Help with A/D math!

[dbodenheimer]

Hi all,

I am building a small system to read a laser distance finder. This sensor outputs a 0-10 volt signal based on the distance from the object.
I am reducing the voltage to 0-5v with a voltage divider so that i get 0-5 v to AD0 on my PIC16F876.

Here are some examples of the data I'm getting back

Distance AD count
42.9 820
36.3 675
33.3 600
19.7 270
11.2 60

What i need to do is come up with a formula to convert the A/D readings into distance. So far I am failing miserably! The sensor can be adjusted to select where 0 is in distance and where 10v is. I currently have it set so that 6" is 0v and 50" is 10V I figure that I will have to provide a calibration routine that samples a near point (known distance) and a far point (known distance).

I REALLY need help with the math here to make this silly thing work.

Thanks in Advance,

Dave

[Acetronics2]

Hi, Dave

You can try :

distance = 8573 + 41*AdResult + ( AdResult / 2 )

with Distance ( use a WORD ) = 1000 * actual distance in " ...

Alain

PS; SURE it is a laser distance finder ??? I'd Better bet a laser pointer on a U.Sonic distance finder :rolleyes: ... overall if measuring as short as 6" ... and no more than 40 or 50" ...

Alain

[dbodenheimer]

Alain,

Wow, works like a charm... now can you tell me how you got to those numbers? <grin>
I'm sure when I install this in the field I will have different numbers, so I will need to know how to recalculate.

You are a lifesaver!

Dave

[Acetronics2]

Alain,

Wow, works like a charm... now can you tell me how you got to those numbers? <GRIN>
I'm sure when I install this in the field I will have different numbers, so I will need to know how to recalculate.

You are a lifesaver!

Dave

Very simple, Dave ... no miracle !

just need to check what is PBP compatible , as an operation, and avoiding any overflow when calculating ...

Have a look here ...
http://curveexpert.webhop.net/

really a very handy tool

Alain

[dbodenheimer]

Got the tool. Yes it looks very handy!
Can you tell me the setup you used to calculate the data i sent?

I want to make sure i understand how to set things up

Thanks!

[lavenatti]

Enter data into “CurveExpert” chart:

X is distance, Y is A/D data

Select ‘Apply Fit”, select ‘Linear” from drop down menu.

A graph will pop up, press the ‘info’ button and you’ll get your equation and the coefficients to use.


I will never admit how long that took me to figure out.

[Acetronics2]

I will never admit how long that took me to figure out.

Hi, Lavenatti

Quite Easy to believe...

Didn't you just swap here input and output values ??? :D

Alain

[thronborg]

Hello
This was very interesting. I have a little bit of the same problem.
I have A/D input 0-5V (0-255)

I want a puls out 0V=1mS and 5V=2mS.

If i take your example with the Laser and run it in CurveExpert i got the formula a-bx.

Where did you get 8573 and 41 from, and what is + ( AdResult / 2 ) means?
distance = 8573 + 41*AdResult + ( AdResult / 2 )

Any help will help.

5272

[HenrikOlsson]

Hi,
How about:

ADResult = ADResult * 10
DelayInMicroSeconds = 1000 + (ADResult */ 100)
PortB.0 = 1
PauseUs DelayInMicroSeconds
PortB.0 = 0
When ADResult is 0 DelayInMicroSeconds is 1000 (1ms), when ADResult is 128 you'll get 1000 + (1280 * 100 / 256) = 1500 and when ADresult is 255 DelayInMicroSeconds is 1000 + (2550 * 100 / 256) = 1996.

It's not perfect but hopefully good enough.

/Henrik.

[thronborg]

Tusen Tack Henrik.
Sorry for the Swedish, i was so happy for the answer AND that you also took your time to described how you did it.

This was exactly what i was looking for.

Thank's
Ole

[thronborg]

My first PIC program works!

Description
A pot connected to an A/D port gives 0-255
A little converting give an output of 1mS when pot CCW and 255 when pot is max CW.
Drives the Servo between 1-2mS
Show the Pot input on LCD line 1, 0-255.
Show the Pulse width in mS on LCD line 2, 1-2mS.

Se Drawing below.

