2003 Toyota Celica T-Sport, 117k miles, 1.8L. 189 HP @ 7800 RPM
Repaint & restoration - my first time spray painting!
I picked up a 2003 Toyota Celica T-Sport as a summer project! Mechanically, it's mostly fine but the paintwork is where the most of my work will be in. It's got 117k miles and was sat parked for 12 years, so it's had quite a lot of damage from the environment.
I plan to fully repaint 8/10 of the body panels by myself, the front and rear bumpers are okay. I will need to:
Sand the car down
Repair rust spots (using rust converter) and fiberglass body filler if needed
Body fill any dents (sanding after)
High Build Primer (sanding after again)
Sand more
Basecoat
Clearcoat
Sand again (you guessed it!) and polish
Interior detail
Few bits and bobs that need replacing
This is my first time ever using a spray gun. I bought a compressor and the full setup I will need. It's going to be a HUGE learning curve to overcome, painting a car is not easy, plus I'm doing it outside. The paint is metallic so it's harder to get right too, but I love making myself struggle so here we go!
Wow. This is HARD.
The last month has come with plenty of pain for this project.
To start with, it took time to learn how to set everything up for paint, as well as mask the car. I watched lots of videos on it all. Oh, and cleaning the gun and area too. I bought two different types of paint thinner, a cellulose-based thinner and a 2k thinner. I am using the cellulose thinner to clean the gun out since it is cheaper.
I spent around 3 days sanding, including sanding rusty areas to bare metal and using rust converter, as well as using body filler.
After a few long days of failing to lay down the high build primer, I spoke to a professional at a local paint shop when I was buying the basecoat. I was using a 1.6mm nozzle, which is recommended for my high-build primer, but it was spluttering everywhere making the finish horrible and coming out inconsistently. He advised that I switch to the smaller 1.4mm nozzle, and told me that it doesn't really matter how much thinner I use to get it to come out of the gun right. It just meant I needed to do more coats. After switching the nozzle and adding more thinner, it laid down REALLY well!
Now, time to talk about the basecoat.
FIRST ATTEMPT
My first panel to tackle was the bonnet, which failed due to sandpiling. Essentially, the paint looked rough and sandy, which is caused by too high of a pressure and the gun being too far away.
No problem. I tried to sand the clearcoat down (I learned now, if the base coat doesn't look right, clearcoat doesn't fix it.), but it gummed up my paper too much. It turns out that the cure time of 48 hrs only applies at the specified temperature ranges, so overnight since it is cold, it is not curing. I will tackle the bonnet later when I have finished the rest of the car, I thought.
SECOND ATTEMPT
This time I was tackling the small rusty area behind the wing mirror, which was very poorly taken care of by a previous owner.
Since it was a small area, I used the spray can of high-build primer, and sanded it down smooth. It always looks so great after primer.
I lowered my air pressure, and moved the gun closer. I was very nervous since my previous attempt went so poorly, but I moved on anyway.
The paint laid down really nicely and it looked great. Then, all of a sudden, it started melting away in one go. As it turns out, the spray can primer was cellulose-based, and unless you let it cure for a few days FULLY, adding a 2k-based basecoat on top will remelt the primer back down. WHAT A DISASTER. More sanding needed to fix this one, and I will stick to the 2k primer.
THIRD ATTEMPT
Primer-ing the rest of the car went really well, as well as redoing the rusty mirror area.
I was all set to do the base coat. This time, I was ready to do the rest of the car in one go, since the basecoat went down so well before melting away.
I lay down my basecoat, but it isn't sticking properly. It was landing wet, not sandy, but it was clumping up as if the surface was hydrophobic. I continued with the light coat, hoping it would get better. It did, but it's not up to standard still. There is still minor sand-piling, which really confused me. If I'm honest, I did not touch the car for the rest of that week and afterwards the rain settled in so I couldn't paint. From my understanding, it was way too hot outside for basecoat, since it was around 25 outside, so it was drying in the air partially still before landing on the panel.
As for why it wasn't sticking to the panel initially, I believe it can be due to three things.
Firstly, I was using the cellulose thinner as a degreaser (I read online that this is okay, but it might not be after all). Apparently, this is not a proper degreaser, so there can be plenty of oils on the paint that were not coming off. Instead, I will use Isopropyl Alcohol since realistically the only oils on the paint are from my hands, and soapy water + isopropyl alcohol will get rid of that.
Secondly, the primer may have needed longer to cure after sanding. When the basecoat lands on it, it might be partially softening the primer, making it not stick. So I will always wait at least 24 hrs next time.
Thirdly, I believe I may need a primer sealer. I have seen plenty of people using sealer and not using sealer, but maybe I need it for my case.
Altogether, I've spent around 70 hours on this project over two weeks and not a single panel is done. I'm still motivated to finish it, and in fact I'm really enjoying the process, but I still have lots to learn until I'm able to figure it all out.
Apprentices were selected to present their designs (following a competitive internal selection process) to an industry panel comprised of colleagues from Nikken, the Children’s Hospital Charity and Finally Agency.
We were tasked to completely redesign the snowflakes in terms of their mounting mechanism, lighting, material selection and manufacturing method.
