Over the past 22+ years, CorrView International, LLC has performed numerous ultrasonic investigations of either very old pipe installed in the early 1900's, or at properties where a combination of old and new pipe exist. That work offers a remarkable demonstration of how low cost foreign imports and excessive environmental controls and regulations over US manufacture have reduced the quality and corrosion resistance of today's steel piping products.
Combined with less tolerant engineering practices, thinner pipe materials, cost cutting, less effective corrosion inhibitors, as well as other factors, most new piping installations can be expected to provide significantly less service life than those built many decades earlier..
First and foremost is the unquestionable superiority in quality and corrosion resistance of pre World War II steel pipe to that manufactured today - whether foreign or domestic. Our ultrasonic testing of steam systems from 1911 have shown a minimal loss of only perhaps 20% from the original wall thickness. Testing of some galvanized steel domestic water risers from 1920 have documented little if any wall loss and no loss of the zinc protective coating. A galvanized steel domestic cold water riser installed in 1890, and only in 2014 finally showing the need for replacement! We have certified 80 year old steam condensate pipe, which traditionally suffers from the acidic conditions of condensate requiring frequent replacement, as useful for another 50 years of service. Condenser water systems from the early 1940's have been documented to corrode at below 1 mil per year (MPY) and offer hundreds of years of future service - even at building locations where chemical treatment was poor or non-existent. In one example of unpainted wrought iron cooling tower pipe, exposed to the environment and constant tower over spray for 45 years, the ASTM pipe stamp was still clearly present - a remarkable illustration to the natural corrosion resistance of wrought iron pipe.
Wrought iron ASTM A 72 pipe, which is well recognized and documented to provide extremely long life due to its internal grain structure and inherently high corrosion resistance, can be found at many older pre-1970 properties. However, it was removed from U.S. production in 1968 and is no longer available.
Any property owner can rest easier where Bethlehem wrought iron pipe is still in use - as it provides virtually unlimited service due to its superior quality and natural corrosion resistance. We have documented 12 in. Bethlehem steel pipe still serving open condenser water service since 1936; having its pipe wall still near new extra heavy specifications of 0.500 in. A recent investigation of a 1965 property having Bethlehem steel pipe installed throughout the HVAC systems, still showed wall thickness in most examples at or exceeding its new ASTM factory specifications after almost 50 years of service.
We consider Bethlehem Steel pipe products as the best ever manufactured. Bethlehem Steel helped build the United States and won two World Wars. Today what remains is a gambling casino, destroyed by extreme government and environmental regulations, excessive union demands, in combination with impossible to compete against low cost, subsidized foreign pipe imports. Now, many decades later, the abandonment of high quality in favor of lowest cost is revealing some staggering results.
Although today's steel manufacturers still meet the same ASTM specification, unknown and unexplained differences exist to produce products which are vastly different in terms of their corrosion susceptibility. Some metallurgical labs and corrosion "experts" have argued that today's steel products are exactly the same as pipe produced in 1910 due to it meeting the same ASTM specification. Simple observation and over 22 years of ultrasonic investigations, however, have overwhelmingly proven otherwise.
Below we show various examples of older pipe serving continuously for as long as 111 years, and still capable of providing another 200+ years of service or more. The remainder of this Internet site documents generally more recent piping systems, and their associated corrosion problems.
In contrast to the high quality of pipe manufactured 50 years ago, we find few new properties able to control corrosion to below 5 MPY today without expending extraordinary cost, supplemental filtration, close monitoring, and added physical effort.
Chemical treatment programs costing $10,000 only 15 years ago now reach almost 10 times that expense. Fully automatic chemical feed and bleed systems are no loner a convenience, but mandatory. Yet, ultrasonic and metallurgical testing at newer installations typical reveals corrosion rates in the 3 - 5 MPY range, with some examples exceeding 15 to 20 MPY. Such high corrosion rates even in examples where the chemical water treatment has been extremely well maintained. We have either directly seen or have been advised of condenser water or process piping systems which have been entirely replaced after only 10 years of service or less, and have found 6 year old large diameter 8 in. and 12 in. main risers salvaged throughout their facilities using emergency pipe clamps. In cases where we have investigated building properties having both new carbon steel and older wrought iron pipe, we have consistently provided remaining service life estimates in the hundreds of years for the wrought iron, as opposed to a few decades or less for the newly installed carbon steel.
