68HC1 | 812A4 | 9S12DP256 | Bootloader | CodeWarrior | D60A | Debugger | DP256 | ECT | EEPROM | EVB | Flash | HC1 | HCS12 | I2C | IAR | ICC1 | Interrupts | LCD | M68KIT912DP256 | MC9S12DP256 | MC9S12DP256B | Metrowerks | Motor | MSCAN | Multilink | PLL | Quadrature | SDI | SPI | Transceiver | XFC

Hi All

Just a quick note to inform that the uTasker project has just switched
its own forum live (9.7.2007). It contains a section especially for
the M9S12NE64.

You are all most welcome to check it out or become an active forum
member.

http://www.utasker.com/forum

Best regards

Mark Butcher
www.uTasker.com

______________________________
Stellaris® MCU Family: New Parts, New Package, New Price.


(You need to be a member of 68hc12 -- send a blank email to 68hc12-subscribe@yahoogroups.com )

I've used the 912D60A for years. It's time to move up to something
better - and that appears to be something like the S12D64 or the
S12E64. [All I really need is reasonable RAM/Flash and SPI
interface. I like the S12UF with USB, but I lose the SPI.]

Aside from being cheaper than the D60A, these let me use both
Colpitts and Pierce forms of oscillator. So I looked at the
suggested PCB layouts (Freescale Oscillator Block Guide) and frankly,
I'm not sure it makes sense (please ignore the fact that I *should*
know these things).

Q1: Are they indeed suggesting a grounded layer immediately under
the uP IC? [I have a ground plane, but it's a layer or two down.]

Q2: It appears they are creating a star by leaving gaps in that
ground plane, correct?

Q3: *If* they indeed want a top surface layer to look like that,
just how the heck do they expect to get to about half the OTHER pins
on the IC. There isn't even any way to slip a bunch of VIAs in there.

Okay, these should be questions I should know the answer to, but just
once I would like one of the crystal to work the first time.

jmk

-----------------------------------------------
James M. Knox
TriSoft ph 512-385-0316
1300 Koenig Lane West fax 512-371-5716
Suite 200
Austin, Tx 78756 j...@trisoft.com
-----------------------------------------------


(You need to be a member of 68hc12 -- send a blank email to 68hc12-subscribe@yahoogroups.com )

Hi,
To share my experience:
Q1: Yes, they do; I followed their suggestion and never had problems with
the oscillator; I have used both configurations on two layers PCB.
Q2: Yes.
Q3: Check with your PCB manufacturer and ask for the smallest vias
they can do; of coarse, you must have enough free space around the micro,
place the vias far enough from micro, and then place tracks up to the vias.
I have managed to place a 48-pin micro (GC16) on a board as small as a
DIP40 socket.
Hope this will help.
Regards,
Yoan

James M. Knox wrote:
> I've used the 912D60A for years. It's time to move up to something
> better - and that appears to be something like the S12D64 or the
> S12E64. [All I really need is reasonable RAM/Flash and SPI
> interface. I like the S12UF with USB, but I lose the SPI.]
>
> Aside from being cheaper than the D60A, these let me use both
> Colpitts and Pierce forms of oscillator. So I looked at the
> suggested PCB layouts (Freescale Oscillator Block Guide) and frankly,
> I'm not sure it makes sense (please ignore the fact that I *should*
> know these things).
>
> Q1: Are they indeed suggesting a grounded layer immediately under
> the uP IC? [I have a ground plane, but it's a layer or two down.]
>
> Q2: It appears they are creating a star by leaving gaps in that
> ground plane, correct?
>
> Q3: *If* they indeed want a top surface layer to look like that,
> just how the heck do they expect to get to about half the OTHER pins
> on the IC. There isn't even any way to slip a bunch of VIAs in there.
>
> Okay, these should be questions I should know the answer to, but just
> once I would like one of the crystal to work the first time. jmk
>
> -----------------------------------------------
> James M. Knox
> TriSoft ph 512-385-0316
> 1300 Koenig Lane West fax 512-371-5716
> Suite 200
> Austin, Tx 78756 j...@trisoft.com
> -----------------------------------------------
>
>

______________________________
Stellaris® MCU Family: New Parts, New Package, New Price.


(You need to be a member of 68hc12 -- send a blank email to 68hc12-subscribe@yahoogroups.com )

At 07:48 7/19/2007 +0300, you wrote:
>Hi,
>To share my experience:
>Q1: Yes, they do; I followed their suggestion and never had problems with
>the oscillator; I have used both configurations on two layers PCB.

