By Donald Kilpatrick Inspec, Inc. Minneapolis, Minnesota Abstract This hands-on workshop will present the benefits of a responsible rooftop sampling program when used in conjunction with full-time or periodic inspections. The presen¬ tation will demonstrate the procedures, required calculations, and interpretation of the results obtained. It will demonstrate the negative effects of the following built-up roof¬ ing workmanship defects: felt-on-felt applications, interply voids, moisture, debris, negative headlap, excessive bitumen temperatures at the point of application, and inter¬ ply continuity and rate of application. Required equipment will be reviewed, sampling protocol discussed, and sample analysis provided. Donald Kilpatrick Don Kilpatrick, a project manager and field technician with Inspec, Inc., joined the company in September 1985 and spent a dozen years in the firm’s materials testing lab. His first exposure to built-up roof sample analysis was through acceptance testing when it was required by the U.S. Air Force. He is fully experienced in roof evaluations, inspections, field testing, nondestructive testing, sample analysis, and roof construction procedures. Kilpatrick is a Certified Asbestos Inspector and a member of RCI’s Interface Peer Review Board, and of ASTM D-8 Committee.
Through periodic or full-time construction observation, we attempt to address the workmanship issues that can detract from the desired quality of the installation. The roof inspector’s role is to make sure that the specified materi¬ als are being installed in compliance with the project specifications. Compliance with published industry and manu¬ facturer guidelines related to the installation of components is also given consideration. During the inspections, it is not uncommon to sample roof system components individually, or the total assembly, as a means to further establish that the products provided are as specified and are being installed in a manner that is consistent with applicable stan¬ dards and tolerances. We sample things to see if they are any good, and if the value represented is to our liking. Through sampling, we can determine if the commodities offered are of lesser value or quality than expected. There are hundreds of recognized standards available that can be used to assist in establishing or identi-fying spe¬ cific desirable characteristics and qualities of roofing materials. Consensus standards, such as those developed by ASTM, set the bar for initial product qualities and performance characteristics. Participants in the process include manufacturers, consultants, building owners and industry members such as contractors and their affiliated profession¬ al organizations. We routinely reference many of these standards in our specifications. The majority of the published standards include descriptive text related to sampling and testing protocols intended to provide the end user with data and performance character-istics of the subject material. Physical properties, dimensions and mass are some of the values that can be used in the subsequent comparative analysis. The industry, through experience and research, has established requirements for roofing products intended as target values for the manufacturers. These target values may be considered by some as the initial benchmarks of quality. The importance of such standards cannot be understated. Without them, a flood of new and continually reinvent¬ ed roof system components would likely appear. With no requirements or performance characteristics established, the initial benchmarks of quality would, in some instances, be forgotten. As consultants and design professionals, we take pride in specifying quality materials for use on our projects. This provides the roof consultant, contractor, and building owner with reasonable assurances that the materials are consistent with established standards and should contribute favorably to the expected long-term performance of the roof assembly. However, the true measure of quality cannot be judged by materials alone. A roof system of good quality is not as simple as “green side up.” The foun-dation of quality cannot be properly represented without acknowledging the importance of installation procedures and workmanship issues. Typically, the end result of any sampling program, pursuant to the recognition of product quality, is based on establishing, acknowledging or improving the characteristics of a single component, or, in the case of a roof assem¬ bly, a combination of components that rely on one another for optimum performance. Most would agree that a responsible sampling program, in conjunction with construction observation, can improve the quality of the installa¬ tion. Many of the test methods available are probably not routinely used or considered appropriate for the purpose of quality control as they relate to the role of the design professional. For example, it is doubtful that many of us have or would intend to sample fiberglass felts, in increments equal to one half the cubed root of the total shipment, as described and required by ASTM D-146, Test Methods for Sampling and Testing Bitumen-Saturated Felts and Woven Fabrics for Roofing and Waterproofing. 