http://www.castbulletassoc.org/view_user.php?id=562>fa38
There is a variety of other factors that effect bullet weight consistency. Here are some factors that I have considered and worked on, but I am sure there are more:
Pot temp variation caused by dropping cut sprues into pot. A thermometer will verify this effect.
Thermal cyclic rate of pot can be modified and narrowed with a PID temperature controller.
Mold temperature is effected by casting cadence, I don't stop to check bullet quality.
Casting cadence maintained by any cooling of the mold, for example wet sponge contact, does cause wide thermal variation in mold temperature very quickly with little control. Adjusting wait time with mold closed after cutting sprue causes a much slower and less abrupt mold temperature change.
Mold temperature changes have a much greater effect on mold fill out and thus bullet weight change than pot temperature effect. Casting cadence constancy that controls mold temperature should be monitored to avoid disturbances.
A mold guide on the pot can be used to control the length of the flow and avoid eyeballing flow length that effects flow temperature and mold fill out.
After adding metal and fluxing I watch the readout of my PID till stable before resuming casting.
Actually use a clock with a second hand to monitor cadence.
These methods and controls generally produce +- 1/2 of 1% weight variation or less for me. Some things can really throw me off. An example of that is when I am casting rifle bullets and then change to another pre-warmed mold like a 1 oz rifled slug or a 12 ga round ball mold or some other mold that definitely requires a different wait time after cutting sprue until opening the mold. I have some notes about that and pot temperatures on the boxes my molds are in, but Murphey's Law won't be denied.
Gary