We'll give it the best try we can. [Can't do any worse than my
Pierce Osc. experiences with the D60A. ]

>Q2: Yes.

Interesting... I've always just used a solid ground plane
layer. Again, I'll give it a try... can't hurt.

>Q3: Check with your PCB manufacturer and ask for the smallest vias
>they can do; of coarse, you must have enough free space around the micro,
>place the vias far enough from micro,

This one is going to require a small bit of compromise. I've been
using PCBExpress, and their "smallest" vias aren't all that small.

Thanks a lot...
jmk

-----------------------------------------------
James M. Knox
TriSoft ph 512-385-0316
1300 Koenig Lane West fax 512-371-5716
Suite 200
Austin, Tx 78756 j...@trisoft.com
-----------------------------------------------


(You need to be a member of 68hc12 -- send a blank email to 68hc12-subscribe@yahoogroups.com )

Hi James,
That layout in the datasheet is a "best practice" layout. Trying
to make the best of all worlds (ATD, Crystal, PLL, etc.)and fit it
all onto a single layer. I agree the Colpitts is not the easiest
design to get working expecially that amplitude controlled version on
that old D60A.
There are several things working to your advantage with the newer
S12s.
First is that the Colpitts was "tuned" so it has more margin causing
it to be not as picky as the old D60A design.
Second thing is you can use the Pierce design if you like. The Pierce
tends to have less stray capacitance across the crystal (which is a
very bad thing in any crystal design).
Going away from amplitude controlled (to the Pierce design) hurts you
in a few ways, first is the part will tend to radiate more expecially
if the Pierce is not optimized (for example if you use a low
frequence crystal with low ESR and do not have any Rshunt). Second is
the added external components needed like Rshunt (resistor between
XTAL and the oscillator circuit).
On a related note, the full swing Pierce is a lot like the old B32
except the voltage it operates under.
Again a bunch of the layout is also trying to deal with other things
like ATD Vdd as clean as reasonable, etc. so depending on your system
requirements you may be able to life with a little more noise on the
ATD, or have more layers in your board design.

To sum all this up you should have much better luck using one of the
newer S12s even with a slighly different layout versus the datasheet.
I would try to keep the layout for the crystal and XFC very simular
to the datasheet but that should not affect you to bad and you could
pull it slightly further away from the package if you need a little
more room expecially if you use a low ESR crystal in the 4 to 8 MHz
range.

Regards,
Darci
--- In 6...@yahoogroups.com, "James M. Knox" wrote:
>
> At 07:48 7/19/2007 +0300, you wrote:
> >Hi,
> >To share my experience:
> >Q1: Yes, they do; I followed their suggestion and never had
problems with
> >the oscillator; I have used both configurations on two layers PCB.
>
> We'll give it the best try we can. [Can't do any worse than my
> Pierce Osc. experiences with the D60A. ]
>
> >Q2: Yes.
>
> Interesting... I've always just used a solid ground plane
> layer. Again, I'll give it a try... can't hurt.
>
> >Q3: Check with your PCB manufacturer and ask for the smallest vias
> >they can do; of coarse, you must have enough free space around the
micro,
> >place the vias far enough from micro,
>
> This one is going to require a small bit of compromise. I've been
> using PCBExpress, and their "smallest" vias aren't all that small.
>
> Thanks a lot...
> jmk
>
> -----------------------------------------------
> James M. Knox
> TriSoft ph 512-385-0316
> 1300 Koenig Lane West fax 512-371-5716
> Suite 200
> Austin, Tx 78756 jknox@...
> -----------------------------------------------
>



