Death Match Approaches...

Monday, March 5. I would like to formally apologize to our speed controls for being upset with them. It was our receiver that was the problem! With a new one, we should be good to go. Competition begins tomorrow, and C.A.N has a new paint job!! I'm getting excited!

OH, SPEED CONTROLS!!!

Friday-Sunday. I worked on C.A.N all Friday and all weekend. My conclusion, the speed controls are EVIL!!! They only turn on 5% of the time. On the bright side, everything moves smoothly in our robot. The problem is we can't move it with our remote. I gave C.A.N a sweet paint job though. I hope to have more pictures up soon, but right now it is 1:02 AM and I really need to un-fry my brain from dissecting my robot.

Oh, Speed Controls...

Wednesday and Thursday. So, our robot is working perfectly!!!!!... April Fool's! Our speed controls won't turn on!!! Other than that, our robot only needs to be attached to the wheels and it'll be done. But there's no point in that if it won't move in the first place. I'm at a loss at what to do.

The End Draws Near!

3/30/10. Monday and Tuesday back from spring break! Now our robot's taking shape. We did something... odd with the grabber. We attached a sort of bridge to it, making it tilt down instead of being flat to the ground. Ingeniously enough, it lengthens our grabber's height enough for it to be able to score in the bucket!!! Now we just need to modify our bridge, now the actual grabbing surface, to be able to grab balls.

Oh joy of joys! I would like to be frank though, I'm not terribly upset by this. In the, oh say, 20 seconds it took for me to go in back and grab the game bucket, Norm somehow managed to destroy one of our motors and tear one of the wires off the other. I'm still confused as to how that happened.

Lastly, nervousness is setting in. In two critical days before competition, Thursday and Friday, Norm will be gone!!! Ahhh!!! This means that I will have the singular task of connecting all our servos, the speed controls, and the battery to the robot and make sure that they are all working properly, on top of securing our gearbox to the wheels. All I can say, is that Norm has a certain confidence in my skills, which is reassuring. I shall prevail!!!!!!

Another update of what we did from day to day

On Wednesday I modified the servos. That's about all we did. Chuck was not there. I started rebuilding the grabber, but there wasn't much time left. Oh, and I helped Alec and Rob's team with their servos, for theirs were malfunctioning.

On Thursday Chuck was back! We finished rebuilding the grabber. It looks a bit uglier than it did before and it doesn't glide perfectly smoothly as it did before, but our tests indicate that it will work well anyway. We also prepared our gearbox for mounting onto the grabber.

Friday was an exciting day. Raghav continued to bother us with questions about whether our car would actually even go anywhere, so we duct taped the wheels to the gear box and drove it around a bit for him to see. After that little adventure, we got our grabber out and started messing with it. Over the last couple days, when messing with the grabber, we had to twist it slightly in various directions so that the gears would mesh. On Friday, we glued hunks of wood to it such that it was permanently twisted properly. These same hunks of wood then proceeded to attach the grabber to the gearbox. We slid the wheels onto the gearbox for testing purposes. Chuck got out the ruler. The whole thing is under nine inches wide. Yay! We knew the grabber would be entirely inside the radius of the wheel (except for some protruding pieces of wood that we would stand off once the grabber was complete) because we checked it periodically as we were designing and building it. It therefore was no surprise when it did indeed fit entirely inside the radius of the wheel, except for some protruding pieces of wood. I took three trips to the sander, and the end result of this whole processes was that the grabber swung freely inside the radius of the wheel but remained close enough to the ground to pick up a golf ball. We tested each element of the grabber with the assembly hanging from the wheels and it all worked as expected. Yay!

I have two remaining worries with regard to this project. The first is that the grabber, though inside the radius of the wheels, is not far enough inside. When both wheels press against the bucket, the bucket will protrude inside that radius between the wheels. It's possible that the grabber will still get caught on that. If it does, we will push the grabber back. If it still does, the motors in the gearbox should have enough torque to push our car back a little bit and force the grabber up. I hope.... They do have a lot of torque with that gearbox after all. My second worry is that the axles will slip inside the coupler. If this happens, I suggest that we clamp the coupler onto the axles. I have several ideas. More on those if we find our axles slipping. We probably will.

It's high time we make another blog update

It's high time we add another picture too. I don't have one at the moment, but we will probably have more stuff to add by Thursday. I am calling just to talk about what we did on Monday and Tuesday. On Monday, I don't remember exactly what we did, but I strongly suspect that we worked on the counterweight design. Our work over the last few days, up to and including Monday, was making tweaks to get the whole grabber assembly to fit within the radius of the wheels. By shortening the grabber itself and rebuilding it once, we have succeeded. Getting the thing to fit within the wheel radius was likely to be the one damning factor in our design; if the grabber smashes into the side of the bucket and gets caught on the way up, it ain't going anywhere. Another thing that we had been working on up until and including Monday was periodically asking around about why the servos did not turn for their full 60 degrees in each direction; they turned only 60 degrees total, that is 30 degrees in each direction. This was quite problematic indeed, for our design depended on getting the full 120 degree rotation from the servos.

