Embedded Files
Practical - Making the zoetrope spin, Strobe effect
Theoretical - Flashing lights causing problems for viewers, Print warping
Technical - 3D printing and failed prints, 3D model exporting
Problem Solving in Ideas 2.2
Colours. The main issue I've already got is how on earth I'm going to add colours to my project - as fun as it is painting models that I've 3D printed, I'm definitely not good enough to achieve the quality I want in this project. However I have thought of one or two ways I can still make it look aesthetically pleasing while still only having one colour.
Dry Brushing
One option I have got is dry brushing. If I print all my characters in a darker colour, I can then go over them with a much lighter colour to add highlights by gently brushing over the model with just a tiny amount of paint.
Bronze Effect
This is something I came across while talking to my tutor about his neighbour's business, modelling and painting realistic models of characters from the horror genre. You can see from the image on the right that there's no colour painting involved, yet it's a lot more interesting to look at because of the bronze effect which gives the model a lot more shape and definition (similar to the effect given by dry brushing).
Washing
Washing is a painting technique that uses diluted paint that flows into recesses and crevices, using model's natural shape to create final effect. There are paints specially designed as washes, however you can also achieve the same effect by diluting regular paints with water. There are a few washing techniques that follow the same principles;
Pre-Washing
Washing
Controlled Washing
Lining
From what I can tell by watching a couple of videos (which I linked at the bottom) the way you 'wash' a model is that you, instead of using normal paints, you use diluted paints or a wash paint that's the same colour as the normal paints you would've used (as shown on the right, which hopefully helps that makes sense). There's also another option - you paint the model with normal paints in it's normal colours and then, once that is completed, add a black wash over the whole of the model to create shadows. As I mentioned earlier I've linked some videos down below to show the effects of this techinque.
Just as a quick sidenote, it's also very possible that I could double this up with dry brushing as well - I could either dry brush the model and then add a black wash, or I could start off by washing the model and then dry brushing it afterwards.
Viewing the Animation
Something I thought of while writing this up is how the animation is going to be viewed. I talked in the research section of this google sites about the different methods used to achieve something called the strobe effect, which is absolutely essential to my project. The problem I've got is that at the moments I'm thinking of using mirrors to achieve the strobe effect. What mirrors do is they show only the current frame of the animation (see video below) whereas with the idea 2 animation, the red kart would be throwing to the frame in front, meaning that using the mirrors would show only the current frame and the kart throwing the shell, but not showing the kart that gets hit by that shell.
An easy way around this though is to simply make the mirrors bigger - rather than having them wide enough to show only one frame at a time, I could make them wide enough to show 2 frames at a time. That way I still don't have to use any technology to achieve the strobe effect, the animation is still well lit and you'll be able to see both the kart throwing the shell and the kart that gets git by the shell.
Another problem I noticed though is that you'd only be able to watch this animation once - after that, the kart that gets hit by the shell would fall out of frame after just 2 loops, and by the second loop you wouldn't even be able to see the kart that threw the shell at it. Also half the fun of this Mario Kart idea was watching the same kart fall back a place again and again and again. If I use the mirrors, I'm just now realising that that wouldn't be possible. I might have to think about other ways to achieve the strobe effect.
Problem Solving in Planning 4.1
My 3D printer chose this week to start playing up - I tried to 3D print a Mario figure to do some test painting, but the entire print just exploded. The problem is there's a really thin layer of filament that gets stuck to the print bed. Thankfully I know someone at my church (and again, there are tutors at college that can help) who has some colleagues that are clued up with 3D printing, and that's been a help and they've given me some suggestions on how I could fix the issue which I am yet to try.
I loaded the above image of Mario into photoshop and changed the colour of the character to see what it could look like. I've obviously chosen red and black as they are the main colours that evil characters are dressed in.
If I were to have the character red, that would be a lot easier than having him black. I want to dry brush my models to get a good contrast between the shadows and highlights, and if I did that I could go over with a more transparent red paint to give it the red colour. However I worry that if I wanted the character to be black, you either wouldn'e see the dry brushing, or the whole model would end up looking more white (as a result of dry brushing). For this reason I'm likely going to paint my character red.
Problem Solving in Production 4.2
While modelling this kart I noticed that I wouldn't be able to stick entirely to the original design of the kart because if I did, there would be a lot of flimsy parts and it's highly likely that the thing would snap. I've provided some images here of the front bumper I designed compared to the original design. You can see how my design is more 'bulky', and although it's different to the original design, it's still very recognisable as Mario Kart which is the main thing.
I also noticed later on that the overhand on the model meant that there would have to have been supports for that area. So, I reduced the angle of the overhand and to be honest, I prefer this look anyway.
