Busdriver said:
The FX120i should lock onto zero even from cold, because it uses zero-tracking to correct drift as it occurs (obviously it needs to sit at/near a stable zero reading from time to time for this to work; some charge-weighing workflows can prevent zero-tracking from working because the balance never 'sees' zero). Like you, I've found that a little drift can occur when the balance is left for a period of hours with a non-zero load on it. Assuming a draught-free, vibration-free and temperature-stable measuring environment this 'off-zero' drift seems to be much worse when the balance hasn't has sufficient warm-up time. I usually try to get mine switched on at least 24 hours before use, but as long as the weighing cycle allows zero-tracking to work then any drift should be corrected before it becomes significant. I've checked charges from my FX120i-based dispenser using a 4DP (i.e. 0.1 milligram) Denver, and the results were all within the expected range, i.e. within +/- half the readability increment of the AND.
I think the true worth of the AND balance is the very fast settling time, responsiveness to very small changes (e.g. when trickling the last few kernels), as much as the fundamental repeatability, resolution & linearity.
..
Particularly for LR shooting. I'm not shooting for group though, so it's not as important in mid-range, as Busdriver points out, but if you can, why not do it? It doesn't cost me any more time in loading.
