Sunday, August 4, 2013


9 weeks have past and we're leaving US next week, awhhhh......
So we managed to finish everything, I did the electrical system and now the one thing I hate most in the world is the hall effect sensor. I would never go for a brushless motor next time. It will just save us so much time in soldering all the electronics stuff. Anyway, the first sensor I did was totally screwed up and two out of three wan't working, so the motor speed was very slow. And I glued guned everything so I needed to do everything all over again T T. So I spent 2-3 hours on the new hall effect sensor to make it perfect and it worked perfectly!!!! The motor was spinning real fast, so fast that the whole body of the kart was suffering from a serious vibration.

The video of the motor spinning could be seen here. And I didn't press the throttle all the way down, so that's only 1/2 of the maximum power we could get. 

One thing we worked on a lot was the alignment of the chain, if we don't align the two sprockets properly, the chain will just fall out as the motor spins. And also the mounting for the motor because of the severe vibration and also the secure the small sprocket so it won't fall off as the motor was spinning. We didn't realize all these things until the motor started spinning real fast and all problems come out. So we needed to make a lot of adjustments along the way and finally we managed to make it work and made a perfect alignment and a good tension for out chain, as you can see in the above picture. Another good trick to see whether your sprockets are aligned is to spin the wheel freely and see if the teeth of the sprockets fall in the middle of the chain. 

So we passed all the inspections. The final race day came but a lot of unexpected problems popped up and we were fixing them along the way. Our first run, the throttle suddenly doesn't work. But we took the seat out and put it back on, it worked again.... Anyway, so we went for another run, and the kart stopped before the fist turn... Turns out to be a battery problem, but that was weird cause we charged our battery overnight and the voltage was ok. So we had to charge it again. But our charger got spoiled and when it hits the limit of the battery, the charger didn't stop and just kept pushing in power....So we blew two battery fuses.... So we replaced the fuse and everything worked fine! I double checked all the wiring and made sure there would be no open circuit.... We were good  to go, BUT!!!!! When jiawei went through half of the first lap, we were stopped by Charles and it turns out we never pumped our rear wheel!!!! NOOOOOOO!!!! And the rear wheel came as an assembly with the sprockets and everything, and the opening to pump air in was slanted in an angle that was so sharp that it looked like it was impossible to pump without taking everything out!!! And with a not fully pumped wheel, the mass of the human body would just deform the wheel and it will just spoil the motor and controller.... So we took it to the bike shop down at IDC but their compressor wouldn't fit the angle either. We then decided to really take everything out and pump it. But then our lovely classmate Xinhui MEOW~ gave us this magical air pump and we managed to find a way to push air inside!!! So everything was fine and we were on our way back to the race place, which is a nearby mit parking lot. BUT !!! Jiawei bumped into the wall as he drove out of IDC and then the chain came out!!! So we needed to loosen everything and put the chain back on. But when we finished that, the rest came back and the race was finished. 

I know I know that was sad and we never got a timing for an entire lap. But it was because that we finished everything quite late so we didn't have much time to test and the problems didn't show until the real race came :p....

But I'm still very proud of our kart and I want to drive it around now!!!!Though one thing we could improve is the steering, cause we're using the free aluminum bearing and it's the same material as the steering rod so there was a lot of friction going on. We should have bought some bronze bearings which were quite cheap actually. And the horizontal rod could be longer if we had enough money so it's easier to control. 

A photo of me with all the wirings I did. It finally worked I was so happy!!!!
More videos to come!!!!

Budget overview:

