logo
Navigation
Home here
One up
JensD
An experiment
Experiment 1
Drop fra Jens kontor
Experiment 2
The experiment (f2)
my result
Experiment 2a elevator
datasheets - on way
back to cansat

The experiment - Results

The experiment took place in the lift between FRB7C C1 and C2.

From top to bottom: from first floor to the cellar.

All data, code and plot used here can be found in data/shield1 in …

See src/index.html

The raw data

3071 lines sampled with 25 Hz (40 msec)

The raw files are

The data visualised

 

raw image

Acceleration

 

raw image

By inspection the lower plot( axis z) shows start of lift going down at 500+ (500*40msec after start).

Stop in cellar is approx 800.

Going up starts at 2000+ and arrival at 2250 approx

x and y axis shows only a short shaken at start/stop.

This is in accordance because z axis is point up from the shield.

SO from here we only look at acceleration in z axis.

Acceleration

Acceleration (in Z axis) is found in

The raw data

 

It can be seen by the nakes eye that there is an offset which we will find and compensate for.

 

We will calculate mean from 800 to 1100: which give 9.3090

Subtracting this from our raw acc z data we get

 

The trip down

  1. It task rougly 12 seconds inclusive acceleratino to take the lift 2 floors down

  2. We will adjust it to 11 second.

    1. This is one of the figures which er not so precise

  3. Sampling with 40 msec we have 25 samples/second.

  4. 12 seconds equal 300 samples

  5. 11 seconds 275 samples

Looking at

 

we magnify

 

Obervations and results

  1. Lift start at approx sample 541

  2. Acceleration stop approx sample 580: 40 samples acceleration - 1 long second

  3. Decelleration starts at approx 852

  4. Travel time around 300 samples or 12 seconds

Summing up acceleration by multiply acc with 40 msec we can estimate velocity after acceleration

  1. sample 541 to 580 final velocity 0.63 m/s

  2. sample 541 to 600 final velocity 0.66 m/s

  3. sample 541 to 620 final velocity 0.65 m/s

adjusting acceleration

We did measure med acceleration to 9.3090 msec2 and gravity is 9.82 ms2.

So we must adjust the acceleration by a factor 1.055

This gives a velocity at 0.65*1.055 msec2 = 0.69 ms

  • 12 sec lift time give a distance around 8.28 m

  • 11 sec lift time give a distance around 7.59 m

Closer inspection

So velocity estimate is around 0.65m/sec

Airpressure

Air pressure in Pascal

 
 
 

Results by inspection

  • left part high approx 7.9m

  • center part low approx 0.4m

  • lift travel is approx 7.5m

Overall result

  • by acceleration 7.6m or a litle more

  • by air pressure 7.5m

Still missing stat etc

qed

Jens xxxxxxxxxxxxxxxxxxxx

xxxxxxxxxxxxxxxxxx

Air pressure

  1. 1 min no movement equals 25*60 samplings

  2. For each calculate mean and std deviation (the last for observing quality)

    1. https://www.cuemath.com/data/standard-deviation/

  3. Do the same for the logged altitude

  4. calculate vertial travel distance (you should have four measurements - are they equal ?)

Accelerometer

  1. Which acceleration axis has been “active”

  2. As for air find the four sections with no travel and calculate average and std deviation

  3. Plot the for set data of movement - Can you setup a simple for for the acceleration and decceleration part for each travel

  4. After acceleratino(start)c and before dec (stop) what is gravity compared to still stand

  5. How long time goes for acc, <travel>, stop

  6. Can we use the accelleration part to estimate the velocity for the after acceleration settle

  7. If so can you by knowing this and travel time estimate the vertical travel distance

  8. If so do you have same velocity up going a down going (uuuhh my english)

Compare air pressure model with the acc model

  1. as title say

your model

aftermath

  • Standard air pressure is 1013 hekto pascal

  • Datasheet for the BMP085

  1. What is the absolute accuracy for the sensor

  2. 1 Bar equal 1013 hekto Pascal. which equals approx 10 m watercolumn (vandsøjle)

  3. 1 m3 air weight 1.2 kg equal 12 Pascal

  4. What is the absolute and relative accuracy for the bmp085

  5. If 1.2 kg air (1m3) equals 12 Pascal what is thiese unaccuracy in meter

  6. How does this comply to your measurements

The image below shows altitude related tp airpressure. It quite linear up to 2-3 km.

 

credit: wikipedia

Experiment - airpressure

https://pilotinstitute.com/pressure-altitude-explained/

See also (same formula bosch producer of bmp

And the code in the bmp085 library:

float calcAltitude(float pressure,float seaLevel)
{
    float A = (1.0 * pressure)/seaLevel;
    float B = 1.0/5.25588;
    float C = pow(A,B);
    C = 1.0 - C;
    C = C /0.0000225577; // or just 44330.0 /JDN
    return C;
}

which is equal to the bosch code

The accuracy goes down when hen altitude goes up.

Jens