The use of more expensive copper in smaller diameter distribution HVAC piping and process loops has become the standard practice today in the effort to avoid the damaging effects of corrosion against carbon steel. Larger diameter 8 in. and 12 in. condenser water piping systems running the entire height of large commercial office buildings have been entirely replaced using extra heavy copper and even 304 or 314 stainless steel - such extraordinary expense expended solely in the effort to avoid the destructive effects of corrosion rarely seen decades earlier.
While low cost foreign produced pipe from Japan, Korea, Mexico, South America, and Eastern Europe has traditionally shown the greatest susceptibility to corrosion, we have not found recently produced American carbon steel pipe products of significantly higher quality in terms of corrosion resistance. Aside from the obvious net effects of stringent U.S. environmental controls and government over-regulation, as well as the competition of low cost foreign steel, we have not been able to establish, nor has any expert been able to explain, a suitable explanation for the obvious decline in American pipe products in terms of corrosion vulnerability from decades ago. Other changing factors acting against steel pipe produced today obviously exist.
For those many reasons, CorrView International strongly recommends that higher corrosion rates should be anticipated regardless of any corrosion control measures planned or implemented. Simultaneous to higher corrosion is a higher volume of rust deposits, and therefore, effective water filtration is an absolute necessity today. It should be noted that standardized corrosivity tests, laboratory methods capable of measuring the susceptibility of a metal to a typical corrosive environment and rating that metal according to an established numerical scale, are available, and offer an excellent prediction or warning to the potential for future corrosion problems.
Hard documentation to the higher corrosion level suffered by newer pipe was provided in a recent series of investigations of approximately 20 U.S. government facilities throughout the United States identically constructed at the same time, and all having the same mechanical and piping expansions subsequently. Constructed in 1960 and having new additions roughly each decade produced a revealing graph of corrosion activity when plotted against its year of installation.
In short, testing at chill water, condenser water, glycol, and hot water heating pipe all showed a near 10 times increase in the rate of corrosion for pipe installed in 2005 as opposed to that installed in 1960. Contrary to the expectation of facility engineers concerned to the condition of the older pipe, thereby the reason for the investigation, ultrasonic testing revealed the older pipe in substantially better condition and far more potential operating problems at pipe installed since 2000.
This condition is clearly shown in the below summary graph with X axis representing the installation date. For each piping service, the older pipe from 1960 is shown at left, and the more recently installed pipe is shown at right
It is not uncommon to find high wall thickness at older properties still exceeding new ASTM specifications. This is due to older pipe having been manufactured generally at or above the wall thickness specified by ASTM, in combination with far lower corrosion rates. In fact, carbon steel pipe can be manufactured having a wall thickness +/- 12.5% of its specified value and still be considered acceptable for installation. No concern is ever raised for pipe which is manufactured over specification, since it offers a great benefit to any building property.
Today, most pipe is produced substantially undersized due to what we believe are closer production tolerances enabling manufacturers to produce a lighter but still approved product within the tolerable limits of the ASTM code. Extensive testing of new steel pipe has consistently found undersized material statistically beyond the possibility of being a random event. With a permissible tolerance of + / - 12.5%, most new pipe we measure is toward the lower boundary of this limit, and very rarely at true ASTM factory specification or above. An FM or UL approval rating does not mean that the pipe is at true ASTM specifications, and ASTM does not monitor piping manufacture.
Above left we show the test result for a section of new U.S. produced 8 in. carbon steel pipe of the ASTM specification A53 Grade B seamless. Its ASTM specification calls for a wall thickness of 0.322 in., which is likely what design engineers are assuming for this addition to an existing condenser water system.
