significant piece; there would be minimal load on it, it would actually be easier to print accurately if it was entirely filled in except where I actually needed the slots, and to set to a 10% fill in the software. This used a lot less plastic and was easier to print since the software made the voids to save on plastic, instead of me guessing what would be easiest, and in this case, being wrong.

In short, 3D printing is really rapid prototyping. Instead of hand machining these pieces out which would have taken days, especially with my mistake and having to redo them, 3D printing easily saved me a lot of time, and also expense, as in this case the amount of plasitc used was minimal.