frequency meter in PBP

[mister_e]

Yes, but why?

[savnik]

Yes, but why?

Because i want to add to code(in pbp) some others things , like to measure ADC.
I try two days now without any result.

[Ioannis]

Hi Niko!

Assembly is much more faster than any other lanquage. The code you posted I think is from an old AN from Microchip, right? I studied few years ago. My opinion is that you will need a top class PIC with the highest XTAL to reach the needed speed. Is it worth really?

Ioannis

[savnik]

Hi Niko!

Assembly is much more faster than any other lanquage. The code you posted I think is from an old AN from Microchip, right? I studied few years ago. My opinion is that you will need a top class PIC with the highest XTAL to reach the needed speed. Is it worth really?

Ioannis

This is not from Microchip , but i have make the an592 from Microchip with a
pic16f84a and prescaler and count until 1500 MHZ.

[skimask]

Because i want to add to code(in pbp) some others things , like to measure ADC.
I try two days now without any result.

Gave up on trying to write one yourself?
You'll never get any fast counting done with a PIC and straight PBP (maybe 250khz), assembly might get you 10Mhz, if you use a 40Mhz PIC and you use one of the TMR inputs.
However, if you take my idea (from the other thread), and use a pre-divider, you'll count much higher.
And check this out....
http://members.cox.net/berniekm/super.html

[Ioannis]

Hi Skimask. If you study the Microchip AN592, you 'll see that a humble PIC 16C55 is counting up to 50 MHz!

It use a clever technique to do that. But in PBP, I think it is quite difficult to obtain.

Ioannis

[skimask]

Hi Skimask. If you study the Microchip AN592, you 'll see that a humble PIC 16C55 is counting up to 50 MHz!

It use a clever technique to do that. But in PBP, I think it is quite difficult to obtain.

Ioannis

Right, using the prescaler, I understand that, 10ns rise/fall time on the RTCC pin...works fine I'm sure (although I've never tried that particular project on that particular PIC).
And that's what I was saying in the other thread about PBP. Use an external 74193 or something similar to pre-scale the count, and PBP should be able to count a fairly high number.

[mister_e]

Gave up on trying to write one yourself?
You'll never get any fast counting done with a PIC and straight PBP (maybe 250khz), assembly might get you 10Mhz, if you use a 40Mhz PIC and you use one of the TMR inputs.
However, if you take my idea (from the other thread), and use a pre-divider, you'll count much higher.
And check this out....
http://members.cox.net/berniekm/super.html

Everything is around the sampling time and the accuracy you want not much. Depending of your PIC characteristic, 50MHZ is easy to do, 1500 MHZ??? i have some doubt, but with external hardware divider, there's no real problem.
For high frequency... who want to use 1Sec time base? Nobody... 100mSec, 10mSec, 1mSec is more than possible.

Nothing hard, nothing clever at all. Set a Timer as counter, Clear it, enable it, do you samppling time,m stop it, read it display it. 4 code line or so.

@Savnik, Which PIC you want to use?

Have a look to T0CKI, T1CKI. This will reveal tons of things.

[savnik]

Depending of your PIC characteristic, 50MHZ is easy to do, 1500 MHZ???

I have use external divider(:256)

Nothing hard, nothing clever at all. Set a Timer as counter, Clear it, enable it, do you samppling time,m stop it, read it display it. 4 code line or so.
.

Yes , i use the below code(from AN592)

asm
enabletmr0      macro
        clrf    TMR0
        bsf     _trisabuf,3
        movf    _trisabuf,W
        tris    PORTA
        endm

disabletmr0     macro
        bcf     _trisabuf,3
        bcf     PORTA,3
        movf    _trisabuf,W
        tris    PORTA
        endm        

        movlw   0x10                ;initialize ddra
        movwf   _trisabuf           ;       /
        disabletmr0
        movlw   0x37                ;set the option register  - B'00110111'
        option                      ;to measure high freq.
        clrw

        enabletmr0                  ;enable tmr0
        call    delay5mS            ;wait for 5mS
        disabletmr0                 ;disable tmr0
        call    getfreq             ;get freq in fhi and flo  
                
;delay1ms, is a very accurate 1mS delay for a 4Mhz clock.
delay1mS
        movlw   197
        movwf   _count1
        nop
        goto    $+1
        goto    $+1
dly1mS        
        goto    $+1
        decfsz  _count1, F
        goto    dly1mS
        retlw   0

;delay5mS uses delay1mS to get a very accurate 5 mS delay
delay5mS
        call    delay1mS
        call    delay1mS
        call    delay1mS
        call    delay1mS
        call    delay1mS
        movlw   4
        movwf   _count1
tweek5mS
        decfsz  _count1, F
        goto    tweek5mS
        return

getfreq
        movf    TMR0,W              ;get the tmr0 value
        movwf   _fhi                ;save in fhi
        clrf    _count1             ;keep track of the toggles
toggle1
        incf    _count1, F          ;inc for first
        bcf     PORTA,3             ;toggle the input
        bsf     PORTA,3             ;       /
        movf    TMR0,W              ;see if tmr0 incremented
        subwf   _fhi,W              ;       /
        btfsc   STATUS,Z            ;yes then skip
        goto    toggle1             ;no then toggle again
        comf    _count1, F          ;else complement count
        incf    _count1,W           ;and increment
        movwf   _flo                ;save in flo
        retlw   0                   ;return
endasm

@Savnik, Which PIC you want to use?

i use 16f88.