The distribution is not random and it is not a chemistry problem. It follows the electric field in the tank: how the part sits, what sits next to it, how far it is from the anodes, and what the part's own shape does to the current arriving at its surface.
It is computable. And once it is computable, it is designable — rack pitch and orientation, hanging point, shielding, auxiliary anodes, or feedback to the designer while the case is still a CAD model and changing it is free.
Anyone who plates watch cases, bracelets, bezels or clasps knows the failure modes by heart. What no one can see, before the rack comes out of the bath, is where on that rack the problem is going to appear — and by how much.
Two identical cases, same bath, same cycle, same chemistry. One hangs at the edge of the rack and one hangs in the middle. They do not come out with the same deposit. The specification is written for a part; the process runs on a rack.
To guarantee a minimum thickness on the least favoured part, you plate everything longer. On gold, rhodium and palladium layers, that margin is not a process detail — it is the cost structure of the line.
Openwork and skeleton designs, interlocking bracelet links, recesses under lugs and horns, sharp edges beside broad polished flanks. Current concentrates on edges and tips and starves anything recessed or shadowed. The more the design is worth, the worse the geometry usually is.
A cosmetic reject or a thickness non-conformity surfaces at final inspection or at the customer's incoming control — when the part already carries its full material, machining, polishing and finishing value.
Simulation is used here to qualify a new reference without burning parts, to justify a rack or tooling change with something more than experience, and to answer a client's thickness question with a distribution instead of an assurance. It is also, increasingly, a way to show a brand's engineering team that you control your process rather than run it.
Simulation is used here earlier: to give plating feedback on a case or bracelet while the geometry is still open, to standardise how a new reference is introduced onto an existing line, and to make process knowledge survive the retirement of the people who hold it.
The fastest way to see whether this is useful is a reference you already know is awkward — an openwork bracelet, a case with deep recesses under the horns, a part where you have always known the corners of the rack behave differently and have never had a reason to quantify it.
Send the geometry and how you currently hang it. We will show you the distribution and where it comes from.