Searching \ for '[PIC] watch crystal and int osc?' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/microchip/devices.htm?key=pic
Search entire site for: 'watch crystal and int osc?'.

Exact match. Not showing close matches.
PICList Thread
'[PIC] watch crystal and int osc?'
2007\06\13@012841 by PicDude

flavicon
face
Perhaps due to my low blood-caffeine level, but I seem to be confused with the
datasheet's explanation of this... On a PIC16F913, can I setup Timer 1 as an
oscillator with a 32.768 watch crystal on the T1OSI and T1OSO pins while the
microcontroller is still clocked from the internal oscillator at 8Mhz?

Somehow in the datasheet, the third sentence in section 6.6 (1) seems to
contradict the previous statements.

Cheers,
-Neil.

2007\06\13@044841 by Jan-Erik Soderholm

face picon face
I can not see what section 6.6 ("Timer 1 Gate") has to do with that...
(DS41250F)

Jan-Erik.

PicDude wrote:
{Quote hidden}

2007\06\13@054255 by Jinx

face picon face
> Somehow in the datasheet, the third sentence in section
> 6.6 (1) seems to contradict the previous statements

Neil, DS41250F says, in 6.4, -

The Timer1 oscillator is shared with the system LP
oscillator. Thus, Timer1 can use this mode only when
the primary system clock is derived from the internal
oscillator or when in LP oscillator mode

So I infer from that you can have the whole chip running
at 32k (LP) or the core at IntRC and T1 at 32k

2007\06\13@061942 by Jan-Erik Soderholm

face picon face
What is not perfectly clear is if the "Timer1 oacillator"
actualy is the *same* osc as "LFINTOSC".

It's also not clear if they with "internal oscillator"
in section 6.4 means "HFINTOSC" (which *I* think) or
something else.

So *I* think that Jinx statement should read :

> So I infer from that you can have the whole chip running
> at 32k (LPINTOSC) or the core at HFINTOSC and T1 at 32k
> (using the Timer1-osc, same as LPINTOSC ??).

Jan-Erik.





Jinx wrote:
{Quote hidden}

2007\06\13@071051 by Jinx

face picon face
> What is not perfectly clear is if the "Timer1 oacillator"
> actualy is the *same* osc as "LFINTOSC".
>
> It's also not clear if they with "internal oscillator"
> in section 6.4 means "HFINTOSC" (which *I* think) or
> something else.

The internal oscillator appears to be selectable between
HF and LF. Both are distinct from the T1OSC pins and
oscillator (Fig 4.1), so I'd say you could use T1OSC with
either HF or LF

2007\06\13@085454 by PicDude

flavicon
face
Perhaps diff versions of the datasheet.  6.6-1 of mine says:
"Setting the T1OSCEN and TMR1CS bits to '1'
will enable the LP oscillator to clock TMR1 while
the microcontroller is clocked from either the
INTOSC or LP oscillator. Note that the T1OSC
and LP oscillators share the same circuitry.
Therefore, when LP oscillator is selected and
T1OSC is enabled, both the microcontroller and
the Timer1 module share the same clock
source."

The first statement there says that I can do this -- ie: separate INTOSC for
the microcontroller and the LP oscillator (with 32.768khz external crystal)
to clock TMR1.

But the next 2 statements say that they are the same oscillator, and hence
both the microcontroller and Timer1 module will have the same clock source.

Cheers,
-Neil.


On Wednesday 13 June 2007 03:48, Jan-Erik Soderholm wrote:
{Quote hidden}

2007\06\13@091618 by Jinx

face picon face
> "Setting the T1OSCEN and TMR1CS bits to '1' will enable
> the LP oscillator to clock TMR1 while the microcontroller is
> clocked from either the INTOSC or LP oscillator. Note that
> the T1OSC and LP oscillators share the same circuitry.
> Therefore, when LP oscillator is selected and T1OSC is
> enabled, both the microcontroller and the Timer1 module share
> the same clock source."
>
> The first statement there says that I can do this -- ie: separate
> INTOSC for the microcontroller and the LP oscillator (with
> 32.768khz external crystal) to clock TMR1.
>
> But the next 2 statements say that they are the same oscillator,
> and hence both the microcontroller and Timer1 module will have
> the same clock source.

To my eye it says Timer1 and LP share the same circuitry, and
INTOSC is entirely separate

> "Setting the T1OSCEN and TMR1CS bits to '1' will enable
> the LP oscillator to clock TMR1

Therefore if you have a 32k crystal on T1OSC pins (which are
also the primary OSC pins), TMR1 will be clocked by the 32k
crystal. You may also choose to run the core on that 32k (if LP
is selected as the core clock source) or use INTOSC for the core,
whilst TMR1 will use 32k

2007\06\13@091826 by PicDude

flavicon
face
On Wednesday 13 June 2007 06:10, Jinx wrote:
> > What is not perfectly clear is if the "Timer1 oacillator"
> > actualy is the *same* osc as "LFINTOSC".
> >
> > It's also not clear if they with "internal oscillator"
> > in section 6.4 means "HFINTOSC" (which *I* think) or
> > something else.
>
> The internal oscillator appears to be selectable between
> HF and LF. Both are distinct from the T1OSC pins and
> oscillator (Fig 4.1), so I'd say you could use T1OSC with
> either HF or LF


I'm looking at section 4.1 (Clock sources) and the associated diagram (Figure
4-1).  There does seem to be a clear distinction between HFINTOSC and
LFINTOSC.  The thing is that LFINTOSC has a separate output path to the LCD,
PWRT, WDT, FSCM, but not to Timer 1.  That's fine, since that's an INTERNAL
low-freq oscillator.  So the LP oscillator is clearly something else --
specifically a low-power mode of the external oscillator.  I'm just putting
this here to be clear/explicit.

At the top of Fig 4-1 is the external oscillator.  Here, it goes to the same
mux that the HFINTOSC goes to, and I don't see a separate path for it to feed
Timer 1.  Now Figure 6-1 shows that external oscillator going to the same
mux, and then going to the Timer1 module -- AFTER the mux.  This implies (or
states) to me that Timer1 would be clocked from either the internal or
external oscillator and the same mux selection would determine the
microcontroller system clock source.  So it seems like I cannot do this.

Sigh.  I think it's time I whip out a breadboard to find out for sure.

Cheers,
-Neil.

2007\06\13@092508 by Nicola Perotto

picon face


PicDude wrote:
> Perhaps diff versions of the datasheet.  6.6-1 of mine says:
>  
Probably Jan Erik want to remark that your datasheet is old and it's
better to update your copy!
Many times I think that they (uChip) for the first edition simply do a
cut & paste from old datasheet and after some years corrects the errors...
Cheers
         Nicola

{Quote hidden}

More... (looser matching)
- Last day of these posts
- In 2007 , 2008 only
- Today
- New search...