> Hi all,
>
> When it says on the chip 04 or 08 for the speed of the chip, is it the
> crystal value or 1/4 of the crystal value.
>
> Thanks in advance.
>
> --
> http://www.piclist.com hint: The PICList is archived three different
> ways. See http://www.piclist.com/#archives for details.
...and what can cause my PIC16F84A-20P to work with a 4MHz resonator but not
a 20MHz resonator?
I have tried two different 20 MHz resonators and the chip doesn't work with
either... 4MHz is just fine. The application is a simple LED turn on and
keep on, so speed should make no difference.
Any ideas? - Programmed with MPLab and the properties for the project
specify 20MHz. XT is specified as the type...
I don't need the extra speed at the moment, but I am curious what I might be
missing. My tools to track this problem are not advanced - no oscillator,
etc... A logic probe is about all I have.
At 04:25 PM 11/03/2002 +1300, you wrote:
>Hi all,
>
>When it says on the chip 04 or 08 for the speed of the chip, is it the
>crystal value or 1/4 of the crystal value.
It's the max crystal value not the MIPS.
Regards...
> ...and what can cause my PIC16F84A-20P to work with a 4MHz resonator but not
> a 20MHz resonator?
>
> I have tried two different 20 MHz resonators and the chip doesn't work with
> either... 4MHz is just fine. The application is a simple LED turn on and
> keep on, so speed should make no difference.
>
> Any ideas? - Programmed with MPLab and the properties for the project
> specify 20MHz. XT is specified as the type...
>
> I don't need the extra speed at the moment, but I am curious what I might be
> missing. My tools to track this problem are not advanced - no oscillator,
> etc... A logic probe is about all I have.
>
> Thank you,
> Fowler
>
At 11:05 PM 10/03/2002 -0500, you wrote:
>...and what can cause my PIC16F84A-20P to work with a 4MHz resonator but not
>a 20MHz resonator?
>
>I have tried two different 20 MHz resonators and the chip doesn't work with
>either... 4MHz is just fine. The application is a simple LED turn on and
>keep on, so speed should make no difference.
>
>Any ideas? - Programmed with MPLab and the properties for the project
>specify 20MHz. XT is specified as the type...
Change the oscillator type to HS for above 4MHz.
Regards...
> ...and what can cause my PIC16F84A-20P to work with a 4MHz resonator but
not
> a 20MHz resonator?
>
> I have tried two different 20 MHz resonators and the chip doesn't work
with
> either... 4MHz is just fine. The application is a simple LED turn on and
> keep on, so speed should make no difference.
>
> Any ideas? - Programmed with MPLab and the properties for the project
> specify 20MHz. XT is specified as the type...
>
> I don't need the extra speed at the moment, but I am curious what I might
be
> missing. My tools to track this problem are not advanced - no oscillator,
> etc... A logic probe is about all I have.
>
> Thank you,
> Fowler
>
> {Original Message removed}
At 11:05 PM 10/03/2002 -0500, you wrote:
>...and what can cause my PIC16F84A-20P to work with a 4MHz resonator but
not
>a 20MHz resonator?
>
>I have tried two different 20 MHz resonators and the chip doesn't work with
>either... 4MHz is just fine. The application is a simple LED turn on and
>keep on, so speed should make no difference.
>
>Any ideas? - Programmed with MPLab and the properties for the project
>specify 20MHz. XT is specified as the type...
Change the oscillator type to HS for above 4MHz.
Regards...
I've run into this problem with a few of my chips. I suspect the major
reason is the extra capacitance from the breadboards I use. One person
I talked to awhile ago measured 20pF extra capacitance from wire to wire
on their breadboard... (I doubt you'll ever find your breadboard that
high, but they still have enough to mess with sensitive circuits)
But I've never needed the extra speed badly enough to really look into
it hard. Use a wire wrap socket for the pic, bend the osc pins out and
up so they don't go into the breadboard, and solder the osc (or another
socket) directly to those pins with a flying lead to ground (or to the
capacitors).
>...and what can cause my PIC16F84A-20P to work with a 4MHz resonator but not
>a 20MHz resonator?
>
>I have tried two different 20 MHz resonators and the chip doesn't work with
>either... 4MHz is just fine. The application is a simple LED turn on and
>keep on, so speed should make no difference.
