Prime Directive of the Knob-Turner: The knob-turner must stare at the correct needle when turning the knob.

This slideshow requires JavaScript.

Notes from lecture:

1. Colorado-Boulder Modeling Framework by the Lewandowski Group
2. Failure is a gift.
   • When I say that your only obligation is to be honest, I am not saying that you must not knowingly lie – that is of course a given. What I am saying is that you must be vigilant against letting your hidden unspoken assumptions mislead or influence your analysis. This is particularly tricky when you try to rely only on your memory, which is notoriously fallible and will often trick you based on your own internal and unexamined biases. The safest (but not always foolproof) way of avoiding this pitfall is rigorous documentation of your thought process in the logbook in real time as you are carrying out your experiment. This is the underlying motivation of most of these bullet points.
   • Scientific Diagrams
3. Colorado-Boulder Modeling Framework by the Lewandowski Group
4. Knobs and Needles
   • Prime Directive of the Knob-Turner: The knob-turner must stare at the correct needle when turning the knob.
   • It is not trivial knowing how fast or how far to turn the knob.
   • It is not trivial knowing which needle to stare at.
5. Knobs and Needles
   • Sometimes one knob is secretly two or more knobs.
   • Sometimes there are secret knobs (classic example: temperature of the room).
   • Some needles are just flashes of light.
   • Most needles are voltages that represent some aspect of your physical system.
6. Models
   • It is useful to make diagrams in the style of Thing Explainer cartoons by Randall Munroe.
   • It is useful to try describing the instrument or experiment using The Up Goer Five Text Editor.
7. Is the agreement “good enough?”
   • What is your prediction for what should happen? This should be specific and clearly articulated.
   • Quantitative – Which needle should you be looking at? What is your expectation for the behavior of the needle as you turn the knob?
   • Qualitative – Where should you look for the effect? What do you think that it will look like? How long do you think it will last?
8. Correlation is not causation.
   • If you observe what you were expecting, keep in mind that it does not necessarily mean that your prediction was verified.
   • You are also obligated to rule out all reasonable alternative explanations with carefully designed tests.
9. How do I construct my initial model?
   • You already double secretly use your own intuition to build models of the instruments that you are using!
   • Your obligation as a Scientist is to examine these previously unexamined internal intuitive models and then reveal them by clearly articulating your underlying assumptions.
   • This is not easy and requires a great deal of open-mindedness, humility, and self-discipline.
10. Model of an Ideal Voltmeter
    • Most inexperienced students do not realize that they are assuming an ideal model of a voltmeter which reports a voltage that is exactly the same as the true voltage applied to the voltmeter terminals.
    • In order words, y = m*x +b, where y is the measured voltage, m = 1, x is the true voltage, and b = 0.
11. Realistic Model of a Voltmeter
    • There is a maximum and minimum voltage that can be measured.
    • The measured voltage is not perfectly linear with the true voltage. For example, y = m*x +b + q*x^2, where q (not equal to zero) represents the lowest order nonlinearity.
12. Realistic Model of a Voltmeter x1000
    • There are statistical fluctuations in the measured voltage.
    • There is an offset in the measured voltage (b is not equal to zero).
    • The measured voltage is digitized – in other words, it does not have infinite precision. Another way to think about this is that there is an order of magnitude scale below which the voltmeter is insensitive to the true voltage input.

-JTS