1 Our acceptance of individual components is typically limited to a cursory review of submittals and checklisting of materials against those required by the project specifications. For the most part, it is assumed that the materials specified and provided meet the performance criteria of the referenced standards as advertised and promoted on the product labels. We trust that the manufacturers are “minding the store” related to quality through testing at the production level. Assuming that, through practical experience, your project documents exclude products that may be considered “also rans,” or those materials with limited, or perhaps less than desirable performance histories, the products on your jobs will be of top quality. It is likely that huddled on a pallet, tightly bound in their factory wrappers, these individ- Kilpatrick-1 ual components are just fine. When the wrapper is removed, all bets are off, as the singular material properties and qualities initially represented can be compromised by the integrity of the installation. The installation of these materi¬ als could be considered the equivalent of a fire and forget weapon. There is but one chance to get it right, and correc¬ tive measures can be costly. The storage, handling and installation of the roof system components are the responsibility of the contractor. The premise of quality, as it relates to the finished roof assembly, is largely contingent upon the contractor’s ability to install the specified components in compliance with the requirements of the project specifications. As the quality of products, system selection and design can vary, so, too, can the means and methods of the installing contractor. The amount of product or system testing to be conducted in conjunction with construction observation is limited only by your budget and “need to know.” Most clients would not be interested in sponsoring a series of tests to deter¬ mine the kerosine number of asphalt-saturated organic felt. While this require-ment, or standard, has a basis, presum¬ ably adopted because of some proven or desirable characteristic relative to performance, we generally would not pro¬ mote this type of testing, the cost of which would ultimately be passed on to the client. The availability of recognized tests that can be done economically, with meaningful results, and the potential to acknowledge or improve the integri¬ ty or quality of the installation are few in number. The following are some of the test procedures and inspection methods, specific to the installation of built-up roofing, that can be used to assist in determining certain qualities of individual components or assemblies. BITUMENS Bitumen, or asphalt, temperature at the point of application is perhaps one of the most important and often over¬ looked elements of a quality installation. Significant industry-sponsored research has been done, determining that EVT, or equiviscous temperature, plays an important role in the contractor’s ability to apply the materials in compli¬ ance with subject tolerances. Operating temperatures of tankers and kettles need to be monitored to assure the proper temperature at the point of application. Different EVT temperatures for like materials have been adopted for applica¬ tions using both mechanical spreaders and hand mopping. The research indicates that operating within the limits of the described EVT range will improve the mopping characteristics of the bitumen and result in more uniform interply applications of materials. Check the temperature of the bitumens at the heated source and point of application. Asphalt temperatures in tankers and kettles should not exceed the finished blowing temperature (FBT) (approximate¬ ly 490° F) for extended periods of time. Overheating can result in fallback, or a change in the softening point of the material. Where applicable, require that the bitumens be installed within 25° F of the EVT. For safety reasons, avoid the heating of asphalt to within 25° F of the flash point. The absorpton of interply asphalt into the receiving substrate, caused by operating temperatures well above the E.V.T., can easily be seen with the benefit of a roof sample test cut. Bitumen temperatures at the point of application need to be monitored. Kilpatrick-2 The picture framing in this new built-up roof system is the result of application temperatures well above the materials’ published E.V.T. Samples of bitumen should be taken on the first day of production at the point of application. Submittal samples from the supplier and chunks off kegs are not representative of materials at the point of applica-tion. Asphalts are graded and categorized by softening and flash points, penetration ratings, ductility and percent solubility in trichloroethylene, all of which are referenced in the body of ASTM D- 312,2 Standard Specification for Asphalt Used in Roofing. ASTM D-36, Standard Test Method for Softening Point of Bitumen (Ring-and-Ball Apparatus),3 is a relatively simple test that can be used to determine if the appropriate grade of asphalt has been delivered to your