Maintenance and Analysis of Water-Glycol Hydraulic Fluids

Roland J. Bishop, Dow Chemical Company G.E. Totten, G.E. Totten & Associates LLC
Tags: hydraulics

Hydraulic fluid performance, including water-glycols (W/G), depends on the chemical composition of the fluid and cleanliness. This article presents an overview of the effect of W/G fluid chemistry on pump wear. An overview of recommended analytical procedures to assure adequate long-term hydraulic and lubrication performance is provided. These procedures can result in substantial improvements in hydraulic pump longevity and performance.

Many industrial applications such as steel making, die casting, etc. require the use of hydraulic fluids that offer greater fire safety than that achievable with mineral oil. One of the most common alternatives to mineral oil for use in these applications is a water-glycol hydraulic fluid. The performance of all hydraulic fluids, including W/G hydraulic fluids, depends on the composition of the fluid and on fluid cleanliness. Although there are numerous references describing analysis procedures for petroleum oil-derived hydraulic fluid, similar references describing the analysis of water-glycol hydraulic fluid are relatively rare.

Fluid Formulation
Water-glycol hydraulic fluid formulations typically contain water (for fire protection), glycol (for freeze point protection), polyalkylene glycol (PAG) thickener, an additive package to provide corrosion and antiwear protection, antifoam/air release additive and dye for leak detection.

Fluid Chemistry and Wear
The performance of a hydraulic fluid depends on the particular additive and concentration used in the formulation. Substances that exhibit a marked effect on hydraulic pump wear, according to the ASTM D2882 test, are water, amines and antiwear additives. The ASTM D2882 test is conducted at 2,000 psi (13.8 MPa) for 100 hours and eight gallons per minute (30.6 L/min) in a Sperry Vickers V-104C vane pump.


Figure 1. Effect of Water Content on Wear Rate

Water content creates one of the most significant influences on hydraulic pump wear rate. Figure 1 shows that wear rates increase with increasing water content. Thus, it is critically important to control the water content of W/G hydraulic fluids if both fire-resistance and antiwear performance is to be maintained.

The effect of the amine, which is present primarily as a corrosion inhibitor, on wear rates is illustrated in Figure 2.


Figure 2. Effect of Amine Concentration (Alkalinity) on Wear.
Relative Wear is Relative to the Lowest Wear Rate at
Optimum Alkalinity.

Wear rates increase with decreasing amine concentration, however, there also appears to be an optimal amine concentration above which the wear rate begins to increase. The magnitude and critical concentration vary with the amine being used.

Fluid Chemistry and Corrosion
The primary function of the amine is to provide corrosion protection. Vapor and liquid phase inhibitory properties of an amine can be determined using the 200-hour Corrosion Test. This test is conducted by aerating the heated hydraulic fluid at 70 ± 2°C in contact with metal test coupons for 200 hours. The test coupons that are immersed in the fluid are: steel (SAE-1010, low carbon), cast aluminum (SAE-329), copper (CA-110) and brass (SAE-70C). Vapor phase corrosion effects are also determined using coupons of cast iron (G-3500) and steel (SAE-1010) which are suspended above the solution. Figure 3 illustrates a typical corrosion test cell.

Figure 3. Corrosion Test Apparatus is Convenient for Corrosion-Inhibitor Studies of Water-Glycol Hydraulic Fluids Under Laboratory Conditions. Immersed Metal Specimens are Separated by a Glass “Z-bar” in the Specific Order Shown. Vapor-Space Test Specimens are Hung from the Top of the Glass Test Cell. Fluid Temperature is Monitored with an Immersion Thermometer, Air is Blown into the Mixture Using an Aeration Tube, and a Cold-Water Condenser is Used to Reduce Fluid by Evaporation.

Optimal corrosion protection depends on the selected amine(s) and amine concentration. Amine concentrations 100 percent above the optimum level will actually aggravate non-ferrous corrosion in the liquid phase and in concentrations, 75 percent below the optimum will aggravate ferrous corrosion in the vapor phase. Therefore, it is essential to monitor the concentration of the corrosion inhibitor and make periodic corrections as needed.

Of course, the selection and concentration of the antiwear additive will affect hydraulic pump wear.


Figure 4. Effect of Antiwear Additive Concentration.
Normal wear is Formulation-Dependent.

Fortunately, as shown in Figure 4, it is possible to formulate a W/G hydraulic fluid so that there will be minimal impact on wear with the inevitable loss of additive over time. Nevertheless, once the critical level is achieved there is a dramatic increase in wear rates with further decreases in antiwear additive concentration.

