Re: [Phys-L] Pendulum again again

["Richard Tarara" <rtarara@saintmarys.edu>]

Not sure this has been mentioned in this iteration of The Pendulum, but this is a great experiment (especially for general education students) to show the importance of the range of data.  I would also do groups to study the different variables (after the whole class came up with what those variables should be), but I split the length study into three groups.  One for lengths up to a meter, another for 1-3 meters, and a last for 3m-10m (we have a stairwell for that group) with a little overlap built in.  When presenting their data and graphs the latter two groups seemed to do well with linear fits and the short length group would be less certain but usually were less certain about the very short data and went with linear.  Then we would combine the three data sets and it was then very clear that the data was NOT linear and we would move on from there but emphasizing that too small a range could produce a false result.  

We also did take amplitude data but went to 3 sets of 3 swings and that group would see a deviation from constant (yes mostly above 30 degrees of swing.)  For the liberal arts students we left that as a mystery to be addressed by the infinite number of monkeys in our theoretical group, but with the science and engineering students we would ultimately resolve this anomaly when we really delved into simple harmonic motion.  Often had 4-5 lab sections with the liberal arts students and while each lab used the combined length data and a preliminary result from the theoretical monkeys to get a value for the gravitational acceleration, I would then combine  the results from all sections  to get a pretty good value of g for that day in that lab!  😉

Richard W. Tarara
Professor of Physics, emeritus
Saint Mary’s College
Notre Dame, Indiana

Free Physics Instructional Software
Windows and Mac
sites.saintmarys.edu/~rtarara/software.html

-----Original Message-----
From: Phys-l <phys-l-bounces@mail.phys-l.org> On Behalf Of Joseph Bellina via Phys-l
Sent: Saturday, March 22, 2025 6:58 PM
To: Phys-L@phys-l.org
Cc: Joseph Bellina <inquirybellina@comcast.net>
Subject: Re: [Phys-L] Pendulum again again

I agree there are many ways to get more precise data but i fear sometimes the additional cognitive load could hinder precise understanding of the concepts. Please remember only 10 or 15% of the students are like us. Another challenge presented by increased precision is that the effect of amplitude will be detected which is fine in a higher order experiment where students understand the difference between large and small effects but for beginners it might be better not to see that. It depends on the pedagogical goals. 
Best
Joe
Sent from my iPhone

> On Mar 22, 2025, at 6:42 PM, John Denker via Phys-l <phys-l@mail.phys-l.org> wrote:
>
> On 3/20/25 04:33, Joseph Bellina via Phys-l wrote:
> > One further point in the event that there are too few or no 
> > stopwatches the experiments can be done by the teacher keeping time 
> > and the students counting swings. I have used a form of this with 
> > students in grades 3 through 13
>
> 1) That's clever. I like that.
>
> 2) How many of them figured out the following?
>  You can reduce the quantization error, especially  for the 
> longer/slower pendulums:
>  Rather than counting "swings" per se, count  the number of 
> zero-crossings plus the number of  turning points, then divide by 4.
>
> 3) Maybe not the 3rd graders, but many of the 13th graders nowadays 
> have smartphones. Here's a web page that functions as:
>  *) a plain old clock,
>  *) and a stopwatch,
>  *) with a powerful split-timer.
>
> It can record an unlimited number of splits, with millisecond 
> resolution.
>
> On a desktop you can copy all the splits to the clipboard, and then 
> paste them into your favorite editor or spreadsheet app.
>
> In an android environment, you can /share/ the splits to the app of 
> your choice.
>
> ============
>
> There's a serious lesson in this: There are a great many situations 
> where you get better results by timestamping things (rather than 
> counting them over some interval) In the research lab there are 
> event-timers all over the place, often with sub-nanosecond resolution.
>
> Back in the horse-and-buggy days, people used counters, which produced 
> a smaller amount of data, which was easier to manage ... but not any 
> more. You don't want less data; you want cleaner data. Managing it is 
> the least of your worries. That's what computers are for.
>
> _______________________________________________
> Forum for Physics Educators
> Phys-l@mail.phys-l.org
> https://www.phys-l.org/mailman/listinfo/phys-l

_______________________________________________
Forum for Physics Educators
Phys-l@mail.phys-l.org
https://www.phys-l.org/mailman/listinfo/phys-l