Yahoo Groups archive

68300

Index last updated: 2026-04-29 00:01 UTC

Thread

dead (?) CPU

dead (?) CPU

2002-09-12 by Jeff

I built up two prototypes of a 68F375 circuit using short-run 
prototypes from one of the dirt cheap manufacturers.  The 3.3V bus 
was shorted to the ground plane at one via (able to withstand one 
amp).

The chip is dual supply -- 3.3V core and 5V periphery, etc.

This means it saw 5V while the 3.3V supply was at 0.4V.

I found and surgically corrected the PCB.  Power is now good, but the 
chip will not respond to the background debugger monitor (PE Micro 
ICD32).

Question -- is it a foregone conclusion that I toasted the chip by 
applying 5V without good power on 3.3V?  I DO have a second board but 
wish to test as much as possible with this one as possible.  I do 
have a tray of parts on order (19 WEEKS???) but have only two samples 
on hand.

respond here or directly to andle@....

Re: dead (?) CPU -- RESET not released

2002-09-12 by Jeff

I have built prototype boards with TSC pulled low as indicated in the 
attached e-mail (with Motorola tech support).

The chips will NOT release reset (it stays low).
All of the pins that I have connected appear tristated, but in single 
chip mode this is to be expected...

I have the CPU configured as follows:
66.667MHz 45/55 duty cycle oscillator chip giving a clean 2X clock 
signal on a short lead.  I did not pin out CLKOUT so I have no access 
to this as a diagnostic.

MODCLK is hard wired low.
BERR is tied low (single chip mode) -- first I used 10K and now it is 
hard wire low...
HALT is tied high through 10K and is high
BKPT is tied high through 4.7K and is high

RST tied high through 1K but measures 60 mV... (will NOT release)

BGACK and BR are tied together and the pair is tied high through a 
10K resistor.  

NOTE that the BR & BGACK nodes are at 1.5V and do NOT appear to be 
tristated.  When I remove the 10K pull-up they remain at 1.5V. 

BR and BGACK should both be inputs... but the manual says they have B-
type output drivers.

Is it possible one of them is coming up as CSE or CS[0] and is 
pulling the other low?  





> 
> Dear Mr. Andle:
> 
> In reply to your Service Request SR 1-35542677 (see details below):
> 
> You are correct.  The 'F375 documentation has some errors in it.  
The TSC pin needs be pulled low for normal operation and pulled to 
Vdd for tristate operate.  Sorry for the incorrect information.  I 
have asked the factory to update this information in the data book.
> 
> ************ Details of your request: ************
> 
> Date Opened :   7/11/2002 11:48:59 AM
> Product :       MC68F375BGMZP33
> Topic :         Technical Request
> Subject:        68F375 erratta?  TSC
> 
> Description : 
> 
> page 4-69 says that asserting a logic low on TSC (I assume this 
means VSS) places the MCU into high impedance mode.
> 
> Please confirm that TSC-->VSS ENABLES outputs.

Re: [68300] dead (?) CPU

2002-09-13 by Charles Melear

Jeff,

What's the storey here?  Do you need a couple of samples for Motorola.  I know some people there and perhaps I can put in a good word for you.  Send me your physical shipping address.

Regards,

Charlie
Show quoted textHide quoted text
  ----- Original Message ----- 
  From: Jeff 
  To: 68300@yahoogroups.com 
  Sent: Wednesday, September 11, 2002 8:51 PM
  Subject: [68300] dead (?) CPU


  I built up two prototypes of a 68F375 circuit using short-run 
  prototypes from one of the dirt cheap manufacturers.  The 3.3V bus 
  was shorted to the ground plane at one via (able to withstand one 
  amp).

  The chip is dual supply -- 3.3V core and 5V periphery, etc.

  This means it saw 5V while the 3.3V supply was at 0.4V.

  I found and surgically corrected the PCB.  Power is now good, but the 
  chip will not respond to the background debugger monitor (PE Micro 
  ICD32).

