I start by sanding down almost an entire stick with some fine 220 grit sandpaper, catching all of the graphite dust in a wide, flat-bottomed plastic container.
The next step is to add two or three tablespoons of 70% isopropyl alcohol. The amount is not critical, you just want enough to make a nice brush-able liquid without clumps. You're after a paint, not a paste.
Mix well with a soft brush, then test the solution on some scrap paper. It should go on a transparent medium gray - not black. If it's too thick, add more alcohol.
Sealed in its plastic container, one batch of graphite solution can last many, many months. If it dries out, simply add more alcohol to reactivate.
Final ThoughtsI'm very pleased with my graphite solution and enjoy how well my short-wheelbase locomotives operate on the layout after applying it. I find it to be particularly useful on turnouts where just a dab between stock rails and points can greatly improve their electrical performance. However, when I first posted this idea on one of the model railroad forums I frequent, I was immediately taken to task for "over-applying" graphite. Dire consequences would ensue, I was warned, if I continued down this dangerous path. There would be short circuits, arcing, loss of traction, and power jumps across gaps with this blasphemous liquid. I was even told that only invisible, microscopic amounts of graphite should ever be placed on the rails, as if it had some magical homeopathic quality that would be destroyed if it was ever used in detectable amounts. I'd like to see the data on that. I have never experienced any of the dire consequences warned about and have only enjoyed happy railroading. It is true that less is indeed more when it comes to graphite, but you do actually need to get some on the rails for it to do any good. What you don't want is a pasty build-up and that is very easy to avoid.
Anyway, I hope that some others will give this solution a try. It works well for me. Thanks for checking in, amigos. Adios for now!