In the roofing industry, several everyday design practices and jobsite tasks can, and typically do, have a major impact on the proper completion of a roof system. These items include the following critical elements:
From the perspectives of a roof consultant specifying the roof assembly and an observer conducting a roof observation site visit, let’s take a closer look at each of these items to better understand their impact on a properly installed roof system and identify best practices for the design and installation of thermoplastic roofing systems. But first, we need some background information on roof insulation assembly selection.
ROOF INSULATION ASSEMBLY SELECTION
Specifying the correct insulation type, R-value, assembly, and attachment method for the project is an important step in the process of selecting a roof insulation assembly. The following are important questions to answer in the process of determining the appropriate insulation assembly and attachment criteria:
I will leave the details of this selection process for another article. Suffice it to say, there are numerous considerations and decisions that affect the insulation selection and attachment for any given roof assembly.
ROOF INSULATION STORAGE AND INSTALLATION
Roofing consultants and observers frequently find that roof insulation is stored directly on the ground or left unsecured, or they observe that the weatherproof covering is not secured and is only partially protected from sun, wind, and rain. The following are among the many reasons to make note of these scenarios and document them in the jobsite’s third-party roof observation reports:
Although warped insulation boards can be replaced after the roof is completed, the repair costs are high and large patches must be installed to accommodate the size of the repair or replacement. The best practice is not to install the damaged insulation in the first place, as stipulated in the roofing manufacturer’s specifications.
INSULATION INSTALLATION
In recent years, a popular insulation and cover board installation procedure has been to adhere both the insulation and cover board with low-rise expanding foam adhesive. This adhesive can be applied directly to a concrete deck in some cases; alternatively, it can be applied over a vapor barrier or substrate board, or over a metal or wood deck. There are a number of other ways in which low-rise foam adhesive can be incorporated into the assembly, such as adhering subsequent layers of flat or tapered insulation adhered over base layers of insulation and cover boards, and crickets adhered over previously installed layers of insulation (see Fig. 2).
The bead size and spacing requirements can be based on a variety of design criteria, the geographic location, and wind-speed requirements. Bead size and spacing requirements vary by manufacturer, project-specified wind rating, and other design criteria. Roof manufacturers specify minimum size and spacing criteria for construction, which do not require specific wind ratings (see Fig. 3).
Roof system manufacturers have varying specification criteria for installation. Many manufacturers require that insulation boards be placed in the wet adhesive before the adhesive “skins” over or starts to dry. Some manufacturers recommend that the worker placing the insulation or cover board into the adhesive “walk in” the boards (walk across the board after placing it in the adhesive), and some manufacturers recommend that the workers roll the board in the adhesive with a weighted landscape roller.
However, I have learned from personal experience that there can be undesirable results when these manufacturer-prescribed methods are followed. When boards are walked in or rolled with a landscaping roller, the foam adhesive may not be held in solid contact with both the substrate surface and the underside of board being placed on top while the foam adhesive is curing. This can be evidenced by the installed board having raised edges and corners, uneven boards, lack of solid adhesion, or loose boards when boards are installed across transitions in slope, crickets, or saddles. Any raised edges or corners of the boards could telegraph through the finished roof surface or result in unadhered areas of insulation or cover board—or both of these problems could occur. You may feel loose boards while walking across the roof. These boards will typically move downward when you step on them, indicating that they are not adhered to the substrate below (see Fig. 4 and 5).
These conditions can be further complicated with other installation errors such as installing fasteners and plates to hold down raised edges and corners. For example, the thermoplastic olefin (TPO) membrane system shown in Fig. 6 was meant to be an adhered TPO membrane over adhered coverboard and adhered polyiso. The contractor’s repair with fasteners and seam plates introduced thermal bridging into this assembly, and the fasteners penetrated the vapor barrier. In my personal experience, if the installed fasteners and plates are installed at the joint between two boards using 2-in. (50-mm) seam plates and covered with TPO membrane patches, there is also a considerable risk that the fasteners and plates may not hold down the loose or cupped boards over the long term. The introduction of insulation plates and fasteners also means this assembly does not qualify for the specified hail coverage under the warranty. Adding metal insulation plates introduces a very hard surface directly below the roof membrane. The roof manufacturer’s hail warranty terms and conditions are very specific in requiring the roof substrate to be adhered and not mechanically attached. These conditions also place the roof system at risk of high-wind damage. The particular system depicted in Fig. 6 was specified to comply with FM 1-90 attachment criteria according to FM Global’s data sheet 1-28.2 Another article in this series will explain how wind-uplift testing can be used to determine whether those criteria have been met.
