Sign in

username:

password:



Not a member?

Search 68hc12



Search tips

Subscribe to 68hc12



68hc12 by Keywords

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

Ads

Discussion Groups

Discussion Groups | 68HC12 | Re: Any issues with driving 9s12's with an oscillator?

Join our technical discussions about Freescale Microcontrollers: M68HC12. (Freescale Semiconductor is a Subsidiary of Motorola).

Any issues with driving 9s12's with an oscillator? - tonalbuilder2002 - May 9 18:06:00 2004

Problems here with my first 9S12 board, was wondering if this sounds
familiar to anybody...

I have a new MC9S12DJ64CPV design that I'm driving with an 8mHz, 5
volt oscillator. The oscillator output reaches the EXTAL pin after
being divided down to 2.5 volts through 2, 470 ohm resistors. There
is some rounding, but the clock looks pretty good. The circuit is
very basic and otherwise exactly like an Elektronikladen "CardS12"
with most of the lines not terminated, but connected to 1 or 2 inches
of traces on a not-yet-populated board.

MODA, MODB, are tied low through 10k's.

XTAL is open.

XIRQ tied hi through 100k.

XCLKS is now tied lo through a small resistor, I have also tried it
tied hi. The Motorola docs make contradictory recommendations for
XCLKS when using an external oscillator.

I have also tried tried shorting together VDDPLL and XFC, as
recommended by Motorola for non-pll operation, no difference.

Basically, *most of the time* I can't get a ComPod12-Pro to
recognize the chip. When I plug in the ComPod12-Pro, I notice that I
get a ~1.48mHz ECLK...is this normal? But a very small percentage of
the time, the BDM does work! Have tried all the varations of
plugging-in/powering up/restarting the program/ComPod target
settings and frequencies. The ComPod works fine with several older
HC912 boards. I've swapped out the CPU a few times (just now
discovered how to use the heat-gun for that, it so easy...). All the
power connections are correct.

Would appreciate any input! WAG's are always welcome here.

Bill T.





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


Re: Any issues with driving 9s12's with an oscillator? - Armand ten Doesschate - May 10 0:35:00 2004

On Sun, May 09, 2004 at 11:06:41PM -0000, tonalbuilder2002 wrote:
>
> Basically, *most of the time* I can't get a ComPod12-Pro to
> recognize the chip. When I plug in the ComPod12-Pro, I notice that I
> get a ~1.48mHz ECLK...is this normal? But a very small percentage of
> the time, the BDM does work! Have tried all the varations of

I have exactly the same configuration with my DP256 : only the clock
divider exists of 2 * 10 K resistor directly driven with a ACT74 @ 4 MHz.
I tied XCLKS high with a 10 K pullup. The 2.5 V clock to the DP256 looked
like a sawtooth but my DP256 works flawlessy.

I experienced the same problems with the bdm : I discovered that my
reset signal came up to slow (I used a TL7705 for this). My solution was
to delay my bdm signal to the BKGND pin for a (very) short time only during
reset, so that the bdm sigal was low when reset was going high and being
stable. My DP256 always entered into bdm afterwards. Armand *******************************************
* choose GNU/Linux : GNU/Linux is Freedom *
*******************************************

----------------------------------

Armand ten Doesschate
Welschapsedijk 141
5652 XL Eindhoven
the Netherlands
tel : (++31) 40 2571 274
e-mail : a_DOT_doesschate_AT_hccnet_DOT_nl


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

Re: Any issues with driving 9s12's with an oscillator? - Doron Fael - May 10 0:49:00 2004

Bill,

PE7/XCLKS needs to be connected to a 2.2Kohm or lower resistance pull-down
resistor to GND, to select the external clock oscillator. The low
resistance is needed in order to overcome an internal pull-up resistor that
is present on PE7/XCLKS during and after Reset.

The 1.48MHz ECLK that you measure suggests that your 9S12DJ64 is in
Limp-Home - Self-Clock mode, and is generating its own low frequency
internally, because it can't detect a valid frequency externally for some
reason (likely because of improper XCLKS settings). When the 9S12DJ64 will
detect the correct 8MHz frequency that you try driving into EXTAL, it will
show a 4MHz clock on ECLK (half EXTAL rate out of Reset, and as long as the
internal PLL is not selected).

Hope this helps,
Doron
Nohau Corporation
HC12 In-Circuit Emulators
www.nohau.com/emul12pc.html

At 23:06 09/05/2004 +0000, you wrote:
>Problems here with my first 9S12 board, was wondering if this sounds
>familiar to anybody...
>
>I have a new MC9S12DJ64CPV design that I'm driving with an 8mHz, 5
>volt oscillator. The oscillator output reaches the EXTAL pin after
>being divided down to 2.5 volts through 2, 470 ohm resistors. There
>is some rounding, but the clock looks pretty good. The circuit is
>very basic and otherwise exactly like an Elektronikladen "CardS12"
>with most of the lines not terminated, but connected to 1 or 2 inches
>of traces on a not-yet-populated board.
>
>MODA, MODB, are tied low through 10k's.
>
>XTAL is open.
>
>XIRQ tied hi through 100k.
>
>XCLKS is now tied lo through a small resistor, I have also tried it
>tied hi. The Motorola docs make contradictory recommendations for
>XCLKS when using an external oscillator.
>
>I have also tried tried shorting together VDDPLL and XFC, as
>recommended by Motorola for non-pll operation, no difference.
>
>Basically, *most of the time* I can't get a ComPod12-Pro to
>recognize the chip. When I plug in the ComPod12-Pro, I notice that I
>get a ~1.48mHz ECLK...is this normal? But a very small percentage of
>the time, the BDM does work! Have tried all the varations of
>plugging-in/powering up/restarting the program/ComPod target
>settings and frequencies. The ComPod works fine with several older
>HC912 boards. I've swapped out the CPU a few times (just now
>discovered how to use the heat-gun for that, it so easy...). All the
>power connections are correct.
>
>Would appreciate any input! WAG's are always welcome here.
>
>Bill T. [Non-text portions of this message have been removed]



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

Re: Any issues with driving 9s12's with an oscillator? - Doron Fael - May 10 2:35:00 2004

Armand,

Please take the following as a constructive criticism. I don't mean to
insult anybody.

