2

Accordingly to the datasheet of HX710B its differential input range is +/-20mV.

The HX710B is used in the air pressure sensor together with sensing bridge MPS20N0040D. The schematics is:

enter image description here

For reading the sensor I use one of HX711 libraries - tried all available libs with the same results. Many people say that HX710B and HX711 have similar communication protocol and the library works indeed.

My observation is:

  1. The pressure change is represented well and propotionally on the MPS20N0040D output in declared range +/-50mV.
  2. The pressure reported by HX710B is saturated (not increasing) after the input voltage difference is more than 12.5mV. This is about a half of declared sensitivity range.

The full declared range of +/-20mV is really needed for my device. To overcome this trouble I tried:

  • to change gain of the HX710B.
  • to change the supply voltage in ranges 3.3V-5V.
  • to shorten resistors on the module separately and both at the time.
  • to increase resistor values from 100 to 1k.

None of the attempts does not expand the input range to +/-20mV.

What else can I try?

I don't need the negative voltages -20-0mV because in my device the pressure is always positive. How to increase the input range even on account of the negative input voltages?

Ivy Growing
  • 335
  • 4
  • 12
  • What is the biggest value you get at saturation? Does it match the 11 mV (0.28 * full scale_23bit@5V)? Is the sign bit OK at negative input values? – Jens Jun 08 '22 at 22:26
  • In byte array of 4 bytes it is 0xFF7FFFFF. However, the MSByte is meaningless. So it is 0x7FFFFF. In negative value input I did not test because in my device the pressure is always positive. – Ivy Growing Jun 09 '22 at 05:57
  • 2
    OK, there may be a fake chip in your board, something like TM7711 or another one which have just +/- 10 mV input range. If you only need relative pressure reading in short intervals, you may add an offset to the bridge with a resistor between one output and GND to move the negative pressure values out of the ADC window. – Jens Jun 09 '22 at 16:40
  • 2
    Double bingo, Jens. 1. Likely it is fake or poor quality ADC. Despite clear marking HX710B it has range of +/-12.5mV - I measured the range with better precision and with negative pressures for the curiosity. 2. Pulling down to ground a pin 4 with resistor of 390k gives exactly -12.5mV on 0 pressure. So I get a range of 25mV total for full range of sensing. If you post such answer I would accept it in your credit. Thank you! – Ivy Growing Jun 10 '22 at 13:32

1 Answers1

2

The chip on this board is not a HX701B, it is a fake or different one.

  1. The used software interface delivers consistent values in the expected range for a 24 bit signed ADC output.
  2. The numerical range matches an input voltage range of +/- 12.5mV
  3. The correlation of input voltage and delivered values seems to be linear.

The internal gain of this chip can not be changed and has a resolution of 1.5 µV / LSB. This would match your needs if you remap the input range from +/- 12.5 mV to 0-25 mV.

If you only need relative pressure readings in short intervals, you may add a DC offset to the bridge with a resistor between IN+ (pin 4) and GND to move the negative pressure values out of the ADC window. This method has no temperature compensation but may be sufficient.

Jens
  • 5,598
  • 2
  • 7
  • 28