MPPT solar battery charger

[iw2fvo]

I have built up a solar battery charger to be used in my Radio Ham station. It is used to keep charge a 27 AH 12 V SLA battery.
The panel power is 50 watt for now. I am NOT a programmer or a software guy and so I wrote the program for the pic18f4523 using a picbasic pro in a very preliminary mode.
The Pic is running at 40 Mhz clock speed and controls the duty cycle of a buck converter at 75 khz frequency.
The whole charger works well : there is a small problem relevant to the status selection that appears when the sky is cloudy (but not always): it jumps between status 0 and status 3.
This is not really a problem but I would like to fix it and make it stable in all the sky condition.
I told before that I am not an expert in programming so I would like if someone could help to improve the software that I wrote: I also think that there could be a lot of simplifications, corrections and others to be improved.
I will try to attach the basic program here. I could post the schematics diagram too but it is hand drawn.
Thanks in advance for any assistance .
Beste regards,
Ambrogio

[iw2fvo]

Demon,
thanks for having respond to my msg.
here is the code.
I will re_draw the schematics ( by hand )in a decent way tonight. I will post it tomorrow.
thnks
Ambrogio

'****************************************************************
'*  Name    :  pic 18f4523                                      *
'*  Author  : [select VIEW...EDITOR OPTIONS]                    *
'*  Notice  : Copyright (c) 2014 [select VIEW...EDITOR OPTIONS] *
'*          : All Rights Reserved                               *
'*  Date    : 28/04/2014                                        *
'*  Version : 1.0                                               *
'*  Notes   : modificare file 18f4523 nella cartella PBP        *
'*          : per il config ... oscillatore MCLR etc            *
'* picbasic pro 2.6  : long used.  12294 BYTES USED.            *
'* 50 WATT SOLAR PANEL, 27 AH 12 VDC SLA BATTERY                *
'****************************************************************
 
' WARNING:
' HSPLL OSCILLATOR HAS BEEN DEFINED IN THE PBP FOLDER > 18F4523 FILE

DEFINE OSC 40    ' 10 MHz XTAL  AND x4 PLL GIVES 40 MHz CLOCK SPEED.
INCLUDE "MODEDEFS.BAS"

'DEBUG DEFINITION> WITH_OUT MAX 232...USE INVERTED=1, 57600 BAUD 8N1

' Set Debug pin port   ________________________________________
DEFINE DEBUG_REG PORTC
' Set Debug pin bit   _________________________________________
DEFINE DEBUG_BIT 6
' Set Debug baud rate _________________________________________
DEFINE DEBUG_BAUD 57600
' Set Debug mode: 0 = true, 1 = inverted ______________________
DEFINE DEBUG_MODE 1


' LCD CONFIGURATION ...HD_4470 LIKE > 4 LINES BY 20 CHRS IS USED HERE

' Set LCD Data port   _________________________________________ 
DEFINE LCD_DREG PORTD
' Set starting Data bit (0 or 4) if 4-bit bus ________
DEFINE LCD_DBIT 0
' Set LCD Register Select port  (R/S)  ________________________ 
DEFINE LCD_RSREG PORTD
' Set LCD Register Select bit   _______________________________
DEFINE LCD_RSBIT 5
' Set LCD Enable port    (EN_DCDC)__________________________________  
DEFINE LCD_EREG PORTD
' Set LCD Enable bit       ____________________________________
DEFINE LCD_EBIT 4
' Set LCD bus size (4 or 8 bits)  _____________________________
'DEFINE LCD_BITS 4
' Set number of lines on LCD     ______________________________
DEFINE LCD_LINES 4
' Set command delay time in us   ______________________________
DEFINE LCD_COMMANDUS 2500
' Set data delay time in us   _________________________________
DEFINE LCD_DATAUS 50
PAUSE 500

'  SOME CONSTANTS TO CONTROL THE LCD.
 
INS   CON 254 '   $FE: INSTRUCTION COMMAND MODE
CLR   CON 1   '   CLEAR LCD , GOTO FIRST LINE, FIRST CHARACTER
DG    CON 223 '   SIMBOLO DEL GRADO °
LINE1 CON 128 '   $80: GOTO LINE 1
LINE2 CON 192 '   $C0: GOTO LINE 2
LINE3 CON 148 '   $94: GOTO LINE 3
LINE4 CON 212 '   $D4: GOTO LINE 4
' eg.: LCDOUT INS, LINE1, "PIBPO TXT=",........

' ADC CONFIGURATION

ADCON1=%00001010   ' AN0,AN1,AN2,AN3,AN4 ENABLED.THE REMAINING PINS ARE DIGITAL
VB_ADC VAR WORD    ' AN0 PIC_PIN 2     ' BATTERY VOLTAGE
VP_ADC VAR WORD    ' AN1 PIC_PIN 3     ' PANEL VOLTAGE
IP_ADC VAR WORD    ' AN2 PIC_PIN 4     ' PANEL CURRENT
IB_ADC VAR WORD    ' AN3 PIC_PIN 5     ' BATTERY VOLTAGE
POT_ADC VAR WORD   ' AN4 PIC_PIN 7     ' POT VOLTAGE FOR MANUAL DUTY INPUT...

' FILTERS          '........................................................
IPT VAR WORD
VPT VAR WORD
VBT VAR WORD
IBT VAR WORD
AVG VAR BYTE
AVG=8               ' MAX AVG=16  4096*16=65536  IN A WORD !    PANEL  data
AVG_B VAR BYTE
AVG_B=4             ' MAX AVG_B=16  4096*16=65536  IN A WORD !  BATTERY data
I VAR BYTE
Y VAR BIT
Y=0
Z VAR BYTE
Z=0
DLY_FLG VAR BIT
DLY_FLG =1
IPF_ADC VAR WORD
VPF_ADC VAR WORD
VBF_ADC VAR WORD
IBF_ADC VAR WORD
'...................................

