[APP]Pocket GForce [Source code released in p20]

Search This thread

Ikari

Senior Member
Mar 26, 2006
98
0
Here you go the 1st program to measure how hard you stump against a wall.

maingps.jpg

MAIN
instant Gforce measure for:
Acceleration/brake (brake is negative)
left/right (left is negative)
Up/Down(up is negative)
Total (always positive)
Speed

GPS
Gps Stats
On/Off

0max.jpg

0-...
Test your time from 0 to whatever you chose
(it uses gforce to start the timer)

MAX
Max Values (i know it needs work)

logoptions.jpg

LOG
Chose what you want to log
Log Size

OPTIONS
Set update speed 50 - 1000ms
Use Km over miles per hour
Use Speed in Acc/Brake over Gforce in Main


http://rjsmsampaio.googlepages.com/grafico.jpg
You can get the gflog.txt in you main memory and open in excel
To clean the log just delete it.
X is left/right
y is acc/brake
z is up/down
t is total

Speed
Altitude
Heading
(all this are obvious)

the GFmax is store in gfmax.txt created in main memory.

How to use it correctly:
the pda must be in portrait whit the screen facing the back of the car. When you open the program it will self calibrate(this don't mean you don't need to have your acc calibrated whit other program).

It should read 0.00 whit the car stop in a level area. You can Turn or incline the PDA a little and calibrate again but it will affect the reading if too much off center.


This is the Beta version so post some feedback, im a full time teacher so updates will be constant but not every day :)

Source code in page 20
Beta vertion 0.25b in page 14
v0.15 0 to x Dynamo
v0.10 Gps integration and KM/Miles options
v0.03 installer
v0.03 maximum Gforce store and logging.
v0.02 New interface, Max working, new formula for Gforce (sorry the first one was wrong a bit) and new formula to calculate Total (thx haggz)
v0.01 First release


http://rjsmsampaio.googlepages.com/Ikari-PtPocketGForce.cab
 
Last edited:

charly_mx

Senior Member
Jan 18, 2007
737
0
got the sofaa

Great App my friend. I have tested on aircraft just when taking off and landing and it is great ! :) didnt use the GPS :)
 
Last edited:

emre_aachen

Senior Member
Jul 26, 2008
82
0
I've been waiting for this program since Ikari has promised to write... Thanks Ikari... will feedback...
 

haggz

New member
Sep 1, 2008
4
0
How do you calculate "total"? Judging from the top picture Id say you use

acc + curve = total

which is wrong. Use

total = sqrt(acc^2 + curve^2) instead :)

Otherwise it looks very nice!
 

cgeboers

Senior Member
Jan 8, 2008
86
1
Schriek
Help needed ?

Good job ! I did send you a mail through the forum mail system to offer some help to enhance it. Did you read this ?
 

paclep

Member
May 20, 2008
18
0
Hello.

Perhaps it should be usefull to offer a option to register some infos such as max acceleration.

Thank you for this app.
 

Ikari

Senior Member
Mar 26, 2006
98
0
haggz; said:
total = sqrt(acc^2 + curve^2)
I'm not sure about that i whant the total not the movement total.

cgeboers; said:
Good job ! I did send you a mail through the forum mail system to offer some help to enhance it. Did you read this ? )
Sorry only read after posting the program :p any way yes i will need help to improve the program to the maximum. TIME/G graphics etc...

paclep; said:
Perhaps it should be useful to offer a option to register some infos such as max acceleration. )
As mention on the first post ins not working at the moment
Any one give me tips to how make a txt file in C#?

As for the rest drive carefully ;)
 

emre_aachen

Senior Member
Jul 26, 2008
82
0
Why don't you integrate the output signal with respect to time and write also the velocity of device? By this way we can also see the current velocity or also write further 0-100km acceleration values....
 

haggz

New member
Sep 1, 2008
4
0
I'm not sure about that i whant the total not the movement total.

But you cant just add perpendicular forces since a force is a vector. Break/acc acts straight forward (backwards is "-forward") while curve acts straight right (again, left is "-right"). Draw 2 arrows on a paper, one for acc and one for curve. Then draw the resulting vector. To find the length (in this case the magnitude, or what you call total) use the Pythagorean theorem

total = sqrt(acc^2 + curve^2)


( http://en.wikipedia.org/wiki/Vector_(spatial)#Length_of_a_vector )
 

diamondhtc89

New member
Aug 12, 2008
3
0
But you cant just add perpendicular forces since a force is a vector. Break/acc acts straight forward (backwards is "-forward") while curve acts straight right (again, left is "-right"). Draw 2 arrows on a paper, one for acc and one for curve. Then draw the resulting vector. To find the length (in this case the magnitude, or what you call total) use the Pythagorean theorem

total = sqrt(acc^2 + curve^2)


( http://en.wikipedia.org/wiki/Vector_(spatial)#Length_of_a_vector )

That's the idea. Adding them is just incorrect. Cause velocity is a vector, it can be expressed as its moduli times a unitary vector. Taking the derivative with respect to time we will find two sources of acceleration: the one caused by change in time of the modulus (acc) and the one due to the change in direction of the unitary vector (curve). As the derivative of a vector is indeed a vector, you should calculate its modulus. Therefore the total should be what haggz said.
Your app works pretty nice despite that. I have liked it a lot. A way to fix this issue is to display both accelerations separately and then if you like to add them just like numbers at least we will be a little bit more pleased :D. Anyway congratulations for your great work. Keep it like that !!!
 

Knochi

Member
Sep 5, 2007
8
0
I don't know how fast you can read out the sensors. But a great application would be a Time/Acceleration Diagram.

Currently we are measuring oscillations of huge masses.
Would be cool if i could monitor oscillations up to 150Hz.
 

Brevi

Senior Member
Mar 16, 2006
141
0
Munich
Looks very nice, but I think a GUI with high contrasts would be better to read while driving.
 

blazer2004

Member
Jan 27, 2008
37
0
Unjar to my computer, transfered to my internal storage, attempted to run the program, but it didnt work. It said I needed a newer version of net or something :confused: any ideas?
 

Ikari

Senior Member
Mar 26, 2006
98
0
Knochi said:
Would be cool if i could monitor oscillations up to 150Hz.
Atm is 4hz but i can test the limit, also depends from the sensor itself.

Unjar to my computer, transfered to my internal storage, attempted to run the program, but it didnt work. It said I needed a newer version of net or something :confused: any ideas?
download and install .NET Compact Framework 3.5 on your pda and it will work.


One more day and i will release a new version whit DB for the maximum values.
About a graphical interface im not sure how to make one cuz we have 3 axis to display plus the Total so im putting that to the side.

If any one have more ideas for this, new options, features or feedback, plz post.
My car is damage so i cant test everything correctly so make your contributing :)

Atm im working in a way to measure peek HP and making a time diagram that can be visualize on excel.
 

cgeboers

Senior Member
Jan 8, 2008
86
1
Schriek
Max reading rate

I don't know how fast you can read out the sensors. But a great application would be a Time/Acceleration Diagram.

Currently we are measuring oscillations of huge masses.
Would be cool if i could monitor oscillations up to 150Hz.
I read somewhere, but can't remember where, that default the accelerator readings are performed every 200 milliseconds, which would mean that you have a max of 5 Hz. It is possible to change the value in the registry by using a program that "tweaks" it, but I don't know what it would mean for the battery life.
Maybe your program could set the readings faster while it is running, but I don't have a clue how you would have to do it.
Never the less, 150 Hz might be a bit too fast for WM6.1, even with a fast processor ?