>
>Any ideas? - Programmed with MPLab and the properties for the project
>specify 20MHz. XT is specified as the type...
>
>I don't need the extra speed at the moment, but I am curious what I might be
>missing. My tools to track this problem are not advanced - no oscillator,
>etc... A logic probe is about all I have.
>
>Thank you,
>Fowler
>
> ...and what can cause my PIC16F84A-20P to work with a 4MHz
> resonator but not
> a 20MHz resonator?
>
> I have tried two different 20 MHz resonators and the chip doesn't
> work with
> either... 4MHz is just fine. The application is a simple LED turn on and
> keep on, so speed should make no difference.
>
> Any ideas? - Programmed with MPLab and the properties for the project
> specify 20MHz. XT is specified as the type...
>
> I don't need the extra speed at the moment, but I am curious what
> I might be
> missing. My tools to track this problem are not advanced - no oscillator,
> etc... A logic probe is about all I have.
A resonator (those 4 pin beauties) or a crystal? I only ask since it a
prior message you said crystal. For a crystal you must use HS mode to get it
to work since the crystal must be driven a little "harder". For a resonator
I'm not sure, I think XT should work? TTYL
Herbert wrote:
> > ...and what can cause my PIC16F84A-20P to work with a 4MHz
> > resonator but not
> > a 20MHz resonator?
<snip>
>A resonator (those 4 pin beauties) or a crystal? I only ask since it a
>prior message you said crystal. For a crystal you must use HS mode to get it
>to work since the crystal must be driven a little "harder". For a resonator
>I'm not sure, I think XT should work? TTYL
No, HS is required for a 20MHz resonator. Been there, forgot that !
Regards...
You haven't looked hard enough. Just about every CMOS clock circuit I've seen
has a feedback resistor across its input/output typically 1 meg to 10 meg. This
ensures a reliable startup. The PIC16f is a CMOS device with a high input
impedance pin and a low output impedance pin (typical of CMOS clock
generators). The data sheet for the 75T202 chip recommends a 10 meg resistor.
The Basic Micro http://www.basicmicro.com/ MBasic manual page 146 recommends a
10 meg resistor across the 16f628 using a resonator. The Motorola CMOS handbook
is infested with clock circuits with this feedback resistor. I have designed
equipment with CMOS logic in commercial/broadcast circuits that use this
resistor. VDO specifies this resistor in their automotive applications in their
instrument clusters. The Piccon at http://www.qsl.net/n6bg/piccon/ shows a 1
meg across the xtal. The Simon Bus radio alert receiver uses a PIC, 4mhz xtal
with a 10 meg across its clock. I could go on and on.....
Rick
> > Try a 1 to 10 meg resistor across the osc pins. I think I saw a datasheet
> > somewhere with this requirement.
>
> I haven't. Care to provide a reference?
>
> ********************************************************************
> Olin Lathrop, embedded systems consultant in Littleton Massachusetts
> (978) 742-9014, EraseMEolinembedinc.com,http://www.embedinc.com
>
> --
> http://www.piclist.com#nomail Going offline? Don't AutoReply us!
> email RemoveMElistservEraseMEEraseMEmitvma.mit.edu with SET PICList DIGEST in the body
Rick wrote:
>You haven't looked hard enough. Just about every CMOS clock circuit I've seen
>has a feedback resistor across its input/output typically 1 meg to 10 meg.
>This
>ensures a reliable startup. The PIC16f is a CMOS device with a high input
>impedance pin and a low output impedance pin (typical of CMOS clock
>generators). The data sheet for the 75T202 chip recommends a 10 meg resistor.
What Olin is saying is that Microchip don't mention it do they?
I would not say a 16F628 oscillator is the same as a standard CMOS clock
circuit.
>The Basic Micro http://www.basicmicro.com/ MBasic manual page 146
>recommends a
>10 meg resistor across the 16f628 using a resonator. The Motorola CMOS
>handbook
>is infested with clock circuits with this feedback resistor. I have designed
>equipment with CMOS logic in commercial/broadcast circuits that use this
>resistor. VDO specifies this resistor in their automotive applications in
>their
>instrument clusters. The Piccon at http://www.qsl.net/n6bg/piccon/ shows a 1
>meg across the xtal. The Simon Bus radio alert receiver uses a PIC, 4mhz xtal
>with a 10 meg across its clock. I could go on and on.....