Instead, we measure a wall thickness of 0.292 in. consistently throughout its length, which represents it being 9.1% undersized. At a typical corrosion rate of 2 MPY, this reduction of 0.030 in. of wall thickness represents a potential 15 years of service lost.
In a second example at left, testing at a section of new 6 in. schedule 40 A53 pipe, with the ASTM specified wall thickness of 0.280 in., produced true wall thickness measurements over its entire length of 0.248 in., and 0.032 in. below specification. For on-site skeptics in this project questioning the accuracy of the instrumentation, physical measurement by dial caliper produced the very same wall thickness dimension.
At 0.248 in., this new pipe is 11.4% under its ASTM factory specification prior to even being installed, and again will provide lower than expected service life. Under higher corrosion conditions more common today, such undersized pipe will provide far less service.
Further testing of two other sections of 6 in. pipe from the same HVAC condenser water expansion project, but from a different U.S. manufacturer, produced similar results with wall thickness dimensions consistently near 0.245 in. - precisely 12.5% under its ASTM specification and at the lowest limit of what is permitted by the ASTM code.
Another example of new 6 in. schedule 40 pipe, installed only one week into a condenser water system and in a completely separate project in another state, shows similarly low initial wall thickness of 0.244 in. as opposed to its defined ASTM thickness specification.
Not only is this section of new pipe undersized, but it is undersized to 12.9%, and therefore technically unacceptable for installation. Actual wall thickness dimensions are virtually never checked, however, leaving the mechanical design engineer, pipe installer, building owner, and facility engineers to all mistakenly believe that the stenciled wall thickness dimension or pipe schedule specification is true.
In this above example of 6 in. diameter schedule 40 pipe into a condenser water system where 2 MPY might be expected as a reasonable corrosion rate, the loss of 0.036 in. of material translates into an immediate loss of 18 years of service life.
In a further example, we can demonstrate that undersized pipe is also provide to heavier schedule 80. This section of 3 in. schedule pipe scheduled for installation for steam condensate service has an ASTM specified wall thickness of 0.300 in. In reality, ultrasonic testing identified a wall thickness of the new pipe, never installed, of 0.266 in.
This is just slightly above the 0.263 in. minimum specified by ASTM, and represents the pipe being undersized to 11.34% of its 12.5% maximum limit. At a typical corrosion rate for steam condensate service of near 1-2 MPY, undersizing this pipe by 0.034 in. means an unnecessary and unanticipated loss of 15 years of service life.
Below, the installation of a new 18 in.condenser water pump header was nearing completion when the ultrasonic investigation at examples of older pipe allowed a quick evaluation to the new pipe still being welded into place. With an ASTM stamp marking the pipe as STD, (standard), as well as a wall thickness dimension of 0.375 in. stenciled on the pipe itself, everyone associated with this piping project likely assumed that 0.375 in. was its true thickness dimension.
Instead, UT testing identified the pipe consistently between 0.332 in. and 0.336 in. across its entire approximate 20 ft. length - remarkably uniform, but lower. At 0.332 in., this new pipe is precisely 12% under its defined ASTM thickness specification; consistent with most new pipe today. Technically, this new 18 in. pipe has a wall thickness more closely associated with 8 in. schedule 40 pipe at 0.322 in.
Ultrasonic testing of the new 12 in. pipe to and from each condenser water pump, again defined by ASTM to have a wall thickness of 0.375 in., instead showed the same thickness values of near 0.330 in., and again being undersized to near its maximum allowable limit.
In another more critical example immediately left, we measure a new 2 in. schedule 40 threaded pipe nipple having an ASTM A 53 factory wall thickness specification of 0.154 in. However, ultrasonic testing shows a wide variance in wall thickness along its unthreaded center of between 0.133 in. and 0.142 in. - significantly lower that its ASTM specification by 13.6 %, and technically below its ASTM minimum allowable limit for installation.
Similarly undersized 2 in. pipe nipples from the same lot were installed and placed into service - now having only 0.063 in. of usable wall before reaching the outer thread cut and a failure condition.