On Tuesday we learned that the wireless control system simply refuses to push the servos beyond 30 degrees in each direction. We spent that day ripping apart our grabber, extracting the servos (that we never intended to make removable, by the way) so we could get them modified for a full 360 degree rotation. Actually this will work better anyway because the servo that tilts the grabber (to specify a bit more, this is not the one that opens and closes the grabber or the one that raises and lowers the drawbridge) will be able to just rest on an outcropping (that we will take a picture of today) instead of wasting energy by exerting a constant, unnecessary force to keep the grabber tilted properly.

Just a lil update

Well we have finished building the dealybob that is going to lift and rotate the grabber. We have bent the copper rod running through the wheels so that the whole mess fits a bit better within the width limits. I have not checked the blog recently for updates by Chuck, but I expect that pictures either have been uploaded or shortly will be.

To quote Mr. Podmers and several other people, "It's never too late for a major redesign." Our redesign is not late, nor is it major, so we're in good shape. The grabber we had built before was originally meant as a prototype, but we decided to include it in the robot anyway and reinforce it as necessary. We now know that decision was flawed and we are correcting our design as necessary. The grabber prototype was fixing to be quite high off the ground indeed.It would be possible, though difficult to grab a golf ball. This is because the gears linking the grabber arms are placed under the grabber arms in our current design. The gears are not shown in the image below, but they would be approximately where the text "grabber arm" is on each side, meshing together behind the golf ball.

One way to remedy the problem of the high-off-the-ground grabber arms would have been to tilt them from this sort of doohickey:

To this sort of doohickey:

However that change would be cumbersome and flimsy. Neither design would likely be able to lift a golf ball anyway, as the gears slip on the copper rods pretty darn easily (this was before we started building keys). Oh, let me take a moment to explain what's going on with those. Gears often slip on copper rods. It's a fact of life and a fact of this engineering class. We have developed a solution to this problem. Ya sand down the copper rod and stick it through a small piece of wood. Then screw an eyelet into the wood so that it presses into the groove in the copper rod. This wood is now locked in place; it's going nowhere. Simply bang on the gear as usual, and to glue it to this piece of wood, as shown in the picture:

Additionally, in our old design the servo was placed high above the grabber like so:


The center of gravity for the old design was high, so the servo that was supposed to rotate the grabber would have a hard time doing what we needed it to, even with a counterweight. The grabber arms have to be close to the ground, so a counterweight that compensates for a high center of gravity would either need to be below the ground or really heavy and so close to the axle that it collides with the various gears hanging out about the dealybob.

Therefore we have decided to scrap our old grabber and build a new one. The new one will have a servo close to the ground, glued directly to the grabber arm, which is glued directly to the gear. There will be no copper rods involved, except for holding gears in place, and therefore no potential slippage in the first place. Keys are too bulky for this sort of work anyway. Additionally, the counterweight will be more practical because it will not have to balance out such a high center of gravity in addition to the weight of the grabber.

3/4/10

Day.... Infinity. It's been a long time since I last updated! I was sick for a long time and then... I sort of forgot. Well, here's what happened. We basically finished our scooping device, only to destroy it to gain a lower center of gravity. Though I must say I did a good job with the axles for our wheels. I also hooked up our motor, it works, but the reverse is tricky; it takes twice as long for it to shift backwards than it does going forwards. We were coming along nicely, but then everything went the way of the exploding stick. Oh goodness, the underlining won't go away!!!!

Mathematics Update

We are indeed using some maths in our design. The servos provide 44 oz*in of torque. We are using three. One of them opens and closes the golf ball grabber. Because the golf ball will slide up a shallow ramp of card stock and the radius is quite small indeed between the servo and the golf ball, it is reasonable to simply assume that the servo will indeed lift the ball.

The second servo is positioned on the end of one of the grabber arms and actuates the draw bridge down which the balls will roll into the bucket. This draw bridge had a weight of 1.2 oz in the preliminary stages and its estimated current weight is 3 oz. Weight is distributed approximately evenly along the draw bridge and it has a total length of about 8 inches. It will be resting on the bucket while in operation so it is not necessary to include the weight of the golf ball in the torque calculations. Nor will the servo ever have to lift it entirely of its own power if all goes according to plan. However, assuming the worst case scenario, that the servo does in fact have to live the draw bridge, the torque on it would be 3 oz * 4 in (average length) = 12 oz*in, well under the rating of 42 at 6V.

