Mouse Droid (MSE-6)

April, 2023

Just in time for Easter, I’ve completed my new Mouse Droid, full of “easter egg” surprise features. I’ve been wanting to build one of these for a LONG time, mine has finally come to life! In the Star Wars universe, a MSE-6 Mouse Droid is used as a messenger, and for cleaning and minor repairs. I don’t recall seeing it do anything more than drive around, and have no idea how it could perform physical tasks, but I’ll take their word for it. 🙂 The completion of this build fills out my collection of other hand-built Star Wars droids nicely, and will give me more options when trooping with the Rebel Legion/501st Legion.

A special ‘Thank You’ to Mr. Baddeley’s design which was the foundation of this build. As you’ll see, I made some major modifications to make it my own, unique droid. Overall, this project took about 3.5 months to complete, with 3D printing happening continuously on a single printer for the first 4-ish weeks. Most other work happening on nights and weekends. As with my D-0 build, I’m going to skip the details of assembly as they’re well covered elsewhere, and I focus this post on my changes/additions.

Looking at this finished product, I’m still astounded that almost every aspect of it was 3D printed. Getting to this end result pushed my skills and my patience. It’s far from perfect, but I’m pleased with the result, and I think others will be as well.

Easter Eggs

With no further ado: Some of the most fun features of this build were ones that I designed and built myself. As enjoyable as it is to use someone else’s design, I find it’s much more rewarding when I add my own touches and make it my own. There are several small and one large Easter Egg that I built in.

Start-up Sound: As with all my builds, I add an automated sound trigger on power-on to help confirm that the audio is functioning as designed. After considering many options, I’m really happy with the start-up sound I came up with: Mickey Mouse saying “Ha-ha, Hi-ya folks!” in his classic, unmistakable voice. Why Mickey Mouse? Well, first this is a “Mouse” droid! Secondly, Star Wars is now owned by Disney, first famous for Mickey Mouse. I thought it was a fun, fitting nod.

Hub caps: I can’t take credit for the idea, but I decided to put Imperial logos inside the wheels of the droid. I took the outter wheel .STLs and modified them to include an embossed logo, which I later painted white. They really stand-out since they’re the only thing that is white on the mostly-black droid. Yes, I know that the MSE-6 droids weren’t exclusively used by the Imperials, but they are most often seen in Imperial settings, so I went with it.

Mouse Droid Song: A very creative artist named Ossi Mykkänen sampled some Mouse Droid sounds and made a catchy, fun mix song out of them! I loved it so much, I added it to my droid. When the normal Mouse Droid sounds get boring, I can turn this on and we can have a dance party!

Grogu!!: Yes, I added several new features including a top that opens/closes, a riser platform, and an animated Grogu that makes adorable baby noises. I expect this will be a crowd favorite. With SO MUCH ROOM in the top of the shell, I knew I needed to do something to make use of the space and enhance the droid. so, this is what I came up with, and it turned out better than I expected. More details on this part of the build below.

Mandalorian Sounds: In keeping with the Grogu theme (and banking on the fact that the Mandalorian series is winning new Star Wars fans by the millions), I thought I’d add some audio of Din Djarin from the series. Fun, random clips that are highly recognizable are randomly played when I press the audio trigger on the RC controller while the Grogu platform is raised. Also, when Grogu rises, I play the theme song from The Mandalorian, and can trigger the audio using one of the RC controller buttons when Grogu is up.

3D Printing

My trusty ‘ol Prusa i3 MK3 was pushed to it’s limits with this build since it required prints that nearly maxed out the print bed. As before, I used a spreadsheet to keep track of all the parts to be printed and the details about each part (# perimeters, infill %, material, layer height, etc.) so I can re-create if needed. This worked really well and I highly recommend it for keeping organized for printing so many different parts over an extended timeframe.

Total print time for this build was 403 hours. Of course, that excluded the prints that failed or needed to be re-printed. I used 5+ rolls of eSun PLA+ (Black) which I found easy to work with. For high-wear parts (gears), I used less than 1 roll of Taulman3D Nylon 645, which has a major learning curve to use correctly. For the tires (or “tyres”, as Mr. B says), I used less than one roll of NinjaFlex Cheetah TPU that I bought years ago, but used for the first time on this project.

