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.

ADS1262: Powering ADS1262 for 10v Load-Cell

Part Number: ADS1262
Other Parts Discussed in Thread: LM27762

I've been using the ADS1262 for a very early stage prototype (I have that mostly working).  I haven't been overly concerned with calibration and power, as I was previously just trying to validate a general concept.

Now I need to narrow down how best to power the loadcell. Loadcells are typically spec'd for 0-10v, but the ADS is +/-2.5v or 0-5v. Can you point me to any resources for how best to handle this? Also, its a battery powered application, which will also power the microcontroller.

For background, my current iteration uses a 7.2v battery through a LM78M05 to provide 5v to the AVDD (pin 6) of the ads1262. That's mostly working, but I'm sure far less than optimal for the loadcell.

  • Hi spanky,

    If you were to excite the load cell with 5V, then the circuit becomes fairly straight-forward to implement...The 5V analog supply (it doesn't matter if its +/- 2.5 or 0-5V, as you mentioned) can be shared and used to power the analog supply, excite the loadcell and also be used as the reference voltage to the ADC. Except for perhap some RC filters on the reference inputs, you can directly connect the 5V reference voltage to the ADC. 

    However, if you decide to excite the loadcell with a voltage greater than 5V, you'll need to divide down the reference voltage to 5V or less before you can connect it to the ADC. The advantage to using a larger supply voltage is that you get a larger output signal from the bridge (i.e. higher sensitivity); however, the disadvantages are:

    • Adding a voltage divider between the excitation voltage and the ADC reference inputs tends to make the measurement a bit less ratiometric. Resistor noise, resistor tolerances, loading, etc. all affect the reference voltage in such a way that the reference and excitation voltages are not exactly proportional anymore.

    • Large excitation voltages increase the power dissipation in the bridge (and also through the voltage divisor), increasing overall system power consumption and possibly causing self-heating effects within the bridge (resulting in additional offset drift, for example).

    Generally unless you need the increased sensitivity, I find the 5V excitation voltage case to be the simplest to implement, and for a battery powered application you'd likely benefit from longer battery life as well.

  • I have a pretty wide load range to measure, but I need to resolve as much detail as possible.My favorite analogy for this project is trying to weigh a chicken on a truck scale.

    It sounds like I'm stuck on the 5v scale either way, so I wouldn't be getting exactly twice the resolution by doubling the voltage (after dropping that back down to a compatible range for the ADS 1262, and associated losses).

    As far as the battery goes, I am open to higher voltages to power the loadcell. I am already using a 7.2v pack to get the 5v input for the AVDD

    Now I'm wondering if I could use a single  3.6v battery to power it on the +/-2.5v range. My loadcell reads push/pull, but I only care about tensile loads (which is why I gravitated towards the 0-5v range). Forgive me for my ignorance (and thank you for your patience), as I am not actually an electrical engineer.

    Can I zero the scale at -2.5v and read the full 5v range on the tensile side?

  • Hi spanky,

    spanky said:

    Now I'm wondering if I could use a single  3.6v battery to power it on the +/-2.5v range. My loadcell reads push/pull, but I only care about tensile loads (which is why I gravitated towards the 0-5v range). Forgive me for my ignorance (and thank you for your patience), as I am not actually an electrical engineer.

    The analog supply for the ADS1262 needs to be 5V. This can be accomplished as a unipolar supply AVDD = 5V, AVSS=0V; or a bipolar supply, AVDD = +2.5V, AVSS = -2.5V.

    If you have to generate these supplies off of a 3.7V battery, then you'd either need to boost the 3.7V to 5V (possibly boosting it higher like 5.3V, and then using an LDO to regulate back down to 5V) OR you'd have to use an inverting charge pump to generate a negative supply rail (such as -3V, and then using LDOs gain to generate the +/- 2.5V rails as well).

    The LM27762 might be a good option for creating a bipolar +/- 2.5V supply rail if you decide to go that route.

     

    spanky said:

    Can I zero the scale at -2.5v and read the full 5v range on the tensile side?

    I'm not exactly sure what you mean here... Are you asking if you can set the common-mode output voltage of the bridge to -2.5V?

    You'll want to zero/set the output common-mode voltage of the bridge to a mid-supply voltage. If you use a 5V unipolar supply, then you'll want the output of the bridge to be 0V differential (when zeroed) with a 2.5V common-mode. As you apply a load to the loadcell, one of the legs of the bridge will increase in voltage while the other leg decreases. By keeping the common-mode voltage centered at a mid-supply voltage you will avoid saturating the PGA outputs into either supply rail.

  • I think I get it now. Either way it needs a 5v span. The ads1262 had an internal source, so I thought it might provide +/-2.5v (I didn't get that far in the datasheet... and probably explains why I was having issues when trying to use that)

    As far as my other question, that's probably not the route I'll take, since 0-5v makes more sense to me.

    Thanks again.