When it came to designing the engine I had a similar problem - if I stuck to the original design then there would have been some very thin parts which would've put the model at risk of snapping. I've again provided an image on the right of the original design. You can see that my design is again very 'bulky', but again it still looks like it's from Mario Kart.
The image on the left is the original design I had for the steering wheel. When I tried to put this model into my 3D printer slicing software (software you use to get the settings right for a 3D print) I noticed that there were a lot of supports on the steering wheel, and taking a closer look at it, I realised that this design was just stupid. There's a lot of air between the parts, and the parts themselves are very thin too, and of course when I 3D print this it's going to be a lot smaller than it appears in Blender. So I redesigned it, as you can see on the right, which works much better for 3D printing.
The image on the left has a layer height of 0.1mm, and the image on the right has a layer height of 0.2mm. The smaller layer height means that I get more detail, and you can see this by comparing the two images. It's most noticeable on the sides and very front of the kart.
I then clicked print and it didn't go too well. This was the first attempted print. This is likely due to a small amount of filament stuck to the print bed which I need to figure out how to get off. Once that problem is sorted though, the prints should work perfectly.
I fortunately managed to solve the issue with my print bed and again loaded the file into my slicing software. The print worked apart from 2 issues - the rear wheels snapped off when I tried removing the supports, and also the steering wheel didn't print properly. This was due to it being too thin and the supports, I assume, not being able to provide a good platform for it to rest on. The steering wheel issue is something I didn't notice until later though, so I only solved the issue with the rear wheels.
I made the platform that connected everything together more round around the wheels so that when the model printed, the wheels would actually be attached to the kart. The only thing I didn't really like about this was the fact that the 'suspension' if you like no longer was attached to the side of the kart, rather than on top of the platform, which doesn't look as nice in my opinion. I also loaded the model into my slicing software in the image on the right.
However the print was successful, apart from the steering wheel issue I mentioned earlier.
This is a comparison between the first and second model I 3D printed, with the first being on the left and the second on the right. I like the design of both, but the one on the right is just too flat for my liking. I'll try and redesign it as I've got to redesign the steering wheel anyway.
You can see here that I started designing the pedals. I also got rid of the steering wheel as it was just too thin and wasn't going to print. It's not an urgent issue as I may be able to 3D print the character holding a steering wheel. If not, I'm sure I'll be able to solve yet another problem and design a steering wheel that can print properly.
I also made the pedals at an angle of 30 degrees which means that supports aren't needed for this area of the model - in my slicing software there is a setting in which I can change the angle, and anything that exceeds that angle of overhang will be supported. I hope that makes sense because I'm not too sure how to properly describe it.
You may notice in the video that I downloaded a 3D model of this kart that someone else had made. This wasn't me cheating or taking parts from it, rather I set it up and took screenshots of it that I could use as reference images. I didn't exactly follow those reference images as I'm aware that because it's a model that someone else had made, it likely wasn't 100% correct. I also had to change bits anyway to accommodate for 3D printing.
Here you can see that the exhausts are slightly different to the ones I made on the standard kart - these ones are much bigger which came with its own challenges. Also because there were 4 exhausts rather than 2 it meant I had to get creative with the modelling. What I did was I put the top exhausts directly on top of the bottom ones, so that when it 3D prints, they don't need any supports as they're already resting on the bottom exhausts.
Once I'd finished modelling, I loaded the model up into my 3D printer slicing software. I noticed that there were some supports underneath the car, so what I did is I went back and made the bottom more flat. In the middle image below you'll also see that there was a support coming in from the left hand side which I didn't want, so what I did was I made it so that the hexagon, where the 'suspension' attaches to the car, had the exhaust resting on top of it, which meant there didn't have to be any supports there. I've added images below of the printed model with these fixes made.
It looks pretty cool and all, but unfortunately it was a lot bigger than I thought. On the right is a size comparison between the Standard Kart model and the B Dasher model. The B Dasher is huge in comparison, which would be a problem for fitting the animation on a zoetrope.
So it was back to blender. What I did is I removed the 'sidepods' from the kart, which meant I could scale everything down apart from the seat - I want to keep the seat the same size so that I may only have to design one character that can fit in both (if it comes to it).
You can see here that I removed the sidepods. I got rid of them so that I could scale the rest of the model down around the seat without the seat intersecting the sidepods.
Thanks to this adjustment, the B Dasher is almost exactly the same size as the Standard Kart. This means that the karts will not only look the right size next to each other, but they also won't collide in the animation due to their size.
I also made the exhausts at an angle of 30 degrees which means that supports aren't needed for this area of the model. You can see the result I had in the slicing software on the right
I finally started to get to the point where I can see everything coming together. With all the 3D models finished, I loaded up a new blender project and started animating. Below is a video of the first animation.
