Problem with DEMOAC

cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

Problem with DEMOAC

Jump to solution
2,870 Views
DakotaHound
Contributor I

Good Afternoon,

 

I recently purchased a DEMOAC board, but cannot seem to get it to operate.  I previously used a DEMOJM board, and wrote C code using CodeWarrior v6.3 and Processor Expert, and everything worked as expected (IO, A2D, PWM, etc.).  On the DEMOAC board, I wrote a simple program, compiled, connected to the board, programmed the board, and it started immediately (it did not wait for me to press the green arrow).  When I select Halt, it stops once, but seems to be stuck in the initialization code.  If I continue (green arrow), it will not stop again (it just ignores the command).

 

I have tried downloading the code from Processor Expert (the initial code with no modifications) and it does the same thing.  Downloading and running this initial code on the DEMOJM works (it doesn't do anything, but it works).

 

Any help or suggestions would be greatly appreciated.  What is the minimum code, using Processor Expert, that will run in the DEMOAC?

 

Thank you.

Labels (1)
0 Kudos
1 Solution
661 Views
ProcessorExpert
Senior Contributor III
Hello,
  
I have reproduced the problem but it is not related to Processor Expert. It is caused by incompatibility of earlier versions of AC256 MCUs with CW V6.3. To avoid the problem it is necessary to program the chip by using debugger (hiwave.exe) from CW V6.2.  
 
Please follow the steps below to upload yout binary(*.abs) file to chip:
 
1. Download and install CW V6.2
2. Go to the prog directory within your CW V6.2 installation and run hiwave.exe
3. Select appropriate connection and derivate
4. Using the "Load..." (CTRL+L) menu command select your binary file - for more details please find attached screenshot
5. Programmation process will start after clicking "Open" button
 
Could you please post here full name of your chip? (e.g PCF51AC256LK, with chip mask M50N)
best regards
Vojtech Filip
Processor Expert Support Team

View solution in original post

0 Kudos
7 Replies
661 Views
ProcessorExpert
Senior Contributor III

Hello,

it seems that the problem is caused by incorrect settings of CPU component (components was formerly called beans).

Therefore could you please post here a simple project to reproduced the problem or at last settings of CPU bean and row where is the problem?

In case that you are running from external crystal, are you sure that its value on demo board correspond to the value entered in the CPU component ("Clock frequency [MHz]" property)?

 

What version of Processor Expert are you using?

 

best regards
Vojtech Filip
Processor Expert Support Team

0 Kudos
661 Views
DakotaHound
Contributor I

Good Morning,

 

Thank you for the quick response.

 

I am using CodeWarrior 6.3 SE.  Although my post states "newbie," I have been using Motorola/Freescale microcontrollers for many years.  This includes the use of many DEMO/EVAL boards.

 

My past experience has been that the default configuration provided by Processor Expert can be programmed into the device, and it runs (it doesn't do anything but sit in the for loop - but it runs).  The DEMOJM board worked this way.

 

For the DEMOAC board, however, I programmed the board with the default startup code, the debugger window appeared, and it began running immediately (the green arrow was gray).  It displays:

 

move.w        #0x2700,sr

 

If I halt the processor, it displays:

 

setReg8(SPMSC2, 0x00); 

or 

lea           _SP_INIT,a7

 

If I start it again, it will not halt.  It appears to be stuck in the initialization code.

 

I tried resetting the watchdog timer in the loop, but that did not seem to make any difference.

 

The startup project is attached.  It should be running with the internal clock at 4 MHz bus speed (well below he limit), with the FLL engaged .  Any help will be appreciated.  What is the minimum stable startup configuration using Processor Expert 6.3?

 

 

0 Kudos
662 Views
ProcessorExpert
Senior Contributor III
Hello,
  
I have reproduced the problem but it is not related to Processor Expert. It is caused by incompatibility of earlier versions of AC256 MCUs with CW V6.3. To avoid the problem it is necessary to program the chip by using debugger (hiwave.exe) from CW V6.2.  
 
