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.

Question about ADS1243

Other Parts Discussed in Thread: ADS1243, ADS1216, ADS1241, ADS1158, ADS1258

We are considering the ADS1243 part for a new design.  Our input signal varies from 0 to 5V.  I am confused about the specs regarding Vref and the analog input range.

If we set REF IN+ to 5V and REF IN- to ground and Vdd = 5V, what is the analog input range?  The datasheet says the full-scale input range would equal +-Vref/PGA for RANGE=0 and +-Vref/(2*PGA).  If we have RANGE=0 and PGA=1, the full scale input range would be +-2.5V.  What exactly does this mean given our Vdd and Vref etc.?

Thanks

  • Hi John,

    Welcome to the forum! The maximum measurement is +/- 2.5V in either range 0, or range 1.  The difference between the ranges is that a 5V reference can be used for range 1.  One approach for measuring your sensor is to use a 2.5 V reference in range 0, and then set AIN- to the reference voltage as well.  AIN+ input is connected to your sensor output.  This will give you the full dynamic range with any sensor output below 2.5V will be negative relative to the the AIN- input, and positive relative to the AIN- when greater than 2.5V.  To get the exact value you will need to adjust the output by recalculating the value using your microcontroller.

    Best regards,

    Bob B

  • Thanks for your quick response.  I am still a bit fuzzy about this.

    From what I gather, if I set REF IN+ to 2.5V and REF IN- to GND and I connect AIN- to the 2.5V ref, then I can input my 0 to 5V sensor output (relative to GND) by connecting the sensor output to AIN+?  In this case a sensor output of 0V would be -2.5V, a sensor ouput of 2.5V would be 0V, and a sensor output of 5V would be +2.5V?

    In our case, the sensor output is ratiometric to the excitation and the sensor excitation must be 5V.  So I would prefer to use 5V for the reference if possible.  If I use the 5V for the reference, how would this work?  Does the 5V reference need to be buffered?

    One more option, it would be a great advantage to power the ADS1243 with Vdd = 3.3V.  We are planning to interface with a low power MSP430 (we are using this on two other products currently) so we need to communicate SPI at 3.3V.  When we use Vdd=5V for the ADS1243, we need to add a level converter between the AD and uP (it would be nice to eliminate this part).  It might be possible to split the excitation voltage of 5V to 2.5V (resistor divider) to use for the reference.  Would we still be able to accomodate a 0 to 5V sensor input in this case? 

    Thanks

  • Hi John,

    Yes, in the case I described 0V from the sensor is considered -2.5V by the ADC, and a 5V sensor output is shown as +2.5V. 

    You could use the 5V reference in range 1, and it does not need to be buffered.  It would be a good idea to place a cap near the reference input pin of about 0.01 to 0.1uF.  You will need to come up with a method for deriving 2.5V at the AIN- input.  To maintain a ratiometric measurement you could use a voltage divider with precision resistors.  Keep in mind that with the buffer off (which you will need to do to measure from 0V to 5V) the input impedance is lower for the ADC.  This means that the voltage divider needs to be low resistance or the input will affect the voltage measurement.

    As far as the supply voltage, you have now hit the limitation of the device in that the ADS1243 doesn't have seperate supplies for the analog and digital.  Running at 3.3V is not an option if you want to measure input signals from 0 to 5V.  For this situation you will need to use some type of level shifting, but it may only be necessary for the digital output pins of the ADS1243.

    Best regards,

    Bob B

  • Thanks again for a clear response.  My desire was to use the AD1243 to read 8 single-ended inputs with a 0 to 5V input range.  It looks like this is not possible with this part.  Is this correct?

    If true, can you suggest other AD parts that fit with my needs?  I can go with 16-bit if need be.

    Thanks

  • Hi John,

    The ADS1243 requires a differential measurement, so one one of the inputs is required to be a common for the others which limits the actual measurable inputs to 7.  The eighth input is the common.

    The ADS1216 and the ADS1241 have inputs and a common input so that you can measure 8 single ended inputs.  The reference is still limited to 2.5V, but there is a separate digital supply that can operate in the digital range that you desire without any level shifting.

    There is also the ADS1258 (and a 16 bit version the ADS1158).  This part will pretty much do everything you want to do, but is a little more complicated as to operation.  The other issue is the input impedance is pretty low, so you will most likely need a buffer.  The one advantage is you would only need one buffer as there is a mux out/ADC in set of pins.  This part will accept the higher reference voltage and has separate digital supply as well.

    Best regards,

    Bob B