SPECIAL PROJECT Design a Robot to do a Real-World Job!

[JimKelly]

Reader Dan M. has an interesting challenge for The NXT Step readers.  He has a manufacturing setup that he wants to replace and is considering using an actual NXT robotics kit.

Right now, the machine in question uses color concentrate that comes out of a large container/hopper.  When the color needs to be changed, the hopper must be cleaned so the colors don’t mix.  Dan would like to replace the large hoppers with much smaller containers that can be changed very quickly. 

Your job?  Help Dan design an NXT device or robot that can replace the color concentrate system by picking two or three of these color pellets at a time from a box or container and transferring them to a funnel system that feeds into a hopper.  Now, when Dan needs to change colors, the hopper doesn’t need to be removed and cleaned – instead, he just replaces the color pellet box that the robot will pick from.  This process will need to run 24 hours per day, 5 days per week.

Interested?  This is a special forum discussion area where you can post comments, questions, and maybe your solution!  Dan has agreed to provide feedback as well as pictures of the final design if one can be found to do the job reliably and repeatedly.  (And The NXT Step will give the selected robot/device plenty of coverage – we’ll definitely want to interview the designer(s) of any selected device.)

[JimKelly]

Pictures and schematics:

[Jonathan]

What size are the pellets (diameter)? Also, let me see if I understand this correctly... all we need to do is make a robot that grabs a few pellets out of a box and drops them into a funnel (which would then drop the pellets into the proper location) continuously? How precise does the robot need to be in grabbing the right number of pellets? How precise does the robot need to be in the frequency it drops the pellets off?

 -Jonathan

[dmccrory]

Hi Jonathan,
The pellets are a bit bigger than 1/8" diameter, and a bit oblong.  I just measured some, a typical pellet is .154" x .130".  Their weight is almost negligible, .00009 ounces each.

Yes, the robot has to pick up two or three pellets from a container (preferably a small clear plastic container) and deliver them to the funnel which will feed them into our machine. 

The system needs an average of 3 pellets every 2 seconds.  If the quantity is higher, then we're using more than we need (these pellets are very expensive per pound) and if the quantity is lower, then the color will not be as vivid as we need.

The same goes for the frequency of time, if too much time passes without pellets, there will be inconsistency in the color of the product coming out of the extruder.

Dan

[Jonathan]

Ok thanks... are you sure you want to limit robots to one retail kit? That might shoot out a lot of solutions, and you could simply require participants to list any extra parts they use and show how you can obtain those parts. You could also take the amount of extra parts into account when choosing the winners.

 -Jonathan *has a really neat solution that needs a few extra parts*

[basicxman]

I have an absolutely great idea, do I have to build it or just explain it?  It is kinda hard to explain so I might just build it.  Could I PM the project holder my stuff so no one else can see?  Sorry but I'm very suspicious of copy cats *shifty eyes

[Robolab 2.9]

Jonathan, I think the reason they are limiting to one retail kit is because Dan only access to one kit. But a couple parts might be ok, you'll have to wait for an official answer though.

Richard

[dmccrory]

Ok thanks... are you sure you want to limit robots to one retail kit? That might shoot out a lot of solutions, and you could simply require participants to list any extra parts they use and show how you can obtain those parts. You could also take the amount of extra parts into account when choosing the winners.

 -Jonathan *has a really neat solution that needs a few extra parts*

I'll consider designs that use extra parts.  Keep in mind that we're looking for simplicity, not complexity.  This is going to be in a real factory setting, and if the need arises, we'll want to easily repair and/or replace. 

[dmccrory]

I have an absolutely great idea, do I have to build it or just explain it?  It is kinda hard to explain so I might just build it.  Could I PM the project holder my stuff so no one else can see?  Sorry but I'm very suspicious of copy cats *shifty eyes

Hi, sorry, but the designer will have to verify that the device has worked continuously for at least 24 hours. 

Dan

[Jonathan]

Ok great... my idea should be pretty simple.

 -Jonathan

[dmccrory]

I have an absolutely great idea, do I have to build it or just explain it?  It is kinda hard to explain so I might just build it.  Could I PM the project holder my stuff so no one else can see?  Sorry but I'm very suspicious of copy cats *shifty eyes

Yes, you can pm or email me the project details, but keep in mind that we plan on posting the winning designs.

Dan

[basicxman]

How are we going to keep batteries going for 24 hours without the rechargeable battery!?!

[dmccrory]

How are we going to keep batteries going for 24 hours without the rechargeable battery!?!

If you don't have rechargeable batteries then I guess you'll have to test it for as long as you can, and then make a judgement call as to whether you think the design is robust enough for the application. 

[Jonathan]

Or you could quickly change the batteries at different intervals of the day (except during the night)... that might get you the 24 hours if you woke up early and went to bed late.

 -Jonathan

[dmccrory]

If any designers want to get some of the actual pellets, pm me an address and I can send some to you.  That may help add some realism to your experimentation.

Dan