3D Printing Nylon requires lots of special care and handling to be successful. In my previous attempt to print with D-0, I got poor results with a lot of spitting. I assumed this was because it was an old roll that had absorbed a lot of water and was no longer usable. So, I bought a fresh roll and was surprised to get the same results. For MSE-6, I decided to give it another go. This time, I bought a cheap Filament Dryer to hold the role while it printed. I also used Capricorn Bowden PTFE Tubing to run the filament from the dryer to the printer hot-end. I custom designed and 3D printed a cap to the filament port of my printer to provide a screw to securely connect the tubing. With the drying running at maximum heat, I was able to keep the Nylon dry enough to print properly. The results were poor at first due to some clogs, but got better as I dialed in the temperature and slicer settings.

One thing I learned about Nylon was that the glue-stick print bed adhesive I typically use with PLA/PLA+ was insufficient to stop the print bed warping. I did start to get better results by cranking up the print bed temperature, but ultimately, I opted to just avoid Nylon for longer parts.

After the challenges with Nylon, I was intimidated to try to master yet another type of filament: TPU. However, the Ninjaflex Cheetah filament is easy to work with using the parameters recommended on their web site. Unlike most other TPU, the Cheetah filament allows high speed printing (thus named “Cheetah”), so I was able to quickly create all 4 tires, and they turned out really well. After some research, I selected a hexagonal infill pattern as I understand this is the ideal option for flexible, yet durable tires. The experience made me realize how versatile my 3D printer is.

There were a few highly visible parts (top Greebles) that I knew would not meet my quality standards coming from my 3D printer, so I opted outsource the work via CloudCraft to someone with an SLA Resin printer. I know, I swore I wouldn’t use this service again, but after shopping around, I realized that they do offer good options, and my previous negatively experience was largely because of my selected vendor, and not their service. My first choice vendor accepted, and later rejected the print stating that the design was invalid (walls too thin?). So, I re-submitted to another vendor and they printed it just fine. The 2nd vendor’s price was significantly higher, but I didn’t care at that point – I was just happy to have the part. The end-result was paint-ready with no clean-up required and looked great. I used Ninjaflex filament to thread on 5 cylindrical stand-offs as per the instructions – when painted, they look like resistors.

Circuit Boards (Side Greebles)

I chose not to 3D print the two side circuit boards. Instead, I found a supplier of a real printed circuit board from Paul (joymonkey) that looked screen-perfect. These were reasonably priced and even came with the daughter boards required for the left side. Following the advice from the https://forum.mousedroidbuilders.club/ site, I used the excellent circuit board designs from TinyP and acquired the required IC sockets and ceramic disk capacitors from Jameco.

I then went onto eBay and bought a bunch of 14 and 16-pin DIPP ICs as close to the specs posted by TinyP as possible. It was at this moment that I realized what I had gotten myself into. Prices for these ICs were much higher than I expected, especially since I was buying in smaller quantities to try to match the shape. Overall, I think I spent about $150 USD on the circuit board + parts. I probably could have gotten them cheaper from an overseas supplier, but I didn’t want to wait 4 weeks for delivery.

Next, I assembled them by first soldering on the IC sockets to the board. At first, I was just soldering 4 pins on each socket, assuming that would be enough to keep it from moving. However, I soon learned that the little pins inside the socket rattle if not connected, and often come out with the IC when removed. Then I attempted to just bend the pins into place, but that still was not good enough. S, I opted to just solder ALL of them. This consumed many hours and quite a bit of solder – frankly wasted since they were not actually making an electrical connection. I briefly considered just gluing the pins, but was concerned it might seep through the pin holes and look bad, so I stuck with soldering. I happened to have a a 16 pin IC insertion tool, which was invaluable for inserting all the ICs.

After much debate, I decided to only paint the ceramic capacitors and the tops of the ICs, and leave the rest of the circuit board untouched. I know most builders will spray the whole face of the circuit board (which is more screen accurate), but I wanted to show off the “real” circuit board assembly and leaving the silver IC pins and circuit board soldering pads unpainted helped achieve that, I think. The paint on top of the ICs hides the printed text which I think would have been distracting.

After briefly attempting to source real aluminum heat sinks, I did ultimately 3D print these and painted them black. I’m glad I did because the aluminum would have only added unnecessary weight and would have been hidden under a layer of paint.

Remote control

Once again, I selected my favourite RC controller for this project: The HotRC DS-600. I love the low cost (especially on AliExpress), ability to easily conceal the controller and the “hackability” of this device (as I outlined in an earlier post). In this project, I tried hard to limit myself to a single controller. To do that, I had to max out the connections and add software to have buttons play multiple roles depending on the mode.