Unlike larger diameter pipe where a 10 % less thickness dimension might not present any immediate threat, this same loss at small threaded pipe having inherently thinner wall thickness has a significant impact. With 0.070 in. cut away during threading, this unnecessary 0.021 in. loss of material can dramatically reduce service life at all but those systems with the lowest possible corrosion rate.
As mentioned, in response to such observed increases in corrosion, many property managers, operating engineers, and plant managers have turned to the use of Type L copper tubing (commonly called pipe) for all smaller diameter run-out distribution lines, and in some cases even for the main risers. This, however, may only be a short term solution to the often more complex corrosion condition, and in some cases may actually complicate an existing corrosion problem with additional and unseen threats.
Important to consider in the substitution of copper pipe for carbon steel is the significant difference in initial wall thickness. Common Type L, ASTM B 88 copper for 3 in. diameter pipe has a standard wall thickness of only 0.090 in., whereas 3 in. A 53 schedule 40 carbon steel has a wall thickness of 0.216 in. Compared to steel pipe having a stress efficiency (a strength rating, not pressure rating) of 15,000 lb./sq. in., B 88 copper pipe only offers 6,000 lb./sq. in. - a physical decrease in strength of 60%. Copper pipe has less than half the tensile strength of steel, and quickly loses that strength at high temperatures.
It is generally recognized that the minimum acceptable wall thickness for copper tubing, under any conditions, is approximately 0.020 in. At higher pressures, this minimum value increases - thereby allowing for less physical wall loss to occur before being judged as unsuitable for further reliable service. Copper pipe, since it exhibits far less physical strength than steel, will fail sooner than steel at the same wall thickness dimension and under the same internal pressure - making it obvious that the greatest threat for any copper piping or components installed in a high rise building property exists at the higher operating pressures of the lower floors.
While the corrosion rate against copper is commonly believed to exist well below 0.5 MPY under all conditions, CVI has well documented that the same corrosive environment responsible for raising corrosion rates against carbon steel past 10 MPY, will greatly increase the copper corrosion rate above its normal value as well. Copper corrosion rates exceeding 5 MPY are therefore not uncommon where an already high steel corrosion rate has been documented.
A further concern relates to the seemingly similar decrease in quality of copper pipe to steel, as well as its often undersized wall thickness dimensions. In the example of new American made 2-1/2 in. pipe shown at left, a field demonstration of ultrasonic testing procedures to project engineers at a construction site revealed dramatically different wall thickness measurements within the same length of new pipe.
With heavier Tyle K copper tube specified due to a known corrosion condition, and having an ASTM defined wall thickness of 0.095 in., our testing identified a wide range in wall thickness from between 0.065 to 0.089 in, and lowest wall thickness a significant 32% below its specification. Measurements were so dramatically low and uneven that observers, as well as the pipe fitters installing the new pipe, questioned the accuracy of the ultrasonic test method.
The pipe was cut and a physical measurement made by standard dial caliper produced the same low wall thickness measurements - with a difference of 0.024 in. at opposite walls. This represents a 25% variance in wall thickness alone; with measurements showing the pipe generally below Tyle L specifications and approaching thin wall Type M dimensions. A close visual examination, again shown by this photo of the subject pipe with substantially lower wall thickness at left, further supported the original ultrasonic findings. Subsequent dial caliper measurement of the remaining copper pipe stock produced similar results.
For new copper pipe being added to a condenser water system plagued with a high corrosion condition, the design engineer's intent to extend service life by specifying heavier (and more expensive) Tyle K copper pipe in reality accomplished nothing. With the service life of all piping being defined by its lowest wall thickness, thin wall Type M copper pipe was instead installed into this project. In the above actual event, the contractor argued that the pipe was acceptable and it was installed.
For older pre-1970's building properties or plant operations, it is unlikely to find anything but schedule 80 or extra heavy black pipe in use for cooling water, steam, or steam condensate service. However, today, schedule 40 is the standard specified. Contrasted to a 0.322 in. wall thickness for an 8 in. section of schedule 40 pipe, schedule 80 provides a significantly greater 0.500 in. of available steel. Contrasting any larger diameter standard grade pipe of 0.375 in., extra heavy pipe offers 0.500 in. of wall thickness.
