Indexer not indexing...a basic PBP problem??

[Archangel]

Hi jellis00,
Don't know if this is a problem or not, I have never seen it done this way before:

C1.Byte0 
C1.Byte1

I have seen it done this way:

C1.HighByte
C1.LowByte

[Jerson]

Jellis

There seems to be some inconsistency in your code. Before the while, you write 7,i
In the while, you do i=i+c1 followed by write 8,i The variable i being a word sized entity would write to 8,9. Is this what you're seeing? Location 8 will always read 0 since the high byte of i would be zero (you're counting till 20 only). Other than this, the only reason for not breaking out of the code will be C1 never catches any counts.

Jerson

[Ioannis]

Also I don't like the pause 200 inside the loop. Are you sure you are not missing any events there?

Ioannis

[Archangel]

Jellis

There seems to be some inconsistency in your code. Before the while, you write 7,i
In the while, you do i=i+c1 followed by write 8,i The variable i being a word sized entity would write to 8,9. Is this what you're seeing? Location 8 will always read 0 since the high byte of i would be zero (you're counting till 20 only). Other than this, the only reason for not breaking out of the code will be C1 never catches any counts.

Jerson

Hi Jerson,
So he would need to do write 8,i.HighByte Write 9,i.LowByte ?
EDIT: thinking about it . . . if he is only counting to 20, why use a word variable at all, should be a byte , Now I get it.

[Jerson]

Hi Joe

The initial write is to locations 7 and 8. The write inside the while loop is to 8,9. So, the location that I assume he's looking at is 7 this obviously wouldn't change from 0.

[Archangel]

Hi Joe

The initial write is to locations 7 and 8. The write inside the while loop is to 8,9. So, the location that I assume he's looking at is 7 this obviously wouldn't change from 0.

Hi Jerson, I get that, if you store your lawnmower in your own garage, looking next door will likely not find it. I am curious though, his statement :

WRITE 5,C1.Byte0 ' Write count into EEPROM location 5,6 for post-run count test
WRITE 6,C1.Byte1

is this valid in PBP ? And my second question pertains to his declaring i as WORD, Wouldn't he write it as var.lowbyte ?

[jellis00]

Thanks to the excellent advice from you guys, I modified the code and it now works in the form shown below. Notice I also changed the processing epoch of the COUNT loop and reduced the Pause, because of one of your comments. With this sampling rate of the sensor pulse input rate I shouldn't miss any events. Even if I miss a couple, this resolution will adequately support the application.

I am also going to continue to use WORD rather than BYTE for the variables since the production version may encounter more than 256 events during the combined epochs.

When I finally understood from your comments that WRITE was overwriting the EEPROM address that I was using to see if the indexer was indexing with a HighByte that was continuously zero, I was able to solve the problem. Actually, the indexer statement was working....I just couldn't see that it was by the way I was trying to monitor it.
Can't tell you all enough how much I appreciate your help in resolving this problem! As a long term Visual Basic programmer, my faith in PBP is restored!

'Registers Settings
TRISA = 0           ' Set PORTA pins to outputs
TRISB = 0           ' Set all PORTB and PORTC pins to outputs
TRISC = 0
ANSEL = 0           ' Set all pins to digital
ANSELH = 0
PORTC = 0			' Preset LEDs off
TRISA.1 =1          ' Set RA1 as input port for meter pulse inputs

'Variables Initialization
 switchstate   var bit ' Stores value of last measured switchstate
 k             CON 60  ' Calibration factor for flow meter...# pulses per gal
 i             VAR WORD    ' Index used in Gallon counter Repeat...Until loop
 C1            VAR WORD    ' Storage variable for count from COUNTER

'Start...normal code here
MAIN:
      'Send PWM pulse to latching solenoid to open valve
        'Generates 10 millisec pulse on RC3 output pin
         LOW PORTC.3            ' Initialize output pulse polarity
         PULSOUT PORTC.3,1000   ' Generate 10 msec pulse to RC3               
  
    ' Valve should be open at this point and water flowing
            
    ' Start measuring output of flow meter at RA1 input pin
    C1 = 0           ' Initialize pulse count variable to zero
    i = 0            ' Initialize total pulse count index to zero
    REPEAT ' Assume it takes k accumulated pulses for required gallons
        ' Count the number of pulses from Hall Effect Magnetic Sensor in 5 sec
        ' epochs and add count from each epoch to total accumulation and then 
        ' shutoff the valve when count meets i>k criteria.
        Count PORTA.1,5000,C1
        'WRITE 5,C1.HighByte    ' Write count into EEPROM location 5,6 during
        'WRITE 6,C1.LowByte     ' pre-production testing
         'Slowly blink LED during water flow  
           HIGH PORTC.0         ' If we get here, Blink the LED once
           PAUSE 100
           LOW PORTC.0
        i = i + C1              ' Add count to total pulse accumulation indexer
     'When i >= k, required gallons have flowed by the meter
     UNTIL i > k
     'WRITE 7,i.HighByte   'Store result as test of code..not in production
     'Write 8,i.LowByte
     GoTo ShutValve  