  Question -- is it a foregone conclusion that I toasted the chip by 
  applying 5V without good power on 3.3V?  I DO have a second board but 
  wish to test as much as possible with this one as possible.  I do 
  have a tray of parts on order (19 WEEKS???) but have only two samples 
  on hand.

  respond here or directly to andle@....


        Yahoo! Groups Sponsor 
              ADVERTISEMENT
             
       
       

  ---------------------------------------------------
  To unsubscribe from this group, send an email to:
  68300-unsubscribe@yahoogroups.com

  To learn more about Motorola Microcontrollers, please visit
  http://www.motorola.com/mcu



  Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service. 



[Non-text portions of this message have been removed]

Re: [68300] Re: dead (?) CPU -- RESET not released

2002-09-13 by Charles Melear

Jeff,

Look for answers below in text.

Regards,

Charlie
Show quoted textHide quoted text
  ----- Original Message ----- 
  From: Jeff 
  To: 68300@yahoogroups.com 
  Sent: Thursday, September 12, 2002 12:09 PM
  Subject: [68300] Re: dead (?) CPU -- RESET not released


  I have built prototype boards with TSC pulled low as indicated in the 
  attached e-mail (with Motorola tech support).

  The chips will NOT release reset (it stays low).
  All of the pins that I have connected appear tristated, but in single 

  chip mode this is to be expected...
  *************************************************************************************
  If you have modclk tied low, you will not be using the VCO / PLL.  So, the chip will run at what ever frequency you are driving into EXTAL.  If you have no clock, you will never release reset.
  ******************************************************************************************


  I have the CPU configured as follows:
  66.667MHz 45/55 duty cycle oscillator chip giving a clean 2X clock 
  signal on a short lead.  I did not pin out CLKOUT so I have no access 
  to this as a diagnostic.

  ***********************************************************************************************************************
  You really need clock out to initially check out your system.  This will tell you whether your clocking scheme is working.
  ***************************************************************************************************************************

  MODCLK is hard wired low.
  BERR is tied low (single chip mode) -- first I used 10K and now it is 
  hard wire low...

  *******************************************************************************************************************************
  You can hard wire BERR low but you will not be able to use external interrupts.  Yes, I know that the manual says you can tie BERR low if you are using the Single Chip mode.  However, all external interrupts will be treated as "Spurious Interrupts".  We won't mention any names about who left this little "funny aspect" of the part in the manual.  It has been corrected in the on-line version to say that you need to release BERR after the release of reset if you want to use external interrupts.
  ************************************************************************************************************************************

  HALT is tied high through 10K and is high
  BKPT is tied high through 4.7K and is high

  RST tied high through 1K but measures 60 mV... (will NOT release)

  BGACK and BR are tied together and the pair is tied high through a 
  10K resistor. 

  **************************************************************************
  NO NO NO NO H... NO.  BGACK is active from the release of reset even in the single chip mode and it will definitely screw up the BDM mode.  BGACK and BR are configured separately and you can get some strange goings-on.  Do not tie these pins together.  However, it is imperative that they both be pulled up.  This is probably not going to cause you any problems right now but it is horrible design practice and it will bite you sooner or later.
  ****************************************************************************** 

  NOTE that the BR & BGACK nodes are at 1.5V and do NOT appear to be 
  tristated.  When I remove the 10K pull-up they remain at 1.5V. 

  BR and BGACK should both be inputs... but the manual says they have B-
  type output drivers.

  Is it possible one of them is coming up as CSE or CS[0] and is 
  pulling the other low? 

  ************************
  YES.  BR and BGACK are configured separately.  BGACK is a really dangerous pin and it should never be left floating.
  *********************************************

  Regards,

  Charlie 





  > 
  > Dear Mr. Andle:
  > 
  > In reply to your Service Request SR 1-35542677 (see details below):
  > 
  > You are correct.  The 'F375 documentation has some errors in it.  
  The TSC pin needs be pulled low for normal operation and pulled to 
  Vdd for tristate operate.  Sorry for the incorrect information.  I 
  have asked the factory to update this information in the data book.
  > 
  > ************ Details of your request: ************
  > 
  > Date Opened :   7/11/2002 11:48:59 AM
  > Product :       MC68F375BGMZP33
  > Topic :         Technical Request
  > Subject:        68F375 erratta?  TSC
  > 
  > Description : 
  > 
  > page 4-69 says that asserting a logic low on TSC (I assume this 
  means VSS) places the MCU into high impedance mode.
  > 
  > Please confirm that TSC-->VSS ENABLES outputs.



        Yahoo! Groups Sponsor 
              ADVERTISEMENT
             
       
       

  ---------------------------------------------------
  To unsubscribe from this group, send an email to:
  68300-unsubscribe@yahoogroups.com

  To learn more about Motorola Microcontrollers, please visit
  http://www.motorola.com/mcu



  Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service. 



[Non-text portions of this message have been removed]

Re: [68300] Re: dead (?) CPU -- RESET not released

2002-09-13 by Jeff Andle

****************************************************************************
*********
  If you have modclk tied low, you will not be using the VCO / PLL.  So, the
chip will run at what ever frequency you are driving into EXTAL.  If you
have no clock, you will never release reset.

****************************************************************************
**************
** The documentation says that with MODCLK low I want 2Fsys...
** HOWEVER, when I removed the 66MHz oscillator
** (no mean feat removing a metal package SMT oscillator
** without even any solder braid!) and replayed it with the
** 16.78MHz SMT oscillator from my 68HC916Y3 version,
** it all works.  Definitely a clock problem but the 66MHz clock
** was very stable...

** The behavior suggests that the manual is in error and what you said is
correct --
** I need a 33 MHz external oscillator to operate at 33 MHz...



  I have the CPU configured as follows:
  66.667MHz 45/55 duty cycle oscillator chip giving a clean 2X clock
  signal on a short lead.  I did not pin out CLKOUT so I have no access
  to this as a diagnostic.

****************************************************************************
*******************************************
  You really need clock out to initially check out your system.  This will
tell you whether your clocking scheme is working.

****************************************************************************
***********************************************
** there are a few pins I *wish* I had pinned out for diagnostic purposes...


****************************************************************************
***************************************************
  You can hard wire BERR low but you will not be able to use external
interrupts.  Yes, I know that the manual says you can tie BERR low if you
are using the Single Chip mode.  However, all external interrupts will be
treated as "Spurious Interrupts".  We won't mention any names about who left
this little "funny aspect" of the part in the manual.  It has been corrected
in the on-line version to say that you need to release BERR after the
release of reset if you want to use external interrupts.

****************************************************************************
********************************************************
** I don't (now), but I will take that under advisement (archive this
e-mail)
** in case i try that later and forget why it failed...


  BGACK and BR are tied together and the pair is tied high through a 10K
resistor.
  **************************************************************************
  NO NO NO NO H... NO.  BGACK is active from the release of reset even in
the single chip mode and it will definitely screw up the BDM mode.  BGACK
and BR are configured separately and you can get some strange goings-on.  Do
not tie these pins together.  However, it is imperative that they both be
pulled up.  This is probably not going to cause you any problems right now
but it is horrible design practice and it will bite you sooner or later.
 ***************************************************************************
** yeah, when I explained the possibility that one is CS[0]
** at startup and that I risked this to save a resistor he kicked me too...
** It will NOT be this way on the pilot production rev's.
** I assume the problems it can cause will all be random failure to start
up,
** NOT sudden failures once operating?
** BGACK and BR bit me on the initial HC916Y3 version of this sensor
controller as well.


  ************************
  YES.  BR and BGACK are configured separately.  BGACK is a really dangerous
pin and it should never be left floating.
  *********************************************
** I took a functional 916Y3 unit and shorted the nodes --
** there was no problem, but as you say, spurious, and that
** is a different integration module as well...
** mental note -- while the trace connecting them is under the chip,
** at least put the resistor back on!

Move to quarantaine

This moves the raw source file on disk only. The archive index is not changed automatically, so you still need to run a manual refresh afterward.