I did my standard mods to the DS-600 including:

• Rewired the trigger button to Channel 6 to make it easier to trigger sounds while driving.
• Carved out the inside a bit to make room for a larger capacity battery
• Upgraded the wrist lanyard
• Cut a view port for the charging light

Next, I configured the receiver with wiring that allowed multiple uses of the buttons (Channels 3, 4, 5 and 6) by wiring them to 2 Arduinos. One Arduino Pro Mini used for triggering audio, and One Arduino Uno for triggering the lid and riser functions. Port 1 (throttle) went to the ESC for motor control, and Port 2 (steering) went to the MG996R Servo. For more details on the wiring, see the “Electronics” section below.

Because of the torque from the motor and the slick tires, I’m finding that the tires spin quite a bit on a smooth floor and it’s a bit difficult to control at high speeds. So, I re-programmed the controller to decrease the range of the throttle. While it reduces the top speed, it does help make it easier to control. So, I won’t win any races without reverting the programming, but I’ll be less likely to drive into a wall or person.

Design Changes

As with D-0, I was REALLY impressed with the design that Mr. Baddeley had produced for this droid. I never imagined that I could successfully 3D print an entire planetary gear box, but he’s done it and it works really well! While there are so many little things about this design that I appreciate, I did have some notes:

• Missing documentation – Lots of good detail to build the base and electronic controls, but no documentation to be found about the shell or other important aspects of the assembly. I did find some archived YouTube videos from Mr. B that were helpful, but some parts I had to figure out for myself. For example, how to orient and attach the piece called “
• Incomplete BOM – The documentation includes a bill of materials that I was able to use to quickly buy the required parts. Unfortunately, it was missing several important items including the list below. I’m adding these here to help others that may have similar challenges.
  • Magnets for holding the top/bottom shell (10mm diameter x 5mm tall rare earth magnets)
  • M5x8mm bolts for mounting the motor
  • M3x15mm bolts for front tire vertical mounts.
  • Power switch
  • T-plug parallel wires for connecting batteries to
  • Various parts for the assembly of the shells
  • Knob for the volume knob on the audio amp board (It’s a minor thing, but I happened to have one and it makes it much easier to use).
• Stand-offs: I opted to add some stand-offs in the design which I think helped a lot.
  • I put 4 stand-offs under the audio amplifier board – it allowed me to run wires under the board for a neater design.
  • I put 4 stand-offs under each side circuit board to accommodate the uneven back of these boards and avoid putting physical stress directly on the soldered joints.
• No sizing on the LiPos – General guidance was given on the battery: S3 LiPo with 5500 mAh, but these come in all shapes and sizes. I would have preferred to see some recommended measurements to ensure I didn’t exceed the tray size. I ended up buying these and they fit with room to spare. 133*43*28mm/5.24*1.69*1.1inch(L*W*H); It was not explicitly stated, but I secured these down with zip ties using the slots provided – it just seemed the correct thing to do.
• Voltage Monitors – LiPo voltage monitors (plural) are recommended. However, since the batteries are connected in parallel, only one is really needed. I used hook-and-loop fastener to attach it directly to the top of the batter to secure it. I also cut off all the unused pins so I would not accidentally connect the plug to the wrong pins.
• Shell Connectors – The outer shells include screw holes to support connectors to aide in the assembly and alignment of the parts. I’m a big fan of improved stability, so I decided to use them. Unfortunately, no STL files are provided. So, I had to do some extensive measurements to figure out sizing and hole placement to create them myself. Mr. B. suggested they could be removed after the glue dries, but I kept them on, and I’m glad I did.

Electronics

First, let’s talk about the MONSTEROUSLY power motor inside this thing. Clearly, this droid was designed for racing because a 24V 100W scooter motor happens to be the same ones that I put inside each leg of my R2D2 and they are powerful! In fact, I had to adjust the max range of the RC throttle to avoid gear slippage. The turn radius isn’t great, so it’s been a bit of a learning curve for me to drive something that isn’t skid-steer like most of my other robots (lots of 3-point turns).

