Midterm Project by Cosette and Eric

As we are both Space Studies students, we wanted to make a space-themed project. In line with this, we decided to use a planetary gear mechanism (similar to mechanism #55) that would orbit multiple objects around a central point.

We created a beginning schematic involving 3 “planet” gears rotating around a central larger “sun” gear within an encompassing “ring” gear. The system would be driven by the sun gear by rotating it with a handle that is shaped like a streaking star. The handle would need to be above everything else in the mechanism. We would what we’ll call a “fidget spinner” piece below the gears that will attach to all 4 internal gears to help them maintain their relative positions. Each planet gear will also have an object mounted to each of them using a dowel rod: a Rice space owl, a James Webb Space Telescope (JWST), and a space boba. There will also be a base plate (a full circle) that will help contain everything with the ring gear.

We wanted our objects to spin with the gears instead of just moving along with them. So that means we would need to place ball bearings in the fidget spinner below them, while fixing the rods to the gears themselves. Similarly, to make the sun gear the driving gear it must be afixed to its dowel rod and spin freely (not directly connected) relative to all the other moving parts. Therefore, the central dowel rod will go through the gear, the fidget spinner, and spin within a bearing in the base plate.

We started out by modeling our gears in geargenerator.com, adjusting it to what seemed like an appropriate size for us. We exported the gear files a placed a 0.5″ diameter hole in the center of each of the internal gears for the 0.5″ diameter dowels we planned to use that will mount our objects to the gears. We decided to use 0.5″ dowels as they seemed much sturdier than the 1/4″ dowels while not being too big.

Using the diameters of the gears (78 mm and 108 mm), we made a fidget spinner in which each leg of the spinner had a hole that was 27.94 mm in diameter (the measured outer diameter of the bearings) and 136 mm from the center of the spinner (sum of the internal gear diameters). The farthest distance of one end of the spinner to the center would be less than the radius of the ring gear (132 mm), so we made it 123 mm as it gave the spinner a nice shape. We also assumed that the fidget spinner would be rotationally symmetric.

Since the bearings were relatively thick, and the internal portion of the mechanism would have the fidget spinner, the gears, and any required spacing in between, we realized that the ring gear would need to be thicker than the internal portion of the mechanism. So for our first low-fidelity prototype we printed 3 ring gears that would be layered together, and one printout of all the other components. We printed the components using the laser cutter from cardboard.

The dowels were slightly larger than the internal diameter of the bearings. Initially, it was planned to insert a dowel into a drill and hold sandpaper against it to uniformly sand it down. Unfortunately, none of the drills were large enough to hold the dowel. Alternatively, we clamped the dowel rods and used a dremel with a sanding belt to sand down the diameters of the dowels until they press-fit into the bearings.

For the low fidelity prototype, we used tape and hot glue to attach the parts together. This way we could also reuse the bearings and dowels for our final prototype.

We had a few realizations while creating this first prototype. Shrinking the size of the fidget spinner (while keeping the holes and their spacings the same) helped the mechanism spin better. The outer edges of the fidget spinner were too large and would hit against the teeth of the ring gear. We had not correctly for the interference of the teeth of the gear, so we shortened the total length from 123 mm to 113 mm. Additionally, the central hole of the spinner, where the dowel is placed, was not centered properly and needed to be fixed. We also noticed there could be potential issues with the gears flexing if they were just made out of one layer, even once we switch from cardboard to wood, so the internal gears would have to be double-layered. The order in which we assembled the mechanism was best when assembled in the following order: ring gear, base plate, fidget spinner, sun gear, planet gears. Also, our components were fairly large so we would have to plan out our prints if we wanted corresponding items to be printed from the same sheet of wood to help precision and uniformity (kerf, thickness, etc.). If we wanted our ring gears to be cut from the same piece of wood, they would need to be on their own sheet. All the internal gears would also need a decent amount of space on their own sheet.

There was a bit of flexing in the ring gear when trying to rotate the gears, but we attributed that to the warping the cardboard had already undergone from testing at that point and the misalignment of the central dowel hole. This would become a harmful assumption.

Another note is that it was difficult to book time for the OEDK machines and it was only going to get more difficult going forward. Many times we would need to ask classmates if we could quickly use the machines in between their cuts. This also meant that sometimes we would need to find alternative solutions to issues that did not involve cutting out new parts.

After making the proper adjustments to the fidget spinner design, we then began printing all our structural/moving components out of wood. We measured the kerf for each sheet of wood that would have a component with a bearing or dowel hole so we could adjust the hole sizes to be press fit. One exception was the fidget spinner dowel hole, we didn’t want that to be tight against the central dowel since it was supposed to turn with the planetary gears.