(You need to be a member of 68hc12 -- send a blank email to 68hc12-subscribe@yahoogroups.com )

At 14:46 7/19/2007 +0000, you wrote:
>Hi James,
> That layout in the datasheet is a "best practice" layout. Trying
>to make the best of all worlds (ATD, Crystal, PLL, etc.)and fit it
>all onto a single layer. I agree the Colpitts is not the easiest
>design to get working expecially that amplitude controlled version on
>that old D60A.

Thanks for the advice. Looks like the S12A will work. I had
originally rejected it because the usual suppliers didn't have any.
[Digikey only had the D and E that looked like they might fill the
bill. Newark had the S12A, but only three. So I broke down and
ordered directly from Freescale. As I said, these are small qty
prototypes for now.]

Again, thanks...
jmk

>There are several things working to your advantage with the newer
>S12s.
>First is that the Colpitts was "tuned" so it has more margin causing
>it to be not as picky as the old D60A design.
>Second thing is you can use the Pierce design if you like. The Pierce
>tends to have less stray capacitance across the crystal (which is a
>very bad thing in any crystal design).
>Going away from amplitude controlled (to the Pierce design) hurts you
>in a few ways, first is the part will tend to radiate more expecially
>if the Pierce is not optimized (for example if you use a low
>frequence crystal with low ESR and do not have any Rshunt). Second is
>the added external components needed like Rshunt (resistor between
>XTAL and the oscillator circuit).
>On a related note, the full swing Pierce is a lot like the old B32
>except the voltage it operates under.
>Again a bunch of the layout is also trying to deal with other things
>like ATD Vdd as clean as reasonable, etc. so depending on your system
>requirements you may be able to life with a little more noise on the
>ATD, or have more layers in your board design.
>
>To sum all this up you should have much better luck using one of the
>newer S12s even with a slighly different layout versus the datasheet.
>I would try to keep the layout for the crystal and XFC very simular
>to the datasheet but that should not affect you to bad and you could
>pull it slightly further away from the package if you need a little
>more room expecially if you use a low ESR crystal in the 4 to 8 MHz
>range.
>
>Regards,
>Darci
>--- In 6...@yahoogroups.com, "James M. Knox" wrote:
> >
> > At 07:48 7/19/2007 +0300, you wrote:
> > >Hi,
> > >To share my experience:
> > >Q1: Yes, they do; I followed their suggestion and never had
>problems with
> > >the oscillator; I have used both configurations on two layers PCB.
> >
> > We'll give it the best try we can. [Can't do any worse than my
> > Pierce Osc. experiences with the D60A. ]
> >
> > >Q2: Yes.
> >
> > Interesting... I've always just used a solid ground plane
> > layer. Again, I'll give it a try... can't hurt.
> >
> > >Q3: Check with your PCB manufacturer and ask for the smallest vias
> > >they can do; of coarse, you must have enough free space around the
>micro,
> > >place the vias far enough from micro,
> >
> > This one is going to require a small bit of compromise. I've been
> > using PCBExpress, and their "smallest" vias aren't all that small.
> >
> > Thanks a lot...
> > jmk
> >
> >
> >
> > -----------------------------------------------
> > James M. Knox
> > TriSoft ph 512-385-0316
> > 1300 Koenig Lane West fax 512-371-5716
> > Suite 200
> > Austin, Tx 78756 jknox@...
> > -----------------------------------------------
> >Yahoo! Groups Links
>
-----------------------------------------------
James M. Knox
TriSoft ph 512-385-0316
1300 Koenig Lane West fax 512-371-5716
Suite 200
Austin, Tx 78756 j...@trisoft.com
-----------------------------------------------
______________________________
Stellaris® MCU Family: New Parts, New Package, New Price.


(You need to be a member of 68hc12 -- send a blank email to 68hc12-subscribe@yahoogroups.com )