Example of current snowflake
Some of the challenges with the current snowflakes include:
The current snowflakes were very heavy, making installation difficult and a safety hazard
The lighting was done via filament bulbs, which was expensive to replace and not reliable compared to new LED technologies.
The sign changes required the whole snowflake to be taken down, reducing flexibility for donors
This group came second in the competition.
The group presented their new snowflake design to a panel consisting of representatives from; The Sheffield Children's Hospital, Nikken and Finally Agency (a marketing organisation working with the Children's Hospital). The group had responded to a brief whereby issues with the existing snowflakes were shared and the apprentices had to both address these points and innovate the design moving forwards.
The group presented confidently and articulately with all members playing an active role in showcasing their design concept and process. The group were scored very positively by all panel members with feedback including;
Blending traditional Christmas themes with modern development and creativity resulted in a really lovely design. Great mockups of the snowflake on a wall, a strong rationale presented for design choices.
It was clear that all members of the team had put a lot of work into these ideas, with all members of the team taking part. Adam was incredibly engaging, warm, entertaining and did a fantastic job of leading the room. This is a really impressive skill, and massively backs up the brilliant ideas presented.
Definitely innovative method of manufacturing, Acrylic usage is easy to machine or laser cut but not sure how feasible the size of the prints would be for the LBPF machine, also need to expand on how the Chip-On-Board LEDs are connected. Excellent start but CAD drawings would need to be improved for manufacture.
Loved the depth they'd gone into thinking about the lighting. Obviously some learnings still to be done but an excellent start.
Congratulations to Martha, George, Haaris, Adam and Hassan who did a great piece of work and represented themselves, their employers and the Training Centre to a very good standard.
It's been around 4 years since I last touched Blender, but my notes from back then came in handy during this project. I remembered what I can do within Blender but forgot how, so it wasn't so bad.
This was a 2-day long project to supplement a longer project for Sheffield Children’s NHS Foundation Trust.
Every year organisations donate towards the hospital and their names are put onto snowflakes, which can be spotted around Sheffield on the hospital buildings. They wanted a complete redesign, and I took some time to make nice renders in Blender for the presentation I will be doing for the staff, showcasing our new design.
What I achieved
I'm most proud of the bricks, which are entirely procedural. So, any configuration and length of wall will be adapted to and it will always look great! It may come in handy on a future project.
There are lots I could spend more time on tweaking, mainly lighting and materials. However, I am happy with the outcome overall and it's re-ignited the desire to make some more renders and animations again.
What it is:
The goal was to alter the way the brake pedal functioned on my sim rig. Previously, the position of the brake pedal was read by a potentiometer, meaning the further you pressed the brake pedal, the more brake pressure was applied in-game.
This has a few disadvantages:
You lose out on a realistic brake feel - in real cars brakes use hydraulics, where the force applied directly correlates to the force the brake pistons apply to the pads to slow you down.
Potentiometers wear out over time, leading to less accurate and sometimes incorrect readings.
Inconsistency in races - since the position of the brake pedal depends on your hips, legs and ankles, applying the same position for the corner can make you less consistent in the way you brake, making it harder to maintain lap times during a race.
What the load cell mod fixes:
Solid state (zero moving parts)
Realistic brake feel - based on force not position
Brake data is no longer proprietary to the pedals, making logging easier for analysis after races
Easily customisable braking curves using code
More consistency and faster lap times
I was very comfortable with this to begin with, and it was interesting seeing how everything worked inside. Originally I attempted an off-the-shelf solution using a third party load cell amplifier, but I had a lot of headache getting that to work so came up with my own solution instead.
This was my first time using an ESP32. Luckily, I was able to code it using ArduinoIDE which I have some experience in. ESP32 automatically connects to the internet and has bluetooth, all in a smaller and faster package than an Arduino. I utilise this functionality to give myself live graphics (see below!) to view my brake pressure
To get the placement right and to connect the conical rubber piece to the load cell I needed an adaptor. It was much quicker to 3D print it myself. Since the part is under compression, the prints should hold up just fine.
I wanted the fitment for the parts to be just right - I struggled mostly here with getting the thread right for the conical rubber piece. It needed specific tolerances to allow it to thread in without too much resistance which weren't standard.
The main two challenges with this project was 3D printing threads and designing a holder for a freeform shape.
By printing the threads large (ISO 22x2) the print quality was better. However, I learned the importance of using more common threads since I needed a specific nut to thread through to correct print imperfections - luckily someone in the workshop had a thread file and this was fixed. A bit of graphite powder from a sacrificial pencil and I was able to spin the whole part around the thread with one hand!
I did not have a model or schematics for the controller so this took a few iterations to get right, but I got there in the end!
Using SR 30 soluble support and ABS model material on the Uprint SE Plus
Using SCA 1200 cleaning station
Something that is print-in-place requires no supports and no assembly to 3D print in FDM printing.
This means all of the mechanisms are contained within the body, and the idea is that you can take it straight off the print bed and use it!
Here I am checking crucial areas such as the hinges – making sure they aren’t printed to fuse
Unfortunately the UPrint SE Plus is an older printer that requires a raft, otherwise this print would have printed completely without supports or assembly