There are some aspects of the reloading process which tend to be significantly more time-consuming than others. Obtaining consistent, to-the-kernel precision on every round is one of those. I've tried a lot of different tools and techniques over the years to save time without sacrificing quality. Stop me if you've heard this one... RCBS Chargemaster, modified with straws and new programming, using an OHaus beam scale and manual trickler to measure the last few kernels, tweezers to add/remove single kernels, then ditching the beam scale for a lab-grade digital scale (e.g. A&D, Sartorius)... Sound familiar?

Not long ago, the AutoTrickler came along. It uses the serial port on the back of the A&D and Sartorius scales to read the weight and trickle powder until the target weight is reached. Brilliant! Really, it performs the same task as the RCBS Chargemaster with a significantly better scale.

## Challenges

The following are a few specific challenges I faced during the development of this project, which might interest you.

First, I thought that I could just turn the motor on for trickling. This created way too much vibration and the powder sprayed out of the tube chaotically rather than pouring out gently. Rather than relying on variable resistors to control the voltage, I changed the software to rapidly pulse the motor on and off at specific intervals. This technique was especially useful when I wanted to drop single kernels very slowly as the target weight neared.

Second, getting the trickler to behave correctly when the powder pan was removed or placed back on the scale was also a little tricky. The naive approach was to wait for the scale to return to zero and be stable. Oddly, as the weight of the pan is removed, the scale counts down from the current weight to the negative weight of the pan, and in doing so, it eventually passes zero and sends it to the Raspberry Pi. So, after I would remove the pan, it would trickle one or two more kernels onto the empty scale and make a mess.

Another problem with this approach was that it didn't actually fit my use-case. I still used my Chargemaster to throw most of the powder into the pan before putting it back on the scale, which meant that the scale wouldn't go back to zero when I put the pan back onto the scale. The solution to these problems was to wait for the scale to read greater than or equal to zero and to have been stable for at least one second. After implementing that fix, it worked perfectly.