job site. Failing that cri¬ teria alone is cause for rejection of the lot, or ship¬ ment, represented by the sample. This can be signif¬ icant, especially when the roof system construction involves structural slope that requires a specific grade of mopping bitumen. FLOOD COAT AND AGGREGATE SURFACING The importance of flood coat and aggregate surfacing cannot be understated. Flood coat and aggregate provide the first line of defense for the underlying felt plies. The bitumen, or flood coat, is the initial waterproofing compo¬ nent of the conventional built-up roof assembly. The exposed aggregate surfacing protects the bitumen flood coat from damaging solar radiation. While there are advantages to the selection of the aggregate surfaced built-up roof, its selection or specification is contingent upon structural capacities and can be driven by budget limitations. Established industry guidelines suggest that flood coat, installed at a rate of 60 pounds per square, with 60 per¬ cent, or 240 pounds of the required 400 pounds per square, of applied aggregate embedded is desired. The contrac¬ tor’s ability to install these materials to subject tolerances is influenced by the size of the job and application tech¬ niques. The uniformity and coverage rates can vary significantly if the contractor is required to use pour cans and shovels to distribute the materials versus hot spreaders and gravel buggies. Regardless of the methods available or employed, it is imperative that this element, which is the primary protec¬ tion for the underlying built-up roof, be applied in a manner consistent with the aforementioned industry standards. Currently, there are no ASTM standards that quantify the applied material as it relates to flood coat and aggregate surfacing, aside from ASTM D-2829 (Standard Practice for Sampling and Analysis of Built-up Roofs).4 That proce¬ dure requires test cuts from finished roof assemblies and is typically thought of as being reserved for use on those roof systems exhibiting performance problems. That leaves the flood coat and aggregate open for interpretation relating to the quantity of applied materials. The following guidelines can help determine if applied quantities are appropriate: 1. When mechanical distribution is used, the aggregate dropping into the hot asphalt should push asphalt, result¬ ing in a wave. 2. A 60-pound per square flood coat is approximately 1/8″ thick. A wet film thickness gauge similar to that used in the protective coatings industry can be used to assist in determining that the applied flood coat is approximately 1/8″ thick. 3. Double flood and gravel means just that. The initial application is provided at the rates previously described. The loose gravel is swept away after the asphalt has cooled, followed by a second application, or the equivalent of 120 pounds per square of asphalt and 700-800 pounds per square of aggregate. This is typically required around mechanical units, roof hatches, or areas where heavy traffic is anticipated. Kilpatrick—3 SAMPLING OF NEW, UNSURFACED, BUILTUP ROOF MEMBRANES Much has been said about the practice of sam¬ pling and analysis of new built-up roof systems at the time of application. References to the perfect square in studies sponsored by the industry suggest that this type of testing lacks credibility and the derived results vary significantly.7 One such study based their findings on samples taken from a test area 5′ x 13′. Reportedly, the construction of the test area was closely monitored using state-of-the-art application procedures. Based on the study, the author determined that the foundation of quality is based on good roofing practices and application procedures, a philosophy with which most would agree. The results challenged the validity of roof sample test cuts when used to establish parameters for quality control. Standard handling and dispensing equipment commonly used in built-up roof system construction. Both the lugger and felt layer are equipped with thermometers. The practice of sampling new built-up roofs is misunderstood and misrepresented. It is not intended to establish the parameters of quality control. It is a means by which to establish some level of confidence that the contractor has the ability to install the specified materials in compliance with accepted industry standards. The aforementioned parameters of quality are self-imposed by the industry and manufac-turers, both of which have adopted and promoted specific application rates for materials. All major component manufacturers publish certain guidelines that are to be followed when installing their products; more specifically, average mopping tolerances, the uniform distribution of applied materials and application tempera¬ tures. It can be said with reasonable certainty that these guidelines are the result of some performance-based criteria, information obtained from samples or systems that did not perform as intended, or, perhaps, research that has been done as the roofing community continues to develop industry-recognized standards. Without the benefit of test cuts, many of these industry standards would be unverified and possibly forgotten. A good example of misrepresentation, as it relates to roof sample test cuts as a means to confirm compliance with project specifications and industry standards, can be found in the roof management program that was, at one time, used by a branch of the military. 