Fluid Contamination and Performance
Hydraulic pump lubrication depends not only on fluid chemistry but also on both liquid and solid contamination. In W/G fluids, the most common liquid contamination is usually petroleum oils, which may enter the hydraulic system from numerous sources. Because petroleum oils are insoluble in the W/G they may be simply skimmed from the fluid reservoir. In practice, removal is often neglected for long enough periods that some of the additives adsorb into the mineral oil and are removed from the working fluid when that oil is skimmed from the surface of the reservoir. Every effort should be made to prevent this form of contamination.

W/G Fluid Analysis Procedures
The following discussion focuses on analytical procedures that may be used to monitor fluid chemistry and physical property variation.

Note: All of the examples in the following discussion are for illustrative purposes only. Because these values are fluid specific, they will vary with supplier. Therefore, the reader should consult his or her fluid supplier to obtain the appropriate recommendations to use with these procedures.

Initial Fluid Observation
The first step of any analysis is to simply observe the sample. Look at the sample in a clear container, such as a clear sample bottle. The sample should be clear without the presence of oil layers or solid debris. If solid debris is observed, a magnet should be used to determine if it is magnetic. Magnetic solids may be a result of either wear or corrosion. Nonmagnetic debris may result from seal erosion or external contamination.

Water Content
Water contained in a W/G fluid can be lost through evaporation during normal hydraulic operation. Water loss increases the fluid’s viscosity. Water must therefore be added back to the system to maintain fire-resistance and to assure proper viscosity and system operation.

The most common methods for determining water content of a W/G hydraulic fluid are refractive index, viscosity and Karl Fischer analysis. Refractive index is the most commonly used and is readily determined using a portable temperature-compensated refractometer that provides readings in degrees Brix.

The water content from the refractometer reading is obtained from a calibration chart such as that shown in Figure 5.


Figure 5. Water Content vs. Refractive Index

The principal limitation of water determination by refractive index is that refractive index is affected by any material, including contaminants, that may be present in the hydraulic fluid. Thus, it is advisable to crosscheck water analyses obtained by refractive index against at least one other analytical method. After the water concentration is determined, additional water should be added if necessary. Some suppliers provide tables such as Table 1, which provide water make-up levels without the use of the calibration plot represented by Figure 5.

Only distilled or deionized water with a conductance of less than 15 µmhos/cm (or a maximum total water hardness of 5 ppm has also been recommended), should be added to a W/G hydraulic fluid system. This is critically important because polyvalent metal ions such as Ca+2, Mg+2, Mn+2, etc. will react with the antiwear additive, usually an organic carboxylic acid, to form a polyelectrolyte complex salt (Formula 1) which appears as a white, soapy solid. This process must be prevented for two reasons. The first is that it will lead to continuous depletion of the critically important antiwear additive. Second, the presence of such precipitates, like any solid material, will increase wear.


The water content of a hydraulic fluid may also be determined by viscosity measurement. One common method of viscosity measurement is to follow the ASTM D445 procedure for kinematic viscosity.


Figure 6. Viscosity vs. Water Content

By using the chart in Figure 6, viscosity vs. water content, the amount of water can be easily maintained within the necessary range for the fluid. Alternatively, a water make-up table based on viscosity, as shown in Table 2, may be obtained from the W/G hydraulic fluid supplier for the specific fluid being used.

The load-bearing capacity of a fluid film depends on fluid viscosity. Oxidative and thermal degradation processes will result in a decrease of fluid viscosity. Thus routine viscosity measurement is one of the best methods of monitoring fluid stability. However, such comparative measurements must be made at the same total water content.

The third, and most unambiguous, method of water determination is by Karl Fischer Titration (ASTM D1744). The advantage of Karl Fischer analysis is that it is a direct measure of water content, while viscosity and refractive index are both indirect measurements which are substantially affected by either contamination (refractive index) or fluid degradation (viscosity).

Reserve Alkalinity (Corrosion Inhibition)
Amine concentration in a W/G hydraulic fluid is designated as reserve alkalinity and is conventionally reported as the volume in milliliters of 0.1N hydrochloric acid (HCl) required to titrate 100 ml of W/G fluid to pH 5.5. A typical titration plot is shown in Figure 7.


