- Training & Events
- Buyer's Guide
Limitations in atomic emission spectroscopy can hinder the ability to monitor large wear particles in gearboxes. This article will explore additional test methods used to monitor abnormal wear of gearboxes. These test methods include direct reading ferrography, the particle quantifier, particle counting and analytical ferrography.
Before these methods can be understood, focus should be placed on the various types of gearboxes while comparing their similarities and differences.
A gearbox is defined as a metal casing that contains a train of gears. They are available in a range of sizes, capacities and speed ratios. Their job is to convert the input provided by a prime mover into an output of lower RPM and, correspondingly, higher torque.
There are multiple gearset configurations that can be utilized depending on application. Some common configurations include spur, helical, bevel, hypoid and worm gears.
Figure 1. Trending Various Wear Modes
The inherent operating conditions of a gearbox can result in increased levels of wear generation. While many gearbox oil samples have a normal iron value of five to 15 ppm, it is not uncommon for a gearbox to have a normal iron value of more than 100 ppm.
This is why it is important to understand the available options for monitoring the large catastrophic wear particles that can be generated during preeminent failure conditions. Figure 1 shows an example of how various wear modes trend as they relate to wear generation.
Monitoring large particles in industrial gearboxes is a two-part process involving detection and analysis. The tools and instruments available in detecting large particles will be discussed first. A brief description of each will then be presented because some of these technologies may be complex enough to warrant a separate session altogether.
Figure 3. The Hall Effect
Atomic emission spectroscopy (AES) is the cornerstone of laboratory testing. Nearly all test slates used to monitor machinery condition will utilize some type of AES testing, whether it is inductively coupled plasma (ICP), atomic absorption spectroscopy (AAS) or rotating disc electrode (RDE). All of these methods are limited to the size of particle that each can detect, with a 10-micron (µm) particle being the uppermost limit.
AES works by superheating the oil sample by either plasma (ICP) or via spark (RDE), which results in the production of light at a given wavelength for each element present. The optics of the instrument (polychromator) converts the light energy into an electric signal. This allows the instrument to make the appropriate readings.
Unfortunately, one drawback that occurs is as particle sizes get larger, the ability to be fully vaporized is reduced until particles larger than approximately 10 µm are not counted at all. If other methods of detection are not employed, a potential problem can easily be missed. Figure 2 shows the diminishing accuracy that is seen with using AES as it relates to large particle detection.
A promising technique in large wear identification is rotrode filter spectroscopy (RFS). With RFS, a porous carbon disc collects large wear debris particles for later presentation to the spark emission spectrometer (Figure 3). Variable porosity can influence the trendability of the coarse, or large, data. A more detailed explanation of RFS can be found in the September 2006 issue of Practicing Oil Analysis magazine.1
Figure 2. AES Accuracy
Direct reading (DR) ferrography measures the amount of ferrous wear debris in an oil sample. The results of DR ferrography are generally given in terms of DL for particles greater than 5 µm and DS for particles less than 5 µm in size.
DR ferrography works by running the sample through a precipator tube and over a high-powered magnet. Larger particles are more quickly attracted to the magnet, allowing them to gather at one end while the smaller particles gather over the exit end. Light is then transmitted through the sample. Photo detectors measure the amount of light passing through the sample, thereby resulting in the DL and DS values.
An advantage of DR ferrography is the information that can be derived from these results. As an index value, simple equations can be applied to find total wear particle concentration (WPC) and the percent of particles that are large.
Performing DR ferrography requires the use of hazardous chemicals, precise dilutions and exact flow rates over the magnet to obtain quality and representative results. While the operators of DR instruments are well trained and understand the importance of proper sample testing, there is still a possibility of operator error given the amount of variables and sampling handling that is involved.
Figure 4. Pore Blockage Particle Counter
Particle quantifiers use the Hall effect to determine the ferrous concentration of an oil sample. The Hall effect is the measurable, induced voltage across a sample under an applied magnetic field and current (Figure 4). In general, the higher the concentration of ferrous wear present, the higher the observed Hall voltage.
Similar to DR ferrography, the particle quantifier gives the measured ferrous concentration as an index value. However, this is where the similarities between the two end.
Particle quantifier results are given as only a single index value versus the two values provided in DR ferrography. There is no separation of size.
The particle quantifier is not sensitive to particle size. When utilized in conjunction with AES, several evaluations can be made. If the particle quantifier result increases along with an increase in AES, it is likely that numerous amounts of small particles are being generated. However, if particle quantifier results increase and there is no change or even a decrease in AES ferrous debris, this indicates the generation of large particles, thereby suggesting an abnormal level of wear.
Table 1. Percentage of Ferrous Particles
Analytical ferrography allows wear particles to be observed by the analyst via microscopic analysis. In this evaluation, active machine wear as well as multiple different modes of wear can be determined. This method has an outstanding sensitivity for larger particles.
