Weight sensor and HX711

[schrodlm]

Hello.
I have been working on my graduation thesis for some time now and now I have encountered a problem.I can't make the weight sensor work.

I use Arduino UNO and the connection between the weight sensor and the HX711 looked like this:

I checked to make sure the resistance of the 
Wheatstone bridge fits and if the seller didn't confuse the colors of wires, but everything fits.

In the Arduino IDE, I used two well-known libraries for HX711.

So I'm sure there is no error in the code, as I used their official examples for calibration.

The weight sensor is only for 1 kg, so I was wondering if it is not

 possible that I didn't manage to overload it.

This is what a serial monitor looks like when using the GitHub - olkal/HX711_ADC: Arduino library for the HX711 24-bit ADC for weight scales library -> Calibration example:

This is what a serial monitor looks like when using the library https://github.com/bogde/HX711 ->
 example HX711_basic_example:

When powering the Arduino Uno, I checked the voltage, it was also on the pins of HX711 A +, A-, E +, E- so 
I don't think it's an error in HX711 either. I tried to measure the change in voltage or current
between the HX711 and the Arduino, 
but the voltage was constant even after the load was placed on the sensor. Same with the current.

Sensor looks like that:

I also checked the pins in the arduino several times. 
I am at the end of my wits so I will appreciate any help. 
Thank you so much for your answers and sorry for any typos as English is my second lanugage.

Calibration.ino (7.13 KB)

[DaveEvans]

Could be one of those HX711 boards where E- is not grounded (should be grounded). See post #23 here: HX711 with 4 equal resistors in a wheatstone bridge config - Sensors - Arduino Forum Fixing that may help but may not solve the problem.

The raw value you are getting (presumably under no load) is about 2^24/2, so that seems to indicate that the load cell output (voltage between A+ and A-) is at or beyond the HX711 limits (see the HX datasheet).

Suggest you measure the voltage between A- and A+ with a good DMM. Is it within the HX711's limits with no load on the load cell, and with full load on the load cell?

If you have the load cell sensitivity spec, you can calculate what the load cell output voltage should be under full load; do that, too, and see if it matches your measurements.

What year in school are you and what degree are you pursuing?

[schrodlm]

First of all thanks for the reply.
I am in a 4th year of highschool and I am hoping to go to a university after this. Degree I've been working on for these past four years is electrotechnical engineering.

So I was able to find out what the limits of A+ and A-. In datasheet it was stated that when 5V supply is used at the AVDD pin, these gains correspond to a full-scale differintial input volrage of +-20mV or +-40mV respectively. In some brief information about the load cell I was able to find from the seller, it was stated that output should be 1.0 ± 0.15 mV/V. Which I don't get because without a load on a loadcell voltage on A pins was about 52mV and with max load it was 56,5 V.

Also
E+=4,71v
E-=0.63v

All are relative to the GND pin.

E+ to E- = 4v
So I'm not sure if E- is grounded.

[DaveEvans]

schrodlm:
E-=0.63v
All are relative to the GND pin.
So I'm not sure if E- is grounded.

Well, since E- is not zero, it is not grounded. So, ground it.

schrodlm:
In datasheet it was stated that when 5V supply is used at the AVDD pin, these gains correspond to a full-scale differential input voltage of +-20mV or +-40mV respectively.

Please read it a bit closer. It is one of those numbers, or the other, depending on which gain you are using. Typical library code examples use 128 gain (on channel A), so your differential limit is probably +/- 20mV.

schrodlm:
Which I don't get because without a load on a loadcell voltage on A pins was about 52mV and with max load it was 56,5 mV.

So, those values are far outside your 20mV limit, which explains your constant raw value. That suggests to me that your load cell has been damaged.

However, the difference between those values, 4.5mV is consistent with the published sensitivity of 1mV/V.

It may be possible to add a shunt resistor to one leg of the load cell bridge to counteract the offset, but the results may not be reliable. If it was my project, I'd try a new load cell.

[MarkT]

A small value trimmer pot on the top or bottom of the bridge can be used to null it out.

For instance this would give +/- 1% or so of adjustment.

Since the trimmer value is so much smaller than the fixed elements of the bridge its tolerance
and tempco are not critical.