8 From the early 1980’s through the early 1990s, they had a program that required samples of the finished roof assembly, flood and gravel and/or cap sheet surfacing inclusive, be forwarded to an inde¬ pendent lab for acceptance testing. The testing was to be done in accordance with ASTM D2829. 5 The results of the analysis were reported on a form that was set up with columns reflecting the required generic identification of com¬ ponents, specified tolerance for applicable quantities and a suggested action to be considered for each element that did not meet the requirements of the project specifications. The suggested actions were listed in a Table of Suggested Actions Based Upon Sample Reports.9 That table was, by all accounts, a brutal document. For example, if the report indicated that the interply mopping weights were less than or exceeded those required by the project specifications and there were voids in excess 5.0 sq. in. within any single interply, the suggested action was to remove and replace the roof system. If that 5.0 sq. in. of void were encountered in a 4″ x 36″ sample, with five 2.0″ headlaps, it would be equal to approximately 1.0% of the interply area. By today’s standards, this would be considered acceptable work¬ manship. It is easy to see why the industry has an aversion to the practice of roof sample test cuts as a means to determine the approximate quantity of applied materials. A responsible sampling program, that is intended to acknowledge or potentially improve the quality of the roof membrane, as it relates to the quantity of applied materials, must be initiated at the face of the work, prior to the introduction of the surfacing. If, for example, a sample of a given day’s work was weighed, and the calculated aver¬ age mopping came in at 15 Ibs./sq., the presence of surfacing materials would all but eliminate the option of installing an additional two plies as a recommended corrective action. While some would argue that a void-free Kilpatrick—4 The handling and storage of the roof system components, ele¬ ments that can contribute to or detract from a quality installa¬ tion, are the responsibility of the contractor. a quality that should not be overlooked or left to chance. assembly with 15 lb. moppings will perform just fine, it is close to one-half of that recognized as the high end of the accepted industry standard for inter¬ ply moppings and is not representative of what would be considered appropriate material quantities. That is the objective of ASTM D-3617, Sampling and Analysis of New Built-Up Roof Membranes,6 currently under revision, to provide a means to sample and analyze built-up roof systems at the time of application, with an emphasis on the quantity of applied materials and the integrity of the membrane components before the introduction of surfacing. The intent is to sample the roof system with construction underway, share the results with the installing contractor so as to either acknowledge or improve the quality of workmanship, or to pro¬ vide a basis for any adjustments that may be required in the application of the materials relative to the desired quantity. Related to built-up roof sys¬ tem construction, the quantity of applied materials is The first quantitative element is headlap, or felt ply coverage. This characteristic of the assembly can be estab¬ lished without the benefit of a test cut and is easily determined with a measuring device configured to represent the 36″ or, 1 meter, width of the applied rolled goods. A specially fabricated template, with attached increments displayed in tenths of inches, works best, as it will later be used in the sampling process. Template size is based on the width of the installed materials. Standard tape measures and even remnants from butt rolls of the installed product can also be used with similar results. The chosen measure is moved across the exposed upper felt edges within the limits of the selected sample area. Deficiencies, or negative headlap conditions, are immediately recognized through this simple process. The measured headlaps, as encountered in the random sample area, are recorded on a worksheet for use in determining the approximate quantity of applied materials. Headlap is but one of the two variables required to per¬ form the necessary calculations. The remaining variable is weight Place the template on the predetermined sample area, or that where the headlaps were measured. With a utility knife, follow the template with repeated strokes on all four sides until such time the blade has cut through all felt plies into the receiving substrate. Carefully push a screwdriver or similar tool into one comer of the sample test area and pry up. This provides the opportunity to pull the sample out of the roof system. The sample is then weighed and the average mopping, or approximate quantity of interply materials, can be calculat¬ ed. Some would stop here and chose not to assess the integrity or quality of workmanship as it relates to characteris¬ tics, such as uniformity and distribution of interply asphalt and voids. To some, a four-ply sample, with extrapolated results reflecting 27-pound interply moppings and 2.0″ headlaps, would be adequate as a basis for acceptance. By simply looking at the sample section, additional information can be obtained. For example, the roofing con¬ tractor chooses to install mopping asphalt at temperatures well above the EVT. It is not uncommon to advise the fore¬ man several times to turn off the burners if the measured temperature at the point of application is 30-50° F above the product EVT. It’s summertime with clear skies and ambient temperatures hovering around 95( F. The foreman argues, “Yeah, but look at that nice bleed through.” The bantering continues, and the guy pulling the felt layer starts to expound upon his personal experience with the three different kinds of steep asphalt. Your repeated attempts and requests for the crew to more closely monitor the asphalt temperatures at the point of application have been ignored and, on some occasions, ridiculed. You may place a call to the project superintendent to explain your plight and con¬ cerns, and he may choose or choose not to respond. Kilpatrick—5 Through sampling, your concerns over application temperatures will be self-evident. Having looked at hundreds of samples, some of which were undoubtedly put together with operating temperatures well above the EVT, the absorption of mopping asphalt into the perlite or wood fiber substrate is easy to see. Keep in mind that 1/8″ thickness of asphalt represents about 60 Ibs/sq. A sample section revealing absorption approaching 1/8″ is an indication that the materials were installed at temperatures well above the EVT. In some instances, we have quantified the bottom mop, or that portion of the interply mopping that interfaces with, and was absorbed by, the insulation substrate, due to elevated temperatures at the time of application, and deducted it from the sample gross weight. In theory, approximately 75% of this mopping asphalt was initially installed as an interply component of the assembly. Using the traditional method, that which does not address applica¬ tion temperatures and absorption, a sample gross weight of 145 lbs/sq., with five 2.0″ headlaps would yield an aver¬ age mopping of approximately 27 lbs/sq. By isolating, quantifying and deducting the bottom mop from the gross sample weight determined to be in this example, 60 Ibs./sq., the average interply mopping is reduced to approximate¬ ly 17 lbs/sq. This information can easily be shared with the contractor to demonstrate the need to comply with the established guidelines related to EVT at the point of application. Now you can explain to the guy running the felt layer your experience with the three kinds of steep asphalt, too hot, too cold and just right. Two more plies please. Freezing the sample with dry ice makes it possible to separate the felt plies and provides the opportunity to quan¬ tify the bottom mop, as described above, and further document the interply characteristics, such as voids, and the presence of moisture or debris. Routine foot and cart traffic across a recently laid-up membrane, that which has not had an opportunity to cool or set, will result in asphalt displacement and voids. TOOLS Communication and documentation are to be considered tools that are to be used as part of the construction observation process. Communicate with the foreman on daily basis. Let him know what your expectations are. Communicate with the building owner’s representative. Be careful with your presentation of issues, real or imagined, associated with what may be perceived as less than desirable performance on behalf of the installing contractor. Respect the contractor’s position with the owner and roofing community as they, too, are continually trying to devel¬ op relationships with existing and future clients. If blatant performance issues are brought to the attention of the fore¬ man and reasonable efforts are not made to correct the problem, document the occurrences in a letter addressed to the president of the roofing company. Staff your projects with qualified individuals. The construction observer should be familiar with all aspects relat¬ ed to the installation of subject materials and applic¬ able industry standards. Provide your construction observation representatives with the necessary tools, resources and training. Be proactive, work with the contractor. On reroofing projects, make test openings at critical details in the existing assembly a part of the con¬ struction observation process. This allows confirma¬ tion of existing conditions related to flashing heights and blocking configuration, as to determine if the intended design will perform as intended and fosters the development of a team effort towards the common goal of a quality assembly delivered on time. This extra effort reflects favorably on both the construction observation entity and contractor. Review of the details in advance provides the opportunity