Computer Model. What am I missing?

  • Context: High School 
  • Thread starter Thread starter swedluap
  • Start date Start date
  • Tags Tags
    Computer Model
Click For Summary

Discussion Overview

The discussion centers around the use of a CAD program called "Interactive Physics" to model a machine that the original poster believes operates continuously under its own weight. Participants explore the implications of numerical methods in computer modeling, particularly in relation to energy conservation and the behavior of oscillatory systems.

Discussion Character

  • Exploratory, Technical explanation, Debate/contested

Main Points Raised

  • The original poster describes creating a computer model of a perpetual motion machine that appears to function continuously, despite expectations that it should fail due to physical laws.
  • One participant suggests that improper numerical methods can lead to energy conservation issues, resulting in unrealistic oscillatory motion.
  • Another participant questions whether the model would eventually stop, implying a need for further examination of the model's behavior.
  • The original poster notes attempts to reverse the machine's direction, observing that it still runs continuously, which raises further questions about the model's validity.
  • A participant points out that numerical errors can obscure real solutions, recommending a smaller time step to potentially improve the model's accuracy.
  • The original poster shares past experiences with numerical modeling issues, indicating ongoing experimentation with the model by adding weights to test its stability.
  • The original poster later claims to have resolved their confusion, suggesting a personal breakthrough in understanding the model's behavior.

Areas of Agreement / Disagreement

Participants express varying views on the reliability of the computer model, with some suggesting that numerical methods may be misleading while others focus on the original poster's observations. The discussion remains unresolved regarding the model's validity and the implications of numerical errors.

Contextual Notes

Participants acknowledge the limitations of numerical methods, including round-off errors and the effects of time step size on model behavior. There is an emphasis on the need for a deeper understanding of the methods used in modeling.

swedluap
Messages
5
Reaction score
0
I'm sure most of you are familiar with the CAD program called "Interactive Physics". http://www.interactivephysics.com/ . I started using this program a couple months ago. I have been, for fun, playing with some theories that I have thought up over the past 15 or so years. I started out by creating some of the typical Perpetual Motion Machines of yesteryear, just to watch them fail for myself, of course, they did fail, quickly. I then started modifying them and adding some of my theories, trying out some of my ideas, and somehow I was able to create a machine that actually works, on the computer anyway. It starts, speeds up and runs contiuously under its own weight. It is very slow, so I doubt it would have any useful application in real life. I thought that maybe friction was not being taken into account, so I added resistance to it, and it still works. Has a working computer model ever been made? If so has that model been built only to fail in real life? What am I overlooking here? Please help me make this machine fail so I can stop wasting my time, or I might be forced to actually build a real life model, and I just don't have the time!
 
Physics news on Phys.org
A improperly implemented numerical method can "fail to conserve energy" ie a pendulum which continues to increase its amplitude of oscillation. A frequent breakdown of numeric methods will result in oscillatory motion which is essentially noise.
 
But it would eventually stop, right?
 
I have even tried kick starting it to run the wrong way, it takes a while but it does stop and reverse to run continuously in the other direction. I still don't see what I am missing here.
 
It can well be that various errors (round off and errors inherent in the methods used) have swamped your real solution. This is not uncommon and is a reason that people doing serious modeling work can rarely use off the shelf products, you need to understand the methods used and how the inherent errors (all numeric methods have them) effect your solution. Sometimes the noise accumulates so your solution blows up other times it will fall into meaningless oscillations.

You may want to take a smaller time step, see what happens. Depending on the method this can help a lot.
 
I have experienced both the "blow up" and the "meaningless oscillations" in previous models. I have just started a new test, I have added a small weight to one side to see if my model can overcome this flaw. I will let you know what happens, but it is running for now. I will be gone for a while, check tomorrow, and thanks for your help.
 
Nevermind, I got my head out of my underside. I figured it out. Thanks.
 

Similar threads

High School Calculate needed dimensions of table - Center of mass, levers etc
  • · Replies 11 ·
Replies
11
Views
3K
Graduate Can humans really create the quantum computer they expected?
  • · Replies 2 ·
Replies
2
Views
2K
High School How do mechanical (weighing) scales work?
  • · Replies 9 ·
Replies
9
Views
3K
Programs What core abstractions connect real analysis, probability, operators?
  • · Replies 1 ·
Replies
1
Views
1K
Sequential Analog Computers?
  • · Replies 10 ·
Replies
10
Views
2K
Programs Computational neurosciences: what to expect contentwise?
  • · Replies 2 ·
Replies
2
Views
1K
High School Understanding Physics for Coding Collision of 2 Balls
  • · Replies 14 ·
Replies
14
Views
5K
Undergrad Feynmann's connection machine model
  • · Replies 1 ·
Replies
1
Views
4K
Undergrad Modeling a Bottle Rocket: Analyzing the Maximum Altitude
  • · Replies 4 ·
Replies
4
Views
6K
Considering Switching from Physics to Computer Engineering
  • · Replies 7 ·
Replies
7
Views
3K