Item Number Model Budget:500
motor  1 turnigy aerodrive 6374-149 76.53
controller  1 KBS48121 149
battery  1 A123 75
rear wheel assembly 1 200x50 Rear Wheel Assembly for Chain Drive Razor E200 (Versions 5-27) 25.99
front wheel 1 200x50 (8"x2") Front Wheel Assembly for Razor E150 23.99
Pedal/Sensor 1 Thumb Throttle with 48 Volt Power Meter 17.99
Ball Joint Linkage Shielded, Steel, 1/4"-28 4 6058K25 11.76
Alloy Steel Threaded Rod Plain, 1/4"-28 1 92580A107 (3ft) 4.78
Aluminum 3/4" steering shaft 1 9056K693 (no use) 7.39
screw 1 92196A636 5.22
Brake lever 1 15.99
Switch (small) 1 3.99
Switch (big) 1 hella 13.29
1/2 aluminum tube (vertical) 1 1658T45 share Andrew (4 ft) 3.5
7/8 aluminum tube (horizontal) 1 9056K723 (1 ft) 7.39
brake cable  1 from charles  5.95
shaft collar 6 6436K14 19.8
bearing  4 6383K16 17.64
nuts 1 91845A031 (free!) 0

Now as I look back, I think about the first night I came to IDC and after the buffet, I saw all those motors, controllers and stuff, and I had not a single clue of what that was all about. But now I just can't believe we built this stuff. Two people with absolutely no clue of what all these things are about (like you need to pump your wheel :p). I still remember myself holding this turnigy motor and asked our TA Justin: what is it? HAHA.....

Everything is machinable. Famous quote of Charles! Thanks for being strict, now I feel more independent and I can actually do things on my own!!!! 

Drag race back at SUTD????

Oh yes, I love MIT :3 Let's bring this culture back to SUTD!!!!

Tuesday, July 23, 2013

6. Last week!

Time flies! We're only left with 3 weeks in US, and this Friday we'll have the rolling frame inspection :3..... So we submitted our waterjet file today and adjusted the seating and put everything together:
kart with no name 
As you can see from the photo, the right wheel is actually not on the kart. It was good at first, but after I sat on it for like 10 seconds, the wood inside the steering part broke... We think the 1/4'' aluminum plates would be fine ;p

This week we also need to finish the electrical system of the kart. Looks complicated. Starting tomorrow!!!


Friday, July 19, 2013

5. Final prototype

front frames not stable 
 Last week before the rolling frame inspection! We finally put the kart together !!!! We still met some challenges and we managed to solve them all - - .

We were arguing about where we should put our knees on the kart. Should it be on the intersection between the front horizontal frame and the vertical frame? Or back off a little bit and avoid that intersection? It turns out that it doesn't really matter. So we are avoiding that part cause that way is more comfortable for the driver. We tried to sit on the kart the other night, and the front frame just kind of sinked a bit... It turns out that we didn't fasten the joints enough. And it's probably because of the wood, and there is no aluminum plate at the top to squeeze the whole structure together.
Our solution was to secure a long horizontal aluminum frame at this critical joint. We cut the battery deck cause it doesn't really need that much space. So we used that part to do the new four joints.

We put on everything, newly adjusted front wheel s, back wheel, motor, batteries and controllers.

Looks pretty good, now all we need is the steering wheel and the seating !!!!

Friday prototype
Friday! We did the steering wheel and the test turns out to be pretty good. It's easy to turn but not that easy. And we also found the bolts and nuts to secure the controller to the battery deck, a way to secure the brake. I think we got the sensor mount too....A little problem was that the steering rod appeared to be a little bit shaky, it turns out that we missed one shaft collar, so we are moving three holes a little bit forward to avoid the shaft collar from hitting the aluminum frame. 

All we are left is the seating, which is quite tricky to secure. We bent some L brackets and the holes are very tricky to drill. which took us quite a long time. SHOULD TRY FINGER JOINTS NEXT TIME!!!

our clean and organized table - -
I've finished all the drilling and assembled half of the kart. Shall do the other half tomorrow and make some slight changes to solidworks files, then our mechanical system is done and we can start doing our electrical system ;p Oh right, haven't put the wooden support for head yet. We'll just use a shaft collar I guess.... 


Tuesday, July 16, 2013

4. Building and more cading

So we tested out our first prototype, and it didn't look good :( The triangular seating was not very comfortable and the joints were kind of splitting in the middle (probably because of the wood we used). But still not very safe. We wanted to do something different, so we started cading again and figuring out a way to make the kart look nice while still maintaing the kneeling down gesture. Jiawei has figured out ways to do the front wheel and also support for the motor. Our ideas evolved from initially a dove shape to a snoopy shape, and now we're settled on the phoenix shape.