With internal operating pressures rarely a deciding factor to pipe selection within the commercial building and process cooling market, the previous use of heavier materials, we believe, has been more intended to counteract the known effects of corrosion activity and thereby provide longer service life.
It is important to realize that decades ago, design engineering for piping systems assumed a reasonable and readily achieved 1 MPY corrosion rate over an intended life expectancy of about 65 years for the typical building property - therefore a theoretical 65 mil corrosion rate or "corrosion factor" was typically applied in piping calculations for open condenser water or process cooling applications. In other words, consulting and design engineers estimated a total loss of only 65 mils or 0.065 in. of pipe over the assumed lifetime of the property. Any additional pipe wall thickness exceeding the corrosion factor and that needed to contain the internal pressure and stresses was simply extra insurance against corrosion losses and future operating problems.
Today, it is not unusual to measure the same loss of 65 mils after only 5 to 10 years of service, and sometimes in as little as two years. But while corrosion activity has obviously increased, the response to this greater loss of pipe has not been factored into the design and planning of modern piping systems - leaving very little if any tolerance for a system wide corrosion rate exceeding a few mils per year. In fact for many piping systems such as fire sprinkler, substantially thinner piping materials are now installed - schedule 10 and schedule 7.
CorrView International has found many building properties constructed in the 1970's and before having clearly benefitted by such engineering decisions - with the heavier schedule 80 pipe wearing over those many decades at low to moderate corrosion rate to approximately schedule 40 today. Some older properties, due to the original use of better quality schedule 80 steel, actually have heavier and longer lasting pipe than their newly constructed neighbors. Where new additions have been made to older properties, it is not unusual to measure much greater wall thickness for the older pipe installed decades earlier. It is an amazing paradox we have well documented in endless ultrasonic testing investigations.
This change in engineering design toward using thinner schedule 40 pipe, and sometimes schedule 20 and schedule 10, is far more obvious and threatening for smaller diameter threaded applications - where the additional loss of metal during the threading process often reduces the life expectancy of open condenser water or process water piping to a decade or less. Threading typically reduces the available wall thickness by approximately 50% - leaving a 0.154 in. thick piece of 2 in. schedule 40 pipe, less its thread cut of 0.072 in., with a true available and working wall thickness of only 0.082 in. beginning day one of service.
For piping systems having a typical 5 MPY corrosion rate, total penetration of the threads will occur within 16 years of installation - guaranteed. In reality, failure usually occurs years earlier. In fact, the use of schedule 40 or standard grade pipe in threaded open water condenser applications does not meet minimum acceptable engineering guidelines for new piping systems, and will typically provide only 15-20 years of service life under good to moderate corrosion conditions. The failure of a threaded schedule 40 piping system in under 5 years is not unusual, based upon our experience.
A growing concern is the recent increase in the use of grooved and clamped constructed schedule 10 pipe in fire sprinkler service. While providing adequate wall thickness initially, schedule 10 pipe has approximately half of the thickness of schedule 40, leaving very little tolerance for corrosion to occur. Where fire pipe is filled and left stagnant over extended time, a small amount of corrosion takes place, oxygen is depleted, and the corrosion process virtually stops. However, where the fire system is frequently drained, or where service extensions, leaks, or repairs bring in fresh water, corrosion rates can reach the level of open water systems, and premature failure is inevitable.
Clamped schedule 10 pipe is seeing greater use in open condenser water piping systems as well - producing disastrous results in even less time.
CorrView International, LLC overwhelmingly recommends using heavier schedule 80 steel pipe in all small diameter applications calling for threaded joints. We also recommend using no thinner than schedule 40 pipe for fire sprinkler or open water condenser systems - whether using threaded, welded, or Victaulic or grooved clamped construction.