First Render (Looping).mp4What I did was I animated everything, and then took each individual frame and copied and pasted it into another Blender project where I laid out every single frame onto a mock zoetrope. I did this animation just to map out where everything would be. You can see that there is no movement in the karts, there's no motion in the character's arm when he throws the shell and it all just looks a bit rubbish. But I was able to work with this well and develop the animation, as it helped me develop my understanding of how an animation works on a zoetrope.
One problem I had was that when the B Dasher was hit by the shell, it was facing in a really odd direction. It genuinely looked like the two karts were travelling in different directions. Because I've got 20 frames, each frame has to be (360/20) degrees apart, which works out to 18 degrees. Because the B Dasher is effectively one frame ahead of the standard kart, because the whole thing is arranged on a circular platform, the B Dasher has to be rotated 18 degrees, something I would never have figured out unless I loaded the animation onto a mock zoetrope first.
Zoetrope Test Render 1 (Looping).mp4(This piece of writing is taken from my evaluation of Week Commencing 28/04)
This week I started animating which was really fun. However I did very little work over the easter holidays and I'm starting to reap the rewards of that now - I notice how far behind I am on my written work - not that I'm that far behind, but it's just a lot to balance with the practical work I have to do as well. However the animation process went well this week. I found it very helpful to get an animation and then take each frame and make a 3D zoetrope in Blender. This helped me understand the zoetrope better and things I need to consider when making an animation for a zoetrope. Things like rather than the kart getting hit and falling straight back, it had to rotate about the centre of the circular platform by 18 degrees (as there were 20 frames I had to do 360 degrees / 20 frames = 18 degrees). I've now finished the original animation idea I had. But there's always a 'however' with this project.
However, when I measured my models and the 3D zoetrope I'd made in Blender, I noticed that the zoetrope was 1 meter in diameter. It got worse when I put my test 3D prints on the platform I'll likely use (that was kindly donated to the college) and realised that they were absolutely gigantic. So I went back into Blender, scaled the zoetrope down to 32cm in diameter (I also scaled down the karts to keep everything in proportion) to match the diameter of the real life platform, and again measured the karts. Now instead of being 8cm long, they were a mere 2.6cm long. When I'm 3D printing something that small, it's very likely that the quality will be awful and that the printer will struggle at that scale. Now to the climax of the however - I might have to make an entirely new animation, with a different amount of frames and with smaller karts than originally intended. One idea for the animation I've had is to just have the standard kart drop a banana behind it, which then hits another standard kart, and then the kart spins around and falls into the slot behind and then the animation repeats. This could potentially be done in slightly fewer than 18 frames, which means that the frames on the zoetrope will be physically further apart allowing the 3D models to be bigger which solves the print quality issue. With the animation also being slightly simpler and the projectile (previously the shell, now the banana) not having to travel as far, I could easliy drop the number of frames of the animation which would solve a lot of my problems.
So I had a problem with the last animation - there's so many different objects to print and I have like 2 weeks to actually get something working so I needed to change that up. So what I've done instead is I've made a slightly shorter animation and now there's only 2 different objects per frame, instead of 3. Basically printing is much quicker and there's less physical assembly etc to do which will save a lot of time.
Banana Zoetrope Looping Final.mp4This is also probably the 5th iteration of this animation - there were a lot of minor issues that I had to change, things like adding mounts to the karts, changing the movement of the karts and the banana slightly etc. Fundamentally the animation has remained the same though.
Measurements
Bad news
Just found out that my 3D printed models at the moment are currently 3X bigger than they actually can be, and they're already pretty small. For a bit of context, I decided to measure up my 3D render of a zoetrope in Blender and then see how big it would be in real life. It was 1 meter in diameter, which is a tad large. Thankfully though a few weeks ago a lady came in and donated a bunch of old animation toys and kits to the college, and in that was a platform for a zoetrope. The platform for the zoetrope was 32cm in diameter, a third of what I had in Blender. So I've had to now make my 3D models 3 times smaller just so they can actually fit on the zoetrope, and when they're that small I'm not convinced the detail will be very good at all. So I've basically got to wait and see how this 3D print goes and if it doesn't work I'm going to have a lot more head scratching to do on top of everything else. One thing I could do is to just do something like a 12 frame animation, and make it so that it's only 1 kart in the animation - sure it's not as impressive as what I had originally planned, but with how little time I've now got, it may be an option. Besides, the main point of this project is to get a zoetrope up and running, so even if the animation is a bit simpler than originally intended, I'll still be happy that I achieved my main goal.
Good news
The 3x smaller 3D print worked, and the detail is surprisingly good at this scale. One thing that really impressed me was that even the 'M' on the front of the kart is still just about visible.