'************************************************* ***************
'* Name : NimSer.BAS *
'* Date : 3/3/2011 *
'* Notes : Move Servo 1 to 2mS depend on POT value *
'* Show valu in mS on LCD display *
'* For standard servo settings 1-2mS set LOW_servo con 100 *
'* and HIGH_servo CON 200 *
'************************************************* ***************
'Define OSC and ADC
DEFINE OSC 4 ' Set internal Oschillator to 4Mhz
DEFINE ADC_BITS 8 ' Set number of bits in result
DEFINE ADC_CLOCK 2 ' Set clock source (3=rc)
DEFINE ADC_SAMPLEUS 50 ' Set sampling time in uS
' Define LCD pins
Define LCD_DREG PORTC 'LCD data port
Define LCD_DBIT 0 'LCD data starting bit 0 or 4
Define LCD_RSREG PORTC 'LCD register select port
Define LCD_RSBIT 4 'LCD register select bit
Define LCD_EREG PORTC 'LCD enable port
Define LCD_EBIT 5 'LCD enable bit


TRISA = %00001001 ' RA0 = A/D input
ADCON1.7 = 0 ' RA.1 = +Vref, Set PORTA analog and left justify result
PORTb.6 =0 ' Prepare RB0 for high-going pulseout

ANSEL = %00000100 ' Set PORTA.2 analog, rest digital
ANSELH = %00000000

' Variables
outpuls VAR WORD ' Variable for the calculated puls out in mS
POT_POS VAR BYTE ' Pot position CC=0, CCW=255
LOW_servo con 60 ' Min Servo pulse 60= 0.6mS 100=1mS
HIGH_servo CON 200 ' Max Span from LOW_servo to HIGH_servo
' Span 100 gives 100+100 = 200=2mS
Pause 500 ' Wait for LCD to start

MainLoop: ' The Loop start here!
ADCIN 0,POT_POS ' Read A/D channel 0 to variable SVO_POS
outpuls =LOW_servo + (POT_POS *HIGH_servo/255) 'Calculate the outpuls in mS
Lcdout $fe, 1, "POT_POS= ", #POT_POS ' Display POT Valu between 0-255 on line 1
LCDOut $fe,$C0, "Puls= ",DEC (outpuls/100),".", DEC2 outpuls,"mS"' Display pulswith in mS on line 2

PULSOUT portb.6 ,outpuls ' Move servo on pin
PAUSE 20 ' Constant 20mS pulses(low) between outpuls
GOTO MainLoop ' Forever
End

5286

[Ioannis]

Nice you got it working. Also very nice graphics in your pdf file!

Ioannis

[thronborg]

Thank's Ioannis. There is a long way to go. But i start to understand what every body is talking about, Loook in the datasheet, 70% of the answer is there.

Continue the good work, I will try.

Ole

[Ioannis]

I see that you have the potential. From your first question to your complete working code there was an impressive progress.

By he way what you used for the graphics?

Ioannis

[thronborg]

Hey
No no i have not write everything by myself, i take some examples here and there and i got help here at the forum. So i am not so smart as it looks.

Actually i use a very old program, Freehand MX. Its still available at Adobes site. Why i use it? Because i have used it for 20 years and its in my fingertips. I made the symbol library myself, if you want it just let me know.

My development kit always go with me.
My MacBook Pro
Proteus
PicBasicPro
MikroC pro for PIC
PIC-kit 2
Programmer
Developer Board CP-Pic training, buyed in Bangkok for 60 euro. se picture.

Great to have everything from Mac and PC in the same Computer.

Ole
5297
http://www.etteam.com/product/01B10.html

[MikeWinston]

Hi Dave,
I've used the DIV32 function alot for stuff like this. In your example I would do something
like this.
50" - 6" = 44"
44" / 1024(10bit) = 0.042968
convert this number to word constant in your program 42968
multiply a/d result by 42968
the very next line must be DIV32 statement (word var = DIV32 10000)
then add your 6" offset (+ 600)

example: (a/d result = 1023)
1023 * 42968 = 43956264 '(32bit result you can't access but DIV32 can)
res = DIV32 10000 '(res = 4395)
res = res + 600 '(res = 4995 hundreth inches or 49.95 inches)

Mike