The third servo is the problematic one. This one rotates the entire grabber assembly, which weighs quite a bit indeed. The unextended draw bridge is about 4 inches from this servo for a torque of 12 oz*in (when extended, it has an average distance of about 7 inches, but again, this servo will not have to hold that weight, for the extended draw bridge will rest on the bucket). The other two servos are placed at approximately 4 inches each from this third servo and at a 90 degree angle with each other, so the worst possible torque will occur when one is straight up and the other is straight out with a torque of 6.4 oz*in. The weight of a golf ball at 3 inches from the servo is 1.6*3 = 4.8 oz*in. The grabber assembly itself weighs about 6 oz and has a skewed weight distribution; the average distance of that thing from the servo is, according to the two-pencil method of finding the center of mass of an object, about 1 inch. The total torque we are putting on this thing so far is... 31.2 oz*in. Additionally, in order to get the 180 degrees of rotation that we need, we must gear the servo in a 3:2 ratio, giving it a torque of merely 28 oz*in. We will add more supports to the assembly, and the servo is clearly not powerful enough to lift it.

Therefore!!

To the servo that lifts the assembly, we attach a second rod in the opposite direction from the assembly with sufficient weight to offset the assembly and allow the servo to do its job.

The motors!

We need as much torque as we can get. When I built a gearbox in quarter 2 to help Mr. Podmers test stuff for this class, eyeballing the axle indicates that with such huge wheels as we are using, speed will not be problematic. Our design requires as much torque as possible, so we can not afford to engineer around speed in the first place. Knowing that the robot will move reasonably fast is quite sufficient for our purposes.

The previous torque calculations hold for the motors that drive the wheels, with the ratio of the radius with respect to the third servo to the radius with respect to the motors conservatively considered to be 8.5/6. Each small motor has a torque of 6.82 g*cm; conversion to oz*in, sending the torque through the gearbox, and multiplying by 2 (because we have 2 motors) gives us a total torque of 65 oz*in, more than enough to lift both the assembly and the counterweight.

2/22/10

Day Thirteen. Lucky day thirteen I guess, a hot glue gun exploded on our table right next to Mr. Podmers on Friday, forgot to mention that. But that was day twelve... huh. Anyway, we attached our motor-to-battery connector wires today, as well as put some finishing touches on our bridge and grabber. We also put on some axles to our wheels with massive amounts of hot glue.

2/19/10

Days Eleven and Twelve. Our grabber is finished! For the most part. We may have to tweak it every so often. Our bridge is nearly finished too! Everything is really coming together nicely.

2/17/10

Day Ten. The wheels on the robot go round and round! We learned how to calculate our motor's rpm today. We then turned around and went directly back to work on our grabber system. We couldn't get too much done with the half class left to us, even less so because the hot glue gun we selected first decided to start smoking. Still, it's coming along quite nicely.

2/16/10

Day Nine. Our robot still does not resemble a robot, but our components are starting to look like... components! We've very nearly completed our "cage" for our lift system, as well as the door through which the balls will fall. Also, our mascot, Herman the hot glue caterpillar, is coming along quite nicely. He's a big help, by the way.

2/11/10

Day Seven/Eight. The internet... has died! Until now that is. I was down from 2/11 till 2/15. That aside, we've been making progress. We started building our grabber, which we've strengthened with a lattice structure. We've even solved our weight problem with a counter weight on the opposite side of the robot. Our diligence is paying off!

2/9/10

Day Six. Our robot's wheels are cut, at least one of them is anyway. We've begun formal sketches, but it's really hard, since we're changing our minds every ten minutes. At this point though, it's better to brainstorm and continually scrap half-finished sketches than have a set sketch and go with it to the end, as it may block out even better ideas later. Like today, Jefferson revolutionized our grabber arm! It'll be incredibly long and house several balls at at time. We'll definitely win!.. I hope.

2/8/10

Day Five. We have completed the gear box! We've chosen configuration D, which has the most torque of all. Jefferson has marked off the wheels and wood components for the wheels. We're really seeing some progress!

2/5/10

Day Four. Our new robot design seems to be a ringer! We're planning on using huge wheels with a suspended body to not only get around fast, but to avoid all the inevitable flipping levers we've been hearing about. Jefferson will be posting formulas and such pretty soon.

2/4/10

Day Three. Mission... not accomplished. Although we got some clarification about routers and battery life, our bucket robot idea turned out to be far too difficult to build. The conveyor belt was the problem. Oh well. We'll sketch up some new ideas tomorrow.

2/3/10

Day Two. Norman was gone today... it's hard brainstorming group ideas alone! Though now we know that there's going to be a bucket, I'm sure we will be able to get some ideas through tomorrow.

2/2/10

Day 1,

We've begun brainstorming, but nothing definite yet. We're thinking of using high speed, lighter robot capable of making many goals very quickly. It'll also be easy to escape other robot's attacks (as we've heard most everyone talking about attack robots). Nothing's set in stone yet, tomorrow we hit the ground running!!