Once the printing process was completed we assembled our gears by glueing them together with wood glue. We initially followed the instructions on the bottle, applying glue to both surfaces that will be adhered together, and spreading the glue out evenly with a brush. However, after doing this with the ring gear we noticed that a lot of the glue would overflow and spill over the edges when the pieces were pressed together. After a little testing we found that the glue stuck just as well without overflowing if we just applied it to one surface. We would clamp our components together while waiting for them to dry.

In addition, the dowels were also cut down to length using a hand saw. Then the rough ends were sanded down using a router, a file, and then sandpaper. If the end was especially rough (a not clean cut), a sanding dremel was used.

When trying to put all the gears together with the bearings, we noticed that we could not put the third planet gear in as the teeth would not align when placed with the fidget spinner. After looking in GearGenerator’s simulated result again, we saw that the planetary gears were not evenly spaced apart, but that two gears were actually slightly closer to each other than they were to the third gear. We calculated the angle separation between each gear based on the number of teeth from the ring gear that were in between the gears. The fidget spinner was then remade with these new angles and refitted.

However, there was still a decent amount of resistance when trying to operate the mechanism. The planetary gears would often be pressed very tightly against the ring gear. This was likely due to the gears being designed to fit perfectly together while we have some tolerances with the cuts, the assembly, and gluing the layers together. Some of the teeth of the ring gear layers were slightly offset from each other. This offset was not consistent in any specific direction or layer so this could be due to a combination of slight bends/warps in the wood and precision of the laser cutter over a large shape. A dremel with a small sandbelt was used on the insides of the ring gear to help fix this issue.

The planetary gears were also shrunk to reduce the tightness of the gear spacing. We tried a few different sizes from 95%-98% of the original size. We printed one of each size to perform a simple fit check and decided on using 96% of the original size. The new gears did make a marked improvement in the movement of the mechanism.

The tightness of the gears also would cause the gears to get pinched upwards. This would slant the gears and cause them to move and fit even worse in the mechanism. To help mitigate against this, we placed a spare of our fidget spinner on top of the gears to help block them from shifting too far upward. The fidget spinner would not be fixed to anything so it wouldn’t interfere too much with movement, and would be held down by the name plate that would be fixed above it to the central dowel.

Next were all the non-structural pieces. In actuality, we did this at the same time we were working on the structural components. All itemsFor the aesthetic items that would be moved by the gears we had three items planned as mentioned originally. We created a model of the James Webb Space Telescope that included the telescope base and the mirror array. The mirror array would be the shape of the JWST but with an additional tab that would fit into a slot cut into the base. The JWST would be painted to look metallic except for the front which would have a vinyl sticker outlining each mirror of the array. The second object, a boba space-man, would use the same laser cut box method we learned in class. A press-fit box creates a rectangular shape, similar to a cup of boba tea, with a hole on the bottom to insert the dowel and a smaller hole on top to insert a smaller dowel representing the straw. Vinyl was used to create the face and boba pearls, as well as the space helmet. The color orange represents Thai tea, and with the black helmet it also represents the colors of Axiom’s space suit demonstration (where Eric interned at). The third item is a cutout of Rice’s owl logo, also with a space helmet for safety, painted Rice’s blue, flying on a cosmic surfboard.

For the metal part, we created a handle with a star shape at the end. This was cut out using the water jet cutter just like in the metal cutting homework. Then the edges were smoothed out with a file and wire brush. Two layers were stacked together to create a more comfortable grip as the sheet aluminum was fairly thin.

All the components were sanded and assembled together. Bearings and dowels press-fit to their holes, gears assembled, and all items placed. Glue was placed around each press-fit hole so the connection would remain strong longer even if the holes started to wear down.

After putting everything together, we noticed that the mechanism had difficulty moving again. After closer inspection it seemed that the base fidget spinner, which is lifted slightly above the base plate, had started to bend from the weight of everything on it. This in turn caused the gears to slant and thus not be positioned properly. This fidget spinner was meant to assist the gears in moving around the sun gear, but has since caused us more trouble than anything else. The reason why the fidget spinner was lifted slightly was because the central bearing in the base plate was slightly thicker than the plate itself. This meant that when the bearing was flush against the bottom of the plate, it would stick out from the other side, lifting the spinner from the middle and allowing the spinner ends to droop down. In response to this, the central bearing was pressed back down to be flush on the inside so the spinner could be pressed flush against the base and be more level, even if it meant the bearing would stick out from the bottom. We also tried manually lifting the planetary gears back up but unfortunately it seemed the spinner had already normalized itself into the bent state.