>Rick
Are any of these examples what Microchip suggest?
I'd say some designers add the resistor out of habit with other micro's.
>Olin Lathrop wrote:
>
> > > Try a 1 to 10 meg resistor across the osc pins. I think I saw a datasheet
> > > somewhere with this requirement.
> >
> > I haven't. Care to provide a reference?
> What Olin is saying is that Microchip don't mention it do they?
> I would not say a 16F628 oscillator is the same as a standard
> CMOS clock circuit.
Figure 14.2 of the 16F628 manual shows a resistor "RF" inside
the chip, across Osc1 and Osc2. Unfortunately nothing else is
said about it
> You haven't looked hard enough. Just about every CMOS clock circuit I've
seen
> has a feedback resistor across its input/output typically 1 meg to 10 meg.
This
> ensures a reliable startup. The PIC16f is a CMOS device with a high input
> impedance pin and a low output impedance pin (typical of CMOS clock
> generators). The data sheet for the 75T202 chip recommends a 10 meg
resistor.
> The Basic Micro http://www.basicmicro.com/ MBasic manual page 146
recommends a
> 10 meg resistor across the 16f628 using a resonator. The Motorola CMOS
handbook
> is infested with clock circuits with this feedback resistor. I have
designed
> equipment with CMOS logic in commercial/broadcast circuits that use this
> resistor. VDO specifies this resistor in their automotive applications in
their
> instrument clusters. The Piccon at http://www.qsl.net/n6bg/piccon/ shows a
1
> meg across the xtal. The Simon Bus radio alert receiver uses a PIC, 4mhz
xtal
> with a 10 meg across its clock. I could go on and on.....
But not one of these references are from Microchip specifically pertaining
to the PIC crystal driver circuit, which was topic of discussion.
Such resistors are often used as negative feedback to keep CMOS logic gates
in the "linear" region. However, the PIC oscillator circuit was designed
specifically for that purpose and needs no such resistor. Adding such a
resistor to the PIC crystal oscillator is merely superstition. You might
just as well wave a dead fish over it during a full moon.
Ahh, the technique which gets used to make the ICD work properly....
> Such resistors are often used as negative feedback to keep
> CMOS logic gates in the "linear" region. However, the PIC
> oscillator circuit was designed specifically for that purpose
> and needs no such resistor. Adding such a resistor to the
> PIC crystal oscillator is merely superstition. You might
> just as well wave a dead fish over it during a full moon.
>
>
> ********************************************************************
> Olin Lathrop, embedded systems consultant in Littleton Massachusetts
> (978) 742-9014, spamBeGoneolinSTOPspamEraseMEembedinc.com,http://www.embedinc.com
>
> --
> http://www.piclist.com hint: To leave the PICList KILLspampiclist-unsubscribe-requestspamBeGonemitvma.mit.edu
*****************************************************************
This email has been checked by the altohiway e-Sweeper Service
*****************************************************************
Actually the salts in the fish flesh cause ionic reactions which in turn
generate electrostatic fields that, while the fish is moving, interact with
the earth's magnetic field which, of course, is affected by the full moon
thus causing ionic bombardment of the silicon substrate.
Thus waving a dead fish over your PIC during a full moon CAN have an
electrical effect on it.
If you've read thus far... only joking! ;-)
Ted Mawson
PS We engineering types do have a tendancy to reject anything that we don't
have a scientific explanation for, don't you think?
>Such resistors are often used as negative feedback to keep CMOS logic gates
>in the "linear" region. However, the PIC oscillator circuit was designed
>specifically for that purpose and needs no such resistor.
Actually all circuits of this type (Pierce oscillator) do require a
resistor, however the resistor is provided internally in PICs. The value is
selected from several possibilities when the oscillator type is selected.
Some micros (Motorola '05, for example) do not have the resistor internally
and it must be provided externally. This might have some slight advantage
in allowing better controlled value (5% vs. 30% tolerance) and higher
value (10M vs maybe 2M).
> Adding such a
>resistor to the PIC crystal oscillator is merely superstition. You might
>just as well wave a dead fish over it during a full moon.
Like waving a fish over it twice, when once has been show beyond any doubt
to be every bit as effective as twice.
UNISYS.COM>
