In May 2004,, Mr. Scott Denny submitted copper tube/fitting specimen(s) to my office for visual inspection. The specimen(s) were removed from the potable hot water circulating system at a residence in Atherton, CA .
The specimen(s) were:
Here's a rather long winded story of a leak problem I had the pleasure to investigate.
1. A 1-3/4 inch piece of 3/4 inch nominal diameter, Type M, hard drawn copper tube connected by soldering to a 3/4 inch wrot copper 45 elbow connected by soldering to a 1-3/4 inch piece of 3/4 inch nominal diameter, Type M, hard drawn copper tube connected by soldering to a 3/4 inch wrot copper 45 elbow connected by soldering to a 2-1/8 inch piece of 3/4 inch nominal diameter, Type M, hard drawn copper tube.
2. A 2-5/8 inch piece of 3/4 inch nominal diameter, Type M, hard drawn copper tube connected by soldering to a 3/4 inch wrot copper 90 street elbow connected by soldering to a 3/4 inch wrot copper 90 elbow connected by soldering to a 1-1/4 inch piece of 3/4 inch nominal diameter, Type M, hard drawn copper tube.
Information provided with the sample indicated the copper tube system was commissioned in 1992. The leak in question occurred in April 2004. Information was apparently not available regarding the velocity or pressure of the water in the affected section of the system. Information provided indicated the typical water temperature was 130F.
Water distributed at the residence is obtained from the San Francisco Municipal Water Utility. No representative chemical composition data for the water supplied was included with the specimens.
Scott Denny
Frank Denny Plumbing, Inc.
May 12, 2004
Page 2…
Examination of the outside surfaces revealed no significant deterioration by the external environment. Basically, the outside surfaces of the tubes and fitting were covered either with solder or a protective tarnish film of reddish brown material that is probably cuprous oxide (Cu2O). In addition, there were deposits of a greenish material that would probably prove to be the remnants of flux residue that had not been cleaned of the assembly after soldering. Specimen 1 also had greenish white deposits the were formed as a result of water leaking onto the tube surface and evaporating, leaving calcium, carbonate, and malachite materials on the tube surface. Finally, a number of the specimens had solder deposits on the outside tube surface in excess of what would be considered “good workmanship practicesâ€.
Only specimen 1 had a pinhole perforating the tube wall. Examination of the inside surfaces confirmed the belief that the tube failures in this tube had initiated on the waterside surface. The grooved and corrosion-product-free areas of attack were considered to be typical for tubes which had been subjected to conditions causing erosion-corrosion (i.e., accelerated attack related to localized high velocity/turbulent water inside the tube).
In those areas where erosion-corrosion had not taken place, the tube wall was overlaid with a tightly adherent film of brown colored products. Analyses of similar products on other tube specimens in the past suggest that this deposit probably consists primarily of copper oxide (Cu2O). In general, the copper oxide was overlaid with a thin friable layer of loosely adherent brown colored products that were clearly evident to be deposits from the water. The belief that the brown products were deposits from the water was supported by the observation that they could be easily removed by gently wiping them with a wet sponge.
Much of the inside surfaces of both specimens was covered with a black tarnish film that would probably prove upon chemical analysis to be cupric oxide (CuO). The presence of cupric oxide on the waterside surfaces indicated that the water circulated had been heated, at least on occasion, to temperatures in excess of about 160º F (i.e. a condition that typically facilitates erosion-corrosion). Examination of the cut ends of the tubes left no doubt that the ends had not been reamed/deburred prior to assembly.
Based on the visual examination of the specimens submitted for investigation, it can be concluded that the failure reported to me concerning these samples had initiated on and propagated from the waterside surface. The cause of the perforation was erosion-corrosion as a result of localized high water velocity.
Erosion-corrosion on the waterside surface of the tube was at least partially facilitated by (1) water velocity flow rates in excess of recommended values for copper tube systems, and (2) unreamed cut tube ends. In addition, the erosion-corrosion process may have been aided by (1) possible presence of abrasive suspended solids ( hydrated hematite, or silica for instance), (2) water pressures in excess of 80 psig, and (3) heating the water circulated to temperatures in excess of 140 Fahrenheit (F).
Scott Denny
Frank Denny Plumbing, Inc.
May 12, 2004
Page 3…
Basically, the flow rate of the water in the affected section of the system was excessive. The role of velocity, temperature, water chemistry, and other factors in the erosion-corrosion of copper is discussed in the paper “Performance and Selection of Materials for Potable Hot Water Service.†A copy of this paper is included.
Bear in mind that this condition is not peculiar to copper tubing, but can affect other materials as well if water velocities exceed their recommended parameters. However, when erosion-corrosion does occur in copper tube, it is readily identifiable by horseshoe shaped pits throughout the inside of the tube, which is the distinctive signature of damage caused by erosion-corrosion.
Although it is virtually impossible to rehabilitate a piping system that has experienced erosion-corrosion related failures short of replacement of affected areas, a number of recommendations can be made for mitigating the erosion-corrosion of copper tube systems.
The temperature of the hot water circulated should be monitored and, if necessary, adjusted to a safe and energy-efficient 130 F.
The pressure in the system should be measured to insure that it does not routinely exceed approximately 80 psig.
Water flow rates should be measured in the affected section(s) of the system and, if necessary, reduced to insure they do not exceed about 8 feet per second (fps) for cold water. Velocity should not exceed 4 to 5 fps for hot water if the temperature is below 140 F and about 2 to 3 fps if the water temperature is above 140 F.
Equally important, plumbing technicians must use industry standard workmanship when installing copper tube systems. For example, cut tube ends must be properly reamed prior to soldering. Adhering to the general guidelines for tube installation, joint preparation, and soldering presented in the enclosed CDA Copper Tube Handbook and ASTM Standard B828 have been known to eliminate many erosion-corrosion concerns.
Should you require any further information or assistance you can contact me at the phone number or address listed on this letterhead.
Best Regards,
COPPER DEVELOPMENT ASSOCIATION INC.
Jim Weflen
Jim Weflen
CDA Western Region Manager