Measuring Roller Coaster Acceleration with a Phone (en)

phyphox
37 followers
36 Video Views·Jan 23, 2025
The phyphox team visits an amusement park to measure accelerations on a roller coaster with phyphox.
In German (auf Deutsch): https://youtu.be/JJzyjK7FnSE
Get phyphox for free on Android and iOS: http://phyphox.org
Support us: https://phyphox.org/contribute
Donatinons: https://phyphox.org/donate
00:00 Intro
00:34 Preparations
03:27 Centripetal Acceleration (Merry-Go-Round)
04:56 Linear Acceleration (Launched roller coaster)
06:08 Barometric Pressure (Drop Tower)
07:34 Full roller coaster run
08:31 Review of some roller coaster sections
09:26 Conclusion
10:17 Outro
Hey folks, this is Sebastian and I'm here today with our PhD students
Marija and Johannes and we're here today at Toverland near Venlo in the
Netherlands to do some smartphone experiments on a roller coaster.
[Music]
Okay so the idea is that we use our free and open source app phyphox to record some acceleration and
pressure data on the roller coasters, specifically Fenix and let's see if we can interpret the data.
In principle you can directly select the accelerometer or pressure sensor in phyphox
but when doing these experiments on a fast ride there's a catch: We have to follow safety rules!
Usually you are not allowed to hold your phone in your hand as it might seriously injure you
or other guests. The rules that have been set by the park and its staff are there for a reason so
please follow them. Luckily, you are allowed to carry your phone in your pocket on most roller
coasters so that's where we have to measure the ride, but to do so we have to prepare a few
things. You absolutely do not want to accidentally trigger anything on your screen during the ride,
but you cannot simply turn off the screen since phyphox needs to remain active. But it can use
the proximity sensor of your phone to disable the screen while it is measuring in your pocket. To
do so go to the app settings from the app's main screen, look for proximity lock and enable it.
You you will not notice a difference at first but whenever a measurement is running phyphox
will turn off the screen as soon as anything gets close to the proximity sensor, like the inside of
your pocket. The next problem is the orientation of your phone if you want to understand the data
you recorded you need to know the orientation of your coordinate axes, but if it's in your pocket
the phone will be at a rather weird angle. To make your life a bit easier scan this QR code from the
plus menu in phyphox to load a configuration that tries to automatically adjust the coordinate axes.
You can save it in your experiment collection so you can just open it from the main screen
without scanning any QR codes in the future. This configuration has a countdown enabled by default
and when the measurement starts it will use the first data points to determine the direction of
gravity and adjust the coordinate axis such that the Z axis points up. This means that
you have to start the experiment put it in your pocket while the countdown is running and assume
the exact position in your seat that you will have during the ride. If you need to change the
countdown duration you can do so in the menu in the top right corner. Also be aware that phyphox
can only determine the direction of down but it is not aware of which direction is forward, so try to
point the Y axis (that is the long side of your phone) along the forward direction. One last tip:
If you have to wait in line before the ride, use that time to do a test run in your pocket.
Losing the data is particularly annoying if you have to wait a long time before riding again.
Everything ready? Then let's see what we get.
Wait, sorry, but this motion already looks rather complicated. How about we let Marija and Johannas
go through a few basic motions first and we come back to the Fenix roller coaster later.
Are you serious? This is what I get to talk about? Yeah, well whatever... So,
most roller coasters will do turns from time to time and the most basic version of a turn
is a circular motion and you can check that out right here on a merry-go-round like this. You
might think at first, that there's not much going on while sitting on merry-go-round,
but let's look at it from above and remember Newton's Laws of Motion. Without any external
forces a body remains in motion at a constant speed in a straight line,
so in order for you to move on a circle instead of a straight line the merry-go-round or, well,
the horse you're sitting on, exerts a force on you towards the center of the circle. This
is called centripetal force and it leads to centripetal acceleration which we can measure.
This merry-go-round here is not particularly fast so the vibrations on the orange Z axis are the
first thing we see but if we look at the other axes we can clearly see that the forward-facing
blue y-axis is barely affected while the green x-axis shows a clear negative signal which is
an acceleration pointing towards the center - the centripetal acceleration. But of course we
can also experience acceleration without a turn I think Marija has found a good example for that
Since Fenix turns and drops right at the beginning let's have a look at the much
simpler motion at first - at least from a physics point of view. This is a launched roller coaster,
so we are accelerated along a single axis. Let's go for it! The launch of this coaster is a perfect
example for a simple linear acceleration at the moment you are seeing noise on all three axes
from small vibrations and a little jerk from the brakes. Let's see what happens next... This really
is the most basic example for acceleration. It only happens along one axis which in this case is
the y axis and indeed when the coaster accelerates forwards we can clearly see that on the blue y
axis and barely anything on X and Z until the track rises and things get more complicated.
But that's enough about acceleration. There's
another sensor in our phones that might be interesting to us.
So far we've only been looking at the acceleration data, however, if your smartphone has got a
pressure sensor you can use that to also record altitudes. And what better place to play with
altitudes than a drop tower? The pressure sensor will measure the atmospheric pressure and it's
sensitive enough to resolve the small pressure changes that occur with altitude. If you go up the
pressure goes down and phyphox can calculate the difference in altitude. Unfortunately, there are
a few technical details to discuss here. First of all, not all phones have that sensor. In that case
you won't get any altitude data but you can still measure the acceleration with your accelerometer.
Or you might notice that your pressure sensor is rather noisy like in this example here. This is
often the case with phones that waterproof as the pressure inside the phones cannot equalize
as fast with the pressure outside. In this case even touching the phone can lead to small pressure
changes inside that dominate your measurements. Other phones might have a much smaller sampling
rate. For example Apple devices only offer about one measurement per second. Averaging our fast
noisy data here in blocks of 1 second each, as shown by the red line, should be similar to what
to expect on an iPhone. Okay, now I think we're finally ready for the roller coaster.
Let's first watch the entire ride, shall we? You might want to pause and review some parts
as there's a lot going on. After the ride we will have a look at some specific parts.
All right, let's review some parts that match what we discussed earlier in the video. Fenix
starts with a drop and of course this is the most pronounced change in altitude during the entire
ride and we can clearly see it in the altitude data from our pressure sensor. Then there is a
banked curve. Just like on the merry-go-round we barely see any acceleration along the blue
forward pointing y axis as it goes at a nearly constant speed but the centripetal acceleration
of that curve is well visible on the other axis. In contrast to the merry-go-roung the
acceleration is now distributed along x and z as the coaster leans into the curve, which
rotates our coordinate system. And finally we get the simple linear acceleration only during the
brake at the end of the ride. Braking is just an acceleration against the direction you're going,
so we see this as a negative acceleration on the blue forward pointing y axis.
With that knowledge you can now go on any roller coaster and interpret its motion - well at least
if your coordinate axes stay aligned with the track. For this one here with rotating carts
we might also need the phone's gyroscope - but maybe that's for another video...
Okay so that's it for this video. I hope you enjoyed it as much as we did making it. Thanks
to Professor Ann-Marie Pendrill, who gave us a lot of tips on how to do these experiments
on the rides. Of course a big thank you to Toverland who allowed us to shoot here. If
you want to learn more about phyphox and maybe some experiments we can do with it check out
phyphox.org or just check out our YouTube channel. Also if you want to help us you can
check out phyphox.org/contribute where we have a list of things that would really help us. All
the links are in the video description. I hope we see you soon in an upcoming video. See you then!