ShutValve: 'Send PWM pulse to latching solenoid to close valve
            PULSOUT PORTC.3,1000   ' Generate required 10 msec pulse to RC3
  
    ' Valve should be closed at this point and no water flowing
    
    ' Put code here to test if flow meter has stopped as an indication that
    ' no water is flowing...is a double check on valve closure.
        C1 = 0                 ' Reset pulse counter to zero
        Count PORTA.1,10000,C1 ' Collect meter pulse input, if any, for 10 secs
        IF C1 > 0 Then
            ' Water is still flowing...close the valve!
            PULSOUT PORTC.3,1000   ' Command valve solenoid to close
        ENDIF
 
    ' Put code here to put Microcontroller back to Sleep State to wait for
    ' next Interrupt
        SLEEP = 65,535 
END

[Archangel]

Thanks to the excellent advice from you guys, I modified the code and it now works in the form shown below. Notice I also changed the processing epoch of the COUNT loop and reduced the Pause, because of one of your comments. With this sampling rate of the sensor pulse input rate I shouldn't miss any events. Even if I miss a couple, this resolution will adequately support the application.

I am also going to continue to use WORD rather than BYTE for the variables since the production version may encounter more than 256 events during the combined epochs.

When I finally understood from your comments that WRITE was overwriting the EEPROM address that I was using to see if the indexer was indexing with a HighByte that was continuously zero, I was able to solve the problem. Actually, the indexer statement was working....I just couldn't see that it was by the way I was trying to monitor it.
Can't tell you all enough how much I appreciate your help in resolving this problem! As a long term Visual Basic programmer, my faith in PBP is restored!

'Registers Settings
TRISA = 0           ' Set PORTA pins to outputs
TRISB = 0           ' Set all PORTB and PORTC pins to outputs
TRISC = 0
ANSEL = 0           ' Set all pins to digital
ANSELH = 0
PORTC = 0			' Preset LEDs off
TRISA.1 =1          ' Set RA1 as input port for meter pulse inputs

'Variables Initialization
 switchstate   var bit ' Stores value of last measured switchstate
 k             CON 60  ' Calibration factor for flow meter...# pulses per gal
 i             VAR WORD    ' Index used in Gallon counter Repeat...Until loop
 C1            VAR WORD    ' Storage variable for count from COUNTER

'Start...normal code here
MAIN:
      'Send PWM pulse to latching solenoid to open valve
        'Generates 10 millisec pulse on RC3 output pin
         LOW PORTC.3            ' Initialize output pulse polarity
         PULSOUT PORTC.3,1000   ' Generate 10 msec pulse to RC3               
  
    ' Valve should be open at this point and water flowing
            
    ' Start measuring output of flow meter at RA1 input pin
    C1 = 0           ' Initialize pulse count variable to zero
    i = 0            ' Initialize total pulse count index to zero
    REPEAT ' Assume it takes k accumulated pulses for required gallons
        ' Count the number of pulses from Hall Effect Magnetic Sensor in 5 sec
        ' epochs and add count from each epoch to total accumulation and then 
        ' shutoff the valve when count meets i>k criteria.
        Count PORTA.1,5000,C1
        'WRITE 5,C1.HighByte    ' Write count into EEPROM location 5,6 during
        'WRITE 6,C1.LowByte     ' pre-production testing
         'Slowly blink LED during water flow  
           HIGH PORTC.0         ' If we get here, Blink the LED once
           PAUSE 100
           LOW PORTC.0
        i = i + C1              ' Add count to total pulse accumulation indexer
     'When i >= k, required gallons have flowed by the meter
     UNTIL i > k
     'WRITE 7,i.HighByte   'Store result as test of code..not in production
     'Write 8,i.LowByte
     GoTo ShutValve  

ShutValve: 'Send PWM pulse to latching solenoid to close valve
            PULSOUT PORTC.3,1000   ' Generate required 10 msec pulse to RC3
  
    ' Valve should be closed at this point and no water flowing
    
    ' Put code here to test if flow meter has stopped as an indication that
    ' no water is flowing...is a double check on valve closure.
        C1 = 0                 ' Reset pulse counter to zero
        Count PORTA.1,10000,C1 ' Collect meter pulse input, if any, for 10 secs
        IF C1 > 0 Then
            ' Water is still flowing...close the valve!
            PULSOUT PORTC.3,1000   ' Command valve solenoid to close
        ENDIF
 
    ' Put code here to put Microcontroller back to Sleep State to wait for
    ' next Interrupt
        SLEEP = 65,535 
END

Hi jellis00,
My question was, . . . actually, answered in this final cut of your code. Thank You