Slideshow: https://docs.google.com/presentation/d/1V0ZO6Rw2LVDoMJ6pKkvZTtCq_IK79hQN9rHWDGmYtgk/edit?usp=sharing

Cost type	Cost	Price	Source	Quantity	Total
Materials	Plywood     Vernier   Wood Glue   Metal Glue   Metal aluminum sheet 11”x 8”   Spray Paint   File kit     Metal Brush   Bearings   Dowle	$7.28 for 1/4 in. x 2 ft. x 2 ft   $35.97     $8.46     $10       $10.23     $5.99   $8     $7.49 for 9 pcs, so It’s $0.83   $2.49   $0.88	Home Depot       Home Depot     Amazon   Amazon     Inchofmetal         Amazon     Amazon     Amazon     Zoro   Home Depot	2   1     1     1   1     1         1       1   4     1	$111.54
Labor	Woodworking Operator	$ 16.58 per hour	https://www.bls.gov/oes/current/ oes517042.htm	7 hours	$116.06
	Prototyping Engineer	$55.93 per hour	https://www.zippia.com/prototype-engineer-manager-jobs/salary/	7 hours	$391.51
Overhead	Facility Cost (Machine Time)	$ 30 per hour   $ 75 for 3 hrs	https://www.techniwaterjet.com/faqs/   https://apps.txrxlabs.org/classes/6/laser-cutting-i/	1 hours   6 hours	$180
	Quality Control	$20.72	https://www.indeed.com/career/quality-control-inspector/salaries/Houston–TX	7 hours	$145.04
Design	Engineering and Development	$44.81 per hour	https://www.zippia.com/development-engineer-jobs/salary/	7 hours	$313.67
	Iterations	No iterations in this case
Misc.	Waste and Scrap	10% of the material   Metal 1.023   Wood 1.45			$2.473
				Total	$1260.29

 

Links

Wood

https://www.homedepot.com/p/1-4-in-x-2-ft-x-2-ft-B-C-Sanded-Plywood-Project-Panel-00101/205748580

 

Vernier

https://www.homedepot.com/p/Husky-6-in-3-Mode-Digital-Fractional-Caliper-1467H/206007130

 

Wood glue

www.amazon.com/Elmers-Products-E7000-Carpenters-Multicolor/dp/B001N7X1UA/ref=sr_1_8?crid=3540Y48ESUA34&keywords=wood%2Bglue&qid=1697558574&sprefix=wood%2B%2Caps%2C114&sr=8-8&th=1

Metal Glue

www.amazon.com/YUE-bonding-Material-Instant-Stainless-Aluminum/dp/B09NCCLCPR/ref=sr_1_1_sspa?crid=1HE3LU2QEPAOK&keywords=glue+for+metal&qid=1697558651&sprefix=glue+for+metal%2Caps%2C94&sr=8-1-spons&sp_csd=d2lkZ2V0TmFtZT1zcF9hdGY&psc=1

 

Metal

Aluminum Sheet 0.063 (Grade 5052)

 

Spray paint

www.amazon.com/Krylon-K05545007-COLORmaxx-Spray-Aerosol/dp/B07LFWZ84M/ref=sr_1_2?keywords=spray%2Bpaint&qid=1697558447&sr=8-2&th=1

 

File set

www.amazon.com/Hardened-Strength-Warding-Triangular-Half-Round/dp/B07WDCBJ1G/ref=sr_1_3_sspa?crid=VOHGV5J9A0M0&keywords=metal%2Bfile%2Bcleaning%2Bbrush&qid=1697006092&s=hi&sprefix=metal%2Bbrush%2Bfor%2B%2Ctools%2C94&sr=1-3-spons&sp_csd=d2lkZ2V0TmFtZT1zcF9hdGY&smid=AL3I8LHZ8J0CO&th=1

 

 

Mini Stainless Steel Brush

www.amazon.com/Brushes-Stainless-Scratch-ToothBrush-Cleaning/dp/B07P6HYVXW/ref=sr_1_3?crid=199VRBIEXHLIJ&keywords=mini%2Bwire%2Bmetal%2Bbrush%2Bstainless%2Bsteel&qid=1697006278&s=hi&sprefix=mini%2Bwire%2Bmetal%2Bbrush%2Bstainless%2Bsteel%2Ctools%2C96&sr=1-3&th=1

Bearings

https://www.zoro.com/tritan-min-ball-bearing-ps-025in-bore-dia-r4-zz-prx/i/G0101206/?utm_source=google&utm_medium=surfaces&utm_campaign=shopping%20feed&utm_content=free%20google%20shopping%20clicks&campaignid=19976937587&productid=G0101206&v=&gclid=EAIaIQobChMIloqx_8L9gQMVtc7jBx3vdjiUEAQYAiABEgJLTPD_BwE&gclsrc=aw.ds

 

Dowle

https://www.homedepot.com/p/6404U-1-4-in-x-1-4-in-x-48-in-Raw-Round-Dowel-10001800/203334060