Poro! from League of Legends
So in the League of Legends community, the poro, commonly seen walking around the Howling Abyss, has been a huge hit and everyone loves its cuteness, it dances around all happily-like when you feed it a poro-snax. I decided I wanted to make one in a plush form with the functionality that it vibrates when you place a poro-snax close to its mouth.
I plan on making this by sewing cloth together and putting stuffing in it to make it into a plush. The only thing problem is that I don't know how to sew. So I'll have to learn how to sew through the process of this project.
To handle the electronics of this project, I plan to use an Arduino Lilypad. This has most of the functionality of a regular Arduino with the added factor that it can be sewn on to fabrics and is washable.
To have the vibration I plan to implement a vibe board, a small board that will vibrate when given voltage.
To have the poro know when the snax is near it's mouth I'll be using a Hall Effect sensor. This is a sensor that detects magnetic fields. It gives off a positive voltage when a magnet facing the correct direction is placed near it. I'm planning on placing a strong enough magnet inside of the poro-snax that will be able to activate the Hall effect sensor through all the stuffing and fabric.
For the design of the poro, I'll mostly be using this website:
I zoomed in on the image of the poro on a screen. I then simply taped a piece of paper onto it and traced.
I then drew a outline of the main body of the poro. This is to allow for that extra space when cutting out and sewing the fabric that the poro will be made of. I left a ruler there for as a scale.
The main basic theory that will be used for the sewing is sort of similar to how clothes are made. Most articles of clothing have a seam in between the parts of fabric, to help if the fabric starts to stretch. This is usually on the inside of the piece of clothing as it juts out. We'll be using this theory of having extra fabric of where we'll be sewing and then flipping the fabric inside out so that the seam is on the inside.
The picture with the outline around the poro is a good illustration of what will need to happen. The image of the poro is where we would sew and the outlined parts would be the sections of extra fabric.
So I was told that the poro would be too small, by my local expert seamstress, my mother. So Time to repeat the process and trace the poro again, only this time larger.
-sorry in advance for the poor picture quality
Since this just a prototype, since i don't know how to sew quite yet, We're using a random shirt we had around the house as our fabric.
I cut the body of the poro out and we secured the image to the shirt with pins and needles. After that we traced the outline of the poro onto the shirt with a red pencil, that shows up on fabric. We then cut the image out with a boarder around it. This is to allow for extra fabric for when it gets stretched and expands due to the stuffing.
For the horns and legs of the poro, I plan to follow the same procedure that I did with the main body:
One thing i did differently was I only did this for blue parts of the leg (the back). I did this because I only had only piece of fabric at the time, and since this is just a prototype I want to see how well I can make it, so that I can make the final product as good as possible.
I cut out the horns and front parts of the feet and saved those paper cut-outs for another piece of fabric.
I want to be able to sew the colors of the feet together before sewing it all together.
So my mom had begun showing me how I should sew the poro. First we marked off where we would want to stuff the poro, it will also be where we will flip it inside and out.
From there the real sewing begins. To make an initial knot, in the beginning, you can loop the thread around your finger, roll the 2 strings together, hold down the loop and then pull, this will most likely make a knot, if not try again. And if it really doesn't work, you can always just tie a regular knot.
The threading part is pretty simple. It has a sort of 2 steps forward 1 step backwards methodology. I kept the poro facing up, I sewed downward then came up. Then went backwards in between both points (entry + exit). After that I sewed forwards again and went backwards again. I repeated this process until the string got short.
The knot at the end can be pretty difficult, you end after the needle pokes up . You then lay the needle on top of the hole. Taking the string that comes out of the hole, wrap that around the needle (via the sharp side), then hold that wrapped part tight against the hole and slowly pull the needle through. Then finally sew it down once more just for extra security.
The last part has been proven the most difficult and so I asked my mom to demonstrate this while I filmed it and she agreed. Sorry in advance for the mixed languages in the video. I mostly got this to use as a reference, hope it helps if anyone plans on making this.
Repeat this process until you near the place that you marked for the opening. Go back and forth over a good section of the poro near the entrance point to strengthen it.
The third image shows what can happen if you don't make the knot tight enough. the thread isn't completely secured and has the chance of being undone. ***BE CAREFUL SEWING WITH THE NEEDLES STILL IN THE PORO**
I had sewn all around the body and left a section to stuff the poro near the back.