We're bending the 1/8'' aluminum plate to make the entire shape, but we're still not sure whether we should split the whole shape into two parts and waterjet them on two separate aluminum plate or just waterjet the whole part on one plate. Still need to figure that out on solidworks.

.........(ONE WEEK PASTS)

We bent the 1/8'' aluminum plate to make the 'seat':

aluminum seat 
It turns out a bit different from what we thought before cause we never considered about the limitation of the metal bending machine..... LESSON LEARNED! There should have been two more bends to make the knee deck so we can just secure the knee deck to the aluminum frames sticking out from the side. So we changed the plan and decided to use the rest irregular pieces of the 1/8'' aluminum plate to make some bent L-shape brackets to secure this seat to the separate knee decks which are then secured to the aluminum frames.... So one of the top things on the list: cut the rectangles and bend the brackets!

And motor support after several changes:

And rear wheel :
Oh yes, and for the cushion, I was learning stuff at the solar ev team downstairs, and we were waterjetting pink foams last Friday night so there were some leftover foams and I borrowed one piece:

Now the only thing left is steering and it's something we're struggling the most. The first few prototypes all broke when we tried to turn and the recent one we did earlier today doesn't turn properly. We added in some washers in between the two round joints to fit in four wooden sticks so that the hole drilled in the middle doesn't make the structure too fragile. For the turning problem, turns out that we need to increase the torque by increasing the distance. So we did this prototype later this afternoon:
And it looks ok, but we still need to do a final test later tonight - -
Brake lever and thumb throttle on our aluminum tube ;p:
Another thing on the list is to bend the aluminum tube for the steering, that's the one in the above picture....

Our budget review till now: 
Item Number  Model Budget:500
motor  1 turnigy aerodrive 6374-149 76.53
controller  1 KBS48121 149
battery  1 A123 75
rear wheel assembly 1 200x50 Rear Wheel Assembly for Chain Drive Razor E200 (Versions 5-27) 25.99
front wheel 1 200x50 (8"x2") Front Wheel Assembly for Razor E150 23.99
Pedal/Sensor 1 Thumb Throttle with 48 Volt Power Meter 17.99
Ball Joint Linkage Shielded, Steel, 1/4"-28 4 6058K25 11.76
Alloy Steel Threaded Rod Plain, 1/4"-28 1 92580A107 (3ft) 4.78
Aluminum 3/4" steering shaft 1 9056K693 (no use) 7.39
screw 1 92196A636 5.22
Brake lever 1 15.99
Switch 1 F48-5595 8.99
1/2 aluminum tube (vertical) 1 1658T45 share Andrew (4 ft) 3.5
7/8 aluminum tube (horizontal) 1 9056K723 (1 ft) 7.39
shaft collar 6 6436K14 19.8
nuts 1 91845A031 (free!) 0
brake cable  1 downstairs?  453.32
So we still have 46.68 dollars left... We still need to get the brake cable and the switch.... And we haven't done the electrical part yet, though it looks like we have everything we need in the lab but still we are saving the money in case something breaks....

Two more weeks before the rolling frame inspection!!!! We were late for the waterjet submission for this week cause our steering problem suddenly propped out so we needed to fix that :p But we've finished the technical drawings and managed to fit the parts on one single sheet (1/4'') Shall wait till next Tuesday!! And hopefully we can put everything together and do a test run this week.

Wednesday, July 3, 2013

3. CADing and prototyping

dimension drawings....
solidworks model

This week,we received all the parts we ordered last week and have been busy CADing on solidworks. We kind of come up with the mechanical structure of the whole kart, how to build it with the materials we have (80/20 aluminum frame, 0.125'' and 0.25'' aluminum plate). After figuring everything out, this is what we come up with:

So we are still adopting the 'kneel down' gesture. And we decided to let the driver (which would probably be me) put her chin on the chin support structure (the tallest thing in the picture) And the person is sitting on the triangular-shpaed structure. We first thought that the two vertical aluminum frame would be a good support for the driver as he makes turns, so there will be some force given to the driver to avoid them from being thrown out of the kart. BUT, our first prototype proved that it would not work....

