projects:mechanical_logic_from_common_materials
Differences
This shows you the differences between two versions of the page.
| Both sides previous revision Previous revision Next revision | Previous revision | ||
|
projects:mechanical_logic_from_common_materials [2013/12/18 20:09] cjoyce |
projects:mechanical_logic_from_common_materials [2013/12/18 20:29] (current) cjoyce |
||
|---|---|---|---|
| Line 1: | Line 1: | ||
| + | ====== Abstract ====== | ||
| + | |||
| + | For our Computer Architecture final, we (Chris Joyce and Brendan Ritter) created mechanical logic gates as part of a lesson plan on computers for middle-school aged children. Detailed instructions on how to build the gates described in this document can be found at the bottom of the page. | ||
| + | |||
| ====== Documentation ====== | ====== Documentation ====== | ||
| Line 20: | Line 24: | ||
| * More popsicle sticks | * More popsicle sticks | ||
| - | From these unassuming materials we were able to make gates for AND, OR, and NOT and came up with an easy method for connecting them, giving us the ability to make any other basic gate. | + | From these unassuming materials we were able to make gates for AND, OR, and NOT and came up with an easy method for connecting them, giving us the ability to make any other basic gate. An example of connected gates is below, in both states -- an OR and a NOT connected to make a NOR gate. |
| - | ==== Why isn’t there a full adder like promised? ==== | + | {{:projects:2013-12-18_20.15.49.jpg?nolink&600|}} |
| - | + | {{:projects:2013-12-18_20.15.59.jpg?nolink&600|}} | |
| - | One of the major flaws in our schedule was our optimistic projection on the time it would take to make a gate. We estimated that it would take 15 minutes, putting a full adder (depending on implementation) at a little over 2 hours. | + | |
| - | + | ||
| - | However, in actuality, each gate, if it was constructed carefully (so as to maximize efficiency of operation), took almost an hour. This meant that making a full adder would take 10 hours. | + | |
| - | + | ||
| - | When you factor in the design time of the gates and the decisions of form factor, prototyping of possible designs and documentation, this was too lofty of a goal in the time remaining. | + | |
| ===== Why did you do it? ===== | ===== Why did you do it? ===== | ||
| Line 41: | Line 40: | ||
| ===== How did you do it? ===== | ===== How did you do it? ===== | ||
| + | We made our gates combining the above materials and puzzling over them for several days until we were able to come up with basic, robust, and repeatable logic gates. | ||
| + | |||
| + | ==== Not Gate ==== | ||
| + | A **zero** for a NOT gate is the input stick fully out. This causes the reciprocating linkage to push the output forward, creating a **one**. | ||
| + | |||
| + | Similarly, when the input is pushed forward in a **one**, the linkage transmits the forward motion into rotational motion and out again as backwards motion on the output, causing a **zero** for output. | ||
| + | |||
| + | ==== Or Gate ==== | ||
| + | A **zero** for either input to the OR gate does nothing so the output is pulled back to make a **zero** also. | ||
| + | |||
| + | Similarly, when either input is pushed forward in a **one**, the input stick pushes the bar forward, causing the output to also be pushed forward as a **one** also. | ||
| + | |||
| + | ==== And Gate ==== | ||
| + | If both inputs to the AND gate are **one**, the rubber band pulls the bar forward making output a **one**. | ||
| + | |||
| + | If either of the inputs is **zero**, the stop at the end of the input stick pulls back the bar connected to the output, dragging it against the rubber band to make a **zero**. | ||
| - | We made our gates combining the above materials and puzzling over them for several days until we were able to come up with basic, robust, and repeatable logic gates. | ||
| We used the Olin principle of spiral learning pretty heavily, because it became pretty clear early that we didn’t really know what we were doing -- so we tried, failed, and tried again. | We used the Olin principle of spiral learning pretty heavily, because it became pretty clear early that we didn’t really know what we were doing -- so we tried, failed, and tried again. | ||
| Line 58: | Line 72: | ||
| * Another gate that would be necessary in a larger construction would be a signal splitter. It would take in one input, 1 or 0 and output two 1s or two 0s. This would allow us to have more complicated circuits. Currently, if you only build in one dimension, you can only connect a gate to one other gate. | * Another gate that would be necessary in a larger construction would be a signal splitter. It would take in one input, 1 or 0 and output two 1s or two 0s. This would allow us to have more complicated circuits. Currently, if you only build in one dimension, you can only connect a gate to one other gate. | ||
| * Decrease throw of gates. This would enable us to make more efficient use of space. The downside of this is it increases the chances of errors or noise in the system occurring. | * Decrease throw of gates. This would enable us to make more efficient use of space. The downside of this is it increases the chances of errors or noise in the system occurring. | ||
| + | |||
| ====== Workplan Reflection ====== | ====== Workplan Reflection ====== | ||
| + | |||
| + | ===== Adherence to Schedule ===== | ||
| We stuck to our workplan fairly well as far as time spent goes -- we actually spent slightly more time than we planned to. It was a little more backloaded than we anticipated, but that happens with all projects, and this wasn't true to an appalling degree, only a small amount. | We stuck to our workplan fairly well as far as time spent goes -- we actually spent slightly more time than we planned to. It was a little more backloaded than we anticipated, but that happens with all projects, and this wasn't true to an appalling degree, only a small amount. | ||
| + | ===== Re-scoping of project ===== | ||
| + | |||
| + | We changed what we did slightly throughout the course of the project, as we realized that our initial goal would involve a lot of manual labor for much less educational reward. We came to the conclusion that, given our knowledge of computer architecture, we would make a larger impact by making gate design blueprints and recommending how they could be used as educational tools. The complex part for a middle schooler would not be the assembly of the gates, it would be the design and instructions to make them. Therefore, in the interest of maximum education for all, we rescoped our project...which leads nicely into our next point. | ||
| + | |||
| + | ==== Why isn’t there a full adder like promised? ==== | ||
| + | |||
| + | One of the major flaws in our schedule was our optimistic projection on the time it would take to make a gate. We estimated that it would take 15 minutes, putting a full adder (depending on implementation) at a little over 2 hours. | ||
| + | |||
| + | However, in actuality, each gate, if it was constructed carefully (so as to maximize efficiency of operation), took almost an hour. This meant that making a full adder would take 10 hours. | ||
| + | |||
| + | When you factor in the design time of the gates and the decisions of form factor, prototyping of possible designs and documentation, this was too lofty of a goal in the time remaining. | ||
| ====== Gate Assembly Instructions & Advice: ====== | ====== Gate Assembly Instructions & Advice: ====== | ||
projects/mechanical_logic_from_common_materials.1387415359.txt.gz · Last modified: 2013/12/18 20:09 by cjoyce
Page Tools
