• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 79 subscribers
  • Views 1490 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.

MSP430I2040 4 channel differential ADC for high resolution measurement

Sudheer Beligere
Intellectual 270 points


Dear Experts,

Planning to use this part for interfacing 4 sets of wheatstone bridge based sensors (1.5mV/V sensitivity), will use external reference ratiomteric to sensor excitation also a two opamp based diff amplifier. I was not able to derive the ENOB for this? suppose if I am working on the lowest sampling rate possible (speed of sensing is not critical), how much effective resolution we could get? has someone used this part for interfacing differential output sensor? I am expecting > 14 bit resolution. 

Thanks and regards,

Sudheer

  • 0
    TI__Guru**** 219997 points
    Hello Sudheer,

    You will want to calculate the ENOB values from the i2040's SD24 SINAD values which are listed in Table 5-22 in the i2040 datasheet. ENOB can be calculated per the equation ENOB = (SINAD - 1.76dB) / 6.02dB

    Where 6.02 is just a conversion factor and 1.76 is the quantization error of an ideal ADC. Therefore if SINAD = 89dB then ENOB = (89dB - 1.76dB) / 6.02dB = 14.49 bits. These values aren't listed in the datasheet itself but it is a very simple equation to exercise.

    Regards,
    Ryan
  • 0 in reply to Ryan Brown1
    Intellectual 270 points

    Hello Ryan, thanks

    is it possible to reduce the sampling speed and improve the resolution? can we get this resolution with all the four channels connected together? what filters do you propose if I need to achieve 16-17 bit of resolution or more.

    Also, if the sensor bridge is DC powered, what resolution can we achieve?

    Thanks and regards,

    Sudheer

  • 0 in reply to Sudheer Beligere
    TI__Guru**** 219997 points
    Hi Sudheer,

    Sampling speed or number of channels does not affect resolution, the largest SINAD results in the best ENOB which was given in the previous example. For DC measurements, additional calculations are needed since it is not characterized. Here ENOB = N - log2(stddev) where N is the number of converter bits (24) and stddev is the standard deviation of a set of data.

    You can set up simple code to save a number of samples in RAM and perform this calculation yourself if you would like. In some internal tests the ENOB has been found to perform better than the AC parameter, closer to ~16.5bits.

    You can also try referencing the oversampling app report (SLAA694): www.ti.com.cn/.../slaa694.pdf

    Regards,
    Ryan
  • 0 in reply to Ryan Brown1
    Intellectual 270 points

    Hello Ryan,

    can I use PGA (16 gain) on all 4 differential sensors? the datasheet mentions yes but is there any limitation observed with this configuration?

    will there be impact of crosstalk on the resolution?

    thanks and regards,

    Sudheer

  • 0 in reply to Sudheer Beligere
    TI__Guru**** 219997 points
    Hello Sudheer,

    There are no limitations that I know of with using the same SD24GAINx setting on all input channels. Additional crosstalk noise will be a result of the board layout and not the MSP430 die. The signal-to-noise + distaortion ration (SINAD) captured in the datasheet details the noise performance of the MSP430..

    Regards,
    Ryan

**Attention** This is a public forum