I wanted to make the prototype as close to the actual poro I wanted to make. This being said, I wanted to include the multicolored parts of it, such as the horns and part of the feet being brown. So I decided to have another color to accompany the color used for the poro's skin/fur for the feet. I plan to sew the colors together first before sewing the sides of the feet together.
So this was a bit more difficult that just sewing the body since the feet have multiple colors.
To do this sewed the felt onto the sides of the feet faces first then sewed the multi-colored pieces together. It took a while, since I wasn't sure how to sew it at first with all the flipping and such, but eventually it worked.
I was also told how felt doesn't really stretch, as regular cloth does, and so the felt doesn't need the extra outer layer, which makes it easier for flipping the cloth inside out.
I determined that placing the stuffing hole in the back might not be the best place for it. This is because of the circuit that I plan to place near the front of the poro, near his mouth. So time to undo that section off thread and then sew the back of the porotype
I decided to stuff the body so that I could start figuring out how large the heart bottom should be. I stuffed the body until it looked large enough. I then used the the pictures and kept comparing it and the image from the internet until it looked close enough.
After stuffing the body, I tried to create the heart by basing the size of the picture and the size for the heart on the real poro until the heart looked about right.
Unfortunately, the poro was a bit mis-sized, so hopefully I'll be able to make the next one a but more accurate with either the initial sewing or through stuffing more accurately.
I took the cut out heart and sewed it directly onto the body of the poro. Though this isn't ideal, it helps give a better sense of what it should look like.
In the future I would like to make the heart attached to the cloth the way the 2 skin/fur cloths are, so that when you run your finger across the surface you can't tell that one cloth is on top of the other. As of now it's not worth it to take apart the already made body skin, but the paper cut outs have been adjusted so that when the 2nd poro is made the body can be made more smoothly.
Because I had never used a Hall Effect sensor before, I decided I should look it up and see if there were any tutorials on it before I tried it. I came across this link:
and decided to use this to try out the Hall Effect Sensor.
It ended up working out! Though the first time around i had the Hall Effect Sensor flipped. I had a feeling it was weird that the light was on the whole time, but after it was flipped it worked perfectly, the code that was used is the same as on the site.
One thing that will need to taken into consideration for later on is that these magnets (1/8" x 1/8" x 1/16" Neodymium Magnets Grade N42), need to be pretty close to the sensor, as seen in the video.
Now that the circuit works with the Arduino Uno, it's time to test the circuit with the Arudino Lilypad. I changed the controller from the Uno the Lilypad and changed the pin for the Hall Effect to pin 11 (since the Lilypad I'm using doesn't have as many pins as the Uno) and positive and negative leads to their respective connections.
I decided to add an external LED because I would need an external connection for the vibe board later on. I hooked up the LED to pin 10.
I made the changes to the pins in the code and it worked!
Since the last test worked, I swapped out the LED with the vibe board to test how it would work. The code did not have to be changed for this. This test also worked. The websites weren't kidding when they said that only the holder of the board would feel the vibrations.
I then tried hooking up two vibe boards so that there would be a larger area that could be affected by vibrations. I first tried to run 2 boards off the same pin but it seems that there wasn't enough power to have both boards vibrate. I then tried to hook the two up using separate pins. This didn't work initially but then they did after a quick battery swap.
The power of the vibrations is reduced but I think that it could still be a viable option the poro, assuming there is enough charge on the battery.
I tried to test the circuit with 3 vibe motors, but the battery doesn't seem to have enough to power 3, looks like 2 is the max.
I had no idea how to attach the legs/feet of the poro, so I went and asked around my family members. My grandma was the one who had come up with a solution. Her method is seen in the first 2 pictures. The method that she used was to cut a hole in the main body of the poro and sew the back hole of the poro to it. This would've been a fine idea but we ran into the problem of the poro's leg wanting to become perpendicular to the body (since both sides of the foot was the same length. She fixed this by sewing part of the inner side to more closely (in order to shorten that side).
Something about this method made me feel like it may not be the best way to do. It certainly is one way, but I wanted to see if I could get the seem a bit cleaner and if I could get the leg to go more in line with the direction of its body.
