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.

PGA117 for volt and current measurement

Other Parts Discussed in Thread: PGA117, PGA113

i need help to calculate volt and current measurement with pga117 ic where i want to measure 0 to +/- 1000 volt dc in one input and other input i want to measure 0 to +/- 10 ampere dc using shunt resistor in application note i am not very clear that how to configure inputs for this requirement so plz help me

  • Hello Shams,

    This could be tricky - what kind of application are you working with that has such large input voltage and current requirements?

    For the current input portion, you can follow the design equations in this TI Precision Design, substituting in your own values for ILOAD(max) and VSH(max), as well as the available gain settings of the PGA117. Be mindful to keep VSH quite low so that the shunt resistor does not dissipate an extreme amount of power at max current.

    The voltage input portion will be more difficult since the ±1000V signal will have to be heavily attenuated. Some kind of hefty voltage divider is required, but a clever combination of resistors will have to be used to prevent exceeding the voltage and current requirements of each component.

    Best regards,

    Ian Williams
    Linear Applications Engineer
    Precision Analog -  Op Amps

  • dear Ian 

    thank you for your reply i am trying to build 6 and 5 digit  multimeter's where my goal is to keep circuit diagram as simple as possiable and cost of product will also remain economical for analog to digital conversion still experimenting on TI delta sigma adc's but result are not as i am expecting according to data sheet but in the mean while results are far more better for linear technology delta sigma adc's and analog front end so let see what i can kind of performance i can get from TI applications engineer help my both products are quantity based products 

    same circuit will be used in two different products one power supplies and other small handheld multimeter's as i mentioned multimeter so very low value shunt will be used instead of multiple shunt due to space is very limited and  purpose to use this pga was to utilise scope gains and inbuilt multiplexer for different signals  while this ic is economical and have high input impedance 10Gohm  so 10Mohm resistor divider can be used but here again confusing question comes that what would be pga117 output at 1 millivolt input using 10Mohm resistor voltage divider on 200 gain selection does this will work while adc input full scale volt is 2.5  so resistor attenuation factor is 1000/2.5=400 and 1 millivolt will become 2.5 microvolt at pga input stage and after amplification it should  500 microvolt  does this will work or not 

    as you said tricky way can you plz provide any example i want to measure ac /dc volt , current and resistor just five things what you suggest for this purpose 

  • Hello Shams,

    If the maximum ADC input is 2.5V, and you plan to use a gain of 200 V/V on the PGA117, then your maximum input to the PGA117 must be (2.5V/200) = 12.5 mV. With an input of 1000V, this is an attenuation factor of 80,000 (-98dB).

    Using this attenuation factor, 1mV of signal input would become 12.5nV at the PGA117 input, and theoretically 2.5µV at the output. However, it is not reasonable to expect the PGA117 to accurately amplify a 12.5nV signal. The offset voltage of the part alone is already in the tens of µV.

    In order to use the PGA117 for this application, the input voltage would need to be significantly higher than the VOS of the part in order to achieve any kind of accurate performance. This would need to be balanced with the resistor divider scaling and PGA gain settings.

    Best regards,

    Ian Williams

  • Dear Ian 

    i think you misunderstood my application 

    how i want to use pga117                                output to pga117

    voltage attenuator off 400             input --^^^^^^----|----^^^^^^-------- gnd

                                                                          10Mohm           25kohm

    on 1000 volt at input it will give 1000/400=2.5 volt on pga117 input so pga117 gain of 1 will give 2.5 volt 

    as input volt become 100 so same voltage divider will give 100/400= 0.25 volt which i will again amplify with pga117 gain of 10 >2.5 

    as input volt will become 10 so same voltage divider will give 10/400 = 0.025 volt which i will again amplify with pga117 gain of 100 >2.5

    and input volt at 1 will give me 1/400 =0.0025 volt which will be amplified again gain of 200 *0.0025=0.5

    i hope you  got my idea how it is used always not what you calculated

              now if you got my point of view then tell me does 2.5 microvolt will be amplified with pga117 or not and what would be the error in these two cases where i am applying 2.5 volt on pga117 vref pin and where i am connecting it to ground i had search this thing on internet but i was not able to find any reference that how does input offset effect when we are using complimentary supply or providing virtual ground to one input of op amp and what if the signal is low then input offset of op amp does this will be amplified or not 

    about pga117 these are written on first page of datasheet 

     Rail-to-Rail Input/Output
    · Offset: 25uV (typ), 100uV (max) 
    · Zerø Drift: 0.35uV/°C (typ), 1.2uV/°C (max) 
    · Low Noise: 12nV/√Hz 
    · Input Offset Current: ±5nA max (+25°C) · Test

    i am trying to amplify 2.5 microvolt to get 500 microvolt where inverting input will be tied to 2.5 volt does this work or not 

    regards

    Shams

  • Hi Shams,

    Your proposed input circuit would work for high-voltage inputs, and the calculated outputs are correct.

    However, if you apply 2.5μV to the input and +2.5V to VREF, your output will be = [Gain*(2.5μV + VOS) + 2.5V]. The VOS of the PGA117 will drown out the contribution of the 2.5μV input signal.

    In order to measure smaller input voltages, you would need to switch to a lower attenuation resistor network such that the resulting input voltage was more significant compared to VOS. This is why handheld multimeters have a rotary switch to select different input ranges. You should not expect to get good DC performance if you use the same input network for 1kV as for 1mV.

    Best regards,

    Ian Williams

  • Dear Ian 

            thank you for verification here i have one last question regarding to pga117 input impedance as data sheet states that input impedance is greater that 10 Gohm and using 10Mohm or 1Mohm  as i am trying to use in my circuit what kind of linearity and gain error is expected with pga117 how to calculate this does this will affect the performance of pga117 or it will be ok to use it in this way

    i am trying to get maximum performance which is possible using pga117 ic 

    Regards

    Shams Iqbal

  • Hi Shams,

    The (typical) gain error of 0.005% and linearity of 0.0015%FSR should not be your biggest concern. Rather, the tolerance of the resistors you use in the input attenuation circuit will have the biggest impact on your error.

    You can use TINA-TI to simulate the behavior of the input circuit and experiment with the real-world values of the resistors. The high-impedance of PGA117 RIN should not have an impact on your accuracy as long as the resistors you choose are of sufficiently low impedance.

    Best regards,

    Ian Williams

    PGA117 Vin.TSC
  • Dear Ian 

                     thank you for your support and help i appreciate your help . Before this post someone from TI  scared me that not to use this ic with ads1259b (email support) where my goal was to directly drive ads1259b input with pga117 output using RC filter to enhance capacitive loading now i am confident that this will work well and about resistors  i will use low noise and low ppm precision resistors because i know that all accuracy depends on these resistors well The (typical) gain error of 0.005% and linearity of 0.0015%FSR  does not matters this is good accuracy its even computable in digital calculation 

    pga117 with pga113 in series will give me gain of (200*200) 40000 and that  what i exactly want  dual channel  multimeter with good zooming options no matter that if i am not able to gain 1 microvolt accuracy by design but at-least i will be able to gain 10 microvolt accuracy by design 

    here i am just wondering that how to compensate input offset error (which is in microvolts )using internal calibration system of pga117 i have no idea that i would be able to compensate this error or not but i will try out this thing 

    i am searching for amplifier which will be able to work in nano volt range so i can measure 1 microvolt accuracy using ads1259b if you had worked with this level of accuracy then plz let me know 

    Regards

    Shams iqbal