I wouldn't recommend the design as you suggest here:

When XCLKS is tied high through a pull-up, the DP256 is expecting to
receive a Colpitts Crystal, which is not what you have.

I would also be concerned of using two 10K resistors for the clock
Oscillator voltage divider. The time constant you get is 5Kohm of
equivalent resistance (the two 10K resistors in parallel), and
approximately 10pF of input capacitance on EXTAL, which yields a 50nSEC
time constant. You are using a 4MHz clock that has a high and a low time of
about 125nSEC. Taking into account that the charge up to a valid logic
level takes 3 - 5 time constants, the EXTAL clock input will not get to
valid clock levels. I think you may have design that is working, but not as
intended and specified in the data sheets.

As a designer of BDMs and emulators, I also don't approve of making any
tricks on the BKGD line like delaying it. BDMs function best when the BKGD
and the Reset lines are connected directly from the BDM POD to the HCS12
pins, with nothing to interfere between them.

I don't know if you have the specified configuration working on one test
board or many production boards, but if I were instead of you, I would look
into fixing these points, so the design works as intended by the silicon maker.

Hope this helps,
Doron
Nohau Corporation
HC12 In-Circuit Emulators
www.nohau.com/emul12pc.html

At 07:35 10/05/2004 +0200, you wrote:
>I have exactly the same configuration with my DP256 : only the clock
>divider exists of 2 * 10 K resistor directly driven with a ACT74 @ 4 MHz.
>I tied XCLKS high with a 10 K pullup. The 2.5 V clock to the DP256 looked
>like a sawtooth but my DP256 works flawlessy.
>
>I experienced the same problems with the bdm : I discovered that my
>reset signal came up to slow (I used a TL7705 for this). My solution was
>to delay my bdm signal to the BKGND pin for a (very) short time only during
>reset, so that the bdm sigal was low when reset was going high and being
>stable. My DP256 always entered into bdm afterwards. >Armand >*******************************************
>* choose GNU/Linux : GNU/Linux is Freedom *
>*******************************************
>
>----------------------------------
>
>Armand ten Doesschate
>Welschapsedijk 141
>5652 XL Eindhoven
>the Netherlands
>tel : (++31) 40 2571 274
>e-mail : a_DOT_doesschate_AT_hccnet_DOT_nl [Non-text portions of this message have been removed]




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

Re: Any issues with driving 9s12's with an oscillator? - Armand ten Doesschate - May 10 11:10:00 2004

On Mon, May 10, 2004 at 10:35:58AM +0300, Doron Fael wrote:
> Armand,
>
> Please take the following as a constructive criticism. I don't mean to
> insult anybody.
>
> I wouldn't recommend the design as you suggest here:
>
> When XCLKS is tied high through a pull-up, the DP256 is expecting to
> receive a Colpitts Crystal, which is not what you have.
>
> I would also be concerned of using two 10K resistors for the clock
> Oscillator voltage divider. The time constant you get is 5Kohm of
> equivalent resistance (the two 10K resistors in parallel), and
> approximately 10pF of input capacitance on EXTAL, which yields a 50nSEC
> time constant. You are using a 4MHz clock that has a high and a low time of
> about 125nSEC. Taking into account that the charge up to a valid logic
> level takes 3 - 5 time constants, the EXTAL clock input will not get to
> valid clock levels. I think you may have design that is working, but not as
> intended and specified in the data sheets.

This was just an experiment : I implemented a collpits configuration
before. Anyway I certainly agree that it shouldnot go into production.
I also fixed the reset circuit ... > Hope this helps,
> Doron
> Nohau Corporation
> HC12 In-Circuit Emulators
> www.nohau.com/emul12pc.html Armand *******************************************
* choose GNU/Linux : GNU/Linux is Freedom *
*******************************************

----------------------------------

Armand ten Doesschate
Welschapsedijk 141
5652 XL Eindhoven
the Netherlands
tel : (++31) 40 2571 274
e-mail : a_DOT_doesschate_AT_hccnet_DOT_nl





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

Re: Any issues with driving 9s12's with an oscillator? - Oliver Thamm - May 10 12:28:00 2004

> XCLKS is now tied lo through a small resistor, I have also tried it
> tied hi. The Motorola docs make contradictory recommendations for
> XCLKS when using an external oscillator. Hi Bill,

low is fine in your case.

Please note that the CRG module data sheet is talking about the
(internal) signal XCLKS while the device guide is describing the
(external) pin /XCLKS. That's the source of the negation - and a lot of
confusion as well :-)

good luck!
Oliver




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

Re: Any issues with driving 9s12's with an oscillator? - tonalbuilder2002 - May 10 20:26:00 2004

Thanks to all of you who helped with this problem. There are some
real experts on this group!

Well, duh! Bringing up a new, expensive board layout is always a
very stressful time and at least I hate to even think that something
might not be right. I hurried up too much in design and tied VDD1
and VDD2 to VCC...the poor 9S12 did not like having its insides
exposed to 5 volts! If I had just checked another 1 or 2 hours
before layout...

But at least I learned about the "limp home" clock and few other
things.

Now everything is fine, and I must say that the ComPod12-Pro is very,
very FAST and a delight to use.

Many Thanks to All,

Bill T.





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