As typical, I diverged significantly from Mr. B’s electrical design to custom build my own. I’m especially happy with the way that the sound system turned out. While I did use the recommended TPA3116D2 Dual Channel Audio Stereo AMP (which is a bit overkill for this application), I went a different path for the sound board and controller. I chose an Arduino Pro Mini and Adafruit Audio FX Board to manage the sound effects. This continues to be my go-to board for robot audio control because of it’s reliability, low cost, and good performance. I’ve used the recommended DF Mini player previously, and found it difficult to work with and lacking in features.

The addition of an Arduino Uno was necessary to control the Grogu easter egg features. I ended up using every available DIO pin on the board, and I was a bit concerned I’d have to graduate to an Arduino Mega if I added any other features. After a LOT of experimenting, I settled on the use of an Adafruit Motor Controller 2.3 Shield Kit to control the 2 linear actuators that open/close the lid. Additionally, I selected some OUYZIGA TB6600 Stepper Motor Controllers (2) to control the two pancake Stepper Motors used to raise/lower the platform. I first attempted to use another Adafruit Motor Controller for the stepper motors, but found that they were drawing a higher amount of amps than they should and I was risking burning the controller out. So, even though they took up more space, the separate stepper controllers gave me better performance and more controlled amp-draw.

Speaking of the pancake Stepper Motors, I actually ended up buying several pairs of stepper motors before I found a pair that met my needs. I needed them to be fast, quiet, and produce enough torque to raise/lower the platform with Grogu on board. However, they also needed to be skinny since they sat on top of the base. the skinnier I could make these, the lower I could mount the platform, and the more room I would have for Grogu. StepperOnline offers a fantastic selection of inexpensive NEMA17 stepper motors via Amazon, and I was able to find this one with the exact specs I needed.

The Arduino Uno also controlled a board of Relays that controlled the functionality of Grogu. I used one relay to turn on/off the 5V power to the Grogu toy. Often, I found that this did not immediately wake the toy, so I added another wire to the tilt sensor inside Grogu’s head which allowed me to accelerate the wake response and get him to start animating immediately (Blue wire). Finally, I added another wire to the plate in Grogu’s head that acts as the capacitive touch sensor. When tapped 3 times, Grogu will perform a “use the force” animation. By adding an inline ceramic capacitor to simulate a human finger (it’s the only way i could get it to work), I was able to rapidly open/close the relay 3 times in order to trigger this function from my RC controller via the Arduino Uno (Yellow wire).

Grogu has a built in speaker and plays random baby noises. I experimented with extending the wires down into the Mouse Droid and running them through the audio amp. Unfortunately, it sounded terrible. It may have been because I did not disconnect the internal Grogu speaker before extending the wire. However, I found that the Grogu-produced sound is loud enough, and is more life-like if it’s coming from him instead of through the Mouse Droid sound system, so I abandon that mod (but kept the wires, just in case).

Modular, polarized quick connect wires are the key to a successful droid build. Nothing worse than reverse-connecting wires in the field and frying something (ask me how I know…). So, I added lots of JST connectors to any part that I may need to remove in the future, to keep things modular. For example, a 4-pin connector between the linear actuator motors (connected to the top shell) and the wires connecting to the Adafruit motor controller. Connectors were also added to all wires between the Grogu lift platform and the mouse base, which came in handy during design and testing.

Finally, I strategically zip-tied all the wires to various parts of the frame to keep them neat and out of the way. The umbilical wire bundle going inside Grogu was a bit of a challenge as it needed to be long enough to reach when the platform is fully extended, and also coil below when the platform is lowered. It’s been a challenge to get it to land where I want, so I’m considering adding a small spring or elastic band to help guide its descent consistently.

Grogu’s Platform

The engineering masterpiece that is Grogu’s platform took quite a while for me to complete. The first challenge was to build a functional mechanism. The second challenge was to fit it all inside the Mouse Droid shell. The rapid prototyping and build of this aspect of the project probably took equally as long as the rest of the build.

I started buy buying a pair of 6061 aluminum bars that I cut to length and used as the platform for the lift. I needed something lightweight but rigid to host all the components of the lift, and this was perfect. I 3D printed custom H-brackets to mount the bars on top of the Mouse Droid lower chassis in order to give the clearance needed. I went on later to drill many holes into this aluminum bar to mount various components.

