Reading ADC then delaying switch

L

lxlramlxl

New Member
A quick overview of what the program should be doing.
Basically on pin C.2 I have the input voltage between 0-5 that is varied using a variable resistor.
Depending on the voltage that is on pin C.2 there will be a delay which will activate the output C.1.
I'm pretty sure that it is my code that has an issue and not the hardware itself.

Code: 
Main:

ReadADC c.2 , b0				; read in pot value

if B0 < 20 then				; 
	wait 0
	
elseif b0 < 40 then
	low C.1
	wait 1
	high c.1

elseif b0 < 60 then
      low C.1
      wait 2
      high c.1
      
elseif b0 < 80 then
      low C.1
      wait 3
      high c.1
      
elseif b0 < 100 then
      low C.1
      wait 4
      high c.1
      
elseif b0 < 120 then
      low c.1
      wait 5
      high c.1
      
elseif b0 < 140 then
      low c.1
      wait 6
      high c.1
      
elseif b0 < 160 then
      low c.1
      wait 7
      high c.1
      
elseif b0 < 180 then
      low c.1
      wait 8
      high c.1
      
elseif b0 < 200 then
      low c.1
      wait 9
      high c.1
      
elseif b0 < 220 then
      low c.1
      wait 10
      high c.1
      
elseif b0 < 240 then
      low c.1
      wait 11
      high c.1
          
elseif b0 < 260 then
      low c.1
      wait 12
      high c.1
      
	
endif 




goto main
 
B

BESQUEUT

Senior Member
To my mind this code is OK.
try debuging the hardware :
Code: 
	low C.1
	wait 2
	high c.1 ' to be sure that the output is OK


Main:

ReadADC c.2 , b0				; read in pot value
[COLOR="#FF0000"]sertxd ("B0=",#b0,13,10)[/COLOR]

if B0 < 20 then				; 
	wait 0
	
elseif b0 < 40 then
	low C.1
	wait 1
	high c.1
Equivalent code :
Code: 
DO
      ReadADC c.2 , b0				; read in pot value


     if B0 >19 then
              low C.1
              b0=b0-20/20
              wait b0	
              high c.1		; 
     endif
LOOP
 
Last edited:
A

AllyCat

Senior Member
Hi,

Try putting in a SERTXD(#b0,cr,lf) in the line after the READADC to see if your hardware really is working.

The code looks alright, but could be simpler as :
Code: 
main:
do
	ReadADC c.2 , b0				; read in pot value
	w1 = b0 * 50				; delay in milliseconds	 
	low c.1
	pause w1					; or could use pulsout c.1,w1
	high c.1					;
loop
Cheers, Alan.

PS: My post was later than Besqueut's while I checked why WAIT bo is an "error" !
 
B

BESQUEUT

Senior Member
lxlramlxl said:
Depending on the voltage that is on pin C.2 there will be a delay which will activate the output C.1.
Click to expand...
With your code if pot>19, C.1 is quasi-always low so it is not possible to see the timing !
Try this (thanks to AllyCat) :
Code: 
main:
do
	ReadADC c.2 , b0				; read in pot value
        [S]w1 = b0 -20/20 [/S]
	w1 = b0*50				; delay in milliseconds	 
	low c.1
	pause w1					; or could use pulsout c.1,w1
	high c.1	
	pause w1
        low c.1
				;
loop
 
Last edited:
L

lxlramlxl

New Member
Thanks for your reply. I will be testing the code tomorrow.
What is the "-20/20" for after reading the value in b0?
 
A

AllyCat

Senior Member
Hi,

Yes, I think your "bug" was that there was no PAUSE or WAIT to define the "gap" between your output "pulses".

-20/20 subtracts 20 and then divides by 20, so I think BQ was trying to ensure a "zero" delay at one end of the pot. But IMHO this is doubly wrong. Firstly, it may produce a "negative" wraparound to a large (positive) number, and secondly PAUSE works in milliseconds whilst WAIT is in seconds (but strangely cannot be used with a variable such as b0), so I believe my *50 is correct.

Cheers, Alan.
 
B

BESQUEUT

Senior Member
AllyCat said:
But IMHO this is doubly wrong. Firstly, it may produce a "negative" wraparound to a large (positive) number, and secondly PAUSE works in milliseconds whilst WAIT is in seconds (but strangely cannot be used with a variable such as b0), so I believe my *50 is correct.

Cheers, Alan.
Click to expand...
I agree.
......................................................
 
L

lxlramlxl

New Member
AllyCat said:
Hi,

Yes, I think your "bug" was that there was no PAUSE or WAIT to define the "gap" between your output "pulses".

