I'm sure I'm not going to change your mind... but you trickle with a Chargemaster and then trickle with the Dandy. I find it very hard to believe that's faster than dumping a dipped precharge into the pan as the Autotrickler is going. The Autotrickler has finished before I have even returned the charging dipper to the powder pot for replenishment. (The Autothrower add-on can automate the pre-charging process.) Even if the need ever occurred, the fastest way to add/subtract a single kernel is with a set of tweezers.
The OP might be interested in the below post from bsumoba who did a side-by-side comparison of the accuracy of a Satorius and the considerably cheaper but eminently capable AD FX 120i. (There is, of course, no reason the Autotrickler can't be used with the more expensive scale.) In short, there is a point where the resolution of the media being measured is the limiting factor rather than the scale's ability to resolve. To be more accurate (ALWAYS just to add back the shouting) you've basically got to cut kernels or pick and choose big from small (to the extent that's even possible). So it is my view that even if one believes that a powder kernel or fractions thereof consistently resolve themselves on the target you're not really going to get a more accurate powder weight (and certainly not ALWAYS) unless you are prepared to deal in fractions of a powder kernel and that's about as anal retentive as I could possibly think of.
The OP can decide how far out on this limb he wants to go. Spend your money where you think it will provide effect. If money is no object then why not buy the more expensive scale and kid yourself you can revolve powder weights to 2 thou gr. The Autotrickler can still deliver +/- a kernel, and fast. A good starting point would be then to take 100 kernels of your favourite powder and measure them on the Sartorious, noting their average weight and SD. You can then come to a conclusion as to what the limiting factor is - a kernel or the scale - and buy a really fine scalpel.
I messed around with the FX 120i last night...My first impressions were that it was bigger than I had expected. Coming from a GemPro 250, I expected something slightly larger than this, but it was almost near the same size as the one I have in my company's stockroom, which is a Satorius and it is a $2K+ industrial/research model. I leveled the unit on my table, made sure there were no drafts even though there are shields provided with the unit, and I left the unit on for an hour, which is 1/2 hour longer than what the manual says to let it warm up.
I borrowed the stockroom Satorius (do not remember the model) that measures 0.002 gr (0.0001g) accuracy and is also calibrated every year. This is used for counting itty bitty parts which I borrowed from work to do this test.
I took all my weights that I have gotten over the years from various scales and measured them on the Satorius. Not to my surprise, all of them weighed in at exactly 20.0000 grams and 50.0000 grams. Who needs a Class-1 weight
I then took (2) of the 50 gram weights and used these to calibrate the FX120i. Finished that and then took a 20gr cal weight and put it on the FX120i. It measured at exactly 20.000 grams. Took an individual 50 gram weight and it measured exactly as well. That was refreshing! ;D
Just for good measure, I took the weights and put them on the Satorius, same thing.
Okay, now I switched to grains on both units, the Satorius and FX120i and used a chargemaster to measure out 56.8 grains of H4831SC. Put this on the Satorius and got 56.826 grains. I removed a kernel to get it to try to get it to exactly 56.800 grains. I could not get it to this number and ended up with 56.804 grains. I suspected if I searched for a "larger" kernel I could get it to exactly 56.800 grains.
Now, I put the 56.804 grain charge onto the FX120i and the unit measured exactly 56.80 grains.
Just to see how the unit would measure a charge, say 56.808 grains on the Satorius, I managed to remove a kernel, added a larger one and finally got it to 56.808 grains. I thought to myself, what a pain! If I was anal enough to want exactly 56.800 grains, that would take a lot of effort!
I took the 56.808 grain charge and put it on the FX120i and I got 56.80 grains. Okay, that is what I expected. I took off the charge, waited a few seconds, and put it back on the scale and it measure 56.80 grains. So it looks like it is stable.
I then changed the charge to 56.812 grains and put it on the FX and got 56.80 grains.
And lastly, I changed the weight to 56.818 grains and sure enough, it measured 56.82 grains.
Keep in mind, for me to get these weights on the Satorius, I had to remove what looked like slightly smaller kernels and add slightly larger ones. If someone is really doing this in real life, I can only imagine that it takes them FOREVER to get a precise 56.800 grain charge.
So what did I get out of this experiment....the preciseness of 0.02 grains is enough for me. There are just too many other variables that affect ES/SD in ammunition like brass, primer and bullet consistency, neck tension, etc. that to worry about kernel sizes to get a precise dead nuts reading of 56.800 grains on a $2K plus unit is probably overkill. The 0.02 grain was enough precision to get all my charges to within 1 kernel, maybe 2 small kernels max from each other.