Lift Mechanism

First, I considered several options to raise/lower a platform inside the shell including a telescoping arm, a scissor lift, and a screw lift. The key challenge was the get the platform as low as possible in order to give as much “head room” as possible for Grogu. I initially intended to limit myself to a single motor for controlling the raising/lowering of the platform, so I bought and prototyped a scissor lift. After some modifications and testing, I concluded that it would not meet my needs for several reasons. First, the motor would be required to raise/lower with the platform, which would need to be on guide rails. Second, the motor would move forward/backward as well, which was challenging to keep stable. Finally, when fully lowered, the scissor lift required too much room, leaving almost none for Grogu. So, I scrapped that prototype and all the investment and moved on.

After much research, I finally looked to my 3D printer for inspiration and decided to attempt to use a stepper motors to drive a screw to raise/lower the platform. I quickly realized the leverage issue and conceded to add a 2nd motor on an opposite corner to give even torque. I bought a pair of inexpensive T8 8mm lead screws and brass nut set. I also use some Actobotics linear rails and linear ball bearings for the non-powered corners to provide guided support while raising/lowering.

I admit that I had to make multiple purchases of the couplers and linear ball bearings to get the dimensions I needed. The set of couplers that came with the lead screws were of the compliant variety that tend to snap when too much resistance is encountered. So, I opted for another option that was a rigid coupler, but found those to be tool tall and chunky. Finally, I found a coupler set that used set screws at Actobotics that were low profile and met my needs. Additionally, the first linear bearings I bought were too long (45mm), and prevented the platform from lowering enough, so I bought the 24mm version which fit better. This re-ordering process added a lot of wait time between prototypes, which was frustrating.

I designed and 3D printed custom brackets that held 1 stepper motor and 1 linear rail and allowed mounting underneath the aluminum bars. Again, lots of rapid prototyping to get the right sizing/spacing for this to work correctly and be rigid enough to meet my needs. I printed two of these and just reversed it in the rear so the stepper motors were on opposite corners of the platform.

After some testing, and replacing the coupler and linear ball bearing, it became apparent that the stepper motor shaft length was prohibiting the platform from lowering all the way. So, I protected the motor with some plastic and tape, and took the shaft to the belt sander and ground down the shaft to allow a better fit. I was nervous about grinding too much, so I took it slowly and measured often. With all those adjustments, I was finally able to get the platform to be almost all the way down to the aluminum bars, giving maximum headroom for Grogu.

After some test fits inside the shell with the lid on, I was finally ready to adjust the lead screws and linear rails to their final height. I was nervous about making this cut, but it turned out to be quick and easy with my angle grinder and a cutting wheel. I was sure to mount the cut side down so the beveled end of each rod was up to make it easier to take the platform on/off.

I considered an option to provide top mount support for the screws and linear rails, but struggled with the placement. I would not want a design that permanently mounted the screws/rails to the top of the shell because it would prevent me from taking the shell off easily. Also, top mounts would prevent me from fully raising the platform. I later decided that top mounts for the screws/rails was not really needed – I’m not going for precision here like a 3D printer. I just want a simple lift mechanism. As it turns out, the 4 post design with 2 screws is quite stable, even when fully extended and driving around. So, I’m glad I gave up on this plan.

Grogu toy

Now that I had a working mechanism, I could effectively measure the head space for the Grogu toy – it was only about 6.75 inches. This helped me make my selection of toys to use. I had 3 main choices for animated Grogu toys:

• Hasboro Galactic Snackin’ Grogu with over 40 sounds and motion
• Hasboro The Child – Animatronic Edition with over 25 sounds and motion
• Mattel The Child Real Moves Plush with remote control

I’m thankful I did this project after the holidays because the prices on these toys were crazy-high and had come down a lot when I was ready to buy! This was a pretty easy choice to make. The Galactic Snackin’ Grogu is 12″ tall, which is much too big for my needs. It also requires interactive props which I could not remotely control. The Mattel toy has a cool remote control, but it was mainly for moving the toy around and I needed a stationary toy. Also it was 11″ tall, and still too big. That left the Hasboro The Child toy – it was 8″ tall and animated its head and 1 arm. It was almost perfect, except some modifications would still need to be made.

Hasboro’s The Child toy has several interesting features. First, when activated it randomly makes baby noises and moves his head, eyes and ears. When tapped 3 times on the forehead, he closes his eyes and raises his left arm to “use the force”. When lay on his back, he simulates sleeping by closing his eyes and playing snoring sounds. He also wakes up when resumed to the vertical position. He often has about 10 seconds between animations and sound which is likely a design to extend battery life.

