Density Gauge Calibration

y=mx+b

b is offset or in the case of calibration - zero

m is multiplier or in the case of calibration - scale/span

If you jump 1 meter up from a 20 meter diving board, or you jump 1 meter up from a 30 meter board, you have still only moved 1 meter vertically in the "up" direction. The 1 meter that you jump "up" is the span, the 20-30 meters that you start from is the "zero" (starting point).

If you don't understand any of this, please go back to school.

The density of water is not zero.

Look at the temperature corrections of the unit and the temperature of the water. There may be an issue with the temperature of the water used to set the hydrometer at 1.000 sp gr. Depends on the type of hydrometer.

Do you have two problems here? Or would putting "changes" in place of "drift away" describe your problem. "Drift" is when you keep all the inputs fixed, but the output changes slowly. In many instruments, span and zero inter-act, you just have to keep adjusting them closer to wanted value....span - zero - span....... while changing input from minimum to maximum, until output minimum and maximum match what you want. There are other words to describe "span", like "slope" or "gain". And other words describe "zero", like "offset" or "bias". If you are talking about an "electronic" gauge, I can show why span and zero inter-act AND how a better (but more complicated/expensive) amplifier makes "offset/zero" and "gain/span" independent.