Please follow the steps below to upload yout binary(*.abs) file to chip:
 
1. Download and install CW V6.2
2. Go to the prog directory within your CW V6.2 installation and run hiwave.exe
3. Select appropriate connection and derivate
4. Using the "Load..." (CTRL+L) menu command select your binary file - for more details please find attached screenshot
5. Programmation process will start after clicking "Open" button
 
Could you please post here full name of your chip? (e.g PCF51AC256LK, with chip mask M50N)
best regards
Vojtech Filip
Processor Expert Support Team
0 Kudos
661 Views
DakotaHound
Contributor I

Good Afternoon,

 

Using the v6.2 CW hiwave.exe worked, although it took several tries (it did not give me the opportunity to select the processor version on the first two tries).  Thank you - I would never have found the problem.  Going back to an earlier version of CW is not something that I would normally try.  Interestingly, the MCF51JM part only works with 6.3 (not 6.2) when using the USB BDM programmer.

 

The device is:

 

PCF51AC256AVLKE

M60N

QAC0817

 

I am somewhat disheartened that this is an older device - I purchased the DEMOAC board a few days ago.  I will likely order a new daughter card for the DEMOAC board.  Is there some way to be sure that the daughter card I order has a newer device?

 

 

 

0 Kudos
661 Views
petr_uhlir
NXP Employee
NXP Employee

There is a problem with initial revision of chip with workaround in debugger. The support in 6.3 intended for final fixed silicon.   To check whether this is the case you can try the following workaround -  you can try to modify the coldfire.ini file (in prog directory, make the copy of current version).

The lines DBG_MODULE_DLL=dbgcoldfirev1.dll should be replaced with DBG_MODULE_DLL=dbgcoldfirev1_ac256.dll for sections related to AC256 and AC128 (like below, only dll name should be changed):  

 

[MCUID7005]
;MCF51AC256A
DMM_MODULE0=FLASH 0x0 0x40000 100 5 1 2 1 1 0
DMM_MODULE1=RAM 0x800000 0x8000 101 5 0 2 1 1 0
CORE=V1
DBG_MODULE_DLL=dbgcoldfirev1_ac256.dll
FCS=cfv1fam.fcs  

 

[MCUID7006]
;MCF51AC128A
DMM_MODULE0=FLASH 0x0 0x20000 100 5 1 2 1 1 0
DMM_MODULE1=RAM 0x800000 0x8000 101 5 0 2 1 1 0
CORE=V1
DBG_MODULE_DLL=dbgcoldfirev1_ac256.dll
FCS=cfv1fam.fcs  

 

[MCUID700E]
;MCF51AC256B
DMM_MODULE0=FLASH 0x0 0x40000 100 5 1 2 1 1 0
DMM_MODULE1=RAM 0x800000 0x8000 101 5 0 2 1 1 0
CORE=V1
DBG_MODULE_DLL=dbgcoldfirev1_ac256.dll
FCS=cfv1fam.fcs  

 

[MCUID700F]
;MCF51AC128C
DMM_MODULE0=FLASH 0x0 0x20000 100 5 1 2 1 1 0
DMM_MODULE1=RAM 0x800000 0x4000 101 5 0 2 1 1 0
CORE=V1
DBG_MODULE_DLL=dbgcoldfirev1_ac256.dll
FCS=cfv1fam.fcs

0 Kudos
661 Views
wrljet
Contributor II

I have similar symptoms with MCF51AC256A and CodeWarrior 6.2.

 

Weird hangs like the previous poster, doesn't seem to get through the initialization.

 

My device came from the sample program, and is marked:

 

PCF51AC256ACFUE

M60N

QAA0811

 

Modifying ColdFire.ini as you explained above worked for me:

 

 

DBG_MODULE_DLL=dbgcoldfirev1.dll should be replaced with DBG_MODULE_DLL=dbgcoldfirev1_ac256.dll

 

 

When I use newer chips with a different mask ID, will I have to switch back to the other DLL?

 

Thanks,

Bill

 

 

0 Kudos
659 Views
ProcessorExpert
Senior Contributor III

 

Hello,

 

I think you should send service request.

 

best regards
Vojtech Filip
Processor Expert Support Team

 

 

0 Kudos