to minimize “time lost” due to incon¬ gruities. Thermometers on handling and dispensing equipment, in some instances, have little to offer as it relates to material tempera¬ ture. Kilpatrick—6 The following includes some of the equipment that should be available for use by those providing the services of construction observation: Thermometer – The thermometer should dis¬ play in degrees Fahrenheit, with a working range of up to 500° F. Pocket styles are available with either analog or digital displays. Some of the digital mod¬ els available are equipped with the ability to store for future reference the minimum and, perhaps more important, maximum temperatures. This feature pro¬ vides the observer with the opportunity to share the results of the given readings with the foreman. Moisture Meter – Moisture meters equipped with probes of varying lengths can be used to deter¬ mine the presence of moisture in roofing materials. The widely used capacitance-type meters require periodic calibration against known standards. It should also be noted that readings obtained do not necessarily reflect actual moisture content. This type of correlation would typically require the oven drying of moisture-laden materials across the spectrum readings obtained. Review of the sample results with the foreman can acknowl¬ edge the contractor’s efforts and potentially improve the char¬ acteristics of the finished built-up roof system. Micrometer – Micrometers can be used to determine if the required gauge of sheet metal has been provided. Charts are available that reflect the relationship of material thickness to sheet metal gauge. More durable, circular standard gauges should also be given consideration. They are made of machine-quality steel, with a series of notches, representative of varying thicknesses of metal. A clean edge of the subject material is slid into the notch that most closely matches the thickness of the sheet metal. A magnet can be used to quickly determine if the metal is G.I. or aluminum. Cell Phone – The ability to call the office of the designer of record and discuss details which may be different than those reflected in the design documents is very important. A crew of a dozen or more roofers may be standing around waiting for direction as to how to best respond to the unforeseen conditions. Camera – Use a camera to assist in building a record of the roof project. Digital cameras can circumvent the costs and delays associated with 35mm photography, allowing the photos from the field to be delivered to the office of the designer in a timely fashion. Sample Bags, Buckets and Cans – Bags and buckets are commonly used to transport materials, such as aggre¬ gate, to the lab for grading tests. Clean one-quart cans with lids are used for bitumens. Retain built-up roof samples representing questionable workmanship or applied quantities of materials for reference and discussion with the pro¬ ject superintendent. They should be retained until such time that all interested parties are in agreement as it relates to remedial repairs and final acceptance. Tool Bag – A bag to carry miscellaneous hand tools, including, but not limited to, tape measure, screwdrivers, chisel, hammer, pliers, utility knife. Scale – Triple beam, spring and portable electronic balances are available to weigh roof samples. Whatever type of scale is chosen, it should be periodically calibrated against laboratory grade balances. SUMMARY Those assuring quality control, through or in association with the service described as construction observation, are mistaken and, perhaps, assuming liabilities that are the responsibility of the contractor. The quality of the installa¬ tion is the responsibility of the contractor. In some instances, the reference or use of the term quality control could be Kilpatrick—7 considered a misnomer. Give consideration to the occasional projects where the contractor is doing all the right things, the inherent characteristics of quality are represented at all levels. On these types of projects, the control of quality is, as it should be, in the qualified hands of those installing the materials. The use of the word control, in con¬ junction with quality, suggests that excessive, or too much, quality is something that we need to control. Individuals engaged in the construction observation process are faced with many challenges. The most basic ele¬ ment of our charge is to document the efforts of the contractor, good, bad or indifferent, as it relates to the installation of roofing materials. Without fail, through the course of the project, difficulties can and will arise. The appropriate response to these unexpected difficulties should be focused on resolve, fostering a spirit of cooperation and union of purpose, pursuant to the common goal of a quality roof installation. REFERENCES 1. ASTM, 1999 Annual Book of Standards, Volume 4.04, Roofing, Waterproofing and Bituminous Materials. 2. Ibid. 3. Ibid. 4. Ibid. 5. Ibid. 6. Ibid. 7. Cullen, Bill C., The Perfect Square: Can it be Built?, Roofing Spec 1987, pgs. 37-40. 8. United States Air Force, AFM 91-36, Built-up Roof Management Program. 9. Ibid. Kilpatrick—8
