MC13213 IRQ problem

[lanceloat]

We design a Evaluation Board referenced on MC13213-SRB schematic. We load the TestMode example application into our evaluation board and get a IRQ problem. when MCU HCS08 set to take the mc13213 modem out of rest and wait for modem interrupt , MCU can't detect the interrupt occur edge.

In this while loop , we fall into a infinite loop because of the check condition "if(IRQFLAG == 1)". We checked  the IRQSC register IRQF bit and fount it never changed

while (u8AttnIrq == FALSE)  {
  if (IRQFLAG == 1)  { //* Check to see if IRQ is asserted */
      u16IrqReg = SPIDrvRead(0x24); //*
      // * Clear MC13192 interrupts and
      // * check for ATTN IRQ from 13192
      // */
     u16IrqReg &= 0x400;
     if (u16IrqReg == 0)  {
         u8AttnIrq = FALSE;
     }
     else  {
         u8AttnIrq = TRUE;
     }
  }
}

We tried use the OSCILLOSCOPE to measure the Interrupt edge in MCU IRQ pin from test point, we found that , the interrupt edge in reference freescale 13213-SRB is droped from 3V to 2V , but the interrupt edge in our evaluation board is droped from 2V to 0V. We don't know if our PCB layout occur problem , can u help us checking if the PCB layout with any problem ?

bellow link1 is Document about mc13213 PCB layout guideline about MC13213 chip
ftp://fantasy.sinosys.com.tw/PCB1.JPG

bellow link2 is Out PCB layout about mc13213 chip
ftp://fantasy.sinosys.com.tw/PCB2.JPG

Does it will cause any undetermined problem ?

thx a lot

[bigmac]

Hello,

If the interrupt input to the MCU is only dropping to 2 volts, this would indicate that something else (perhaps another output) is loading that connection.  Assuming there is not an error in the PCB layout, the most likely causes are a short between adjacent PCB traces, or a mis-aligned SM component.  If you have an external pullup resistor on the interrupt pin, check that the resistor value is correct, and not an extremely low value.

Does the problem occur on more than one of your assembled boards?  If so, this would be indicative of a layout error, unless perhaps there is a consistent alignment error.

The cause of this problem will need to be found.

Regards,

Mac

[lanceloat]

Regard,

Thx for replay and we probably found the reason about our problem.

The solder issue about the Internal connect pin for test point .

Really Thx a lot

Rogner

[fantenna]

Hello,

Could you elaborate on what you found out about the problem and your solution?  We are currently having exactly the same issue. 

Many Thanks.

[peg]

Hello fantenna,

It sounds like this person had a problem experienced by quite a few people here.
This was solder shorts asssociated with the row of pins in the middle of the device.
At least one person got around their problem by following my advice to make a hole where these test pins exist (at least on prototypes) and got around their problem.
Do a search on this forum for this.

[fantenna]

Many thanks for the hint which saved us a lot of collective headaches.  The test pins were definitely the problem.  We are removing the soder pads for these pins in the next edition.

Best Regards.