A topic near and dear to my heart, I am often tasked with answering the question; “What went wrong?” As we all know, there is no one answer and as our buddy Ray Drago has oft been heard to say, “it depends.” Seriously, it is paramount to know what you are working with prior to placing a gear system into service. Especially if the gear design is new to the application, whether a completely new design, a redesign, or an upgrade to an existing design. When you place a new design into service for the first time, you are verifying the results of your development work and determining whatever adjustments it may need. To be able to effectively determine the correct changes, if any are needed, requires the results of the first run to be well documented.
Without knowing the precise geometry of the gear under test, there is no way to work backward from effect to cause. The interaction between any two gears is a function of the contact geometry and the lubricant layer managing the energy transfer between them. As we also know, all our analysis techniques are based on constant tooth face geometry, equivalent load distribution and a fully formed elastohydrodynamic shear layer. Deviations in any one of those parameters significantly alters things such as service life, efficiency, NVH response — the list goes on. Not to say that small deviations in geometry, etc., are not part of the final production implemented physical product, however they must be quantified and understood. If you think about it, this is how we determine the appropriate quality designation. Certainly, it is common to start product development with higher than hoped for quality requirements as these add cost to every component. As the robustness of the full product becomes more completely understood, the cost of quality may be able to be reduced, gears included.
So, what do we do? My recommendation is that before any newly designed (or redesigned) gears are placed on test or into service, they are thoroughly vetted via a full GMM (Gear Measurement Machine) analysis. The gears will be measured, inspected, and the results analyzed. As we know, the pitch point (pitch circle, pitch diameter) are all defined as the instantaneous point of contact between two mating teeth where there is no sliding motion. This “diameter” can be different for every tooth combination. Which in turn makes this “pitch point” a function of the two teeth in mesh at any given moment. The concept of increasing quality (higher AGMA 2000 Q numbers, or ISO A numbers) is to reduce transmission error (TE) and improve gear tooth conjugacy. Measurement, on the other hand, is the quantity at which any object can be compared to a known value. Finally, inspection is the process by which one checks the measurement against the requirement to determine whether it is fit for function, or whatever the criteria that has been established and is expected from the part.
The first level result of our inspection will be to verify suitability of application for whatever our first piece test or use will be. The second level use of the inspection results (GMM or other) will be to be a baseline to compare test results with requirements; generally, service life, NVH, etc. The basic composite checks, like profile evaluation and/or lead evaluation, etc., are a good start and can be used to tell us a lot about the gear tooth. Remember, there are four attributes that we evaluate in terms of a gear and the pair. Relative to the gear as an individual piece we evaluate the shape; usually the involute profile, the size (height, width, etc.), and the spacing (pitch, tooth-to-tooth runout, etc.). The fourth attribute is of the pair, generally mesh interaction at a defined center distance, that sort of thing. The trend in the industry is to move away from quantifications that involve the pair in mesh (backlash, runout, etc.) and define the gear data chart specific to only the gear it references. We see this in the move away from parameters such as SAP and EAP and toward SOI and TIF, for example. To that end, we now recommend that attributes such as the aforementioned backlash and runout, etc., are better placed on the assembly drawing of the gearbox.
The measurement categories of elemental and composite address and evaluate all tooth surfaces, whether in tight mesh with a master gear or in comparison to the ideal gear (generally the mathematical representation). When we look at either the basic or complex attributes, we are looking at individual attributes of the gear tooth. The difference between basic and complex is the measurement tool used. Basic inspection elements use either measurement tools and/or gauges typically developed to assess one attribute of all tooth parameters at a time, whereas complex inspection elements use measurement systems that perform multiple measures in one setup and are usually conducted via computer-controlled gauge techniques. These usually are defined as GMM systems and are very precise, accurate, and repeatable.
Okay, so we have the basis of the measurement techniques and a very brief overview of the tools at are at our disposal. Now what? As this article started, we will use some or all of these various tools to quantify the various attributes of importance prior to placing the gear into service. We strive to know exactly what we are dealing with when we conduct a test or runoff of a new design or gear development. The plan entails measuring all aspects of the gear tooth form, spacing, and pair development. This information can be used to either determine what went wrong during the manufacture of the gear or as a means to explain the results obtained from testing.
We use GMM results to also record history such that we can build a history of all our designs, as well as develop family representations of gear designs to be used during new concept development. Certainly, with enough tooth surface representation data, one can build a model of each tooth and then analytically “mesh” it with its mate to better understand all sorts of tooth-to-tooth interactions. All of this and more can be done as more time is committed to the analysis. It can also be very overwhelming.
Now fortunately for all of us, AGMA offers two highly valued courses on gear measurement, inspection, and result interpretation. I encourage you to look them up, sign up, and I will see you there.