The rest of the pictures document my attempt to make a better connection with the leg and body. The front leg is a bit bigger than the back leg. I knew I wanted to have the leg more in line with body, with this I thought it might be a good idea to cut the hole on the side of the leg instead of using the hole in the back and then sewing that to a hole in the body. This proved to be a more difficult method of attaching the leg, but I think it makes the seem a lot smoother and makes the leg attached the way I was intending, in line with the body.
Some things that could improve this method is to seal off the back hole in the leg once the side hole is cut and to make sure the poro's leg is at the right height, It seems that it would be too low along the body after stuffing has been added.
I decided to move the hole in the body back to the back of the poro. I realized that if you were to change the batteries for the electronics, you wouldn't want to rip open the poro's face every time.
I remembered how other stuffed animals with electronics, I had seen, had a sort of Velcro opening so that you could get into the stuffed animal and change whatever part was necessary.
I decided to model this with the 3rd rendition of the body.
I sewed the horns like the feet, except that I knew I would not need to make a hole in the side of horn as I did with the feet. I could just leave off one of the sides so that I can attach the horn to the body using that hole.I sewed the pieces together in accordance to the white paper outline, except for one side, and then trimmed the sewed sides down, sincee felt doesn't require much of an extra "safety" layer.
I marked where I wanted the horns to go. I then cut a straight line . I pushed the horns through and sewed along the lines so that horn and body would match up. This was similar to how the feet were sewn on, just in a different location and with a different piece.
I had shown the poro to a few people who know about the League of Legends and a few of them had suggested putting a heart so that there would be a visual indication that the poro has detected the snack.
I decided to try testing this out and I found that the lilypad would not be able to power both the motors and a single LED.
Because of these results I've decided that although having an LED heart would be nice, having the motors vibrate is of a higher priority.
I wanted to draw out the final circuit since it's can be hard to tell from the previous pictures. This also makes it easier to see for when I'm implementing it into the actually poro instead of just on a breadboard.
Because of the inability to power 3 motors at the same time I started to think of other ways I could get it seem like they were on. I thought back to a 4 number 7 segment display I learned how to use during my Freshman year at WPI. What I learned was that only one of the digits is lit up at any given time, The display is just switching from one digit to the others faster than the human eye can detect.
I decided to adopt this methodology and try cycling through the motors. Since the lilypad could power 2 motors as once I decided to alternate them so that 2 motors would be on at any given time. I would then cycle through these sets fast than could be detected to give the illusion that they were all being powered at the same time.
I wanted to try putting in multiple Hall Effect sensors. I wanted to try this so that there would be a larger range and area that the snack could be in order to activate the Hall Effect sensor. When I tested this out it seemed to confuse the lilypad and the motors. It would turn on for an extended time than it was supposed to as well as making the motors turn on when they weren't supposed to.
I decided that though this would've been a good Idea, it needed more fine tuning than I wanted to do on this, and so the idea was scraped.
I wanted to place all of the circuit on a piece of felt and then place that inside of the poro skin. Before I did that, I was having trouble deciding where I wanted the motors. I decided on placing 1 motor in between the legs on both side and one closer to its forehead. I figured this way there would be somewhat of an even spread of vibration throughout the poro.
I tried to mark where I wanted each part of the circuit onto a piece of felt, according to its position with the poro. I marked out where the parts -the Hall Effect sensor, the motors, the Lilypad and the battery- would be.
I was able to sew down the battery and the lilypad and using the conductive thread to make the connections for the positive and negative leads. I then used conductive thread again to connect the Hall Effect sensor the the necessary spots.
To secure the connection, I threaded conductive thread through the loops of Lilypad and battery,
To make it easier to sew the Hall Effect Sensor and resistor I made their leads into loops, with pliers, and then soldered the loop so that it's secure and won't break off.
I tested the connections with and LED and it worked!
I probably should have done this sooner, but I tested to make sure that the Hall Effect Sensor can detect a magnet through a few layers of cloth. It ended up working but, this was definitely a step that should've been done sooner.