VP VAR LONG
IP VAR LONG
VB VAR LONG
IB VAR LONG
WP VAR LONG
WB VAR LONG
IBSIGN VAR BYTE    ' BATTERY CURRENT SIGN

DUTY VAR WORD      ' PIC PWM DUTY CYCLE CONTROL 
DUTY=150           '  START  AT MINIMUM DUTY
GOSUB SET_PWM      
DUTY_S VAR BYTE    ' DUTY CYCLE STEP for MPPT 
DUTY_S=1           ' MUST BE WIDE ENOUGH TO OVERCOME PANEL DATA NOISE ! 
                   ' IF MPP IS NOT FOUND USE 2, 3, 4 OR 5...
LD_EN VAR PORTC.1  ' PIC_PIN 16   LOAD POWER MOSFET CONTROL PIN
LOW LD_EN          ' DISABLE LOAD POWER MOSFET
           

EN_DCDC VAR PORTC.3     ' PIC_pin 18   DC/DC CONVERTER ENABLE
low EN_DCDC             ' DISABLE DC/DC CONVERTER

LOAD VAR BIT
LOAD=1

MODE VAR BYTE
MODE=0

STATE VAR BYTE
STATE=0
' STATE 0= OFF
' STATE 1= ON
' STATE 3= FLOAT
' STATE 4= BULK

DV VAR BIT         ' PANEL DELTA VOLT USED FOR HILL CLIMBING
DP VAR BIT         ' PANEL DELTA POWER
VP_OLD VAR LONG    ' OLD PANEL VOLTAGE
WP_OLD VAR LONG    ' OLD PANEL POWER

' SCAN FOR MPP ......................

WPMAX VAR LONG
WPMAX=0
VPMAX VAR LONG
VPMAX=0
IPMAX VAR LONG
IPMAX =0
VBMAX VAR LONG
IBMAX VAR LONG
WBMAX VAR LONG
DUTYMAX VAR BYTE
DUTYMAX=0


' BACK LIGHTING    ......................................................

BCK_LT VAR PORTA.4     ' PIC_PIN 6
FOR I=3 TO 0 STEP -1   ' LCD BACK LIGHT TEST AT POWER ON
    HIGH BCK_LT
    pAUSE 100
    LOW BCK_LT
    PAUSE 100
NEXT I
HIGH BCK_LT            ' LEAVE BACK LIGHT ON AT THE END ...  


' EXTERNAL POWER  ........................................................

EXT_PWR_RLY VAR PORTC.0    ' PIC_PIN 15 OUTPUT CONTROL FOR EXT_PWR SOURCE RELAY 
HIGH EXT_PWR_RLY           ' DISABLE EXT PWR_RELAY   HIGH= DISABLE                                    
EXTPWR_OK VAR PORTC.4      ' PIC_PIN 23  INPUT > POWER_LINE STATUS; 1=OK
U_VOLT VAR BIT             ' EXT POWER MONITOR 
U_VOLT=0                   ' EXT POWER GOOD
U_FLG VAR BIT      ' BATTERY UNDERVOLTAGE FLAG
U_FLG=1 
EX VAR BIT
EX=0

' SPLASH SCREEN:

Lcdout INS,line1, "Ambrogio 06/04/2014 "    ' LCD CHECK
Lcdout INS,line2, "PV PIC18F4523       "
debug 13,10,13,10
DEBUG " STARTING H E R E ! ! ! ! "          ' SERIAL LINK CHECK
PAUSE 1000
LCDOUT INS,CLR                              ' TURN LCD OUT

' PUSH BUTTONS PIN ASSIGNEMENT  

PB_1 VAR PORTE.0       'PIC_PIN 8  STEPS THROUGH THE MODES
PB_2 VAR PORTE.1       'PIC_PIN 9
PB_3 VAR PORTE.2       'PIC_PIN 10  TOGGLE LOAD MOSFET 
                  

'********************************************************************
MAIN:  ' ************************************************************
'********************************************************************

' PB_1 SELECTS THE MODES ...
' mode =0 is THE normal OPERATING MODE
' mode =1 pot input controls the duty cycle
' mode =2 scan duty cycle to search for the MPP CONDITION
' mode =3 TOGGLES the duty cycle to see the system response caracteristics

' MODES SELECTION....................

IF PB_1=0 THEN          ' MODE SELECTION...
   MODE=MODE+1
   IF MODE >=4 THEN
      MODE=0
   ENDIF
   
   IF MODE=1 THEN
      LCDOUT INS, CLR
      HIGH EN_DCDC
   ENDIF
   
   IF MODE=2 THEN
      LCDOUT INS, CLR
      WPMAX=0         ' reset maximum registers
      VPMAX=0
      IPMAX=0
      VBMAX=0
      IPMAX=0
      WBMAX=0
      DUTYMAX=0
      HIGH EN_DCDC         
      DUTY=150          ' THIS IS MY MINIMUM DUTY CYCLE
      GOSUB SET_PWM
      PAUSE 1000
   ENDIF
   
   IF MODE=3 THEN
      LCDOUT INS, CLR
      LCDOUT INS, LINE1, " DC STEP TEST " 
   ENDIF
   
WHILE PB_1=0: WEND
ENDIF


' LOW VOLTAGE PROTECTION ...and load control..................................

IF ( VB<11800 ) THEN       ' BATT LOWER THAN 11.8 Vdc
   U_VOLT=1                ' SET UNDER_VOLTAGE FLAG 
   LOW LD_EN               ' DISABLE LOAD
   LOAD=0
   U_FLG=1
ELSE
    IF ( VB>12600 ) THEN   ' IF VBATT MORE THAN 12.6 Vdc
       U_VOLT=0            ' RESET UNDER_VOLTAGE FLAG
       IF U_FLG=1 THEN
          HIGH LD_EN       ' ENABLE LOAD
          LOAD=1
          U_FLG=0
       ENDIF
     ENDIF
ENDIF

IF (U_VOLT=1) AND (EXTPWR_OK=1) THEN
   LOW EXT_PWR_RLY             ' ACTIVATE EXT POWER RELAY
   EX=1
ELSE
    IF U_VOLT=0 THEN
       HIGH EXT_PWR_RLY        ' RELASE EXT POWER RELAY
       EX=0
    ENDIF
ENDIF

IF U_VOLT=0 THEN
    IF PB_3=0 THEN
       if LOAD=0 THEN
          LOAD=1
       ELSE
          LOAD=0
       ENDIF
       IF LOAD=0 THEN
          LOW LD_EN
       ENDIF
       IF LOAD=1 THEN
          HIGH LD_EN
       ENDIF   
    WHILE PB_3=0 : WEND
    ENDIF
ENDIF   


' MODES MANAGEMENT BY PUSH BUTTONS ...........................................

