Oscillator question and measurement

I only had a PIC in a QFN package but wanted to make a protoype with it. What I did was strip a 24 gauge wire and solder each individual strand to each pin. I'm never doing that again but.....

It seems slower than what it should be. I want to determine if it's the hardware or software so I used this:

Code:

---

Start:
TOGGLE portd.1
PAUSEUS 1
GOTO start

---

I have a 20 MHz crystal and DEFINE OSC 48. When I measure the frequency, am I supposed to get 1MHz? I get about 374 KHz. One thing that may be causing this is, to stabilize the soldering, I put hot glue on all of the pins not realizing that it's probably putting a little resistance between the oscillator pins.

Any ideas?

Hi,

Get rid of the line "PAUSEUS 1" and tell us what happens.

some JELT pcb " Tropicoat " protective varnish makes real good glue ... did you also try nail lacquer ??? ( NO metallic colour !!! - of course -)

a simulator run would show your program NEVER will run @ 1 Mhz ... as TOGGLE and GOTO commands need some time to execute

we also must suppose your Pic is a 18F series for the Pauseus1 to be equal to 1µs ...

Now, Hot glue doesn't disturb the oscillator in any way ...

Alain

Quote:

---

Get rid of the line "PAUSEUS 1" and tell us what happens.

---

I didn't have a scope handy and the meter I had isn't able to calculate high frequencies. I'll be able to try that tonight. What should I expect to see with just a toggle command?

Quote:

---

we also must suppose your Pic is a 18F series for the Pauseus1 to be equal to 1µs ...

---

Yes, it's an 18F4550. The program works but the CAN data isn't flowing smoothly compared to the previous devices I made. All of the pinouts are different, and wanted to rule that out, but I can't even see how that would cause slow CAN data in the first place.

a little MPSIM ( MPLAB ) simulation - supposed to be relatively true - gives à 3.2µs half -cycle to your test ... around 160 khz output frequency !
BTW ... I'm not sure the DEFINE OSC 48 and a 20 Mhz Xtal to be a good combination ...

best result @ 1.8µs half cycle given by

Code:

---

Portb.1 = ! Portb.1

---

instead of TOGGLE + PAUSEUS1

Here are the results. Are these typical?

Code:

---

start:
pauseus 1
toggle portd.1
goto start

---

375 KHz

Code:

---

start:
toggle portd.1
goto start

---

1 MHz

Code:

---

start:
Portb.1 = ! Portb.1
goto start

---

This does nothing

Code:

---

start:
portd.1=1
portd.1=0
goto start

---

3 MHz

Quote:

---

Originally Posted by Christopher4187

Code:

---

start:
Portb.1 = ! Portb.1
goto start

---

This does nothing

---

Sure it does not show anything on portD1 ... :rolleyes:

Alain

I know. I was too lazy to edit it yesterday. My original comment still stands.

ohhh ... may be the RMW issue ...

due to a 18F chip ... I should have written:

Code:

---

start:
latb.1 = ! latb.1
goto start

---

on test with the EP5 tomorrow morning ...

Quote:

---

Originally Posted by Christopher4187

I didn't have a scope handy and the meter I had isn't able to calculate high frequencies. I'll be able to try that tonight. What should I expect to see with just a toggle command?

---

I think the problem is that you are not configuring the fuses of the PIC correctly. Are you also setting the OSCON register? I use a 4MHz crystal with my 18F4550 and to make it work the first time it was a pain.

Results of the test ...

output freq is exactly that expected ... and that MPSIM shows ....

111.1 Khz with a 8 Mhz Xtal
277.7 Khz with a 20 Mhz Xtal

sooo ... getting 1 Mhz output looks a bit too much !!! - gives 72 Mhz @ Clock - 18Mhz Xtal and PLL x 4 ; that looks real hard overclocking for such chips ...
MAY BE POSSIBLE, but with very light loads on the output ...

Alain

Results of the test ...

output freq is exactly that expected ... and that MPSIM shows ....

111.1 Khz with a 8 Mhz Xtal
277.7 Khz with a 20 Mhz Xtal
666.5 Khz with 48 Mz @ Clock
833.1 Khz with ... 60Mhz ( poor USB clock ! )
looks a bit over 1Mhz with ... 80Mhz ( poor ... etc )

Code:

---

#CONFIG

    __CONFIG    _CONFIG1L, _PLLDIV_3_1L & _CPUDIV_OSC1_PLL2_1L & _USBDIV_2_1L 
                            ;              ;                      ; USB clock source comes from the 96 MHz PLL divided by 2
                            ;              ; [OSC1/OSC2 Src: /1][96 MHz PLL Src: /2]
                            ; 1/2 prescale


    __CONFIG    _CONFIG1H, _FOSC_HSPLL_HS_1H & _FCMEN_OFF_1H & _IESO_OFF_1H
                            ;                  ;              ; Oscillator Switchover mode disabled
                            ;                  ; Fail-Safe Clock Monitor disabled
                            ; HS oscillator, PLL enabled, HS used by USB
    __CONFIG    _CONFIG2L, _PWRT_ON_2L & _BOR_ON_2L  & _BORV_2_2L  & _VREGEN_ON_2L 
    __CONFIG    _CONFIG2H, _WDT_OFF_2H
    __CONFIG    _CONFIG3H, _MCLRE_ON_3H & _LPT1OSC_OFF_3H & _PBADEN_OFF_3H & _CCP2MX_ON_3H
    __CONFIG    _CONFIG4L, _STVREN_ON_4L & _LVP_OFF_4L & _ICPRT_OFF_4L  & _XINST_OFF_4L & _DEBUG_OFF_4L
     
    __CONFIG    _CONFIG5L, _CP0_OFF_5L & _CP1_OFF_5L & _CP2_OFF_5L & _CP3_OFF_5L
    __CONFIG    _CONFIG5H, _CPB_OFF_5H & _CPD_OFF_5H
    __CONFIG    _CONFIG6L, _WRT0_OFF_6L & _WRT1_OFF_6L & _WRT2_OFF_6L & _WRT3_OFF_6L
    __CONFIG    _CONFIG6H, _WRTB_OFF_6H & _WRTC_OFF_6H & _WRTD_OFF_6H
    __CONFIG    _CONFIG7L, _EBTR0_OFF_7L & _EBTR1_OFF_7L &_EBTR2_OFF_7L &_EBTR3_OFF_7L
    __CONFIG    _CONFIG7H, _EBTRB_OFF_7H   

       
#ENDCONFIG

trisa = 0
trisb = 0
trisc = 0
trisd = 0
lata = 0
latb = 0
latc = 0
latd = 0

Start:
Latb.1 = ! Latb.1
GOTO start
END

---

some surely will throw me everything that can fly .... LOL !

sooo ... getting 1 Mhz output looks a bit too much though !!! - gives 72 Mhz @ Clock - 18Mhz Xtal and PLL x 4 ; that looks real hard overclocking for such chips ...
MAY BE POSSIBLE, but with very light loads on the output ...

I did it ... so any idiot can do it too ...

the 4550 did not die ... really nice chip, isn't it ???

Alain

PS: this "timeout" really is a shame, Lester ....