S12 / MagniV MicrocontrollersのトピックMM912H634 low-side drivers issue

MM912H634 low-side drivers issue

daviddai — Tue, 07 Jul 2015 01:33:18 GMT

Hi,

I have a stange issue for MM912H634 low-side drivers test,

Firstly, I use B_LSCR_LS1 and B_LSCR_LS2 to control the relay with below code and 9V voltage,

Everything is ok, the relay can work as I expected.

B_LSCEN = 0x05;

D2DCTL0_D2DCLKDIV1 = 1;

D2DCTL0_D2DCLKDIV0 = 1;

D2DCTL1 = 0x8F;

D2DCTL0_D2DEN = 1;

B_WDR = 0b01000111;

B_WDSR = 0xAA;

for(;;)

{

B_LSCR_LS1 = 0;

long_wait();

B_LSCR_LS1 = 1;

long_wait();

B_LSCR_LS2 = 0;

long_wait();

B_LSCR_LS2 = 1;

long_wait();

_FEED_COP(); /* feeds the dog */

}

Then, during the relay is working, I artificially simulated a voltage drop(9V --> 3.6V), relay stopped.

Strange issue happened, relay stopped forever, even I re-make the code and re-flash MCU, re-Power on,

I checked the MCU LSx rigister and found no change as below even I run Step over the code.

I do not believe the haredware is destroyed.

Can you please tell me the possiable reason?

Thanks a lot!

Best Regards

David Dai

Re: MM912H634 low-side drivers issue

RadekS — Wed, 08 Jul 2015 15:56:52 GMT

Hi David,

I suppose that it could have something to do with bg1p25sleep trimming at analog side.

Please try following recommended initialization procedure:

#define IOBYTE(address) (*((volatile unsigned char*) (address)))

#define IOWORD(address) (*((volatile unsigned int*) (address)))

#define TRIM0 IOBYTE(0x80C0)

#define TRIM1 IOBYTE(0x80C1)

#define TRIM2 IOBYTE(0x80C2)

#define TRIM3 IOBYTE(0x80C3)

D2DCTL1 = 0x8F; // IRQ enable, max timeout

D2DCTL0 = 0x80; // D2D enable, 4Bit, !Stop in Wait, D2DCLK=BUS

D2DSTAT0 = 0x80; // Clear D2D Errors

MMCCTL1 = 0x01;

B_CTR0 = TRIM0;

B_CTR1 = TRIM1;

B_CTR2 = TRIM2;

B_CTR3 = TRIM3;

B_CTR2 = 0x1E;

MMCCTL1 = 0x00;

B_LSCEN = 0x05;

…

I hope it helps you.

Have a great day,
RadekS

-----------------------------------------------------------------------------------------------------------------------
Note: If this post answers your question, please click the Correct Answer button. Thank you!
-----------------------------------------------------------------------------------------------------------------------

Re: MM912H634 low-side drivers issue

daviddai — Thu, 09 Jul 2015 08:40:43 GMT

Hello RadekS,

Thanks for your help, I think I have to use much much time to read the MCU SPEC carefully!

Thanks!

Best Regards

David Dai

Re: MM912H634 low-side drivers issue

RadekS — Thu, 09 Jul 2015 09:51:25 GMT

You are welcome.

If I understood correctly, it works now… Right?

Unfortunately, I am not sure if reading MCU SPEC will really help, in this case. I never thought that trimming value of internal reference could have any influence on low-side drivers. But it has. It is simply analog die and it needs set trimming values before use.

However I suppose that this issue with low-side drivers is described in errata:

http://www.freescale.com/files/analog/doc/errata/MM912_634ER.pdf

See “Unexpected setting of VROVC Bit and disabling of Low-side Driver(s) at VSUP Power On”

It seems that recommended loading values according reference manual is slightly different:

B_CTR0 = TRIM0;

B_CTR1 = TRIM1;

B_CTR2 = TRIM2 & ~B_CTR2_SLPBG_LOCK_MASK;

B_CTR2 |= B_CTR2_SLPBG_LOCK_MASK;

B_CTR3 = TRIM3;

Errata recommend do not lock CTR2 values by lock mask as workaround

// B_CTR2 |= B_CTR2_SLPBG_LOCK_MASK;

In that case trimming values must be loaded at every MCU reset.

Note: I already met with similar issue and using template code from delivered CD helped me.

Code from my last reply is just copy-paste from this template code.

I hope it helps you.

Have a great day,
RadekS

Re: MM912H634 low-side drivers issue

daviddai — Fri, 10 Jul 2015 01:26:20 GMT

Hello RadekS,

Yes, it works well.

Also I found the VROVC Bit is abnormal, it always is 1.

Now after adding the new code, it's ok.

Thanks a lot!

Best Regards

David Dai

Re: MM912H634 low-side drivers issue

RadekS — Fri, 10 Jul 2015 07:29:32 GMT

Hi David,

You are welcome.

I branch your new question about True-Time Simullator to separate thread for better traceability:

https://community.freescale.com/thread/359003