We laser cut the joints with plywood, and cut the aluminum frames. We made our first prototype of the chassis last night:
bike from the recycle bin, could be useful in the future?
parts we've ordered ;)

Jia wei working on the annoying screws
Everything looks so tiny when we actually make it....Anyway, lessons learned: don't design a structure that needs too many joints to connect. The screws are very annoying and the thing with those 80/20 aluminum frame is, if you mess up one screw, you may have to unscrew the rest and do it all over again....And the frames all have sharp edges so be careful! One problem we are running into right now is at some parts, the joints are not strong enough to hold the whole structure. We're guessing it's probably because of the property of the plywood and we didn't use the proper setting when we laser cut so some parts are burned a little bit and the whole wood is kind of crispy. Maybe it will get better with aluminum plate, if not then I guess we'll just have to buy steel plate....ALSO, always remember to delete those crosses on your dxf file before sending them to lasercut.....Or you may ruin the whole joint. I'll post a picture of our first physical prototype later tonight or tomorrow. We're also exploring different methods of joining parts together....And Jiawei drilled the holes on the gear to fit our motor and adjusted the holes on the washers, so we'll probably assembly the rear wheel today or tomorrow. Front wheel we'll just need to wait for our second BOM to arrive :p

Monday, June 24, 2013

2. Design and selection of parts

This week, we start to shop for the parts we need for out little go karts. We submitted our order last Friday, cause delivery takes time. 

A123 Battery 2S2P 

2S2P battery configuration gives us 24x80=1.92KW, while 3 batteries in series gives us 36x`40=1.44KW. So we'll go with four batteries and pay 75$ for the extra one. Both Jiawei and I agreed that motor is the most important part, so we were looking for those motors that can supply a super large amount of power. I was shopping around hobbyking, then I found some powerful TURNIGY ROTOMAX motors that claims to provide 2KW+ of power. We were very happy and decided to go with that. But Charles pointed out that the power would be a total waste for our go kart cause the maximum power our go kart can provide is only 1.92KW. And although on its data sheet says the maximum current is only 70A, in reality it would suck hundreds of amperes of current and just blow the fuse in the battery. So no ROTOMAX - -. Charles suggests SK3 line, we previously thought that it was not powerful enough, but it turns out that it is actually enough according to our calculation later. For our calculation.....both of us read the dcmotor pdf file so we tried to solve the problem in a systematic mathematical approach. We first wrote a nice EXCEL sheet and tried to fit everything in, and all the mathematical equations as constraints, like solving an optimization problem you know. BUT I think we complicated things too much, all the graphs, analysis, complicated equations...... Two days scratching our head, we decided to take the easy way out, for a rough estimation, we didm't take in the frictions and air resistance. Kv and Kt are negatively proportional  to each other, so lower Kv gives us a  higher acceleration for a fixed gear ratio. Cause gear ratio should be between 3 and 5 to keep things simple, so we set it as 5. With a system voltage of 24 V and a 149rpm/V motor, motor speed with no load =3576 rpm. Our wheel speed is 715.2 rpm with a gear ratio of 5. The wheel we choose has a radius of 0.1 meters, that would give us a vehicle speed of around 7.5m/s < 9.7 m/s. Gear ratio X maximum torque X rotational speed of the wheel = F X speed of the wheel. F=ma and maximum torque= Kt X maximum current. With the 1 minute current 120 A controller, our maximum torque became 7.68, approximate total mass to be 90 kg, we get an acceleration of 4.3 m/s2. Not bad. But with all the friction, we're still not sure how fast it will go, but it's the best we can get within the budget. So...But we 're still not sure about whether to use four batteries or three batteries, cause it seems that 3 batteries in series has higher voltage: 36v, and the max current would be 40A. That way we'll have a higher speed of 11.2 m/s, well that's a much higher speed! But we're still not sure how the power part is playing out > <  For controller we spend 149$ on the 120A KBS line, cause we want to increase our max current, then we still have space to adjust if it's too high and the fuse blows. 

