Wednesday, August 17, 2011

Shell Eco-Marathon Asia 2011 Technical ver.

Haha! Here's finally a post about me. Not Google, chrome or MS.

Anyway, while everyone else are enjoying their long semester holidays last June-July, I've participated in a prestigious yet very interesting and exciting competition - Shell Eco-Marathon Asia 2011!

Nah, that's just a demotivational poster I made. Haha! It's not about running 40km non-stop. It's about building your dream car!!!

Ok... Enough of those annoying posters I've made. Anyway, Shell Eco-Marathon is about building the car that saves the most fuel i.e. built a car that can travel the furthest distance per liter petrol. The competition was held from 6-9 July 2011. This is the ASIA level competition, so we're competing against 120 teams around Asia region!

WARNING: This blog post may contain a lot of engineering related stuff, you may skip the last half if you want to and proceed to the next post about the event only. I just wanna share the knowledge with my engineering friends after all.


And believe me, it's TOTALLY NOT EASY to build your own car at all. No one can really teach you what to do, because you don't learn them in your textbooks at all. You'll start from scratch, do your own research and solve every problem that you've come across by yourself. Well, perhaps some suggestions from friends, advisors and lecturers. As a preview, let's see what we've built!

"UPM Eco-Team"  and the car "Grevolution"!!!

Well, it's not really that great, but to achieve that with just 6 weeks of effort is really something!! Sadly, we get the budget from Uni just 6 weeks before the competition. Other teams started like 8 months-1 year before that.

We're a team of about 20+ people. Most of the team members are from the Mechanical Engineering Department. The rest are from the Faculty Rekabentuk. And I'm the only one from Chemical Engineering Department!

It was a compact 6 weeks of researching something we've never learned about and putting all those ideas into something real. Mainly the competition is divided into 2 category.
1. PROTOTYPE CATEGORY - where you build a 3-wheeled vehicle, for the sole purpose of saving the most fuel (i.e. not really suitable for daily driving).
2. URBAN CONCEPT CATEGORY - where you build a 4-wheeled vehicle that might be used as concept future car.

Then for each main category, there's small classes, mainly divided into
1. INTERNAL COMBUSTION ENGINE CLASS - i.e. your normal grass cutter, motorcycle and car engine that runs on petrol
2. E-MOBILITY CLASS - i.e. Electric, Hydrogen, Solar.

Our team is participating in the PROTOTYPE - INTERNAL COMBUSTION ENGINE, the largest category!

However, things are not that simple even if you are using normal engine. There's a lot of choice for the fuel your team can use. Besides conventional RON95, there's Diesel, Ethanol and FAME. Our advisor choose ETHANOL and that decision was made before I were even invited to enter the team. (I were invited only after they've obtained the participation rights from Shell, meaning submitted form approved.)

And what's more, it's going to be 100% pure Ethanol!!! (Your petrol are made up of mostly alkane) When I first heard about about using Ethanol as fuel, the first thought that pop up my mind was....

LOL... The problem with using Ethanol as fuel for normal motorcycle engine? Here's a 3-page report I wrote during the research. In layman term, Ethanol just DON'T BURN LIKE PETROL!! For those who can understand, it has high ignition temperature and worse still, its energy content is just half of petrol. I'll be back for the ENGINE part to solve the problem. For now, we'll go for the CHASSIS of the car. 

Back to the car construction, you'll need to build the frame or bone of the car - CHASSIS. The team used steel and welded them together to create it. I'm just so impressed by the Mechanical Engineering Team! They're so great at handling those machines!! MEMET is the boss in this aspect!!

~Chassis of "Grevolution"~

They're so good at handling those machine that I bet they can cut their nails like this~

Warning: Professionals Only!!

Anyway, besides the chassis, you need to work on the BODY (the cover of the chassis) as well. Well, at first, it was designed to be something like this...

That's an awesome design work from the students of Faculty Rekabentuk!! However, in the end, due to several complications, that it turned into something like this...

TROLLING~~ Lesson learned? There's always more than one approach to a problem! Just do whatever you can. In fact, I like the black design a lot! It's very cool, so cool that you have a 360 degree view in the car because it's 100% transparent plastic + cool tinted! People outside might not see you, but you can see them from the inside! And most important, the whole car is just 60kg max! That's really LIGHT~

Alrighty, back to my department. I'm in the Engine department. I'm also incharge of most of the calculation related task, be it energy or measurement. I'm also doing the car system and programming.

Back to the ETHANOL part, there's a lot of workaround for that problem, basically you'll need to MODIFY THE ENGINE. But with just 6 weeks of work period and with zero experience, it's just somehow impossible to create something good. To win this competition, the most important aspect will be the ENGINE. Your engine should be very efficient so that it won't waste energy.

At first, no one really did any calculations and I wasn't there the first week due to course trip. And the team just abruptly decided on the using Honda GX35. But that's one hell of a grass cutter engine, dude! How is that going to be powerful enough to move the car?! Of course, upon mounting the engine to the chassis, the car won't even budge.

However, nothing is really lost in the process. Using that small 35cc engine, I've learned A LOT about engine mechanism, such as how the CARBURETTOR functions, how the piston works, how does pressure variance affect fuel pumping, how does air-fuel ratio affect the efficiency of the engine and how POWERLESS it is using ETHANOL as fuel without modifying the engine. The energy output is just 60% of using petrol. And here's how gear ratio will help to increase the torque.