IF MODE=0 THEN         ' NORMAL MODE  ........................................
   GOTO NORMAL_MODE
ENDIF

IF MODE=1 THEN         ' POT CONTROLLED DUTY CYCLE............................
     
     LCDOUT INS, LINE1,"POT CONTROLLED DUTY:"
     GOSUB ACQ_POT     ' ACQ TAKES ABOUT 2.3 mSEC
     DUTY=((POT_ADC/39)+149)   ' POT CONTROLS DUTY 0>4096 = 150 to 254
     GOSUB SET_PWM
     GOSUB GET_PANEL_DATA
     GOSUB GET_BATTERY_DATA
     LCDOUT INS, LINE3,DEC VP/1000,".",DEC1 VP/100,"  ",DEC IP/1000,_
           ".",DEC2 IP/10,"  ",DEC WP/1000,".",DEC1 WP/100," ", DEC DUTY,"  "
     LCDOUT INS, LINE4,DEC VB/1000,".",DEC1 VB/100," ",IBSIGN,DEC IB/1000,_
           ".",DEC2 IB/10, " ",IBSIGN,DEC WB/1000,".",DEC1 WB/100,"  "
     GOSUB SERIAL_OUT
     GOTO MAIN
ENDIF   

IF MODE=2 THEN         ' SCAN DUTY CYCLE FOR MPP SEARCHING.................
   
   DUTY=DUTY+1         ' INCREMENT DUTY CYCLE
   GOSUB SET_PWM
   PAUSE 10            ' NOT REALLY NEEDED .. LCDOUT PROVIDES TIME DLY
   GOSUB GET_PANEL_DATA
   GOSUB GET_BATTERY_DATA
   
   IF WP> WPMAX THEN   ' STORE MAX READINGS
      WPMAX=WP
      VPMAX=VP
      IPMAX=IP
      VBMAX=VB
      IBMAX=IB
      WBMAX=WB
      DUTYMAX=DUTY
   ENDIF
   
   IF DUTY >= 254 THEN
      LCDOUT INS, CLR
      LCDOUT INS, LINE1, " MAX POWER :        "
      LCDOUT INS, LINE3,DEC VPMAX/1000,".",DEC1 VPMAX/100,"  ",_
             DEC IPMAX/1000,".",DEC2 IPMAX/10,"  ",DEC WPMAX/1000,_
             ".",DEC1 WPMAX/100," ", DEC DUTYMAX
      LCDOUT INS,LINE4,DEC VBMAX/1000,".",DEC1 VBMAX/100,"  ",_
             DEC IBMAX/1000,".",DEC2 IBMAX/10,"  ",DEC WBMAX/1000,_
             ".",DEC1 WBMAX/100    
      DUTY=150
      GOSUB SET_PWM
      while PB_2=1 :wend       ' WAIT FOR PB_2 TO BE PRESSED
      
   ENDIF
     
   lcdout ins, line1, "MAX PWR SEARCHING:"
   LCDOUT INS, LINE3,DEC VP/1000,".",DEC1 VP/100,"  ",DEC IP/1000,_
       ".",DEC2 IP/10,"  ",DEC WP/1000,".",DEC1 WP/100," ", DEC DUTY,"  "
   LCDOUT INS, LINE4,DEC VB/1000,".",DEC1 VB/100," ",IBSIGN,DEC IB/1000,_
       ".",DEC2 IB/10, " ",IBSIGN,DEC WB/1000,".",DEC1 WB/100,"  "
   
   GOSUB SERIAL_OUT
   GOTO MAIN
   
ENDIF   

IF MODE=3 THEN    'DUTY STEP CHANGE RESPONSE TEST > SET TWO DUTY VALUES
                  'SET PAUSE FOR STEP TIME_WIDTH ,,,,,,,,,,,,,,,,,,,,,,,,,  
                  ' system response time to duty change is about 15 msec
                  ' measured with a scope
   STEP_DUTY:
   IF Y=0 THEN
      HIGH BCK_LT      ' CONNECT TO SCOPE TRIGGER INPUT
      DUTY=170         ' FIRST DUTY VALUE
      GOSUB SET_PWM
      Y=1
      PAUSE 50         ' WIDTH
      GOSUB GET_PANEL_DATA
      GOSUB SERIAL_OUT
   ENDIF
   
   IF Y=1 THEN
       low bck_lt
       DUTY=175       ' SECOND DUTY VALUE  > DUTY_step is 5 now
       GOSUB SET_PWM
       Y=0
       PAUSE 50       ' step time WIDTH  IS 50 MSEC
       GOSUB GET_PANEL_DATA
       GOSUB SERIAL_OUT
       GOTO MAIN
    ENDIF
ENDIF

NORMAL_MODE:    '*************************************************************

' BATTERY STATE CONTROL ......................................................

GOSUB GET_PANEL_DATA         ' 2.3 msec x avg
GOSUB GET_BATTERY_DATA       ' 2.3 msec x avg_b


SELECT CASE STATE       ' MODIFIED TIM NOLAND PROGRAM IN c LANGUAGE... ( ! )

     CASE 0    ' OFF  ***  * * * * * * * * * * * * * * * * * * * * * * * * * *
     
         Z=Z+1
         IF Z>=254 THEN  ' wait sometime for the battery to settle down
            DLY_FLG=0
         ENDIF
         DUTY=150       ' MINIMUM DUTY CYCLE.
         low EN_DCDC         ' DISABLE DC / DC CONVERTER
         GOSUB SET_PWM
         debug dec z,13,10         
         IF DLY_FLG=0 THEN
             IF VP > VB+1500 THEN      ' day light is coming ...
                IF VB < 13800 THEN
                    DUTY=150
                    HIGH EN_DCDC
                    GOSUB SET_PWM
                    PAUSE 1000
                    STATE=1 ' >> ON
                ENDIF
             ENDIF
         ENDIF
         
     CASE 1    ' ON  ***  * * * * * * * * * * * * * * * * * * * * * * * * * *
         
         HIGH EN_DCDC    
         IF (VP < VB) and ( IP<80 ) THEN     ' it is dark 
            DLY_FLG=1  : Z=0
            DUTY=150
            GOSUB SET_PWM
            STATE= 0  ' >> OFF
         ENDIF
         
         IF VB > 14800 THEN        ' MAX BATTERY VOLTAGE IS 14.8 VDC
            STATE=2   ' >> FLOAT
            DUTY=150
            GOSUB SET_PWM
         ENDIF
                 
         IF ( WP < 5000 ) THEN
            DUTY=254
            GOSUB SET_PWM
         ENDIF
                 
         IF WP> 5000 THEN
            DUTY=240
            GOSUB SET_PWM
            STATE=3   ' >> BULK
         ENDIF
         
     CASE 2   ' FLOAT  *** * * * * * * * * * * * * * * * * * * * * * * * * *
         
         HIGH EN_DCDC         
         IF VB > 14200 THEN      ' KEEP FLOAT VOLTAGE @ 13.8 TO 14.2 VDC
            DUTY=DUTY-1          ' DECREASE DUTY
            GOSUB SET_PWM
         ENDIF
         IF VB < 13800 THEN
            DUTY=DUTY+1          ' INCREASE DUTY
            GOSUB SET_PWM
         ENDIF
    
         IF DUTY >= 254 THEN
             DUTY=200
             GOSUB SET_PWM
             STATE=3  ' >> BULK
         ENDIF
         
         IF (VP < VB) and ( IP<80 ) THEN
            DLY_FLG=1 : Z=0
            DUTY=150
            GOSUB SET_PWM 
            STATE=0 '  >> OFF
         ENDIF
         
     CASE 3   ' BULK
         
         HIGH EN_DCDC
         IF (WP < 5000)  THEN
            STATE= 1  ' >> ON
         ENDIF
         
         IF WP >= 5000 THEN 
            GOSUB MPPT
         ENDIF
         
         IF (VP < VB) and ( IP<80 ) THEN
             DLY_FLG=1  : Z=0
             DUTY=150
             GOSUB SET_PWM
             STATE=0  ' >> OFF 
         ENDIF
         
         IF VB > 14800 THEN
            DUTY=150
            GOSUB SET_PWM
            STATE=2  ' >> FLOAT
         ENDIF
         
END SELECT


' DATA OUTPUT TO LCD, SERIAL AND RADIO.......................................

GOSUB DIS_LCD         ' execution time is 15 msec

GOSUB SERIAL_OUT      ' execution time is 25 msec

GOTO MAIN             ' main loop execution time is 65 msec if avg=8 and
                      ' avg_b=4

'============================================================================
'========================  SUBROUTINES  =====================================
'============================================================================


ACQ_PV:    'ACQUIRE PV ANALOG DATA      ... 2.3 msec ........................

    ' READ ANALOG 0 VP   .........................
    ADCON0=%00000000    ' SELECT AN0, VP
    ADCON2=%10100110    ' ACQ TIME AND FOSC SELECTION
    ADCON0=%00000001    ' ON, ACCESO
    PAUSE 1             ' INPUT RESISTANCE AND CAP FOR DC ACQISISTION 
    ADCON0=%00000011    ' ON + GO
    WHILE ADCON0.1=1    ' WAIT FOR CONVERSION COMPLETED , DONE
    WEND
    VP_ADC=(ADRESH<<8) + ADRESL   ' ADC RESULT  12 BIT 
    ADCON0=%00000000    ' TURN OFF, 
    
    'READ ANALOG 2 IP    ........................
    ADCON0=%00001000    ' SELECT AN0, IP
    ADCON2=%10100110    ' ACQ TIME AND FOSC SELECTION
    ADCON0=%00001001    ' ON, ACCESO
    PAUSE 1             ' INPUT RESISTANCE AND CAP FOR DC ACQISISTION 
    ADCON0=%00001011    ' ON + GO
    WHILE ADCON0.1=1    ' WAIT FOR CONVERSION COMPLETED , DONE
    WEND
    IP_ADC=(ADRESH<<8)+ ADRESL   ' ADC RESULT  12 BIT 
    ADCON0=%00000000    ' TURN OFF, 

RETURN

ACQ_BATT:    ' ACQUIRE BATTERY ANALOG DATA   ........ 2.3 msec ................

    ' READ ANALOG 1 VB  ..........................
    ADCON0=%00000100    ' SELECT AN0, VB
    ADCON2=%10100110    ' ACQ TIME AND FOSC SELECTION
    ADCON0=%00000101    ' ON, ACCESO
    PAUSE 1             ' INPUT RESISTANCE AND CAP FOR DC ACQISISTION 
    ADCON0=%00000111    ' ON + GO
    WHILE ADCON0.1=1    ' WAIT FOR CONVERSION COMPLETED , DONE
    WEND
    VB_ADC=(ADRESH<<8) + ADRESL   ' ADC RESULT  12 BIT 
    ADCON0=%00000000    ' TURN OFF, 
    
    
    'READ ANALOG 3 IB   ........................
    ADCON0=%00001100    ' SELECT AN0, IB
    ADCON2=%10100110    ' ACQ TIME AND FOSC SELECTION
    ADCON0=%00001101    ' ON, ACCESO
    PAUSE 1             ' INPUT RESISTANCE AND CAP FOR DC ACQISISTION 
    ADCON0=%00001111    ' ON + GO
    WHILE ADCON0.1=1    ' WAIT FOR CONVERSION COMPLETED , DONE
    WEND
    IB_ADC=(ADRESH<<8)+ ADRESL   ' ADC RESULT  12 BIT 
    ADCON0=%00000000    ' TURN OFF, 

RETURN

ACQ_POT:    'ACQUIRE POT ANALOG VOLTAGE   ....................................

    'READ ANALOG 4 POT
    ADCON0=%00010000    ' SELECT AN0, VPOT
    ADCON2=%10100110    ' ACQ TIME AND FOSC SELECTION
    ADCON0=%00010001    ' ON, ACCESO
    PAUSE 1             ' INPUT RESISTANCE AND CAP FOR DC ACQISISTION 
    ADCON0=%00010011    ' ON + GO
    WHILE ADCON0.1=1    ' WAIT FOR CONVERSION COMPLETED , DONE
    WEND
    POT_ADC=(ADRESH<<8)+ ADRESL   ' ADC RESULT  12 BIT 
    ADCON0=%00000000    ' TURN OFF, 

RETURN

DIS_LCD:     '.................... 15 msec ...................................
    
    IF STATE=0   THEN   LCDOUT INS, LINE1,"OFF  "
    IF STATE=1   THEN   LCDOUT INS, LINE1,"ON   " 
    IF STATE=2   THEN   LCDOUT INS, LINE1,"FLOAT"
    iF STATE=3   THEN   LCDOUT INS, LINE1,"BULK "
    IF U_VOLT=1  THEN   LCDOUT LINE1,"LOAD> OFF > UV"
    IF LOAD=0    THEN   LCDOUT LINE1,"LOAD> OFF     "
    IF LOAD=1    THEN   LCDOUT LINE1,"LOAD> ON      "
    IF EXTPWR_OK=1 THEN LCDOUT INS,LINE2,"LINE> AVL    " 
    IF EXTPWR_OK=0 THEN LCDOUT INS,LINE2,"LINE> NOT AVL" 
    IF EX=1        THEN LCDOUT INS,LINE2+14,"EXT_PWR"
    IF EX=0        THEN LCDOUT INS,LINE2+14,"       "
    LCDOUT INS, LINE3,DEC VP/1000,".",DEC1 VP/100,"  ",DEC IP/1000,_
           ".",DEC2 IP/10,"  ",DEC2 WP/1000,".",DEC1 WP/100," ", DEC DUTY
    LCDOUT INS, LINE4,DEC VB/1000,".",DEC1 VB/100," ",IBSIGN,DEC IB/1000,_
           ".",DEC2 IB/10, " ",IBSIGN,DEC2 WB/1000,".",DEC1 WB/100,"  "       
    
    

RETURN

SERIAL_OUT:  ' .................. 25 msec.....................................

     DEBUG "VP ",DEC VP/1000,".",DEC1 VP/100,"  IP ",DEC IP/1000,_
           ".",DEC2 IP/10, " WP ", DEC WP/1000,".",DEC1 WP/100,_
           " DC ",  DEC DUTY, " VB ",DEC VB/1000,".",DEC1 VB/100,_
           " IB ",IBSIGN,DEC IB/1000,".",DEC2 IB/10, " WB ",IBSIGN,_
           DEC WB/1000,".",DEC1 WB/100," M ",DEC MODE," S ",DEC STATE,13,10
           
RETURN

 
MPPT:        '...............................................................

    ' PANEL DATA CHANGE
    
    IF WP>WP_OLD THEN         ' POWER INCREASES
       DP=1
    ELSE 
         IF WP<WP_OLD THEN     ' POWER DECREASES
            DP=0
         ENDIF
    ENDIF
        
    if wp=WP_OLD THEN RETURN        
        
    IF VP>VP_OLD THEN         ' VOLTAGE INCREASES
       DV=1
    ELSE 
         IF VP<VP_OLD THEN     ' VOLTAGE DECREASES
            DV=0
         ENDIF
    ENDIF
        
    ' HILL CLIMBING.... FOR MPP SEARCH
    
    ' CLIMBING FROM LEFT
    
    IF DV=1 AND DP=1 THEN
       DUTY=DUTY-DUTY_S     ' DECREASE LOAD TO INCREASE P_VOLTAGE
    ENDIF
    
    'DESCEND TO THE RIGHT
    
    IF DV=1 AND DP=0 THEN
       DUTY=DUTY+DUTY_S     ' INCREASE LOAD TO REDUCE P_VOLTAGE
    ENDIF
    
    ' CLIMBING FROM RIGHT
    
    IF DV=0 AND DP=1  THEN      
       DUTY=DUTY+DUTY_S     ' INCREASE LOAD TO REDUCE P_VOLTAGE
    ENDIF
    
    ' DESCEND TO THE LEFT
    
    IF DV=0 AND DP=0 THEN   
       DUTY=DUTY-DUTY_S     ' DECREASE LOAD TO INCREASE P_VOLTAGE
    ENDIF
      
    GOSUB SET_PWM
    
    WP_OLD=WP
    VP_OLD=VP

RETURN


SET_PWM:  '...................................................................

    IF DUTY >= 254 THEN      ' MAX DUTY FOR IR2104
       DUTY=254
    ENDIF
    
    if   DUTY <=150 THEN     ' CONSIDERING ABOUT 12 VDC OUTPUT..
         DUTY=150
    ENDIF
    
    HPWM 1,DUTY,75000       ' 75 KHz CLOCK FREQUENCY @ DUTY....
    
RETURN


GET_PANEL_DATA:     ' ........................................................

' GET PANEL DATA

    FOR I=1 TO AVG              ' VP AND IP FILTERING LOOP
        GOSUB ACQ_PV            ' ACQ_PV TAKES ABOUT 2.3 mSEC...
        IPT=IPT+IP_ADC
        VPT=VPT+VP_ADC
    NEXT I
    
    IPF_ADC=IPT/AVG
    VPF_ADC=VPT/AVG
    IPT=0 : VPT=0
    VP=(VPF_ADC*7200)/1000      ' MILLIVOLTS
    
    IF IPF_ADC>2052 THEN        ' ACS_712 BIAS IS ABOUT 2.5 VDC... 12 BIT adc
       IPF_ADC=IPF_ADC-2052
    ELSE 
         IF IPF_ADC =< 2052 THEN   
            IPF_ADC=2052-IPF_ADC
         ENDIF
    ENDIF
    IP=(IPF_ADC*6790)/1000     '  MILLIAMPS
    WP=(IP*VP)/1000            '  MILLIWATT
RETURN


GET_BATTERY_DATA:    '........................................................

' GET BATTERY DATA 
    FOR I=1 TO AVG_B
       GOSUB ACQ_BATT      ' ACQ_BATT TAKES ABOUT 2.3 mSEC
       VBT=VBT+VB_ADC
       IBT=IBT+IB_ADC
    NEXT I
    
    VBF_ADC=VBT/AVG_B
    IBF_ADC=IBT/AVG_B
    IBT=0 :VBT=0
    IF IBF_ADC>2053 THEN
      IBF_ADC=IBF_ADC-2053
      IBSIGN="-"
    ELSE 
      IF IBF_ADC=<2053 THEN
         IBF_ADC=2053-IBF_ADC
         IBSIGN="+"
      ENDIF
    ENDIF
    VB=(VBF_ADC*7000)/1000          '  MILLIVOLTS
    IB=(IBF_ADC*6790)/1000          '  MILLIAMPS
    WB=(IB*VB)/1000                 '  MILLIWATT
    
RETURN

[Demon]

... there is a small problem relevant to the status selection that appears when the sky is cloudy (but not always): it jumps between status 0 and status 3...

Are you sure it isn't a debounce issue?

Maybe add 1 second PAUSE to rule that out?

Robert

[iw2fvo]

Thanks Robert,
I will put a delay ( 1 sec or more ) when changing status.
Need to wait for next cloudy days to check if OK .
Do you see any other area om the program that could be improved ?
Ambrogio

[Demon]

Honestly I have trouble concentrating so it's kinda easy for me to lose track of logic.

I try to help with the simple things.

Robert

Edit: your program is huge, have you considered breaking it down?

I'd make a copy, reduce all the subroutines to only display a status on LCD, then add logic one section at a time.

[iw2fvo]

Robert,
I am attaching the schematics digram for the charger and two captures.
I am sorry : it is hand drawn. I am not an electronics expert nor a software man !
This morning I have activated the delay as suggested, I am waiting to see the results in the next days.
The whole device works well and the MPPT is tracked well too.
Thanks for all.
regards,
Ambrogio

[Demon]

Wow!

Wish I could contribute something.

Now you're making me think of adding a solar panel to my alarm project.

Robert

[iw2fvo]

Robert,
i woluld like to attach a photo of the LCD and a screen shot of the output.
I am not able to attach any .jpeg : any direction for me ?
Thanks
Ambrogio

[Demon]

Test attachment using picture icon on posting toolbar:

...

[iw2fvo]

Robert,
they are here.
Ambrogio

[Demon]

Thanks Robert,
I will put a delay ( 1 sec or more ) when changing status.
...

I'd put it here temporarily:

   IF MODE=3 THEN
      LCDOUT INS, CLR
      LCDOUT INS, LINE1, " DC STEP TEST " 
   ENDIF

   PAUSE 1000     ' Test if problem is debounce issue

   ' Not necessary during test ... WHILE PB_1=0: WEND

ENDIF

Robert

[Demon]

... there is a small problem relevant to the status selection that appears when the sky is cloudy (but not always): it jumps between status 0 and status 3...

Oh great, I misread. I thought it was jumping between MODEs. Aaargh, sorry.

Gonna look through logic later. I have errands to do during the day.

Robert

[iw2fvo]

Robert,
i think it ias a problem related to panel power wp, panel voltage vp and duty cycle when switching among the various status...
It needs to find out the proper values that controls the switching...
tnks,
Ambrogio

[Demon]

Robert,
i think it ias a problem related to panel power wp, panel voltage vp and duty cycle when switching among the various status...
It needs to find out the proper values that controls the switching...
tnks,
Ambrogio

Yup. I can't see bad GOSUB structure, nothing that stands out anyways.

The STATE can evolve from 0 to 3 in a single pass of that SELECT CASE. I can only assume the readings from your hardware makes it jump unexpectedly.

Are you able to debug in real time and step thru your code?

It's about the only way I can think of the visualize what is going on. I see you dump a lot of information but I'm unable to follow that manually through your logic.

I prefer to dump like this:

MAIN --------------------------------------------------
STEP 1, list of variables.
SUBROUTINE XYZ ----------------------------------------
STEP 1, list of variables.
STEP 2, list of variables.
SUBROUTINE ABC ----------------------------------------
STEP 1, list of variables.
...etc...

I find it easier to follow in the code this way. I place the STEPS in relevant places, like before & after IFs, etc.

Robert

[iw2fvo]

Robert ,
I agree with you.
The point is the data input to the state machine.
In particular sun condition it plays back and forth between the two status because of the data input.
I have to change / find out some different duty / panel power / panel voltage to be used to control the staus and avoid oscillations.
It does not happen frequently: just in some particular case and in a cloudy day.
The delay applied to the program as stated reduce the oscillation frequency.
Anyway it is not a real problem since it happens when the charging current is already low because of the low sun condition.
Thanks for all.
regards,
Ambrogio

[EarlyBird2]

A very impressive piece of code! I have been programing for over 40 years and never seen a novice produce such elegant code before.

The issue is in your select case section. If case =0 then within this code case is set to 1 then case 1 code is run and again Case is set to 2 then Case is set to 3. Here is the main issue Case 3 is bulk charging because the battery is low as you would want but if a cloud comes over then Panel Voltage is less than Battery Voltage and Case is set to 0. Once the cloud goes then the Case will again revert to 3 bulk charge until the next cloud. Obviously depending on how dark the cloud is there is a possibility that the Panel Voltage can be above Battery Voltage changing to below Battery Voltage and then above as the cloud passes.

This explains what is happening. My only question is why are you switching the Panel on and off?

All my solar panels are permanently connected to the battery regulator. The panels do have a diode in line.

If you do not switch the panels on and off you can take out all of the switching code and solve the problem.

[theFob]

Hi iw,
I'm impressed of your work.
I played once with mppt converters and i know how crazy is.
the attachments show dc simulation of 21V@8A photovoltaic. May be will be useful for you. I have dozens simulations of buck, boost and sepic converters, but i didn't finish the work because the 'client' give up. Now i have 3kg prototype but no panel to play
Allow me to ask you what mppt algorithm you use ?
Some of them are:
* perturb and observe (P&O - sign dP)
* incremental conductance (dP/dV=0)
* constant voltage or current (CV,CI)

May be i'm too lazy to read all your code and find out, sorry.

I suggest reading of this:

Comparative Study of
Maximum Power Point
Tracking Algorithms
D. P. Hohm and M. E. Ropp*,y
Electrical Engineering Department, South Dakota State University, Brookings, SD 5700-2220, USA

It helps me a lot, amazing study.

[iw2fvo]

Thanks EarlyBird2,
Yes , the system works as you described.
The point is that sometime it jumps or oscillates around OFF state. I think your suggestion to do not turn off the converter ( see EN_DCDC ) could solve the problem.
I will try it durnig next week. I will post the results.
Thanks again
Regards,
Ambrogio

[iw2fvo]

Thanks TheFob,
I think that the trick to get a stable MPP tracking is that the "Perturbation" must generate a delta power that is greater than the " noise Power ".
If the VP and IP acquisistion by the PIC ADC has some noise, then the power has a noise too. The perturbation created by " DUTY_step" must generate a delta power response bigger than the noise power.
If the hardware is good then and the noise low we could use a duty_step=1 . If you have a sensible noise you could increase it to 2, 3 , 4 ...
The piece of program is this:

MPPT: '................................................. ..............

' PANEL DATA CHANGE

IF WP>WP_OLD THEN ' POWER INCREASES
DP=1
ELSE
IF WP<WP_OLD THEN ' POWER DECREASES
DP=0
ENDIF
ENDIF

if wp=WP_OLD THEN RETURN

IF VP>VP_OLD THEN ' VOLTAGE INCREASES
DV=1
ELSE
IF VP<VP_OLD THEN ' VOLTAGE DECREASES
DV=0
ENDIF
ENDIF

' HILL CLIMBING.... FOR MPP SEARCH

' CLIMBING FROM LEFT

IF DV=1 AND DP=1 THEN
DUTY=DUTY-DUTY_S ' DECREASE LOAD TO INCREASE P_VOLTAGE
ENDIF

'DESCEND TO THE RIGHT

IF DV=1 AND DP=0 THEN
DUTY=DUTY+DUTY_S ' INCREASE LOAD TO REDUCE P_VOLTAGE
ENDIF

' CLIMBING FROM RIGHT

IF DV=0 AND DP=1 THEN
DUTY=DUTY+DUTY_S ' INCREASE LOAD TO REDUCE P_VOLTAGE
ENDIF

' DESCEND TO THE LEFT

IF DV=0 AND DP=0 THEN
DUTY=DUTY-DUTY_S ' DECREASE LOAD TO INCREASE P_VOLTAGE
ENDIF

GOSUB SET_PWM

WP_OLD=WP
VP_OLD=VP

RETURN

Thanks for the interest,
Regards,
Ambrogio
IW2FVO
North Italy

[theFob]

Thanks iw,
I suppose the first chart explains your code. I found it in the net. But i'm still out of focus.
Why is the measurement of dV? As is shown on the simulation, the voltage of the panel
is almost constant up to 90% of available current. So there is "constant voltage" and "constant current" methods. The voltage remain the same even at night, of course with almost no load. The photons flow become electrons flow, i.e. available current. So the panel is kind of current generator.
And of course, measured dV will be small and inaccurate, may be the noise is from there.
I used hardware circuit similar as algorithm in second chart. So i'm interested of your work. I will try to find my circuits, but unfortunately they are on almost dead HDD ...

Regards

[theFob]

ok, i answered myself. I did simulation again.
The voltage have little change indeed and can be used to determine the direction of duty. But this can be done without this too.
Iw, if i find some time i will make Proteus simulation with your code.
Will be nice to check some other algorithms before the fire of soldering iron



PS I don't know how to resize the images, sorry

[iw2fvo]

Thanks theFob,
I am interested in seeing the simulation of my program when you have time .
The mpp_tracking code I am using works well . My program allow to find out the MPP manually using the potentiometer. It also allows to scan the duty cycle to search the MPP : this is automatic.
The normal mode of the program continuously tracks the MPP. There is a very minor difference among them.
Thanks again,
regards,
Ambrogio
IW2FVO
North Italy

[theFob]

Hi iw,
I'll be glad to do this Had no such fun from long time.
I need the nominal voltage and maximum current values of the panel or just the model.
I noticed the lack of general decoupling diode in your circuit. Is needed for reverse current protection. May be the panel have such internally.
Also a small input filter will be good idea, because the input current of the buck is trapezoidal and is very stressful for the panel.
So the preferred converters are boost (when is possible) or sepic. The input currents there are triangle in continuous mode.
I allowed myself to check your CQ on the net. How far you've come with SG3525 inverter ? I guess it will work with this converter .... ?

Regards

[iw2fvo]

TheFob:
panel data are:
max power 50 w
open circuit volt: 21.6 Vdc
short circuit current : 3.04 A
Max pp volt: 17.6 Vdc
Max pp current : 2.84 A
There is a reverse protection diode ( see D1 on buck converter schematics ).
There is 220 uF at the buck input . Could it be too small ?
The panel keeps the SLA battery charged and so I do not have any inverter to produce line alternate power. I just need 12 VDC nominal.
CQ: >> I have a fully working homemade SDR rx tx for the HF. I have to re_install the antenna that went out completely during the last windy thunderstorm.
I plan to do that at the beginning of the next month.
Thanks.
73, IW2FVO.
Ambrogio

[theFob]

No, you have no panel protection diode. there must be a diode right after the panel, before the point C. It is essential.
If the sun goes down and you have no load at the moment, the input capacitor will return current which can destroy the junctions in the panel.
Diode D1 protect only the buck from the battery. Also the input capacitor is good to be bigger. At least 2200uF in my opinion.
Plus ~1uH inductance after the diode - will help a lot against noise. Few turns on the fly.
SDR sounds interesting, it is something new for me.

