Surface Roughness Ra to Rz Calculator

A bearing will run cool. A hydraulic valve will seal. An aerospace part make it through its first flight.

You’d expect all these functions is driven by something more than surface finish. But that’s what defines success. Polishing a seal face too smooth remove the valleys that hold oil. Turning a shaft too rough create minute stress risers.

What Ra, Rz, Rt and RMS Mean

Four numbers tell the story: Ra, Rz, Rt and RMS. They’re four numbers that sound interchangeable, yet describe different characteristics on the same surface, the difference between a warranty claim and functional success.

The Arithmetic Average is Ra. It is simple to calculate. Tells you average depth of your valleys and hills. Doesn’t count the tall peaks or deep scratches.

Roughness Depth (Rz). Averages the 5 deepest valleys & 5 tallest peaks within sample length. Gives you a better sense of how surface feels.

Trace Roughness (Rt) measures distance from the highest peak to the lowest valley in a trace. Measures how far from the highest peak to the lowest valley there is in that trace. Use this when even one defect will ruin a seal.

Root Mean Square (RMS). Squares the deviation first then averages it. It is usually around eleven percent bigger than Ra on machined profile. None of these are incorrect. Each answer a different question regarding surface topography.

These numbers vary based off each process. Because of their sharp peaks, for example, the Rz-to-Ra ratio for a ground surface will be greater then for a turned surface. Deep valleys created in honing raise Rz, yet Ra remain moderate. Plateauing lapped surfaces once more compress the Rz-to-Ra ratio.

The material also matters. Graphite flakes present in cast iron increase surface porosity and tend to widen Ra versus Rt separation. Aluminum, however, shears cleanly and behaves more consistanty. Also, the cutoff length selected on profilometer is important. A shorter cutoff disregards waviness included by a longer cutoff, changing all derived values.

There is no single multiplier that works for every part. To know what Ra value to go after, you have to understand the process that formed the surface. What type of material is it? Who is inspecting it and how? For example, a cosmetic cover can have same Ra target as a bearing journal. Yet they behaves differently at their peaks and valleys.

This is why calculator makes those tweaks on your behalf. There is no way to know if my shaft with a 1.6 µm Ra would read more like 7 µm or 11 µm Rz. These are the predictable errors.

The engineers make a callout in an old drawing and just copy it over, not even thinking to check the process. The inspectors fail to verify cutoff length in the spec. It doesn’t match what the supplier has. The purchasing person sends them a print reading: Ra 0.8 maximum, with no mention of peak height or lay direction. Surprise! Lapped parts don’t pass a leak test.

Failures of math? No. Failures to speak same language in the supply chain?

Real surfaces aren’t neat; they have profile shapes, tolerance bands, and cutoffs. Polishing can produce rounded peaks while turning can create sharper feed marks that behaves differently under load. Parts concerned with fatigue will be more sensitive to highest stress riser rather than average roughness. Use function over one magic number. It makes it all less mysterious and more helpful.

Roughing up the surface isn’t an exercise in reaching a certain number. Roughing up the surface is a matter of picking the proper texture for the task at hand. Make sure both metrologist and machinist are measuring the same thing. You should of gotten this part correct, then the calculator will be a bridge between the drawing board and the shop floor.