EPS@ISEP | The European Project Semester (EPS) at ISEP

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision		Previous revision
report [2019/06/23 14:44] – [8.7 Tests and Results] team4report [2019/06/26 16:32] (current) – [8.7 Tests and Results] team4
Line 2547: Line 2547:
 <WRAP centeralign> <WRAP centeralign>
 <figure HSschematic> <figure HSschematic>
-{{ ::homestation1_schem.jpg?600 |}}+{{ ::SchematicNew.jpg?600 |}}
 <caption>Home Station Schematic Drawing </caption> <caption>Home Station Schematic Drawing </caption>
 </figure> </figure>
Line 2564: Line 2564:
 ==== - Tests and Results ==== ==== - Tests and Results ====
 In the following, all parts of bGuard are installed, tested and evaluated. In the following, all parts of bGuard are installed, tested and evaluated.
-=== - Crash Test Home Station ===+=== - Pressure simulation of the Home Station in SOLIDWORKS ===
  
-The material for the home station concept is going to be ABS. This material is strong enough to hold a force of 100 N on top without bending too much or breaking it. The reason for 100 N is because the model has to fall from two meters without breaking. However, the model is only 58 g so when it falls from two meters, it will probably only reach a force of 50 N. This is the reason of a safety factor of two what makes 100 N. A SOLIDWORKS test has been carried out to prove this. See **Figures** {{ref>tension_1}} and {{ref>verpl_1}}. The images show that when there is a force of 100 N on the top of the ABS. The displacement {{ref>verpl_1}} of the material is only 0.386 mm on the most loaded point. Normally, you can't consider this to be anything. The tension in the product shows the same result.+The material for the home station concept is going to be ABS. This material is strong enough to hold a force of 100 N on top without bending too much or breaking it. The reason for 100 N is because the model has to fall from two meters without breaking. However, the model is only 58 g so when it falls from two meters, it will probably only reach a force of 50 N. This is the reason of a safety factor of two what makes 100 N. A SOLIDWORKS test has been carried out to prove this. See **Figures** {{ref>tension_1}} and {{ref>verpl_1}}. The images show that when there is a force of 100 N on the top of the ABS. The displacement {{ref>verpl_1}} of the material is only 0.338 mm on the most loaded point. Normally, you can't consider this to be anything. The von Mises stress equivalent in the product shows the same result.
  
 <WRAP centeralign> <WRAP centeralign>
Line 2745: Line 2745:
 <figure temp30> <figure temp30>
 {{ :temphum30.png?nolink&1000 |}} {{ :temphum30.png?nolink&1000 |}}
-<caption>Temperature and Humidity measurements for 30°C and comparison to reference device and error and uncertainty calculation</caption>+<caption>Temperature and relative Humidity measurements for 30°C and comparison to reference device and error and uncertainty calculation</caption>
 </figure> </figure>
 </WRAP> </WRAP>
Line 2753: Line 2753:
 <figure temp28> <figure temp28>
 {{ :temphum28.png?nolink&1000 |}} {{ :temphum28.png?nolink&1000 |}}
-<caption>Temperature and Humidity measurements for 28 °Cand comparison to reference device and error and uncertainty calculation</caption>+<caption>Temperature and relative Humidity measurements for 28 °Cand comparison to reference device and error and uncertainty calculation</caption>
 </figure> </figure>
 </WRAP> </WRAP>
Line 2760: Line 2760:
 <figure temp26> <figure temp26>
 {{ :temphum26.png?nolink&1000 |}} {{ :temphum26.png?nolink&1000 |}}
-<caption>Temperature and Humidity measurements for 26 °C and comparison to reference device and error and uncertainty calculation</caption>+<caption>Temperature and relative Humidity measurements for 26 °C and comparison to reference device and error and uncertainty calculation</caption>
 </figure> </figure>
 </WRAP> </WRAP>
Line 2771: Line 2771:
 </WRAP> </WRAP>
  
-The required accuracy of the temperature is ± 3 % and of the Humidity ± 5 %. Therefore the values are acceptable.+The required accuracy of the temperature is ± 3 % and of the relative Humidity ± 5 %. Therefore the values are acceptable. 
 + 
 +<WRAP centeralign> 
 +<figure temphumkurve> 
 +{{ :bildschirmfoto_2019-06-26_um_16.35.06.png?nolink&1000 |}} 
 +<caption>Temperature and relative Humidity tendency curve and synthesising graph</caption> 
 +</figure> 
 +</WRAP> 
 + 
 +The measured values are calibrated software-wise by calculating the average absolute error and subtract it from the measured value. Doing that the precision of the sensor could be improved as it can be seen in **Figure {{ref>temphumkurve}}**.
  