Figure 7. Determination of Reserve Alkalinity by Acid Titration

Two breaks in the titration curve are observed because two chemical components are actually being analyzed: the amine carboxylate (Formula 2), how the carboxylic acid used as the antiwear additive in the aqueous fluid, and the excess “free” amine (Formula 3). Thus, acid titration also provides a method for quantifying the concentration of antiwear additive.

Changes in corrosion inhibitor concentration may also be monitored by pH. It is recommended that the pH of the W/G system be greater than 8.0.

Fluid Degradation
W/G fluids may oxidize when subjected to high operational temperatures, such as those encountered during a heat exchanger failure. If the temperature and times are sufficient, then low molecular weight acids, such as formic acid may be produced. The presence of formic acid is particularly harmful because concentrations higher than 0.15 percent may lead to excessive wear as shown in Figure 8.

Figure 8. Two-dimensional Contour Plot - Effect of Formic Acid and Reserve Alkalinity on ASTM D2882 Wear Rates of a Conventional W/G Hydraulic Fluid. Wear Rate is Affected by Both Formic Acid Content and Alkalinity.


The presence of low molecular weight acids, especially formic acid, may result in increased wear. Ion chromotography is the preferred method to detect acids in the fluid.

Ferrography
It has thus far been shown that hydraulic fluid quality and performance depends on fluid cleanliness and chemistry variation. On occasion, it is necessary to troubleshoot fluid performance in malfunctioning systems. In addition to the chemical and physical analyses described, it is often valuable to analyze any wear debris. Ferrography is one of the principal wear debris analysis methods. It can be used to determine the concentration and distribution of wear particles contained in the hydraulic fluid.

With a combination of direct reading ferrography and ferrographs (photo-micrographs), the identity of various contaminants and solid particles can be discovered. This includes carbonaceous material, rust, copper, soft metal (such as aluminum, zinc, etc.), as well as severe and normal wear.

It has been shown that W/G hydraulic fluid performance, like all other hydraulic fluids, depends on both fluid cleanliness and fluid formulation chemistry.

Water, antiwear additive and corrosion inhibitor concentrations must be monitored to assure optimum fluid antiwear performance. Recommended analytical methods include:

  • Visual observation
  • Water content
  • Reserve alkalinity
  • Ion chromatography (formic acid content)
  • Ferrography

While analysis by ion chromatography and ferrography are specialized procedures and may be conducted as required, analysis for water content, reserve alkalinity, viscosity as well as visual observations are critical and must be conducted regularly (usually by the fluid supplier).

If the hydraulic system is properly maintained and fluid performance is adequately monitored, excellent long-life hydraulic and lubrication performance with water-glycol fluids is achievable.

References

  1. Forgeron, E. and McCormick, N. Production Engineering (Cleveland). April 1981. p. 82-85.
  2. Skoog, P. Hydraulics & Pneumatics. November 1991. p. 41-44.
  3. Skoog, P. Foundry Management & Technology. November 1990. p. 40-41.
  4. Totten, G. SAE Technical Paper Series, Paper Number 921738. 1992.
  5. Totten, G., Bishop Jr., R., McDaniels, R., Braniff, D. and Irvine, D. SAE Technical Paper Series, Paper Number 941751. 1994.
  6. Zingaro, A. Hydraulics & Pneumatics. April 1994. p. 37-40.
  7. Zingaro, A. Hydraulics & Pneumatics. December 1994. p. 25-26.
  8. Fitch, E. and Hong, I. The FRH Journal, 6. 1986. p. 41-51.
  9. Fitch, E. and Hong, I. The FRH Journal, 6. 1986. p. 53-61.
  10. Xuan, J. The FRH Journal, 6. 1986. p. 63-68.
  11. Ou, R. The FRH Journal, 6. 1986. p. 69-75.
  12. Holiday, L. Ionic Polymers, Chapter 6. New York: John Wiley & Sons, 1975.
  13. Rossiter, W., Brown, P. and Godette, M. Solar Energy Materials, 9. 1983. p. 267-279.
  14. Tessmann, R. Proc. Natl. Conf. Fluid Power, 34th, 32. 1978. p.179-183.
  15. Ciekurs, P. Ropar, S. and Kelley, V. "Prediction of Hydraulic Pump Failures Through Wear Debris Analysis." Naval Air Engineering Center Report, NAEC-92-171. July 19, 1983.
  16. "Wear Particle Atlas -Revised." SOHIO Predictive Maintenance Series. Standard Oil Inc.

About the Author
About the Author
Create your own user feedback survey