Shown above is how a 4-stroke engine works~

To test the efficiency of the engine, I've developed a few manual methods to test various parameters. No one can really teach you such things, and I tried my best to apply what I've learnt, mostly trial and errors. We don't have advance instruments to do so at all... Aren't we suppose to be a Research Uni?  We thus do everything crudely and manually. With the help of the team, we've managed to determine engine performance systematically!

Tested Performance Curve of GX35

And in the end, we decided to use Honda EX5 Engine. Of course, it's all based on calculations this time. It's also easier to modify since we can get spare parts easily and so we can destroy it as much as we want!!

And hell yeah, we've bought a completely new motorcycle just to get the engine!!

First thing first, test the Engine's performance!!

To those interested in the technical stuff, here's something you need to do if you want to use Ethanol as fuel in the conventional engine:

1. Increase the compression ratio
The reason is already stated in the report about ethanol. For chemical engineers, you'll learn this in your Thermodynamics. To do this, you'll need to modify the Block and Piston. (In textbook, you learn about compression ratio, but no one tells you how to reduce it. Only MAT REMPIT knows~) The details are difficult to explain without you seeing the engine. The concept is that you reduce the volume of the burning compartment to increase the compression ratio. We've doubled the compression ratio from 8:1 to 15:1, which is totally awesome!!

2. Adjust the air fuel ratio
Ethanol need less air for combustion in a stoichiometry sense when compared to petrol. If you use too much air, the air will cool down the engine so it'll be very difficult to burn. Too little air, incomplete combustion occurs. So it's a very tedious work of adjustment to achieve fine tuning. You'll want the engine to combust on the lean side, but not with too much air. The calculation for fine tuning can be done, using ODE and working with mass and energy balance. However, to do so, you'll need to determine the specific composition of the inlet and exhaust of the engine. Since we do not have such precise instruments to measure them, calculations are pretty much useless. So, we're just doing this tuning based on the sound of the engine (Mat Rempit Style~~~)

3. Change to hotter spark plug
Simple reason, since Ethanol is difficult to burn, stronger and hotter spark helps to achieve complete combustion.

4. Adjust the spark timing
Ethanol is difficult to burn. So we'll just make advance the timing, making the burning to occur earlier, so the fuel will have enough time to burn. This is very delicate work. And we didn't manage to really complete it due to the lack of time. But it was a good try after all.

5. Modifying the piston
If you've seen how the engine works in the animated gif before, you'll notice the valve is constantly moving up and down. Since we've cut the Block to reduce the overall volume, the piston will now knock against the valve when the valve is open, due to shorter distance. That's dangerous. The way to avoid knocking will be to cut the piston as well. And frankly, this is totally PRECISE ENGINEERING. We're working on 0.1mm scale, because the piston is only 2mm thick and if you punch through it... BYE BYE!! Total salute to EDMAN and REZA for completing this task!!

Punched 2 holes on the surface of the piston. Drawn using CATIA.

That should be most about engine modifications. Thank god we have PEDO on the team!! I'm doing all the calculations stuff, while he's the one directing the mechanical stuff. He knows so much about motorcycle that he did take out the whole engine, disassemble it into gears and screws, then put it back together again!! Without him and of course the Mechanical Team, my theory is just theory. He's not an Engineering Student though. Maybe he was a MAT REMPIT??

After that, you'll need to determine the performance of the car, so to calculate the km/L petrol. It's a bit complicated. It's related to writing out force balances, then solve the ODE individually while trying to maintain most of the parameters in the equations, so that we can find the constants based on test result and adjust the parameters to suit the situation. Physics students might find this simple, but I took nearly 3 days to derive it because I've never learn it before.

Mainly, the technique we use in the race, which most teams applied is called "Burn and Cruise". The vehicle starts from zero speed, then accelerates up to maybe 50km/h. That's the "burn" part because the engine is running. Then the engine is shut down so that the car can "cruise". When the car reaches a minimum speed, say 30km/h, the engine is turned on again. The cycle is repeated. This saves fuel.... a lot!

The derivations I did are here. While the discussions on various parameters are here. In short, I've derived the following equation to be used. I might not be correct, but that's the best after I tried. Anyway, my calculations says that we'll do at least 170km/L petrol. During the competition, our best was 190km/L. That's close enough~!

Haha! It's getting more technical?? Anyway, ABDUL and I tried to include an automated system for such "Burn and Cruise" process. The system will detect the vehicle speed and will turn the engine on and off automatically. We've managed to complete the circuit, wiring and programming, however it was not used at all in the end, because the time is not enough for us to install it. Well, nothing is lost, I've learned a lot on Electronic Engineering and C language at least.

Arduino Uno

Yeap, we used the board above, the Arduino Uno. It's basically a programmable board, with input and output. The circuit that I've designed is the one below.

~Image generated using Fritzig~

The code I've written for this circuit is available here. Just download and open it with notepad++ or something. It has basically a Hall Effect sensor to detect the speed of the vehicle, 2 relays to on and off the engine and some signals inputs and outputs.

It's kinda sad that we didn't get the chance to install it after completion. But it's ok, since I've learned something. Special thanks to SECK HOUNG and YIK SHAN for giving me guidance on the circuit and the coding.

Enough said on the technical part, we shall proceed to the next blog on the EVENT!!! If you're interested in the calculations or other technical stuff, feel free to comment here!! I'll try my best to answer.

Definitely, this event has enabled me to know a bunch of experts from the Mechanical Engineering!! They taught me a lot and I'm so glad I joined this!! It's a chance for me to apply what I've learned to reality, instead on pens and papers.

If you ever get the chance to join this competition, DON'T EVER MISS IT!!!

No comments: