Thanks very much for your response.  Exactly the type of thing I'm 
looking for.  I have much to learn.  i'll take a look at some of those 
templates you mentioned.  Comments below.

Olin Lathrop wrote:

> email@micahcarrick.com wrote:
>
>> Crystal Oscillator = 8.000 MHz
>> Tosc = 125nS
>
>
> Yes, but the timers are incremented every Tcy (if so set up), not 
> Tosc.  Tcy
> is 4x Tosc = 500nS.
>
>> Prescaler = 1:16
>> PR2 = 199 (0xC7)
>
>
> 500nS x 16 x 200 = 1.6mS
>
>> Therefore, my period is 1.6mS (625 Hz).
>
>
> OK, I guess you figured out about the Tcy part but didn't show it.


Yeah, I used Tosc because the data sheet used it in it's formulas.  It 
used (4 x Tosc x TMR2_Prescaler) in the formula for determining the 
period and just (Tosc x TMR2_Prescaler) in determining the duty.

>
>>          __CONFIG _CP_OFF & _WDT_OFF & _BODEN_OFF & _PWRTE_OFF & 
>> _XT_OSC &
>> _WRT_OFF &
>> _LVP_OFF &       _CPD_OFF
>
>
> You might get away with XT oscillator mode at 8MHz, but you should 
> probably
> be using HS.  For the 16F877 XT was only recommended up to 4MHz.  
> Maybe that
> has changed for the 16F877A, but you might want to check it.

Good call and thanks for pointing that out!  I've only used f628, f627, 
and f84 in the past.  This is a first time using something higher than 
4MHz and I had no idea.  Thanks.

>
>> ;***** VARIABLE DEFINITIONS
>> w_temp   EQU     0x71        ;variable used for context saving
>
>
>
> Learn the right habits now.  Don't use absolute mode and don't define 
> hard
> variable addresses in the code.
>
>> status_temp EQU  0x72        ;variable used for context saving
>> pclath_temp EQU  0x73        ;variable used for context saving
>
>
> If you do things just the right way in the interrupt routine, you 
> don't need
> the STATUS and PCLATH save areas to be in the global RAM.  After a 
> while you
> will find that these 16 bytes are precious and anything that doesn't 
> really
> need to be global usually shouldn't be.  See my interrupt routine 
> template
> QQQ_INTR.ASPIC at http://www.embedinc.com/pic.

I'm just taking this from the template from microchip.  Although I'm not 
using any ISR yet, I will be once I get the PWM working properly.  So 
I'll take a look at that template of yours.  Again, this is where I'm 
really "wet behind the ears".  In the past I typically worked in C and 
am only recently trying to get more aquainted with the chips using asm.

>
>>          ORG     0x000       ;processor reset vector
>>          nop                 ;nop required for icd
>>          goto    main        ;go to beginning of program
>
>
> This will only work if MAIN is in the first page, and then only on reset
> where PCLATH can be counted on to contain 0.  You really should set 
> PCLATH
> to the high address byte of MAIN before the GOTO.

Again, from the template provided by microchip.  I'm not sure I 
understand what you're saying here.

>
>>          ORG     0x004       ;interrupt vector location
>>
>>          movwf   w_temp      ;save off current W register contents
>>          movf    STATUS, w   ;move status register into W register
>>          movwf   status_temp ;save off contents of STATUS register
>>          movf    PCLATH, w   ;move pclath register into w register
>>          movwf   pclath_temp ;save off contents of PCLATH register
>
>
> You probably want to add CLRF STATUS and CLRF PCLATH here to select known
> banks and page.  Again see my interrupt routine template.
>
>> ; isr code can go here or be located as a call subroutine elsewhere
>
>
> Don't forget to set PCLATH before any CALL.
>
>> main
>>          clrwdt              ;clear watch dog timer and prescaler
>
>
> Why?  WDT is disabled.
>
>>          movlw   0xC7        ;1. Set the PWM period
>>          movwf   PR2
>
>
> Huh?  And where did C7h come from?  First, if it is a magic number it 
> should
> be clearly documented.  Second, it shouldn't be burried down in the code.
> If it's a tweak constant it should have a symbol up top with clear
> documentation on what it does.  Third, if it's a derived constant as this
> is, derive it from the original factors.  In this case it should be 
> derived
> from the oscillator frequency and your desired PWM frequency.  Especially
> since this is dependent on oscillator frequency, you should let the
> assembler do the math.  Imagine what a pain it will be to update your 
> code
> to a different oscillator frequency as it is now.  When done right, 
> you only
> change one number and rebuild.
>
> For an example of this, see my TIMER2_USEC macro in STD.INS.ASPIC.
>
>>          bcf     STATUS, RP0 ;select bank 0
>
>
> No, this only guarantees banks 0 or 2.  Imagine inserting some 
> instructions
> above that leave RP1 set.  Now the code below it will break.  This is a
> dangerous way to set banks.
>
> There are better ways of manging banks, but that's for another 
> discussion.
> I'm going to quit here since I see a lot of the same things I mentioned
> above happening below.

Coming from an f628... so, would simply putting a ...

bcf STATUS, RP0
bcf STATUS, RP1

...be enough? 

>
>
> *****************************************************************
> Embed Inc, embedded system specialists in Littleton Massachusetts
> (978) 742-9014, http://www.embedinc.com


-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist