Tips and techniques for specifying surface texture parameters


ZYGO met with Dr Mark Malburg from Digital Metrology and Carl Musolff from Musolff Consulting to discuss how to get the most out of surface parameters. Their years of frontline experience provide valuable insight into precision manufacturing processes and explain why choosing the right type of metrology and examining the correct parameters are essential to controlling the manufacturing process.

There are currently many different ways for designers to specify surface finish parameters for thetheir parts. Some specs are good, some not so good. Some are sophisticated, others quite basic, and while all are used by manufacturers today, it’s fair to say that some are less useful than others.

Alternative specifications
At the less sophisticated end, design engineers may have already manufactured, specified, or designed a component with a similar function or in a similar environment to the new product under development. So, saying, they just copy the old technical drawings, assuming that they are close enough, and thus use the same surface texture settings and boundaries.

When the environment is a little more demanding, design engineers can push the boundaries, which is a pretty haphazard approach that sometimes works and other times it doesn’t. In other cases, they may call on the capacities of external suppliers. For example, if the design is a shaft with an o-ring in it, the obvious route is to talk to o-ring specialists, as they have considerable experience regarding the surface texture parameters that will work.

For some applications, the decision may be to do modeling. It doesn’t work very well when looking at roughness – because modeling abilities aren’t able to handle roughness – but modeling is better when looking at ripple and shape. This is especially the case if you are looking to manufacture two complex components, for example gears, bearings or camshafts. It is possible to model interfaces and develop very good estimates of surface texture profiles that will work.

Testing is probably the most demanding and “high end” way to specify surface texture parameters. If a functional test can reproduce the environment and operating conditions under which the final design will go, then parts can be measured first and then measured in that test. As a result, surface texture parameters can be selected to separate functional parts from non-functional parts, or those that perform well and those that may fail early. Testing can take a lot of time and resources, but once it’s done, product designers have a set of parameters that you can be relatively comfortable with.

In the real world, most specs are carryovers. Musolff’s comments, The most common parameters for the roughness of the surface texture are Ra and Rz, and they are not very good at defining the function. But they are used, I will not say everywhere but almost everywhere. These are the parameters that people know and are used to, and therefore they are included on a drawing whether they work or not, whether they define a function or not.

Malburg adds: “This is also what I encountered. Drawings are often simply copied and pasted from a previous version, and there is no knowledge in the specification. I’ve had clients with tolerances from the 1960s and 1970s, but the processes have changed and they can’t meet those limits, and if they did, the surface wouldn’t work. The copy-paste and the copy-paste of a handwritten note become the contract! “

Are the surface settings working?

This leads to the problem at the heart of specifying surface texture parameters. How do we know that the parameters used are the right ones to use?

Musolff says, “At a very basic level, you can see that the surface texture settings are working by looking at them. If you look at a system where there is wear between one part and another, you may see bands or scratches on a surface that is going to be wavy. With roughness, you won’t see any scratches, the wavelength is not big enough, but if you see scratches, you’d better include ripple settings in your specification. I think it’s a matter of using logic. For example, if you have an anti-liquid seal, chances are that if it is leaking, it probably isn’t a roughness issue, especially if the liquid is viscous enough like an oil. If you have a leak, it takes a certain area for this fluid to pass, which tells you more about the ripple than the roughness.

Malburg comments: “When asked ‘are the parameters the right ones’, often they are not. The surface finish parameters that tend to be used are easy to measure and available. For example, if you are trying to control friction, there is no friction parameter, so you will be looking at roughness instead. If you are trying to control the appearance, there is no appearance number, so you will try to find a number that you can measure that relates to it. I think we’re getting better at the settings that better describe the features. But often we measure what is cheap, affordable and traceable. We measure the height because we can get a national standard for the height even though the height does not matter for the friction.

The role of good metrology is to work creatively when no parameter can describe the surface. The options available are quite extensive, and the key is to get additional information about the gauges being used. Design engineers need to determine what comes out of a metrology system and be able to analyze the data effectively.

Malburg comments: “You can see so much in data, but one of the most important lessons to learn is the ability to separate data from noise. Can I trust my measurement? When we look at an image of a three-dimensional surface, do we see the surface or do we see noise in the measurement? Often times there are irregularities that are not real, so an engineer has to ask, “Can I trust the data?” This comes with a little practice. There are, however, tools that are easy to adopt. For example, if you are measuring a surface from left to right, rotate the surface and measure it from right to left and see if it’s the same shape. Measure slower, measure faster, measure with different magnifications, but start trying to separate the measurement from the area.

Musolff adds, “Also ask, ‘If I see these characteristics in my measurement, does that make sense for this surface? If you think of a lapped surface, for example, the process should not generate various random high peaks. If you measure this area and get a lot of random high peaks, you are measuring something and it is not the area. You are probably measuring debris. You did not clean this surface well enough before your measurement. So use logic. Do what I see and how does this part make sense when seen together? The question of whether you can believe what your instruments are telling you raises the whole question of uncertainty. Do you have noise or other factors entering your equipment that have a significant noticeable impact on your data? There are methods to measure the uncertainty in your system, “what error can there be?” “,” How wrong can my answer be? These are often not well understood or applied well, so this is another area that people should be familiar with.

Malburg concludes, “The key is to think of your surface as a surface with a well-scaled profile graph and a nice 3D rendering that you can move around. The most powerful tool we have today is not just a static image, but the use of the ability to rotate your data and see how features relate to each other. I often find it difficult when a company contacts me and says “here is some data for our surfaces that are not working and they give me a list of numbers and not pictures”. Not datasets that I can look at, but instead a column of RA roughness values. Such figures are not descriptive. An analogy I like to use is that I could come home from a gig and tell you it was 105 decibels. But this is also the case for a fire engine. The best advice is to go to the concert, be there, and listen to it.

Or in the case of surfaces, the best advice I can give is to start with an image, look at the surface, rotate it, enlarge it, shrink it, and then start filtering the data to see what wavelengths are present, which shapes our present, and blast your way from the knowledge of the function. How does this shape interact with another shape? Not what is my number? So being aware that surfaces are not numbers, but rather surfaces are shapes, and incorporating shape information into conversations via graphics and models is a game-changer. This is the greatest benefit we can offer people in our field.


Comments are closed.