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.

ADS1256: INA188

Part Number: ADS1256
Other Parts Discussed in Thread: INA188, , LM7705, OPA365, INA225, REF5025

Hi,

I was using INA188 to measure the battery voltage in differential mode. And give the output to a single ended ADC.

Now I plan to use ADS1256 SDADC in differential mode to measure my battery voltage. Can I just take off INA188 and feed in the battery terminals to SDADC pins ? If yes then do I need any signal conditioning circuitry.

Thanks

Regards

Sushant

  • Hi Sushant,

    Most likely you should be able to connect your battery directly. In general, adding signal conditioning stages has the effect of adding noise to the signal you are trying to measure; therefore, I tend to avoid additional signal conditioning when it is not necessary.

    May I ask what is the maximum battery voltage you are trying to measure? If the ADS1256's internal PGA gains are sufficient for your purposes, then I see no reason for an additional gain stage. Additionally, no external level-shifting circuitry is required since the common-mode range of the internal PGA extends to AGND, allowing you to connect the negative side of your battery to ground and still measure the battery differentially without saturating the PGA output.

    Best regards,
    Chris
  • Thanks Chris.
    The battery voltage will be between 0 to 5V.
    My concern is that with high resolution adc without rc filter conditioning , I will have large sdadc step variations even for constant battery voltage. How do I get rid of this. I don't want to load my controller with signal processing algorithms.

    Regards
    Sushant
  • Hi Sushant,

    I would add an RC filter on the ADC inputs. When referring to "signal conditioning" circuitry, I was only considering active components, but an RC filter is recommended for anti-aliasing (noise filtering will be primarily accomplished through the ADC's digital filter).

    To reduce the signal noise without providing additional post-processing filtering in your MCU, I would recommend running the ADS1256 at the slowest data rate that you can, as this reduces the overall signal & noise bandwidth.

    One potential issue I see with a 0-5V signal is that you will not be able to use the ADS1256's internal buffer for signals above 3V. With the buffer disabled you will have a higher input bias current (load) connected to the battery. Unfortunately, the way to avoid this would be to either have a signal conditioning stage (supplied with >5V), or to divide down the input voltage (which also places a load on the battery).

    So I guess my next question is how much of a load can you tolerate on the battery?

    Best regards,
    Chris
  • Thanks for your feedback Chris. I am planning to use ADS1256.
    I dont want to use resistor divider.
    I have a doubt:
    What is meant by "you will not be able to use the ADS1256's internal buffer for signals above 3V." If I give ref vtg of 2.5V and use gain 1, then I shall be able to measure 2*2.5/1 = 5V input. Isnt this right ?
    May be I donot understand the use of buffer in this context. Could you please clarify...
  • Hi Sushant,

    There is an internal buffer stage before the PGA that can be enabled to increase the ADC's input impedance. It is recommended to enable the buffer to prevent loading of the voltage your are trying to measure:

    However, when this buffer is enabled the input common-mode range is limited to the input range of the buffer. The range of allowable input voltages is then +/- 3V...

    I'm not sure if the unbuffered input impedance of the ADS1256 will place too much of a load on your battery or not. If the loading is an issue, then you'd probably need to provide an external amplifier with supply rails greater than 5V and less than 0V to allow for a linear 0-5V buffer output. For the negative supply voltage you could use something like the LM7705, but you would still need another voltage greater than 5V for the positive rail.

    When choosing an amplifier, note that going by the "output swing to rail" specification does not tell you the linear output range of the amplifier, for this you need to pay close attention to the conditions listed for the open-loop gain specification. The following example is taken form the OPA365 datasheet, note that the op-amp can drive to within 10 mV of the supply rail, but the 10-100mV range is likely non-linear as the open-loop gain is not guaranteed to be high in this range.

    I hope that helps,
    Chris

  • Thanks That was useful. I now wonder if I could use it as a bidirectional current sensor (replacing INA225). ADS1256 says it can go to +-2Vref/PGA. But I dont see applying any negative voltage to the IC. (or does it.. !!). So how can it (if it can ) compute negative voltages in differential mode.
  • Hi Sushant,

    Yes, you could use the the ADS1256 to measure current. Usually this is done by forcing current through a resistor and measuring the voltage drop across the resistor.

    The"negative voltages" that you can measure are not actually negative (in absolute value), but rather the differential voltage between positive and negative inputs can be negative. For example, a -3V differential signal will have VINP = 0V and VINN = 3V, such that the difference VINP- VINN = -3V. Therefore, as along as your signal swings around some mid-supply voltage (1.5V in the case that the buffer is enabled), then you can take advantage of the negative input range.

    Best regards,
    Chris
  • Okay Thanks.

    Consider this setup:

    I have buffer enabled, give referance voltage of 2.5V, PGA of 16 and connect a current shunt of 15mOhm, and pass current through it. My understanding is:

    1. Max current I can pass will be (3/16)/0.015 = 12.5A.

    2. I can measure bidirectional currents of 12.5A in both directions. ie: +12.5A and -12.5A. 

    Hope my understanding is correct.

    3. in line with discussions earlier for measuring battery voltage I will have buffer disabled and connect an external RC circuit with opamp buffer. Request your view on this.

    4. Considering that I am interested in high accuracy and precision of current and voltage measurements need your suggestions for supporting circuitary if any I need to include for above measurement cases.

    Regards

    Sushant

  • Hi Sushant,

    1. Your calculation is correct, but read on to #2 as you may need to reduce this value...

    2. I think you will need the +/- 12.5A current to result in an input voltage that swings +/-1.5V around a 1.5V common-mode voltage.  In this case you will probably need to fix one side of the shunt resister to 1.5V and then the maximum voltage swing will be reduced by half, so the current swing will in turn be limited to +/- 6.25A.

    3. This is fine.

    4. Other components to look for would be:

    1. The front-end amplifier: check out this list as a starting point: Precision amplifiers
    2. A 2.5V reference source. I would recommend the REF5025 or any of the REF6x25 devices.
    3. Low-noise LDO for the power supplies of the ADS1256.

    I hope that helps,
    Chris

  • Okay Chris. Thanks.
    Your explainations have helped me to understand better. A last question:
    " I think you will need the +/- 12.5A current to result in an input voltage that swings +/-1.5V around a 1.5V common-mode voltage. In this case you will probably need to fix one side of the shunt resister to 1.5V"
    Is is necessary to have common mode voltage of 1.5V ? What if I do not tie any one side of shunt to 1.5V. In that case, If zero current flows then the voltages at both ends will be same and hence the commonn mode voltage shall be zero. Is this of any concern.

    Regards
    Sushant
  • Hi Sushant,

    You'll have to show me what you mean with a simple schematic...

    The input voltage will have to be correctly biased with respect to the ADC's common-mode input range. If the input signal is floating then you'll run into issues with the PGA saturating. Biasing one side of the resistor to 1.5V gives you some headroom for positive and negative currents. If instead you bias the resistor to 0V, then likely you only be able to support positive current measurements.

    Take a look at this Excel calculator for some guidance on the ADC's input range: www.ti.com/.../sbac145

    Best regards,
    Chris