Wheels! To simplify the searching process, we decided to use the same wheel given to us for free for all the three wheels. Razor E150, and it turns pretty smoothly so.... The wheels on Mcmaster are super expensive by the way. We chanced upon this rear wheel assembly, which includes a band brake and a sprocket for chain drive, and it's only 3$ more expensive than a single wheel. PS: we are still sticking to the chain drive plan. 

The batteries and controller would be mounted on a aluminum plate under the seat, with the motor stabilzied behind the controller. We've started doing bits and pieces in solid works. Most of the parts are done already but still need to assemble them. And we still need more aluminum frame to build our chassis, planning on ordering coming Tue. and we'll ahve the 2D drawings of the parts to be cut on aluminum plate ready by tomorrow's class so we can get the stuff back this Friday and start building things next week ;)

Sunday, June 16, 2013

1. Sketching and Idea Generation

For the first week of our EV project, we generate ideas about the rough shape and outline for our go-kart. We read through the websites and the blogs posted by MIT students before, the following are the sketches we did along the way.

From the very beginning, we've decided to use two non-driven wheels in front, one driven wheel at the back. Cause two driven wheels would mean two motors and two controllers. Plus three wheels compose a rather stable structure. When I did this very first sketch, I was thinking more of the aerodynamic shape of a race car or a bullet. So the driver would be kind of lying on his back as he drives the car. This idea was then abandoned during our first discussion cause we feel that the driver would risk a quite high chance of being thrown out when turning, and since the driver is lying down, he cannot really see what's going on in the front, but mostly the sky. 

We then evolved our second idea, in which we lift up the back of the 'seat' so that the driver has a clearer view of what's going on in the front. My teammate Jiawei then proposed that both of the previous ideas make the kart too long, which would inevitably increase the overall weight of the kart. And the gesture of the person still makes it quite possible for the person to be thrown out when turing, cause the driver's center of gravity is quite high.

So we changed the gesture of the driver and decided to make him lean forward, which is kind of like riding a motorcycle. We also tried to ride on the go-karts built by MIT students before and it works like a charm. At first I was thinking more like a motorcycle except the whole body of the kart is lowered down a bit so the center of gravity is low as well. But still, another question would be how to prevent the driver from being thrown out of the kart when turning as we all saw what happened to one of the teams in the Youtube video :p Both my teammate and I liked the kart made by Nick and amwang (I didn't find the kart's name in their blog), which would be kind of secure for the rider to ride on. So we decided to adopt the same gesture. We increased the distance between the two front wheels to increase the stability and also widened the body of the kart a little bit (where the driver sits) so the rear wheel could fit under the kart. We also moved the steering wheel to the very front of the kart so its rotational angle won't be limited when turning. Basically we moved the upper body of their kart backwards to hide the wheel and thus prolonged the part with which the driver would have physical contact. So the driver is still able to lean forward and not reach out of the kart.

Here comes our final sketch for the first week, it looks like those little horses that children would ride on.

We're going with chain drive instead of gear drive so there won't be a part sticking out from one side of the rear wheel thus leaving no space to put the driver's feet. Chain drive is also easier to build, so.... We're probably using the 8 inch wheel provided to us as the rear wheel. As for the design of the steering wheel, we compared the methods of the previous go-karts and finds the chibikart has the best maneuverability, so we are going with that for now. For the curvature part, we are still not sure how to do that. Jiawei pointed out that maybe we could just buy one of those plastic horses for children to ride on, then secure it to the aluminum frame. But will it be kind of heavy? And not very customized for our design. We're also not sure how well plastic and aluminum come together so.... Or maybe stuffed animals? a long pillow? a sofa cushion foam? Or the wood laser-cutting patterns used by the other group before?

The top view and side view are attached below:
*All of the three drawings are in scale. 1'':0.5cm

Top View

Side View 

Note: We haven't decided on the wheel size yet. In the above picture, the rear wheel looks much bigger compared to the two front wheels...which may not be true as we proceed to next week. And we've decided to move the wheel out from the chassis a bit cause it doesn't look very safe intuitively.  We'll look into that later. Btw, we are planning to put all the batteries, controllers etc. on the chassis right in front of the rear wheel.