For many decades, building and plant engineers relied solely upon the use of chromate based chemical additives to provide the required corrosion protection of steel piping systems. With even the most inferior application methods, often nothing more than an unmeasured scoop of chromate powder dumped into the cooling tower sump at irregular intervals, corrosion rates could often be maintained at or below 1 MPY. Biological fouling was a rarely encountered problem due to chromate's inherent toxicity. Such trouble free operation ended, however, in the mid-1980's - with the prohibition of all hexavalent chromate use in open water circulating systems in the United States.
Today, molybdate, phosphate, and other U.S. EPA approved chemical inhibitors rarely equal the effectiveness of past chromate treatments. Molybdate, which was considered less effective than its predecessor chromage, is now being replaced by less effective phosphates due to excessive molybdate costs. Though offering impressive corrosion suppression in bench test or laboratory settings, non-chromate programs rarely provide similar results under real world conditions. They are suggested by some authorities as being substantially ineffective in stagnant, low flow, or dead ended piping areas. During our years of ultrasonic pipe testing, we have identified numerous examples where the highest corrosion rates have been found exclusively at those areas having the lowest flows. We also attribute much of such higher corrosion rates to rust settlement and secondary corrosion effects.
Non-chromate treatments offer no microbiological or fungal control - thereby placing increased emphasis on the use of alternating biocide chemicals. Unfortunately, biocides themselves have had their effective half-life reduced to approximately 6 hours by U.S. federal and state environmental authorities. The result - a legal limit of the amount of biocide one can apply over a given period of time, as well as a legal limitation over its strength, effectiveness, and the time it will remain active.
And yet, microbiological activity has been found to play an increasing role in metal corrosion - MIC being the most serious and potentially destructive piping threat known. The alternative, using oxidizing biocides such as chlorine, bromine and ozone, all offer excellent microbiological control at the trade off of greatly increased corrosion and pitting. The common overuse of oxidizing agents such as chlorine and bromine on a weekly or daily basis have been known to produce severe pitting of steel and copper pipe, and to quickly remove the galvanizing coating from cooling tower pan surfaces.
Together, the combination of all the above named factors has placed a higher priority on corrosion control which did not exist 25-30 years ago, and which has raised the issue of chemical treatment, and the monitoring over its effectiveness to new levels of importance within those involved in building and plant engineering, maintenance, and operations.
Virtually all advanced piping failures we have seen have been traced back to an ineffective or lacking water treatment program at some point in the building's history - most notably due to poor initial clean out and start-up procedures. Quite clearly, the first six months of operation are critical, and can mean the difference between long and reliable operation, or substantially fewer years of corrosion inflicted problems. To the inherent limitations of the chemicals must be added the experience and reputation of the water treatment professional. A lack of knowledge, experience, and professionalism, or a primary interest for profits can be equally as damaging to a piping system as a lack of chemical protection entirely.
A less reputable water treatment company may be discovered and replaced, but the effects of even six months of sub-standard service can initiate a lifetime of corrosion headaches for an office building or plant operation. A further limitation may also be the budget constraints of the property or plant itself, or the need to meet a contract limit established by a previous water treatment vendor. The failure to timely recognize and effectively address a corrosion problem will place even further demand upon any future chemical treatment contractor and available control products.
For many of the worst corrosion caused failures we have investigated, total reliance on the information provided by corrosion coupons kept even the most obvious clues to a corrosion problem, such as thread leaks and rust deposits, from being further investigated. Although capable of providing some information relating to whether corrosion activity may be increasing or decreasing, corrosion coupons results have nothing to do with the wall loss occurring at the pipe itself, and therefore are virtually worthless as a corrosion measurement and monitoring tool.
Ultimately, some building property or plant operator is faced with not only the difficult task of correcting or replacing a piping system, but perhaps a responsibility for the actions or inactions of those years prior as well. Constant and accurate corrosion monitoring is therefore critically important.
© Copyright CorrView International, LLC
P.O. Box 8513Landing, NJ 07850 www.corrview.com Ph: 973-770-7764 Fax: 973-770-6576
For More Information
New content is regularly added to the site, when time allows.
For the latest updates and other advisories you are invited to join our mailing list.