I tried to use the same method to get one of the motors hooked up with the Lilypad. I thought that it would be okay to stretch the connection all the way from where I want the motor to where the Lilypad is. This seemed like a good idea at first, but upon trying it, I found that the motor did not have much output. I debated what could've been the problem and came to the conclusion that it might be due to the length of conductive thread being used since it has a resistance of 3 ohms/inch of thread. I figured that this is creating too much resistance and therefore lowering the amount of current going to the motor which also lowers the power output.
My solution to this is to try and make the connection shorter by moving the 2 points of connection closer to one another.
I decided to make an "inner shell" of felt for the electronics of the poro. This way the lilypad and the back 2 motors would be closer. Hopefully this lowers the resistance and increases the power needed to power the motor sufficiently.
I got all the connections sewn on properly and then ran into some issues with the conductive thread. I was thinking since I could operate 2 motors at one time, that maybe I could make the alternating motors run only 1 motor at any given time as well as one LED. I tested this out and found that one strong motor gave off more of an effect than all 3 motors being alternated around.
Because of this I tried out the circuit and code so that ONLY one motor would be powered, as well as an LED, whenever the magnet is placed near the Hall Effect Sensor. This started to work for a while but for some reason it seemed to either drain the battery OR make it so that the motor did not function after a short period of time.
The intial design didn't work out the way I was hoping and so, I hit a wall. When that happens it's time to search for answers and research more.
I was looking around the internet and found this instructable describing someone's journey of making a foot massaging slippers.
There were some helpful insights I took away from reading it:
I'm going to try to implement these pieces of insight into the circuitry of the poro. As well as putting the Hall Effect Sensor in the tongue of the poro and trying to implement an LED as well as the 2 motors, since the Lithium battery will be able to power more voltage than the AAA battery.
I also noticed that the stuffing started to get into the pins and parts of the Lilypad. So to unsure they aren't interfered with by the stuffing I'm going ot place all the big electronic parts (lilypad, battery, motors, hall efffect) on the outside part of the inner shell.
I found that only 2 of the 3 motors I had were decently strong. Because of this I decided to only power 2 motors. I then hooked up the 3.7 V battery to be sure it could power both of the motors at the same time.
I wrote a simple program that basically just turned on the motors. After uploading the code onto the lilypad and turning it on, with only the battery as its power source, the motors started to vibrate!
So it turns out the wires were connected very thinly with the leads of the battery. This led to the connection being undone. I ended up stripping the wires a bit and then soldering them back on to the leads of the battery.
I only have the ground wire coming off shown here but in the end both the positive and negative leads became undone.
After soldering the wires back on, the connection between the battery and the wire seemed to be even stronger than it was before.
I revised my circuit plan and layout to be more detailed and accurate to the current setup (with the 3.7 V battery and the 2 motors.
I tested the circuit out with alligator clips and it worked! all that was left to do was sew it onto a piece of cloth so that it can be placed inside the poro.
So I realized in my circuit how I wanted to place the Hall Effect Sensor in the tongue of the poro. This requires me to know how the tongue would look and where it would be. To do this I decided it would be best to just make the tongue, since it doesn't seem like a very difficult part to make.
I just estimated the approximate size of tongue and made it. I think I might've placed the tongue too high up on the body, but I think it's okay for now.
Also I'm not sure if I should close the tongue shut before sewing it onto the body, that way the stuffing in the tongue doesn't interfere with the stuffing in the body. This might also help secure the Hall Effect Sensor. The only downside would be that It would be a lot harder to maneuver around the extra piece of felt that is there.
I used the prototype poro to estimate where I would want everything. I wanted the Lilypad and battery near the opening in its back so that it could be turned on/off, reprogrammed, or have its battery replaced.
I started to sew down the Lily pad since that is the "center" of the circuit as everything needed to be hooked up to it. This time around I decided to sew it down using regular thread, as opposed to using the conductive thread like I did the first time around. This was to help limit the chances of a short circuit. Though the pins with the securing conductive thread (from last time) were not programmed to output anything, it is still better to be safe than sorry.
After I sewed the Lilypad down, I began to work on the battery pocket. This would simply be another piece of cloth sewn on so that the battery could be placed between it the pink felt in order to secure it from moving around too much.
I positioned the vibe motors in a way that was perpendicular with the center of the cloth. This way it motors will be more evenly spaced once placed into the poro, I worked my way around to getting all the parts sewn in, following the diagram I had made, to sew underneath or sew "in-and-out" to accommodate future connections.
