RayLivingston:
The load cells are alright. Your setup is wrong. Check any commercial Weighing Scale that has used these kinds of load cells and make your setup accordingly.
RayLivingston:
Because he has the load cell connected to the amplifier incorrectly, or has the load cell mounted upside-down.
Im sorry i've found always the same scheme of connection wich is this one. https://www.instructables.com/id/Arduino-Bathroom-Scale-With-50-Kg-Load-Cells-and-H/ I really dont know in wich other ways connect them.. i've checked thousend times
E+ E- A- A+ and they are all correct and also black and white..
please let me know if there is another way to connect those cells
1. Draw a 6"x6" square box using pencil on your Table.
2. At the four corners of this box, mount your four load-cells with screws as per Fig-1.
Figure-1:
3. Make connections among these load-cells so that you get positive strain.
4. Place a transparent plate on the four load-cells; be sure that the plate touches all the load-cells.
5. An then carry out all the tricks discussed in this thread.
GolamMostafa:
1. Draw a 6"x6" square box using pencil on your Table.
6'' are inch right? is it important the distance? becouse then i will use it in a 50x50 cm (19,685 inch) square for mesure the weight of a bee hive
GolamMostafa:
3. Make connections among these load-cells so that you get positive strain.
how do i do that? i try to invert black connection and white connection right?
thanks a lot 
ilteo85:
6'' are inch right? is it important the distance? becouse then i will use it in a 50x50 cm (19,685 inch) square for mesure the weight of a bee hive
Yes, it is 6" (inch) for test purpose.
how do i do that? i try to invert black connection and white connection right?
It is in the link of your Post#20.
GolamMostafa:
It is in the link of your Post#20.
Sorry i didn't found anything..the only thing i've found is that
Your calibration factor may be very positive or very negative. It all depends on the setup of your scale system
and the direction the sensors deflect from zero state
doesn't that mean that i could use a negative calibration factor to have the right results?
i've tryed to invert black and white but had anyway a smaller numer when i pu some weight..
Do i need to invert E+ WITH E- AND A- WITH A+
if it's usefull i post here the result i get even if it's decreasing with weight..
500g
16:29:09.005 -> C31DA
16:29:09.944 -> C3178
16:29:10.925 -> C325E
16:29:11.930 -> C3282
16:29:12.953 -> C324C
16:29:13.938 -> C32BC
16:29:14.930 -> C329C
16:29:15.951 -> C3270
16:29:16.942 -> C32A4
16:29:17.960 -> C32B4
16:29:18.926 -> ===averaging process=========
16:29:18.963 -> Average Count = C325C
16:29:18.997 -> C32E8
16:29:19.955 -> C32AA
16:29:20.939 -> C32CA
16:29:21.929 -> C32E4
16:29:22.951 -> C3318
16:29:23.941 -> C327A
16:29:24.963 -> C327E
16:29:25.951 -> C32C8
16:29:26.943 -> C330C
16:29:27.942 -> C32F0
16:29:28.941 -> ===averaging process=========
16:29:28.976 -> Average Count = C32CE
1 kg
16:30:09.030 -> C3336
16:30:10.005 -> C32DC
16:30:11.000 -> C334E
16:30:11.988 -> C3338
16:30:12.998 -> C32FA
16:30:13.983 -> C335A
16:30:15.003 -> C333A
16:30:15.983 -> C32E6
16:30:17.013 -> C335E
16:30:18.001 -> C3350
16:30:19.002 -> ===averaging process=========
16:30:19.039 -> Average Count = C332C
16:30:19.039 -> C3348
16:30:19.984 -> C3364
16:30:21.004 -> C3386
16:30:21.999 -> C3398
16:30:23.004 -> C3338
16:30:24.009 -> C335C
16:30:24.990 -> C3370
16:30:26.016 -> C3340
16:30:27.000 -> C3344
16:30:28.011 -> C32BA
16:30:29.000 -> ===averaging process=========
16:30:29.035 -> Average Count = C334E
1.5 kg
16:31:19.103 -> C3204
16:31:20.058 -> C3228
16:31:21.063 -> C322E
16:31:22.059 -> C31FA
16:31:23.045 -> C325E
16:31:24.073 -> C31B8
16:31:25.056 -> C3194
16:31:26.076 -> C31B6
16:31:27.073 -> C3132
16:31:28.065 -> C3184
16:31:29.052 -> ===averaging process=========
16:31:29.089 -> Average Count = C31D7
16:31:29.123 -> C31B4
16:31:30.067 -> C31E8
16:31:31.081 -> C3262
16:31:32.063 -> C323E
16:31:33.054 -> C3234
16:31:34.080 -> C3214
16:31:35.061 -> C327A
16:31:36.073 -> C3296
16:31:37.055 -> C31E8
16:31:38.053 -> C3254
16:31:39.061 -> ===averaging process=========
16:31:39.095 -> Average Count = C322E
I ve setup everything as raccomanded it's a square of 20 cm 
here is the pictures
Thanks a lot
Also the calibration factor of the example in the link of post #20 is negative
float calibration_factor = -7050; //-7050 worked for my 440lb max scale setup
I found also this https://www.tindie.com/products/lpaseen/load-cell-bridge-for-hx711-pcb-only/ where seems to be inverted white and black couples.. but here seems even more random the value that i get.. i begin to think that is not working my load cell?