-20/20 subtracts 20 and then divides by 20, so I think BQ was trying to ensure a "zero" delay at one end of the pot. But IMHO this is doubly wrong. Firstly, it may produce a "negative" wraparound to a large (positive) number, and secondly PAUSE works in milliseconds whilst WAIT is in seconds (but strangely cannot be used with a variable such as b0), so I believe my *50 is correct.

Cheers, Alan.
Click to expand...
Do you know how the ADC reads the voltage? Is it in millivolts or volts? e.g 1 volt would read as 1000 (mV) or 1
This is significant due to the *50. If it reads in Volts it'd be just 50ms but if it reads in mV then it'd be 50seconds (Based on 1volt)
 
L

lxlramlxl

New Member
BESQUEUT said:
255 = Vref (Vref is probably 5V)
Click to expand...
Sorry about this, but where you do you get that information? I can't find it in the help documents (I might of just missed it)
And in picaxe itself you cannot right click the +ve pin to check the value.

Edit: I'm guessing its 255 since that is the maximum value that can be given on the pin?
So going on that basis 51 would be 1volt?
 
Last edited:
B

BESQUEUT

Senior Member
lxlramlxl said:
Sorry about this, but where you do you get that information?
Click to expand...
Picaxe Manual section 2 page 29 : adcconfig

The default Vref+signal for the ADC is the power supply (V+)

+ Appendix 5 & 6 : Picaxes variations
Voltage Range (V) for 20 X2 = 1.8-5.5

With Picaxes M2, you also have FVR ; see fvrsetup...

Your power supply is probably 5V, but can be anything from 2V to 5V...
lxlramlxl said:
So going on that basis 51 would be 1volt?
Click to expand...
YES
 
L

lxlramlxl

New Member
BESQUEUT said:
Picaxe Manual section 2 page 29 : adcconfig

The default Vref+signal for the ADC is the power supply (V+)

+ Appendix 5 & 6 : Picaxes variations
Voltage Range (V) for 20 X2 = 1.8-5.5

With Picaxes M2, you also have FVR ; see fvrsetup...

Your power supply is probably 5V, but can be anything from 2V to 5V...
YES
Click to expand...
Thanks alot for your help. I will be testing the program tomorrow. This is what I have
Code: 
main:
do
	ReadADC c.2 , b0			; read in pot value
	w1 = b0/10*196			; delay in milliseconds (Divided by 10 then multiplied by 196 to make it more accurate.)
	low c.1
	pause w1				; or could use pulsout c.1,w1
	high c.1	
	pause w1
      low c.1
	
		;
loop
If im right this should give me 1 second delay per volt.
 
S

srnet

Senior Member
lxlramlxl said:
I don't mean to be ignorant but isn't it the same?
Click to expand...
No, work through your example, assume say b0 = 128.

When you do a 'b0/10' you end up with 12 as the 0.8 is dropped, since this is integer only maths.

Now do it the way BQ suggested.
 
L

lxlramlxl

New Member
I just did it with 255.
255*196/10=4998
255/10*196=4998

51/10*196=999.6
51*196/10=999.6

I'm pretty sure the maths is correct. Unless its a way the picaxe works with decimals (I think it doesn't work with decimals)

So am I right thinking 51/10 = 5.1
Picaxe reads as 5
5 * 196 =980
 
hippy

hippy

Technical Support
Staff member
N / 10 * 196

By dividing first, 0 to 9 gives a result of zero, 10 to 19 gives a result of 196, 20 to 29 gives a result of 392.

By multiplying then dividing you will have fewer jumps, a much smoother progression as the pot value changes. For example 15 would give a result of 294
 
B

BESQUEUT

Senior Member
BESQUEUT said:
To be accurate with integer math, it is better to write :
w1 = b0*196/10
Click to expand...
With your code, w0 will be by 196 increments,
with mine, 19 or 20 increments.

lxlramlxl said:
I just did it with 255.
Click to expand...
255*196/10=4998 : OK
255/10*196=4998 KO : Integer result is 4900

51/10*196=999.6 KO : Integer result is 980
51*196/10=999.6 KO : Integer result is 999
lxlramlxl said:
So am I right thinking 51/10 = 5.1
Picaxe reads as 5
5 * 196 =980
Click to expand...
YES
Also, be aware that :
355*296=39544
This is because w0 is a 16 bits number, so when result > 65535 you have to substract 65536...
This is true for intermediate calculations, so :
355*296/10=3954
No problem with readADC, but be carefull if using readADC10...
 
Last edited:
You must log in or register to reply here.
Top