The dissection of an adorable toy did cause me a moment of pause, and I admit that I did not get over the creepiness of seeing the innards of this toy during this project. After a bit of experimentation, I found I was able to use a Dremel cutting tool and remove the bottom 1.25″ of the rigid plastic bottom which included the battery box. Luckily, I would not need the battery box because I intended to feed power through an umbilical wire bundle. I used a belt sander to ensure a flat bottom. After reverse-engineering the power switch, I was able to determine how to connect the wires in a way that allowed me to control the power functions. Next, I experimented with the circuitry and added 2 wires into the head to control the sleep tilt switch and capacitive touch sensor (as discussed above). I hot-glued all new wires into place to help mitigate strain.

I use a pair of L-brackets to bolt the front half of the toy shell to the platform, and positioned it over a large hole that allowed the umbilical wire to hang down from the inside. I then attached the back half of the toy shell using the provided screws which provided a stable base. I had to make an alteration to his robe by popping the back seam and adding hook & loop fasteners to allow easy on/off since pulling it down over his head was risky and difficult. With Grogu now a bit shorter, the cloak kind of billows a bit around him, but I think it kind of looks like he’s sitting, so I didn’t bother altering the length of the robe.

Top Doors

Next, the roof of the Mouse Droid would need to be replaced with hinged doors that automatically opened/closed. For this, I started with the original top design and rapid prototyped some 3D printed hinges. I had previous experience with 3D printing a hinge that was very successful, so used that as a foundation. After any attempts, I was able to get a good-looking, functional design that mostly concealed the function. While the print was running, I slid in some carbon fiber rods inside the hinge and let the print job run to completely hide the hinge. I also designed the 2 halves of the roof to have a small overlap to allow a snug fit. To re-enforce this closure and to keep it from rattling around when the Mouse Droid moved, I added small, embedded rare earth magnets that work perfectly.

Next, I bought some micro linear-actuators to force the doors open and closed. The first set I bought was a 12V 2″ model, but it was too long, so I had to purchase a second smaller set that was 12V 1.2″ model and it was a better fit. To make this work, I had to custom design and 3D print special braces to fit inside the top shell to give a platform for these actuators to push/pull from. The key was to solve the puzzle that positioned the new components of the top shell so they did not interfere with the components of the bottom shell and fit together. The key was to mount the linear actuators vertically and very close to the hinge where it could move a small distance and result in a wide door movement. I found this excellent online calculator by Fergelli Automation that really helped me figure out this complex geometrical problem.

In the process of finding the best mounting point for the brackets, I ended up needing to adjust the position of the Stepper Motor Controllers to avoid a collision. This was a tight squeeze, but I was finally able to position everything so that it fit nicely without touching. This was probably the most stressful part as it required a lot of iterations and rework.

To ensure the actuators were secure, I used small bolts to affix the braces to the shell, and then used 2-part epoxy to weld them together. I custom designed and 3D printed a small bracket that helped with wire management for the linear actuators. I bolted and epoxied this bracket to the interior shell wall as well. I terminated the two pair of control wires for the linear actuators into a 4-pin JST connector for quick and easy disconnects to allow me to easily remove the top lid and work on the inside.

Mounting Electronics

Mounted to the pair of aluminum rails are:

• 2 Stepper Motor controllers
• 2 brackets, each holding a stepper motor and linear rail
• an Arduino Uno + Motor shield
• A Relay board
• Support brackets for the Mouse chassis

Unfortunately, there was just not enough room to mount all of these things! So, I had to get a bit creative and think 3-dimentionally. I custom designed and 3D printed mounting boards for the Arduino Uno and Relay Boards. I bolted the electronics onto the plastic mounting boards (which help prevent shorts). Then, I stacked the relay board underneath the Arduino Uno mounting board and was able to snugly fit both stacked on top of each other. To my surprise and delight, I found that it just fit perfectly with the top shell on, and I had to slightly bend the wires connecting to the Arduino DIO pins so they didn’t hit the front of the shell. I hot-glued the pins in place to prevent them from popping out.

In the picture above, you may notice the stepper motor sitting almost on top of the downward facing speaker. This was the main reason I went through 3 different stepper motors – trying to find one thin enough to fit. I was a bit concerned that the magnets in the stepper motor would interfere with the audio. However, after some testing, I found no issues as long as I kept the gain on the amp at a reasonable level.

