+- phyphox Forums (https://phyphox.org/forums)
+-- Forum: Discussion (https://phyphox.org/forums/forumdisplay.php?fid=1)
+--- Forum: Experiments and Ideas (https://phyphox.org/forums/forumdisplay.php?fid=4)
+--- Thread: approximation (/showthread.php?tid=240)
approximation - Jessica Landeira - 04-02-2019
Good afternoon! Me and my team are part of a University in Brazil. Where we studied Metallurgical Engineering. We are doing several experiments in the Experimental Physics class using Phyphox. We would like to know if the application makes approximations when calculating the experiments? How accurate is the resolts shown in Phyphox? What is the margin of error of the application?
We are trying to run the outputs using the application and comparing with the conventional way in the physics lab. We have not come to any conclusions yet. What has been perceived by you guys?
RE: approximation - d.girardi - 04-03-2019
Sorry, Jessica, there is no easy answer to your questions. Each brand uses a different type of sensor. It's is impossible to define a generic error or accuracy of the software because phyphox use the data coming from the sensor. The best way to estimate the errors is by letting your phone over the table and measure the standard deviation of the sensor. Each sensor (accelerometer, gyroscope, magnetometer) has its own error and accuracy.
RE: approximation - Jessica Landeira - 04-03-2019
(04-03-2019, 07:35 PM)d.girardi Wrote: Sorry, Jessica, there is no easy answer to your questions. Each brand uses a different type of sensor. It's is impossible to define a generic error or accuracy of the software because phyphox use the data coming from the sensor. The best way to estimate the errors is by letting your phone over the table and measure the standard deviation of the sensor. Each sensor (accelerometer, gyroscope, magnetometer) has its own error and accuracy.
Thanks for your help. In science we can never be too carful. Me and my team will apply your suggestions.
RE: approximation - Sebastian Staacks - 04-04-2019
On most Android phones you can also find out which sensor is in the phone. In phyphox this will be available with the upcoming version 1.1.0 (or in the current beta version under "Device info" in the top right menu on the main screen), but you can also just use a third-party app like "Elixir 2" (not affiliated, but that's what I used before). Usually you will find the brand and model for most sensors and then you might be able to find datasheets for them. However, be careful about applying the info from the data sheet directly as phyphox does not get the data directly from the sensor but from the Android API. The manufacturer might run the sensor in a different mode (many sensors support different ranges and acquisition rates with different resolutions and/or noise), apply some filtering or do some calibration before handing the data to the system.
If you are interested in the exact operations done by phyphox in each experiment, I would suggest, that you have a look at our editor (http://phyphox.org/editor). There you can open any experiment configuration and see what it does. Even more details can be found in our source code if you dare to dive into it (https://phyphox.org/source/).
RE: approximation - Jessica Landeira - 04-04-2019
(04-04-2019, 11:43 AM)Sebastian Staacks Wrote: On most Android phones you can also find out which sensor is in the phone. In phyphox this will be available with the upcoming version 1.1.0 (or in the current beta version under "Device info" in the top right menu on the main screen), but you can also just use a third-party app like "Elixir 2" (not affiliated, but that's what I used before). Usually you will find the brand and model for most sensors and then you might be able to find datasheets for them. However, be careful about applying the info from the data sheet directly as phyphox does not get the data directly from the sensor but from the Android API. The manufacturer might run the sensor in a different mode (many sensors support different ranges and acquisition rates with different resolutions and/or noise), apply some filtering or do some calibration before handing the data to the system.
If you are interested in the exact operations done by phyphox in each experiment, I would suggest, that you have a look at our editor (http://phyphox.org/editor). There you can open any experiment configuration and see what it does. Even more details can be found in our source code if you dare to dive into it (https://phyphox.org/source/).
thanks Sabastian! Is incredibel the achievements already provided by the app. I will talk to my grupo and professor sow we can collaborate more with phyphox in pro of the science field! We are going to write a paper at the end of the period.
RE: approximation - Jessica Landeira - 04-16-2019
(04-04-2019, 11:43 AM)Sebastian Staacks Wrote: On most Android phones you can also find out which sensor is in the phone. In phyphox this will be available with the upcoming version 1.1.0 (or in the current beta version under "Device info" in the top right menu on the main screen), but you can also just use a third-party app like "Elixir 2" (not affiliated, but that's what I used before). Usually you will find the brand and model for most sensors and then you might be able to find datasheets for them. However, be careful about applying the info from the data sheet directly as phyphox does not get the data directly from the sensor but from the Android API. The manufacturer might run the sensor in a different mode (many sensors support different ranges and acquisition rates with different resolutions and/or noise), apply some filtering or do some calibration before handing the data to the system.
If you are interested in the exact operations done by phyphox in each experiment, I would suggest, that you have a look at our editor (http://phyphox.org/editor). There you can open any experiment configuration and see what it does. Even more details can be found in our source code if you dare to dive into it (https://phyphox.org/source/).
Sebastian, in the spring experiment, what represent the unity (u.a) in tha correlation tab?
RE: approximation - Sebastian Staacks - 04-17-2019
u.a. stands for "arbitrary units". This is typically used when the exact unit is not relevant or not known, for example if you do not have the exact calibration of a tool but only analyze relative changes in any case. In this case we have decided to not use the exact unit because it is mostly irrelevant for an autocorrelation and it might be extremely misleading as students might mistake the autocorrelation for a time-based measurement. Also, we are adding the sensor input here, which makes its quantitative interpretation even more difficult.
The (analystical) autocorrelation is defined as
![[Image: 08ebc33ac065bbdc0ab02e2d10261a7064711976]](https://wikimedia.org/api/rest_v1/media/math/render/svg/08ebc33ac065bbdc0ab02e2d10261a7064711976)
so the unit would be the unit of x squared. In this case, the numerical autocorrelation works similarly and would have the unit m²/s⁴ (the exact implementation can be found here: https://github.com/Staacks/phyphox-android/blob/master/app/src/main/java/de/rwth_aachen/phyphox/Analysis.java#L1318). This by itself is difficult to interpret as a physical value. The autocorrelation is measure for "how much the function resembles itself" and the most sensible thing to do would be to normalize the result by the maximum value at τ=0.
On top of this, we do not know along which axis the user will create an oscillator. So, for this experiment to work on any axis, we are not calculating the autocorrelation of a single axis (or on the absolute value, which would give trouble as the direction is lost) but of the sum of all axes. Therefore, the interpretation is even more difficult if you are not oscillating along one of the three axes.
So, to sum this up. Technically, the unit is m²/s⁴, but usually you should not interpret it like this, but instead just look at the "self-similarity" relative to the first peak.