Shop OBEX P1 Docs P2 Docs Learn Events
PLL problems - uneven power pins? — Parallax Forums

PLL problems - uneven power pins?

PhilldapillPhilldapill Posts: 1,283
edited 2009-04-23 16:24 in Propeller 1
Correct my understanding of this, but from what I've read, the main reason for the PLL failure in the Propeller is uneven power pins? i.e., one Vss might be a little higher than the other.

I have never had a PLL problem with the DIP chips - only with the QFP. The QFP and QFN chips have 4 Vss and Vdd pins, so is this the reason those chips seem to fail more often? Are these chips more susceptable to failure because they have a higher chance of different voltage levels on the pins?

I can't find the thread, but I remember seeing chip say something about the Prop II having many power pins in order to minimize any PLL failure... That seems counter intuitive.

Or am I totally wrong on this?

Comments

  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2009-04-23 01:33
    The only reason for any chip to have different levels on its supply pins is a poorly-designed PCB layout. Bringing out more Vdd/Vss pins will help to even out any internal differences in voltage levels, since doing so can take advantage of heavy PCB conductors, which have less resistance than the internal power busses. But this requires careful layout and bypassing on the part of the PCB designer to be successful.

    -Phil
  • Beau SchwabeBeau Schwabe Posts: 6,568
    edited 2009-04-23 16:24
    Internally the chip can have different VDD or Ground power if only one external VDD or GND has been connected. This is due to IR drop over the distance of the actual power routing within the chip.

    In the PropI you have Power and Ground that's roughly 100um wide that goes around the perimeter of the Chip. Since the Chip is about 6.8 mm a side (6800um), and Power and ground only enters the Chip at two opposite ends, the IR drop from one end to the other if you did only connect Power and Ground to one side would be an estimated total of 10.6 Ohms.

    Given that the sheet resistance is about 78mOhms per square... there are 68 "squares" on a side (6800/100 = 68)

    68 squares X .078 Ohms = 5.3 Ohms ... remember, since you have two wires (Power and Ground) this would effectively be 10.6 Ohms


                 5.3 Ohms    5.3 Ohms
               o---/\/\---o---/\/\---o 
               |                     |
    Power1 <---o                     o---> Power2 
               |                     |
               o---/\/\---o---/\/\---o
                 5.3 Ohms    5.3 Ohms
    
    
    Rtotal between Power1 and Power2 = 5.3 Ohms
    
    



    When Power and Ground are applied to BOTH ends, the worst case scenario IR drop goes down to 2.65 Ohms (or 5.3 Ohms total for both Power and Ground) at the extreme distance away from power or ground.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Beau Schwabe

    IC Layout Engineer
    Parallax, Inc.
Sign In or Register to comment.