 === - Pulse Measurement === === - Pulse Measurement ===
Line 2845: Line 2854:
 </WRAP> </WRAP>
  
-In the following the graph of the reference medical device Silvercrest SPO55 with an accuracy of ± 2 BPM in the range of 30 – 250 BPM Pulse. The error and uncertainty calculations are done the same as explained in the temperature and humidity measurement chapter before.+In the following the graph of the reference medical device Silvercrest SPO55 with an accuracy of ± 2 BPM in the range of 30 BPM to 250 BPM Pulse. The error and uncertainty calculations are done the same as explained in the temperature and humidity measurement chapter before.
  
  
Line 2923: Line 2932:
 </WRAP> </WRAP>
  
-For calibrating the sensor a sound with a constant level is created with the function generator and the speaker. While creating the sound the sound meter measures the decibel. The microphone measures at the same the voltage level. Assuming that the sound meter decibel (x) is linear to the voltage level (y), the output sound level in dB is calculated with this formula:+For calibrating the sensor a sound with a constant level is created with the function generator and the speaker. While creating the sound the sound meter measures the decibel. The microphone measures at the same the voltage level. Assuming that the sound meter decibel is linear to the voltage level, the output sound level in dB is calculated with this formula [(Sensitivity)]:
  
 <WRAP centeralign> <WRAP centeralign>
 \begin{equation} \begin{equation}
-    L= x +10*\log_{10}({\frac{mic}{y}}) (dBA)+     Output Sound Level= Sound Level Meter +20*\log_{10}({\frac{Sensititvity Sensor}{Output Sensor}}) (dBA)
 \end{equation} \end{equation}
 </WRAP> </WRAP>
-If the microphone value (mic) is similar to the voltage level (y), the logarithm is zero and the sound level is equal to the decibel of the sound meter measurement (x).+ 
 +If the microphone value is similar to the voltage level, the logarithm is zero and the sound level is equal to the decibel of the sound meter measurement. This offset added to the referenced formula.
  
 A baby cry has a frequency range from 336.9 Hz to 502 Hz [(babyschrei)]. Due to that, the frequency stability of the MAX4466 is tested. In **Figure {{ref>extfreq}}** the noise level in dBA overtime for the two extreme frequencies can be seen. The reference graph is laid over the ones of the microphone. The lower the frequency gets the bigger is the deviation between these graphs. Due to the fact that our target is to create a cry detection and not an exact sonometer the deviations can be accepted. Because of that, we accept a deviation of ± 5 %. Within that range, a cry is still reliably detected. As an example, both extreme values are shown with the acceptance range in **Figure {{ref>extfreq}}**. It can be seen that the measured extreme values for frequencies of 100 Hz and 500 Hz compared with the sound level meter with the acceptance range of 5 %.  A baby cry has a frequency range from 336.9 Hz to 502 Hz [(babyschrei)]. Due to that, the frequency stability of the MAX4466 is tested. In **Figure {{ref>extfreq}}** the noise level in dBA overtime for the two extreme frequencies can be seen. The reference graph is laid over the ones of the microphone. The lower the frequency gets the bigger is the deviation between these graphs. Due to the fact that our target is to create a cry detection and not an exact sonometer the deviations can be accepted. Because of that, we accept a deviation of ± 5 %. Within that range, a cry is still reliably detected. As an example, both extreme values are shown with the acceptance range in **Figure {{ref>extfreq}}**. It can be seen that the measured extreme values for frequencies of 100 Hz and 500 Hz compared with the sound level meter with the acceptance range of 5 %. 
Line 3034: Line 3044:
  
 Also bGuard mobile application is one of the things that need further development. The team had an idea of what the app should look like and what data it should provide but were only able to design the interface. The application is now merely a concept. Moreover, the team was not able to develop a real app in the given time. For example, now bGuard is limited to sending an audio file to the Smart Pillow but in the future, the team would like to expand this to a communication feature which allows the parents to talk to the baby.  Also bGuard mobile application is one of the things that need further development. The team had an idea of what the app should look like and what data it should provide but were only able to design the interface. The application is now merely a concept. Moreover, the team was not able to develop a real app in the given time. For example, now bGuard is limited to sending an audio file to the Smart Pillow but in the future, the team would like to expand this to a communication feature which allows the parents to talk to the baby. 
 +
 +Additionally, the tests can be improved. The simulation in SOLIDWORKS by taking into account the calculation of the force (based on the real weight of the home station including all the components) as also the pressure. Now, the tests has been done with assumptions of real life examples.  Also, the tests of the sensors and their aim can be improved.
  
 Furthermore, the team would make our bGuard comply with the regulations of being a medical device. This would help to position the bGuard as a trustworthy device and help make bGuard to a market leader.   Furthermore, the team would make our bGuard comply with the regulations of being a medical device. This would help to position the bGuard as a trustworthy device and help make bGuard to a market leader.  
  
-Additionally, the team would focus on using smaller and more precise electronic components. This way they will take in less space and become more accurate results. The given budget made these things difficult to realize during the project.+Finally, the team would focus on using smaller and more precise electronic components. This way they will take in less space and become more accurate results. The given budget made these things difficult to realize during the project.
  
 ===== Bibliography ===== ===== Bibliography =====

Trace:

QR Code
QR Code report (generated for current page)