Networks: Paths and switches

Today Ben is in class, back from his trip to the UK. So now that we have our director back we can run through our idea and make sure that he's ok with it too. He's supportive of our stirling engine idea, so that's good.

We drew up a basic diagram of our network. We'll be doing it with four points to show how it works.



Basically, at each point you have one stirling engine. Which brings up the packages in its containers. As it rotates, the package is dropped on to a track which branches out into three tracks going to the three other points.

At each point you have three tracks sending packages out, and an additional three tracks receiving packages from the three other points.

To get two different directions at one point there are two options to do this:




One is to attach two additional gears on the other side of the main gear so that the directions can be opposite to each other. Option two is to simply stack an additional stirling on top of the first one, but facing a different direction.

Personally I think there's a better way to do that, but will be thinking a bit more with the group tomorrow. At the moment the visualisation of outgoing and ingoing packages as lines is a bit fuzzy and spaghetti like. So tomorrow we'll be marking out these lines and points with chopsticks and string to see where our tracks are all gonna go. It would also make it clearer for us too.

About the branching out of tracks, there's a way to do it.



The idea that we have for it at the moment is that we have these cylindrical capsules that have a notch on the left, right, or no notch at all. The notches will activate a pivot switch, redirecting the capsule's track to a different one that leads to its destination. The one without a switch remains on the original track towards its destination.

It's still just an idea though, because we're not entirely sure the specifics of how we're going to get these pivots to work yet.


Here are the things we need to keep in mind:

-The pivots/switches have to be sensitive to work properly

-Need to work out how to integrate the switches on the track

-Must have something before the branching of tracks to slow down the flow of packages. This is to avoid jamming the path with too many packages.

-Track has to be designed to keep capsules in place so that the triggers for the switches would work

-Capsules must have sufficient weight to gain enough momentum, since the tracks will be sloping, letting gravity help the capsules move.


After talking with the group we went separate ways to find more information on how we could do this.

For me, I think our tracks should all be straight slopes going down to the destination. It's the switches that I want to find out more about to get the branching out to work properly.

Yesterday I blogged about this marble machine video I found. In it, they've got this weight-sensitive flip flop mechanism that I think would go well with our switching. I have a fair idea on how it works already, so I went to find something that would slow down the flow of packages.

This is what I found (Denha, 2010):



I thought this was pretty cool. I like how it directs the marbles on to different paths and then stores the marbles until there's another to set them all down to a different route. I'm not sure how much it can slow down the flow of marbles, but it's definitely interesting. I might be worth testing out, especially because it's not that difficult to make.



__________________________________________________________________

Reference List


Denha. (2010, December 22). RS-FF + AND logic [Video file]. Retrieved April 30, 2012, from http://www.youtube.com/watch?v=Huspz20MTDE&feature=related