Hehe tnx gardner for the info :). But I have not used an A/D converter. That is still to be used....
I am using a load cell; giving output voltage for the weigh on it.
The instruments used is an oscilloscope (Voltage peak+) and a very good power supply. Even the weights 1kg are exactly 1kg since I made a spent some time putting exactly 1kg of sand.

And yes i think that the last point is bogus...
Ill hook up the equipment again and take other readings but I think it is quite a good almost linear graph.

**Another question is this, when you wrote :**
*** grams = 17.299 volts^2 + 4068.3 volts - 488.45***
**How did you work that out??**

With excel I managed to get this summary:

SUMMARY OUTPUT

Regression Statistics
Multiple R 0.999765814
R Square 0.999531683
Adjusted R Square 0.999505665
Standard Error 0.031847119
Observations 20

ANOVA
df SS MS F Significance F
Regression 1 38.9645237 38.9645237 38417.49623 2.01014E-31
Residual 18 0.018256302 0.001014239
Total 19 38.98278

Coefficients Standard Error t Stat P-value Lower 95% Upper 95% Lower 95.0% Upper 95.0%
Intercept 0.122532035 0.01375446 8.908531055 5.12634E-08 0.093634987 0.151429083 0.093634987 0.151429083
X Variable 1 0.000241939 1.23436E-06 196.0038169 2.01014E-31 0.000239346 0.000244533 0.000239346 0.000244533

Meaning that it is quite a good almost linear graph, but still I need to be accurate in this since I am refilling a cars a/c tank.

tnx again