I tested the connections with the 2 motor program from before, with the changes in pin number to account for the motors being connected to different pins this time around, The results of the test were successful! The motors were vibrating the way they were supposed to which means all the connections are properly made.
I positioned the Hall sensor where I would be placed in the tongue (and marked where the tongue would be) and slowly but surely sewed the circuit down following the diagram of the circuit. It too a while but eventually the circuit was all sewn in. I then tested the circuit by uploading the original circuit onto the Lilypad and it worked!
I then took non conductive thread to secure the sections that I had to extend on one side in order to make room for other connections to be formed.
I then put clear nail polish over the knots of the circuit, as the Adafruit video had suggested. Hopefully this solidifies the knots well enough that I won't need to worry about any thread fraying and creating short circuits. I have to be honest, I was a bit unsure of it this would work, but it worked beautifully! All the knots were secured tightly!
I needed to figure out a good size for the poro snax. To do this I folded a piece of paper to the dimensions listed in the pictures. I used this as guidelines to base the snax off of.
I then got some brown felt and cut out brown pieces of felt according to the dimensions. I also cut a long strip out of the brown felt as well, for the outer ring of the snax.
In the last picture you can see what I envision the snax to look like, a cylinder with a circular top and bottom as well as a single strip that wraps all around.
I came to the realization that it probably isn't the best idea to have to pull apart the poro's mouth every time you need to swap out the battery or access the electronic. To remedy this problem. I moved the hole to the back, where it was originally.I also sewed up the opening near the front part of the poro.
I wanted to get a lot of the parts done quickly and planned on using a sewing machine to speed up the process, but first I had to learn how to use it as every machine is different.
This machine was a lot more sophisticated than the one I had at home and it a lot easier to use (mainly for the fact that it doesn't go as fast and can be controlled more easily. I tested the machine while using a random piece of felt and tried to sew in the shape of the tongue.
After I was satisfied with the "test" piece of felt I moved on to the sewing the actual pink tongue.
I thought it would be a good idea to sew the blue parts of the feet first. I would then sew the black parts afterwards.
This turned out to make things more difficult as I would need to stick the black parts inside of the space between the blue sheets and make sure to sew the curve that the foot has.
I took the lesson I learned from the last step of sewing the blue parts together, of NOT to do that, and changed my method on making the feet.
I sewed the black parts onto the blue parts this time. This allowed me to be accurate when making the curve that separates the blue and black sections of the foot, as well as making it easier to make as a whole. This is because I would be able to sew the 2 parts together (to make a single foot) much more easily than before.
I then cut a straight line along part of the feet to make a hole. I made sure to make the holes on the right sides (opposite sides for each of the same feet (front and back)).
I then sewed a white footpad directly onto the bottom of the foot to account for the actual footpads as seen in the image of the poro.
I wanted to find out how many magnets would be good to place on the poro snax. I placed them around the circular part of the snax in order to see how the spacing looked and I determined that 8 magnets should be good.
From there I made a mark on the strip at every 1/8th of the strip. This is so that I know where to sew the battery onto the strip.
I also made a spiral shape using the sewing machine on the top circle, this is because I've seen it in many of the other reference photos of a poro snax.
I sewed the tongue part onto the electronics sheet. I did this by flipping the tongue inside-out and then sewing the tongue onto the sheet near the seam that was already on the tongue (the one that is used to keep the tongue together).
I then flipped one layer of the tongue over both the electronics sheet as well as the other side of the tongue. This surrounded the electronic sheet (the part with the Hall Effect sensor) inside the tongue.
I began to create the ruffles that can be seen as part of the poro's mouth.
I did this by making a paper tracing and trying to figure out the best shape for the ruffle. I determined a shape and started to sew it together, The result can be seen above. The bumps were not well defined and needed to be fixed. So I decided to give it another shot.
I decided to make the "humps" of the ruffles larger in hopes of having them more defined after being sewn and flipped inside out. I left the straight section unsewn so that the ruffles could be stuffed.
Even after I had sewn and cut the ruffles out they still weren't as defined as I would have liked.I asked chrisg about this and she mentioned how it could be because the fabric is pulling on certain points in multiple directions. We discussed methods of fixing this problem and she mentioned how you could cut some slits around those points on the excess fabric so that the fabric could expand more.
