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.

ADS1120: DVDD supply current much higher than datasheet spec

Part Number: ADS1120

Hi experts:

My customer is designing with ADS1120. To filter out the noise they use a serial 220ohm resistor between 3.3V supply and DVDD pin. The DVDD voltage is initially 3.3V after power up and before configuration the device, however, once the SCLK is pulsed (with 5MHz frequency), they found the DVDD voltage dropped to about 2V. That means the DVDD voltage is in fact 1.3V/220ohn = 6mA, while the datasheet spec is only 95uA for IDVDD. Why?

Here is the schematic:

The device seems function properly. They have tried to short the 220ohm resistor and ADC works properly. I have also confirmed with engineer that the filtered out, 3VISO1 is only used to drive DVDD pin and 1M pull up resistor, no other load on it.

Question is why the DVDD current of device is so high?

Thank you!

John

  • Hi John,

    The CMOS devices have a crossover region where for a very short period of time both the PMOS and NMOS transistors are conducting.  The current shown in the datasheet is based on an average over time.  It does not show instantaneous current where larger currents can occur.

    Adding ferrites and resistance in line will affect operation. The ferrites will attempt to choke these fast instantaneous currents.  Also, large resistance in the supply path will add a lowpass filter, but again will add disturbance in the supply.

    There may also be an assumption of operating current in a static operation and current when the device is operating.  The output driver circuits for the digital could potentially drive more current than the average current of the digital operation.

    If the voltage drops too low (2V for example) the device operation will be uncertain as the minimum operating voltage is 2.3V.  I would recommend either using the resistor or the ferrite but not both.  If the resistor is used, I would recommend using a value less than 20 Ohms.

    Best regards,

    Bob B

  • Hi Bob,

    Thank you for your reply.

    But I do not think it is due to fast instantaneous currents. Pls note customer has used two bypass capacitors across DVDD and DGND (0.1uF+2.2uF). I think the two bypass caps would be enough to supply the fast instantaneous currents, and the ferrite+220R resistor only supply the average DVDD current (95uA). Considering the average current is very low and the ferrite can only isolate very high frequency noise, so I think using a 220ohm resistor is a good design. 

    Yes, using a lower resistance can make the device work. I just want to understand why the digital supply current does not match the datasheet.

    Regards,

    John

  • Hi John,

    This is always difficult to explain as this discussion also takes place on the analog side as well and can be an even larger issue.  There is the  assumption that the supply caps will be able to hold a sustained charge with the ability to recharge the cap to maintain the same level of charge quickly.  The IC designer makes the assumption that the device power supplies are low impedance supplies.  At some point the current demands are strong enough that the resistance and inductance can choke the current which prevents the cap(s) from charging back up quickly.  So no matter the actual current draw, this is not a good practice to limit the supply current.  I have seen instances where a device will start to power and collapse the supplies until a brown-out condition occurs and places the device back to a POR condition.  This process can repeat multiple times and may not even come completely out of the POR state.

    As to the actual current draw of the device in this particular situation, there is the assumption that the current is exceeding the datasheet values.  As I said, this is an average current over time.  As to conditions such as oscillator startup, clocking and driving an output, these are conditions that may cause the current to increase for short periods of time.  As an Ampere is a measurement of Coulombs per second, the total amount of charge movement may increase for periods of time and decrease for periods of time.  I have seen current demands of short duration into the 10's of mA.  This would be enough to collapse the supply where the recharge will not be fast enough even with a 2.2uF cap.

    It may be best for customer to watch the power with and oscilloscope to see the action.  If they have a current probe that would be even better to see the current draw of the device over time.

    Best regards,

    Bob B