Team 2085 Design Questions

[jmarsh24]

This is our 3rd year participating in the competition and our robot is coming along rather nicely. Questions that are plaguing our team right now is on traction control... The team has decided to do a idler wheel with an encoder along with encoders on the gear boxes. But people are thinking we will get false readings if we are trying to turn and the idler wheel wont be giving the right values that we need. I was looking for the teams that have gotten their traction control to work, to see if they could help us out. We are thinking we might need 2 idler wheels perpendicular to each other, allowing us to get the vector speed between the 2 spinning... Any help would be great.

Secondly, we are looking to do an optical sensor to start our ground conveyor system. We are trying to figure out which sensor would be the best for this. The two we are going back and forth on are a laser trip sensor, or a light sensor, also if teams who have used sensors like these and know places that have good adaptable or reliable sensors would be awesome.

Thirdly, we are designing our conveyor system, and at the moment our conveyor belt is a solid piece, and we are looking to do urethane belting and are trying to design the grooves in the rollers. We have 1/4 inch orange mcmaster belting and are wondering what has been a good grove or groove shape that veteran teams have found successful. I will be posting pictures of our robot soon!

[EricH]

Originally Posted by jmarsh24

Secondly, we are looking to do an optical sensor to start our ground conveyor system. We are trying to figure out which sensor would be the best for this. The two we are going back and forth on are a laser trip sensor, or a light sensor, also if teams who have used sensors like these and know places that have good adaptable or reliable sensors would be awesome.

This one should be pretty easy... <R02-D> bans all types of lasers.

For a light sensor, one way would be to place reflective material across an opening from it. If you aren't using the camera elsewhere, though, you could use it there.

[dtengineering]

Sharp makes a neat range of IR rangefinders such as this one http://www.robotshop.ca/sharp-gp2d12...-sensor-1.html You can order them in a range of sensitivities and so long as you are looking at something that is not particularly reflective (such as diamond plate) in the IR spectrum then you can usually get a good distance reading by simply reading an analog input.

In fact, now that you mention that, we should probably strap a couple of the ones we have on the shelf on to the front of the robot so we can do some rangefinding on trailers when we come in to dump on them.

As for the idler wheels, you are allowed to use omni wheels... VEX robotics makes some nice little ones (and an encoder to match up with them) and you could always use the AM omnis... they would run parallel to your drive train at all times.

Can't help you with the round belting question, though... this is our first time using it, too. So far all we've learned is that the 3/16" hollow core falls apart really, really easily, while the 1/4" hollow, 3/6" solid and 1/4" solid are darn near indestructable at their joints.

Jason

[jmarsh24]

Is there a certain shape we should use in our pvc guides?

For the traction control, has anybody else thought about creating a mouse set up of omni wheels to get correct turning values?

[MrForbes]

See if you can find some pictures of 842's robot Carmen, they found what looks like a good low-buck easy-to-build way of making the belt roller/pulleys, using rings of PVC pipe on a smaller ABS pipe. The belts fit between the rings.

[Greg Peshek]

As far as the groove/ separation thing, something we figured was an easy solution and are implementing right now.. is just putting the urethane belting directly onto the ungrooved PVC and using some old pneumatic tubing as guides to keep them in place.

The pneumatic tubing is just wrapped around the piece of PVC, wood screwed in at the ends and then we used CA (cyanoacrylate) glue to keep the middle of the pneumatic tubing from sliding anywhere.

It's a fairly easy and cheap solution. If you have a lot of left over urethane belting, you could always use those as guides too. Just go through the same process I described but with the belting. Of course the belting is probably more valuable than old pneumatic tubing.

-Greg

[jmarsh24]

For the sensor, which would be the most accurate and precise instrument to use, an ultra-sonic or infrared sensor?

[Bob Steele]

Originally Posted by squirrel

See if you can find some pictures of 842's robot Carmen, they found what looks like a good low-buck easy-to-build way of making the belt roller/pulleys, using rings of PVC pipe on a smaller ABS pipe. The belts fit between the rings.

We are using the same arrangement exactly... white PVC rings on Black (ABS) Styrene pipe.

We actually ran our shooter using this system at 3500 RPM last night and it worked great... although you have to be careful with alignment at this speed.
If you are collecting you won't need this speed and the ring system should work fine.

We did purchase the mini-welding kit from Fenner... for the tubing...
It is about 1/2 the cost of the fancy one from MC and works great...and I believe it is worth the price...

Here is the link... I ordered directly from Fenner and also got the belting for about 70 cents per foot instead of the 90 cents from MCarr...

http://www.fennerdrives.com/eagle_po...e_mini_kit.asp

This is the link to it... around $250 but worth it if you are using a lot of belting.

Good luck with your design!!

[=Martin=Taylor=]

Originally Posted by jmarsh24

Is there a certain shape we should use in our pvc guides?

For the traction control, has anybody else thought about creating a mouse set up of omni wheels to get correct turning values?

In '06 we just cut strips of wedge-top and pop-riveted it onto a PVC roller. The wedge top provided traction while also serving as guides.

One problem with using plastic spacers is that the objects will lose traction as they pass over that point. Might not be a problem.

Or you could use timing belts