I knew I was going to sew this onto the the electronics sheet and could be sewn as one single piece and then sewn on as a whole.
I stuffed the ruffle and then sewn it shut. This could then be sewn directly onto the electronic sheet.
I decided not to base the eyes on a stencil. I figured I could just eyeball the shape of the ovals needed.
So I cut out 2 black ovals and 2 white ovals. I took some white thread and sewed the white ovals directly onto the black oval, so that it seemed to make an eye.
I then marked where the eye should be on the body, based on the stencil of the body.
I sewed the horned together, much like I had sewed the horns on previously. I sewed a seam along the lines of the horns except for straight side.
The difference from this pair of horns and the previous pair, is that I only used 1 type of horn, instead of the 2 different kinds that was seen in image from the league of legends forum.
I attached the legs very similarly to the previous prototype. I sewed the feet a good distance away from the centerline and made sure they were faced the right way (as I previously learned I had to pay attention to).
I cut some slits on the top part of the poro. Then i sewed the edge of the horn onto the edge hole.
Similarly to sewing the white part of the eye to the black part, I sewed the black oval directly onto the places that seemed best placed for the eyes of the poro
I decided to stuff the feet now, as I thought the it would be a lot more difficult to stuff once the electronic sheet was placed in.
I then secured the mouth onto the body.
I cut a hole where the mouth should be and then stuffed the tongue and ruffle in the hole. I then carefully sewed the mouth parts onto the edges of the hole.
It looks pretty cool at this point!
I then tried to flip the whole thing inside and out, in order to get the electronic sheet inside the poro. I tried to secure the sheet onto the poro skin; one at the top hump, one near the back, and one near each of the vibe motors. The only one that seemed to be useful is the one I placed near the top hump of the poro.
I then stuffed the poro, trying to get all the parts stuffed well.
It looks even better and realistic! :D
Although I stuffed the poro, I'm still stuck with the hole that I left so that you could access the electronics.
I was debating between using Velcro and snaps, but i decided that Velcro was my best bet. I just sewed it them directly onto the edge of the hole.
I wish I had some more time to test out the best method for attaching the Velcro because it didn't end up looking very good aesthetically. If I had time I would make it so that the Velcro is more hidden, perhaps making it so that one side goes over the other instead of having them meet in the middle as I did here.
Sometimes the vibrations weren't as spot on so I had to sew in some connections to make the electronics sheet stick better to the poro skin.
Some of the problems that could still arise are most likely due to conductive thread crossing and touching, creating short circuits.
The solution would be to either spread out the sheet well so that the connections don't touch, find a way to secure the sheet to keep it spread out, or potentially putting another sheet on top of the electronics sheet that prevents wires from touching (almost like a shield).
Here are some photos for the poro!
It's completed as of now. the circuit board can be accessed in the back and the circuit can be turned on and off without opening the poro.
The Velcro makes the poro look a bit bad but other than that I think it looks pretty good.
Sadly the vibrations from the vibe motor are not strong enough to be seen on video, so you're just going to have to trust me when I say it worked.
I was struggling on putting the magnets on the strip of felt that would go around the poro snax, since the only needles I have are metal.
I was looking around and I found these fabric pads. These were perfect! They covered the magnet and allowed me to sew the pads onto the strip, creating a sort of pocket for the magnets. I got all the magnets sewn on and they worked well after I made 2 passes of sewing around the magnet.
I realized that the snax is a bit too large for my little poro so I wanted to make it a bit thinner.
I did this by sewing the the top piece closer to the magnet than i first intended. I redrew a line that was closer to the magnets and sewed along that line and along the edge of the circle.
For the other side of the snax, I basically followed the same procedure, sewing along the edge of the circle, except this time I sewed most of it and then stuffed it before I finished sewing the whole top.
The 2 magnets at the end of the strip happened to line up and it allowed for the end of the strip to be sealed off and didn't require extra sewing to close off the snax.
And here it is! the complete poro and poro snax!
This was an awesome learning experience and was awesome to make!
Some things that I think could be improved are:
Overall I'm very pleased with how it came out and learned a lot from this project with regards to sewing and electronics!