ADS1261: Input Voltage Range for Strain Gauge

Hi Ismail,

You are correct that the typical output of a 10-V excited bridge would violate the input range on the ADS1261. Therefore, you would need an external amplifier / INA to level-shift the bridge common-mode, typically to mid supply (AVDD/2). For the reference, you could do something similar, or you could use a simple resistor divider as shown in the first image below. This configuration with the two resistors offers the closest approximation to a ratiometric reference. However, any mismatch in error between the amplifier stage on the input and the resistors on the reference will reduce the ratiometricity of the measurement.

You could also consider using the ADS125H02, which can accept up to +/-15V directly using a +/-18V supply. Then you would not need the signal conditioning circuitry on the ADC inputs, though you would still need the resistor divider for the reference.

Another option would be to bias the bridge with +10V and -5V, instead of +10V to GND. This bipolar, asymmetric supply puts the bridge output common-mode voltage at 2.5V, so you would not need the signal conditioning circuitry on the ADC inputs, though you would still need a resistor divider for the reference. This is shown in the second image below. This also assumes your bridge or load cell can support an excitation voltage of 15V.

Ensure that in all cases you use precision resistors for the reference divider. Any gain error or drift in these resistors directly shows up in the reference voltage, but not in the input, reducing the effectiveness of the ratiometric reference.

High voltage unipolar supply (+10V)

High voltage bipolar supply (+10V, -5V)

-Bryan

Hi Ismail,

Yes, you are correct, something like a +7.5 / -2.5V supply would work as well. That image is taken from an application note that we are going to release soon that covers this topic, and I just didn't change the values.

Take care with a +6V / -4V supply however, as the bridge common-mode voltage will be around 1V in that case. This is much closer to the input limitations on the ADS1261 PGA (and most general INAs), which prefers a voltage around AVDD/2 as I mentioned. You can use our ADS1261 calculator tool to help visualize if your settings will work: https://www.ti.com/lit/zip/sbac200.

The image below shows this tool in operation. As you can see, a 10mV signal centered around VCM = 1V still enables the maximum gain of 128. However, a larger input signal will reduce this range. Feel free to use this tool to check if your system parameters will work.

-Bryan

Hi Ismail,

Yes, as shown in the images from my first post, the reference voltage is divided down into the input range of the ADC using a resistor divider. You might consider adding a placeholder for a buffer between the resistors and the ADC VREF inputs, though I would assume the ADS1261 internal VREF buffer should be sufficient (specified at 30Mohm). The resistors should be high accuracy, low drift, as any error in the resistors directly causes a measurement gain error.

You could also use an amplifier to level-shift the signal. Again, choose high performance components to maintain high measurement accuracy.

Let me know if this answers your question.

-Bryan

Hi Ismail,

Yes, this is possible in the bipolar, asymmetric supply situation. I would recommend targeting a voltage that is slightly less than the full range of the ADC input. So, for example, 4.9V for the ADS1261 where the maximum VREF voltage can be 5V. You would then scale the resistor divider accordingly to generate these voltages based on the value of the (+) and (-) excitation voltages.

Our application report on this topic should be released in the next few weeks. I will post a link to this thread when it is live on TI.com.

-Bryan