I have just started working on MSP432P401R evaluation board. I am curious to know is it possible to measure bipolar analog voltage measurement using internal 14bit ADC & reference?
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.
>For MSP432P401R, as well as most other micro-controllers, all analog signals must be within the power-rails. That is, between AVcc and AVss.
So you need virtual ground for input signal
I agree with you & appreciate your reply. But I wish to know as in case with single supply OPAMP/ADC for bipolar + 2.5V measurement can internal 2.5V reference be used as biasing so that input doesn't exceed rating of the device?
>can internal 2.5V reference be used as biasing so that input doesn't exceed rating of the device?
Never. Reference is designed to only source very small currents. You shall read datasheet, reference section.
Device (inputs) shall be protected using (tsv) diodes
Thanks for the reply. I guess better to describe my requirement in detail.
In my case the input signal is from 4 channels +10V. I was planning to use simple resistor divider for attenuating the signal to +1V or +2.5V followed by buffer. But since -1V or -2.5V level is exceeding the input limit this must be centered on the mid scale of power supply. In this regard I wish to know, what can be simplest method to measure +10V signals using minimum hardware on MSP432P401R ADC?
>May I know did anybody of you measured +10V signal using MSP432P401R ADC? If yes can you share your experience with it ?
Your idea is right - bring voltages down to ADC range using resitive divider. If your voltage source is guaranteed to not exceed specified voltage, then you are more or less done. Otherwise protect ADC input using clamp diodes. If you measure slow signals then you would want to add capacitor between ADC input and uC analog ground which together with R-divider will act as low pass filter. Refer to some ADC tutorial, chapter "antialiasing filter".
Regarding virtual ground - again use resistive 1:1 divider of _reference_ voltage, amplify it using buffer amplifier - because reference usually is weak current source. If by any means you want to avoid using buffer opAmp - then only solution would be to use VCC as ADC reference, placing virtual ground in a middle of VCC with resistive divider and some capacitors.
bnilesh said:did anybody of you measured +10V signal using MSP432P401R ADC?
I would start with a suitable opamp to buffer the ADC Vref voltage, say 2.5V.
I would take four 2kohm resistors and create two voltage dividers from the buffered reference to ground.
Then connect the differential ADC inputs to this divider ( +1.25V each input)
Then connect a 10Kohm resistor to the positive ADC input.
The other end of the 10 kohm resistor can measure +/- 10 VDC.
Note that the 0VDC voltage reading will be offset by 1.25V + any resistor errors.
Your SH sampling speed may need to slow down for the 1K source impedance.
Peter
Or try TI INA159 High-Speed, Precision Gain of 0.2 Level Translation Difference Amplifier
The gain of 1/5 makes the INA159 useful to couple ±10V signals to single-supply analog-to-digital converters (ADCs), particularly those operating on a single +5V supply
Thanks a lot Peter for so detailed input & introducing INA159. I will keep that in my records.
Let me correct if required, in case INA is operated with 3.3V supply no protection diode will be needed whereas in case of potential divider as mentioned is used schottky diode might be preferable for protection.
If you permit I will like to know more about the relation between :
1. source resistance relation with the sampling speed of ADC
2. Effect of low impedance, like in mentioned example it will be around 1Kohm (2Kohm || 2Kohm) on ADC measurement.
Peter Dvorak said:Your best sources are the MSP432 spec sheet www.ti.com/.../slas826a.pdf
This ap note has good info :www.st.com/.../CD00004444.pdf
AN1636 UNDERSTANDING AND MINIMISING ADC CONVERSION ERRORS
peter
Thanks a lot for so informative application note!
**Attention** This is a public forum