During the slide preparation process, the sample is run over a slide placed at a set angle over a magnetic source. As the sample runs, ferrous particles are arranged on the slide in a given pattern related to the size of the particles. Larger particles remain at the entry point of the slide while smaller, submicron particles are deposited at the exit end of the slide.
Once dried, the slide is placed into a high-powered microscope for the analyst's evaluation. This evaluation allows more for the particle identification (qualifying the wear) than determining the amount of wear (quantifying). However, it is relatively easy to observe whether a slide contains more debris than what is typically found in a specific type of component.
Even though this method is partial to ferrous debris due to the magnetic capturing of particles, identifying some nonferrous debris is possible. This will typically take place when the nonferrous particles are large and fall out of the oil while passing over the slide, or when the nonferrous particles get trapped by ferrous particles already settled out and held in place via the magnetic field.
Analytical ferrography requires extensive analyst training to properly interpret the results. Evaluation of a ferrogram can differ between analysts due to the subjectivity of the analysis. The level of training and experience will also impact the accuracy of the ferrographic evaluation.
Several techniques are employed during ferrogram analysis that make this level of examination far from entry level. Some of these techniques include the heat-treating of slides as well as chemical microscopy. A complete understanding of the effects of these techniques is required to fully utilize the tools available.
The overall cost of analytical ferrography makes it prohibitive to use as a routine test. The high cost is due to the lack of automation and the amount of time it takes to perform the preparation and examination. In order for a single sample to receive this analysis, it could take from 25 to 45 minutes from start to finish. This is in addition to the remainder of the tests that must be performed.
Micropatch, also known as filtergram, is the study of wear particles deposited onto a filter. Similar to the ferrogram, the micropatch is efficient at capturing large particles (depending on the size of the filter pores). Industry-standard filters used for this test are .8 µm, making these effective for observing wear of all sizes.
The micropatch is used for applications that involve nonferrous wear as the leading indicator of adverse conditions. Examples of these situations include worm gears, turbo machinery (with bronze or babbitted bearings), stainless steel, ceramic bearings or even on-site evaluation.
One drawback of micropatch analysis is that the ability to employ heat treatment is extremely limited and requires special patches to be used. Additionally, it is harder to distinguish between different types of metals and there is a poorer resolution of transmitted light.
The level of training required to perform micropatch study is similar to analytical ferrography because a complete understanding of wear modes and effects must be present.
Particle counting in industrial gearboxes will tell the same story as particle counting in a hydraulic system or pump application, that of cleanliness. When establishing an oil analysis program proactive in controlling contamination, particle counting is a vital component to the routine test slate.
There are two types of particle counting that are generally performed: optical and pore blockage. Optical particle counters work by a sensor measuring either the amount of light lost or the amount of light scattered when crossed by a particle, which is then related to a specific size of particle. A pore blockage particle counter works by a transducer measuring either the amount of pressure increase or flow decay that occurs as the oil sample passes through a sensor containing a set number of specifically sized pores. In this method, larger particles that cannot pass through the sensor are trapped while smaller particles get trapped in the open gaps of the larger particles (Figure 5).
Particle count results are then used to derive the ISO cleanliness level. ISO 4406:99 reporting structure gives the cleanliness as a three-digit value (18/16/13). It is also occasionally acceptable to report the cleanliness as a two-digit value (16/13). These digits correspond to the number of particles detected at the >4, >6, >14 micron levels for the three-digit code and the >6, >14 micron levels for the two-digit codes.
An additional form of particle counting that can be accomplished in conjunction with pore blockage is ferrous particle counting. This method allows a determination of the percentage of particles that are ferrous in nature. Table 1 indicates how having knowledge of the percentage of ferrous particles can indicate two completely different conditions.
Using particle count data in conjunction with other routine tests can also help to differentiate between a dirty gearbox and a gearbox filled with excessive contamination.
Other methods of measuring large particles are available; however, due to time and cost, these methods are not readily deployed as part of a standard routine test slate. These methods include scanning electron microscopy and X-ray fluorescence spectroscopy.
The various methods of testing for larger wear particles in gearbox oil samples have now been introduced. To gain the best value for the price, a test slate should include tests that will allow for the best detection of what the analyst is looking for. While there are always exceptions and special cases, as a general rule, the following should be included in the wear monitoring of gearboxes:
Contamination Monitoring: AES, particle quantifier, particle count with analytical ferrography as an exception test when an adverse condition is recognized.
No Contamination Monitoring: AES, DR ferrography with analytical ferrography as an exception test.
Additional tests that should be included in both scenarios include viscosity, water percent, acid number and FTIR.
Given the current desire to increase reliability and decrease spending, it is important to understand how to properly utilize the technologies available. With oil analysis, several different aspects of machinery wear can be monitored. Many single tests performed in oil analysis have limits; however, when using the correct combination of tests, oil analysis is a powerful predictive maintenance tool.
1. Malte Lukas, Robert Yurko and Daniel Anderson. "New Rotrode Filter Spectroscopy Method." Practicing Oil Analysis magazine, September-October 2006.