Here is a link to the Arduino sketch files I used –> link

Finishing

My least favorite part of any build is the finishing. Rushing early steps (as I tend to do), leads to obvious reveals of flaws in the final steps. I seem to fall into this trap every time, and this was no exception.

The main focus for finishing were the upper and lower shells. I started with sanding using 80 – 150 – 220 grid paper. I purchased an inexpensive detail sander because I saw other builders using them and thought it might reduce the effort. However, even with the mostly straight, flat parts of this build, I found it difficult to get good results. So, I reverted back to hand-sanding. What did help a lot were these rubber sanding grips which I bought years ago. They were able to help me sand down the rounded transitions really effectively.

After sanding, I filled in the gaps using 3M’s Bondo Glazing Spot Putty, having successfully used it on previous robot builds. This required a few applications with sanding in between, but did effectively fill the gaps. I think I should have sanded a bit more because they are still slightly visible after the final coat.

Next, I did multiple coats of Rust-Oleum 2-in-1 filler and Sandable Primer to build up a nice even finish across the whole surface. I followed this with several coats of Rust-Oleum Universal Premium spray paint in a Black Satin finish. Despite knowing better, I applied some coats during a period of high humidity and paid the price with a bubbled finish that I had to sand off and reapply. As a result, I bought an inexpensive Hygryometer to use as reference before any future spray paint attempts.

I took some time to do some detail work on the wheel hubs by applying some white paint on top of the embossed imperial logo. I slid paper around the inside to act as a mask, and they worked really well. This white paint really pops compared to the mostly-black body of the Mouse Droid. It was intended to be a subtle thing, but now it’s one of the first things folks notice.

Finally, I prepared all the painted pieced for a clear coat. I used a left-over can of SprayMax 2K Clear Satin. I LOVE the final result of this stuff, but many precautions were required when applying because it is pretty much poison. Long pants/sleeves, gloves, eye protection, respirator – I felt like Darth Vader. I did 1 light coat and 2 heavy coats with 10 minutes in between. I did it outside for maximum ventilation and was surprised to see how it attracted every insect in my yard. I had to be on patrol while the coats dried to prevent an insect from getting stuck in the paint. I did a light sanding coat with 600-grit after to knock down the high points, and it was done!

Conclusions

The first reaction that I had when I started printing this was: This thing is HUGE! Then I started doubting myself – Mr. Baddeley knows his stuff and has screen-accurate designs, so maybe the droid really is that big? After consulting with Wookieepedia, I learned that the Mouse Droid is 0.3m tall (about 12 inches). This design is a towering 0.4m tall (about 16 inches), and is significantly larger in all dimensions. Seen below next to some standing Stormtroopers, it’s obviously much smaller than mine. Do I really care that it’s bigger? No. I figure that this extra room really allowed me to add the Grogu easter egg and would not have been possible otherwise. Yes, it’s a bit more cumbersome to transport and maneuver, but I love it!

New skills/experiences earned with this build include properly 3D printing Nylon, printing with TPU, and 3D printing gears and building a gear box. Yet again, by studying Mr. B’s designs, I’ve gleaned some clever new tips and tricks that I’ll use in future custom projects. I’m also delighted with the pace and turn-around time I’ve been able to achieve with these droids. 3D printing is truly a fantastic process. Previously, I was consuming 9-12 months on a droid, which has now reduced to 3-4 months. Additionally, I have plans to buy another (larger) 3D printer soon, so that should accelerate future builds even more.

Mouse Droid seems a lot more durable than some of my other robots, so I’ll feel more confident bringing it to a troop. I’ll still need to avoid the kicking and grabbing children, but I won’t have to worry so much about a minor accident. I am a bit worried about people touching Grogu, especially anyone who knows the toy is touch-activated. However, this droid is more than capable of outrunning most children, so there’s always that option. I’m still working on a plan to transport the Mouse Droid. Someone suggested a rolling toolchest, which would probably work well, but would not fit into my current vehicle. For now, it will have to be gently placed into the back seat and carried.

Looking ahead, I do have a few minor things I’d like to address. First, there is a high-pitched noise that comes from the motor when it runs – I’ve attempted to dampen this with some foam, but it didn’t help. Next, the relay clicking is quite loud and I’d like to dampen that sound as well. Finally, I can sometimes hear the differential gear slip for sudden accelerations – I need to open that up and see what’s going on.