• State Resolved
  • Locked Locked
  • Replies 2 replies
  • Subscribers 46 subscribers
  • Views 768 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

ADS114S08: Excessive input current on selected input

M Julier
Prodigy 20 points

Part Number: ADS114S08

I am using the ADS114S08B in single-end inputs, 0-5V range with reference=0V and full scale=5V, with PGA bypassed.

The negative input is AINCOM, while the positive input is successively connected to AIN0 to AIN11, each one with a duration of 1 ms.

With the scope, I see that each time an input is selected, an input current is flowing to it: the ADC is sourcing about 150 µA.

According to the datasheet, the input current should be "0.5 nA typ.", for input voltages between 0.1 and 4.9V. Some input voltages are between 0 and 0.1V, but this current exists even when the voltage is 0.3V. Input signal AIN11 is +5V (signal "-15V" is disconnected), and I am afraid AIN0 might be >5V.

I have checked that:
- PGA is bypassed
- positive and negative reference buffer bypassed are disabled
- IDAC (excitation current sources) are disabled and disconnected
- VBIAS is disconnected (which implies disabled)
- burn-out current sources are disabled

The current really seems to come from the PGA or the core ADC: in register SYS, if I set SYS_MON=001 (disconnect PGA inputs from multiplexer and short them), the input current disappears (but the ADC gives a zero value).

This input current makes the converted values imprecise, when they are not low-impedance: this is the case for AIN0 (2k), AIN3 (>10k), AIN4 (2k), AIN8 (ISENSE_ANA have 1k impedance).

What can be the reason and is there a solution?

  • 0
    TI__Guru** 113765 points

    Hi M Julier,

    Welcome to the E2E forum!  Often times when excessive current is being drawn one of the inputs is exceeding the absolute maximum input voltage.  This could be related to grounding between the source and the ADC ground.  It could also be related to an over-voltage condition on one of the inputs.  If the ESD input protection diodes start to turn on, then current will begin to flow.

    You say that you think AIN0 might be greater than 5V?  What is causing this suspicion?  Is the op amp powered by bipolar supplies or by a supply greater than 5V?  I suspect that the op amp is powered by a voltage greater than 5V being that you have a voltage divider on the op amp output.

    The schematic doesn't have reference indicators.  Is you circuit a prototype or do you have a PCB?  How are you determining the current (is this an indirect measurement across the input filter resistors)?

    Can you send me the exact configuration settings you are using so that I can try and replicate what you are seeing?  Do you see this behavior on every input?

    I would try to simplify and only connect one input source at a time to see if there is one particular source that may be causing the issue.  I would also suggest that each input be measured at the ADC relative to analog ground.  If I have a little more information I can try and replicate to see if I can help to determine why the current is greater than expected.

    Best regards,

    Bob B

  • 0 in reply to Bob Benjamin
    Prodigy 20 points

    Hello Bob, and thank you for answering so fast.

    Indeed, I realized that AIN0 could be >5V when I was writing my question. I have just checked, and indeed, when the op amp has its input disconnected, the output is +10.5V (it is powered in 0V/+12V), which sends about 2.5mA to the internal clamp diode. After all, having just 150µA on the input is not so bad in these conditions.

    Adding a pull-down on the input before the op amp seems to solve entirely the problem.

    To answer you: I sent an unrouted version of the schematic because I had it readily available, but the board was routed with no modification. It is in prototyping phase, currently with a system in EMC tests, and it will undergo some minor changed, in which I will include the pull-down.

    Regards