[iw2fvo]

Thanks.
I have learn now the need for a panel diode. I will move D1 to the panel input. I will also install a 2200 uF low ESR at panel input and after the diode.
The RF choke is a good idea too.
Ambrogio

[EarlyBird2]

Thanks Ambrogio for this very interesting and informative thread.

When I first read this thread I wondered why make your own Solar Regulator when I buy them for 20€ but obviously now I realise the models I use are not MPPT. The MPPT version costs 250€ so making your own is not only an interesting project but saves lots of cash.

I also realise that your design allows for Solar or Mains power which makes it even more interesting but I could not work out how the program/micro does the switching between the two power sources.

[iw2fvo]

Steve,
the program controls both the load disconnection and the external power .
You can set the voltages at your choice.
Here is the piece of code. The whole code is not optimized : I am not a software expert as I told at the beginning. There are a lot of comments on the program.
I think that a software guy could improve the code in a significant way. BTW it is still working at my home.
regards,
Ambrogio
iw2fvo

' LOW VOLTAGE PROTECTION ...and load control..................................

IF ( VB<11800 ) THEN ' BATT LOWER THAN 11.8 Vdc
U_VOLT=1 ' SET UNDER_VOLTAGE FLAG
LOW LD_EN ' DISABLE LOAD
LOAD=0
U_FLG=1
ELSE
IF ( VB>12600 ) THEN ' IF VBATT MORE THAN 12.6 Vdc
U_VOLT=0 ' RESET UNDER_VOLTAGE FLAG
IF U_FLG=1 THEN
HIGH LD_EN ' ENABLE LOAD
LOAD=1
U_FLG=0
ENDIF
ENDIF
ENDIF

IF (U_VOLT=1) AND (EXTPWR_OK=1) THEN
LOW EXT_PWR_RLY ' ACTIVATE EXT POWER RELAY
EX=1
ELSE
IF U_VOLT=0 THEN
HIGH EXT_PWR_RLY ' RELASE EXT POWER RELAY
EX=0
ENDIF
ENDIF

IF U_VOLT=0 THEN
IF PB_3=0 THEN
if LOAD=0 THEN
LOAD=1
ELSE
LOAD=0
ENDIF
IF LOAD=0 THEN
LOW LD_EN
ENDIF
IF LOAD=1 THEN
HIGH LD_EN
ENDIF
WHILE PB_3=0 : WEND
ENDIF
ENDIF

[EarlyBird2]

My rules for programming in order of importance.

1 It works
2 It is easy to follow
3 It is easy to modify

My definition of a good program is

It works

Which yours does except for one minor detail. So the question is whether to rewrite the program or not to solve this cloud issue? Personally I would rewrite the program because I would learn by doing so. I only wish I had the hardware.

[iw2fvo]

As far as the cloud issue I wikll modify the program to improve stability.
I will do the modification to the code tomorrow and I will then test it for some days expecially when cloudy.
If ok I will post the mod later on. The mod will involve the ON STATE code only.
Thanks
Ambrogio
IW2FVO
North Italy

[iw2fvo]

I am posting the picbasic program B_4523 mainly modified to avoid oscillation among the ON state in particulary cloud sky condition.
I did minor changes as reported in the .txt program headlines.
Thanks to all of you for the suggestions provided.
Regards,
Ambrogio
IW2FVO
North Italy

[iw2fvo]

Good day,

I connected my charger to the PV and I left it for about 10 days having good results in all the sun conditions available in that period of time.
The final code together with the photo of the prototype implementation is attached to the present mail .
Regards,
Ambrogio
IW2FVO
North Italy

[Demon]

Congratulations!

Robert

[iw2fvo]

Hi to all of you in the forum.
My MPPT charger is working since last June with no problem.
In this period I have no possibility to go on with my electronics hobbies.
I would like to know if someone has reproduced this project and has introduce some improvements.
Just to know !
Thanks
Regards,
Ambrogio

[harryweb]

HI IW2FVO,
I plan to use your schematics and parts of your code for my project (just ordered 2x100w PV, now need battery)
I would like to know what do you mean (schematics) replace D1 with ideal diode ?
It could take some times to make my project because of the season.
Congratulation for your work
Regards - 73'
F1SLU
Herve

[iw2fvo]

Hi Herve,
D1 introduce a voltage drop that impact on the whole system efficiency. This loss can be calculated an depends on current and diode performance in terms od forward voltage drop.
I used an ideal diode circuit to replace D1 during my last test at the bench only. I have not introduce it on my final circuit because of the space on the board.
You could find out a lot of these circuits here :
https://www.google.it/search?q=ideal...L&ved=0CCAQsAQ
You can also google " ideal diode solar " for additional info.
I do suggest to calculate the power saved and evaluate if really worth. It is up to you !
I will really start without it.
Best regards,
iw2fvo, Ambrogio

[harryweb]

Hi Ambrogio,

I just finished to install my solar panels 2 weeks ago. (Yes... I'm a turtle...)
As I doesn't make any charger I bought a cheap MPPT solar charger from Ebay (china) and it's a fake MPPT (just PWM).
So I've made the PCB for my real MPPT solar charger. Using most same parts as your schematics (just using other equivalent parts and not ACS712 but a maxim current sensor)
I have to modify your source program. :-)
I just would like to ask you, if you're happy with your charger/program, or did you make any changes ?
I have 220W solar panels.
I will give some news about the MPPT charger asap, I just received 5 PCB from... china today.... Have to solder now.

regards and 73
F1SLU Hervé
Best regards

[iw2fvo]

HI Herve,
It is still working well and no additional modification was introduced to the program.
I introduced an IDEAL DIODE found on the Ebay in place of the diode: it is OK.
I think the state machine portion of the program could be improved sometime...
My system has no PCB: do you have a spare one for me ?
All the best,
73, Ambrogio

[harryweb]

Hi Ambrogio,

My schematic is a bit different, so my PCB too ! As I have 220W panels some choice have to be done, and I also want to use serial to log on my computer. I think to make a python script to do that. I will upload picture of the schematics and PCB here but next week because I have it at work.
OK for the ideal diode. May be I will buy some of them ...
73, Hervé

[iw2fvo]

Thanks,
I will like to see it .
73, Ambrogio