This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

SN74AVC4T774 Current Leakage

Other Parts Discussed in Thread: SN74AVC4T774

I'm working on a new design which included the use of 2 SN74AVC4T774.  They work fine under normal operating conditions, however when I power down VCCB 1.8V -> GND the chips consume 2mA from VCCA which is at 3V.  Normally around 25uA or so is being pulled from 3V when VCCB is 1.8V.  

If I remove these chips from my board, the extra 2mA disappears so I know that the issue is related to these chips.

To clarify the problem scenario,

I have a modem which is being powered by a 3.8V step-up regulator, the modem provides a 1.8V reference.  I use the 1.8V reference to supply VCCB and I supply VCCA with 3V from an LDO which also powers an MSP430.

I would like to shut down the modem to save power, in doing so I turn off the 3.8V regulator which causes the 1.8V reference to go to GND (I've checked this 1.8V with an oscilloscope and it does as I expect it to).

When I turn off the modem I would expect the current consumption of my board to be at 2mA, but instead it is closer to 4mA+.

I've attached a schematic in PDF format to hopefully give you a better idea of what is going on.

/cfs-file/__key/communityserver-discussions-components-files/151/2337.Schematic.pdf

Things I've tried already,

Jumper OE to GND so that the chips are always on.  I removed U15 and R49 in doing this. 

Removed all of the 1M pull-down resistors (I didn't want floating inputs, but It was pointed out to me that maybe this was causing an issue).

Placed a 6.8K pull down resistor on the 1.8V line just to make sure it was at ground and not floating after turning the modem off.

In software I have made all the lines leading to the chips from the MSP430 outputs at 3V, outputs at 0V, all inputs, all inputs with MSP430 internal pullups enabled. This doesn't seem to change the consumption at all.  Obviously I can't control the modem side, but they are all at GND when I check them with an oscilloscope.

I've also been talking to an FAE from U-Blox the company which provides the modem, as they used this exact chip in their system integration manual but so far we haven't been able to solve the problem.

  • Hi Drew

    Increased current consumption might be caused if the inputs are not in defined high (vcc) or low  states(GND). Please ensure that there not no floating inputs.

    Also, is it possible to reference the OE to Vcca through a pull-up resistor?

    Thanks

  • Hello Jennifer,

    I've made sure none of the inputs are floating, via pull-down resistors on the Modem side of the circuit as well as making sure that all the pins connected to the device on the MSP430 side of the circuit (the net labeled pins on the schematic) are outputs and set to a specific value.  I've verified this with an oscilloscope.  Originally VCCA was referenced through a pull up resistor, but I've since just put it straight to GND to keep it always enabled.  Either way the issue occurs when the 1.8V side is powered down.

    Best Regards,

    Drew

  • Hi Drew

    Can a debug be done by connecting the 3V to another supply, instead of the LDO to see if this issue still exists?

    One other aspect to consider is the power up sequence. This can also lead to increased supply currents. Power up Vcca first. Vccb can be ramped along with or after Vcca. I understand that you see this issue only during power down of Vccb, but I wanted to point this out in case there was multiple toggling of supplies in your application.
  • Jennifer,

    I moved the 3V Rail to a Bench Power Supply, still the same issue if the 1.8V Rail is powered down.

    On my application board the battery feeds a 5V step up and the 5V step up feeds the 3V LDO, so 3V is always on and ramped up before the 1.8V.  The 1.8V Rail is the only one that is being toggled.

    Best Regards,

    Drew

  • Hi Drew
    One thing worth looking into is using the bench supply to power the 1.8V. I think you have diligently done the process of elimination on all other sources that might cause this leakage.
    Thanks