This is my current setup, mounted on a rigid metal plate. The problem is that the measurements vary a lot depending on where I put the mass on the plate ( middle, corners..), up to 50% between different positions. I flipped the scale upside down and was placing a known mass on each cell one by one, and noted that they don't read the same raw value (again up to 20-30% difference between cells)
I'm thinking that the solution is to get 4 more hx711s and some resistors, since each cell is a half-bridge, create 4 wheatstone bridges and then in the arduino code calibrate each one seperately.
Like this setup, 4 times:
Does this sound plausible? Is there an easier way to fix it, and can other users tell me if it's normal for these sensors to have such different sensitivites? Or maybe I screwed something up in the wiring or I have faulty sensors?
Are the four loadcells equidistant from the center of the metal plate?
What is below the loadcells, how are they mounted?
What are the loadcell max weight?
What weight are you trying to measure?
Yes, they're equidistant. They are rated 50kg max each, and I'm trying to weigh masses in the range of 10-50kg. The plate is pretty rigid and only very slightly bends when something heavy is placed on it, so I really don't know what could mechanically be wrong with the setup. The aditional wires are for the lcd on the other side.
What’s that black cap on the (hidden in this pic) point-load side of the cell? Does it distribute the load on other bits of the cell, or is the load concentrated properly as a point load on the nib in the center? If there is force on other portions of the cell, the center bar where the strain gauges are mounted won’t strain as designed.