Even better news
That was with a 0.4mm nozzle, but I've recently bought a 0.2mm nozzle. With the new nozzle being 0.2mm it means that it can output smaller amounts of filament which means I can get much more detail at a smaller scale. The only downside is that the models would take roughly twice as long as they would if I were printing with the 0.4mm nozzle, but I can just set the models to print while I'm at college and they'll be ready by the time I get home. The reason I actually bought the nozzle was because, although the quality was really good at this small scale with the 0.4mm nozzle, the exhaust pipes managed to end up being more oval shaped, and I thought there was a very high chance there would be inconsistencies with each model, but of course I don't run that risk with using a smaller nozzle.
Below is a comparison between the same kart printed with a 0.4mm nozzle and with a 0.2mm nozzle. The image on the right is quite poor quality but I'm hoping that you can still see the difference in quality.
0.4mm nozzle
0.2mm nozzle
However even this print wasn't perfect. When I printed all of them, because of the shape of the wheels and the angle of the exhaust pipes, it meant that there were still some supports. Because of how small it is, when I took the supports off it just naturally pulled of some of the filament on the main model as well leaving a pretty ugly finish. So I went back into blender and changed the exhaust pipes so that they are now sitting at a 30 degree angle (similar to what I did with the B Dasher model) and made the bottom of the wheels go straight down, rather than being rounded. Of course I still had to have a hole to put the mount in and I changed this also to a 30 degree angle which again means it doesn't need supports.
Here's the results in the slicing software - supports show up as green making them very easy to spot, and you can see here that there are no supports needed anywhere.
The beauty of this as well is that to print 9 karts at this scale without any supports uses up less filament than printing just one kart at the bigger scale with supports. This is great because the whole zoetrope is going to weigh pretty much nothing which means I don't have to worry about burning out the motor on whatever I'll use to spin it (likely a record player).
The Platform
The images on the left show the material that I used. Would love to tell you what it's called but I completely forgot but I do remember that I had the choice of this material or another one, but I went with this one because the other material was more prone to warping and could get damaged more easily. This material is great because it's solid, thick and also not too heavy, meaning that it won't burn the motor out on the record player.
Well it wouldn't burn it out if I was actually going to use a record player which provides a nice segway into the 8 millionth problem of this project. After 3D printing all the karts and all the mounts to the correct size, carefully measuring up the size of the platform in Blender and going through the whole process of getting it ready for the laser cutter, and finally making the platform which seemed like the last little hurdle of this project, I discovered that I made the thing too big to fit on a bally record player.
On the left you can see our lovely platform, and you might say that it seems like a pretty reasonable size for what we need.
And then on the right is how big I was actually supposed to make it.
Thankfully my tutors once again came in to the rescue and we have a couple of solutions. One is that we can simply just put something underneath the platform to prop the zoetrope up a bit so that it doesn't hit the needle on the record player. Another is that we use a turntable which college have. The turntable is just basically a spinning circle, nothing to get in the way of the platform. So it's not the end of the world and I should still be able to get the thing spinning and 'viewable.' Now I've just got to hope and pray that the mounts actually fit in the slots because if they don't I have no clue what I'll do. If I really really really wanted to get this thing to fit on a record player, I'd have to go and re-size the whole zoetrope (platform, karts, mounts, everything) and make sure it would actually fit onto the record player, 3D print everything again, and then go and make the platform again. That means in total I would've printed 45 karts just to get to the final product and that's not even including all the karts I made previously that were 8cm long instead of 2.6cm.
Assembly
I took the day of college to paint my models at home, and before I started painting I made sure that all the mounts would actually fit into the holes that I'd made in the platform (which I also took home). Thankfully they did all fit, however I noticed that it was hard to get all of them to stay at exactly the same height and to be perfectly flat. A solution I came up with was to have something either side of the kart while the glue was setting (I used super glue to set the karts and bananas in place), so that it would be perfectly flat. I managed to find some bits of wood that were funnily enough part of a pre-made zoetrope platform and they were the perfect height for what I needed. Here's an image to show what I did;
Because I used the same 2 pieces of wood for each kart, I knew that they would all be at the same height and perfectly flat, meaning that when I played the animation with the strobe lights and all, there wouldn't be any 'jolts' in the animation.
So with the zoetrope completed, the only thing left to do was spin it and somehow make a strobe effect. Because I'm not actually presenting the zoetrope for about another month yet, I don't have to worry about concerns with epileptic viewers, so we just used a strobe light that was donated to the college. We used a roll of sellotape to prop the zoetrope above the needle of the record player which meant everything was finally ready to go. Also the platform not fitting directly on a record player was almost a blessing in disguise as the record player's needle would've been right in the way of the animation.
Page updated
Google Sites
Report abuse