[Unclenick]

Thickness matters because the temperature is not instantly uniform throughout the brass. As you say, it takes time for the core of a sample to approach its surface temperature. In steel annealing, the general rule of thumb for an oven, where the heat source is self-convecting hot gas, is to allow one hour per inch of thickness for the core to be close enough to temperature. This is called heat-soaking. The value of 70:30 brass thermal effusivity (a measure of the ability of a material to exchange heat with its surroundings) is about 1.6 times greater than 4140 steel. In addition, the thermal diffusivity of brass (a measure of how quickly heat will move within the brass once it gets there) is about 1.3 times greater. From those numbers, I would expect brass to need about half as much soak time as steel does per inch of thickness to heat-soak. That may seem like too little difference looking at the thermal conductivity alone, but what happens is the greater conductivity drops the temperature more at the interface with the hot gas or other heat sources, so a chunk of the speed advantage is self-limiting. Then you have to work out from the specific heat capacity and density of the material, how much heat has to move to make a given temperature change. That's what those numbers do.

All that is part of what makes induction heating attractive. The heat is generated internally in the metal. Exact uniformity depends on uniform field strength from the coil at different places in the metal, but, generally speaking, with the right coil, you can make the heat more uniform throughout the piece during a given exposure time. On the other hand, case neck brass is so thin, that the temperature drop to the core is less significant than it is with a thick piece of metal. Based on the rule of thumb we looked at, 25 seconds for a 0.014" neck wall in an oven, and less with the torch, which won't allow as much of a gas temperature gradient at the surface.

As we've been over before, in ascending order of the combination of time and temperature required to achieve them, the stages of annealing are:

1 Recovery
2 Recrystallization
3 Grain Growth

To stress-relieve to prevent case splitting, all you need is recovery. This is using heat to provide triggering energy that allows dislocated atoms to find their way back into the crystal lattice. But there is a complication: The more numerous and extreme (greater work hardening) the dislocations are, the less time and temperature it takes to initiate this process. For example, a 50% work-hardened piece of brass requires about ten times longer exposure to a given temperature for recovery than a 90% work-hardened piece does. Same for recrystallization (for which it is believed recovery is a required pre-cursor). So when you target a particular temperature for a particular amount of time for highly work-hardened brass, it can stress-relieve all your very hard brass completely and do practically nothing to your half-hard brass.

As a practical matter, this means that if you set up your annealing time and temperature to anneal a very hard case and use it for every load cycle, for most of those cycles, it won't do anything. When the work hardening finally builds back up enough over several load cycles for that time and temperature combination to start recovery again, the brass will go back to the stress-relieved state, and you have to start re-working-hardening it over again. So it won't be the same hardness every load cycle. It will, instead, go through a series of hardnesses during which each piece with the same load history will track each other's degree of hardness, but they won't be the same as they were on the previous cycle or will be on the subsequent cycle.

If you want the brass the same every load cycle, you have to find a time and temperature combination great enough to get recovery completed after one load cycle, then repeat every load cycle. Just be aware that any other piece of brass you anneal at that time and temperature that starts with a greater degree of work hardening will go further in that time and temperature, so its hardness won't match the other brass, either.