On projects where my firm is the designer of record, we regularly recommend that the low-rise foam-adhered boards be temporarily ballasted
using pails filled with adhesive, cinder blocks, or other portable ballast available on jobsites, such as buckets partially filled with concrete, old toolboxes filled with concrete, or bucket or boxes of fasteners or attachment plates. Temporary ballast is meant to provide uniform compression of the foam adhesive beads, spreading the foam beads out and ensuring that the top board remains in contact with the foam adhesive and substrate while the adhesive is curing. Figure 7 presents an example of installed results when the adhered insulation and coverboard are temporarily ballasted while the foam adhesive is curing. There is minimal visual evidence of raised, curled, or cupped roof boards.
When mechanically attached roof insulation is specified or required by the designer of record for projects where hail warranties are not required, there is also a risk that nonstandard or noncompliant installation techniques may be used. Fastening patterns must be established or specified to meet building code and FM requirements, as well as the roof manufacturer’s warranty prerequisites.
Common defects found and issues to identify include the following (refer to Fig. 8 and 9):
MEMBRANE SELECTION
The most common thermoplastic single-ply membrane types are TPO per ASTM D6878, Standard Specification for Thermoplastic Polyolefin Based Sheet Roofing3; polyvinyl chloride (PVC) per ASTM D4434, Standard Specification for Poly(Vinyl Chloride) Sheet Roofing4; and ketone ethylene ester (KEE) PVC per ASTM D6754, Standard Specification for Ketone Ethylene Ester Based Sheet Roofing.5
Most thermoplastic membrane sheets are available in several thicknesses such as 45 mil, 50 mil, 60 mil, and 80 mil; even thicker membranes are available when “fleece” or other similar materials are laminated on the underside of the membrane.
Thermoplastic membrane sheets are internally reinforced. Formable flashing is generally not reinforced so that it can be molded to fit penetrations, inside and outside corners, pitch pans, and complex penetrations such as angle iron or I-beams.
The sheets typically come in white, which is very reflective and, in most cases, meets the reflectivity criteria defined by the Cool Roof Rating Council (CRRC) Product Rating Program Model6; ANSI/CRRC S100, Standard Test Methods for Determining Radiative Properties of Materials7; Energy Star8; and LEED.9
These membranes also come in other standard, regularly manufactured colors, including tan and gray. Custom colors can be ordered; check with the roof manufacturer for the minimum order quantity. Reflectivity ratings by color may vary by manufacturer.
Along with membrane thickness and reflectivity, the following are other factors to consider when selecting membrane products for a project:
MEMBRANE INSTALLATION ISSUES
Rooftop Staging/Loading Points
Rooftop loading access points are an important area to observe closely. On some commercial roofing projects, there can be other trades accessing the roof and loading materials and debris onto and off the roof. In many cases, these trades use the same ground-level staging areas used by the roofing contractor because those areas are conveniently located relative to the location of the jobsite crane or material lift (see Fig. 10).
Some means of protecting the roof assembly from damage should always be in place at loading areas. Loose-laid insulation covered with plywood or similar is a good protective measure. Plywood sheathing over a slip sheet or extruded polystyrene insulation board may be used.
However, we have all been in situations where protective measures were not taken. In these cases, there can be visual evidence of damage on the roof membrane such as scratch marks, punctures, debris, indentations into the roof assembly, damaged flashings, and crushed insulation. This damage should be marked, dated, and temporarily repaired when found (see Fig. 11). Permanent repairs should be completed once the client, owner, or general contractor has determined that all loading and unloading activity has ceased.
Most commercial roofing manufacturers specify that no more than 10 patches should be installed in any one roofing square or 100-ft2 (9.3-m2) roof area. When more than 10 patches are located in any 100-ft2 (9.3-m2) area, one large patch must be installed to cover or replace the damage. The decision whether to repair, cover, or replace damaged membrane depends on the severity of the damage, as defined by the roof manufacturer.
It is recommended to repair scratch marks in the roof membrane as well as cuts and punctures. These scratches can grow or expand through
expansion and contraction forces in freezing-and-thawing cycles as well as normal expansion and contraction with temperature fluctuations. Single-ply membranes are flexible and experience dynamic movement during their life spans. As the membrane around the scratches expands and contracts, the scratches can penetrate through the reinforcement and bottom layer of membrane.
Heat-Welded Thermoplastic Field Seams
In my experience, the procedure to heat-weld thermoplastic field seams has several important steps. The generator used to power the robotic seam-welding equipment is of utmost importance and must match the continuous wattage output required in the welder manufacturer’s specifications (see Fig. 12). This generator also must not be used to power any other equipment while it is powering the robot welder. Additional power drain on the generator by hand welders, screw guns, or other electrical tools could cause power surges and power drops, which can be detrimental to field seam-welding quality (see Fig. 13). The specification on the continuous wattage varies by welder manufacturer. Please refer to the welder equipment manufacturer’s technical data for specific generator wattage requirements for hot-air welding equipment.
Other factors that will affect field seam-welding quality are sunlight, wind, shade, ambient temperature, and humidity. It is imperative that the designated robotic welder operator is thoroughly trained and familiar with the equipment being used.
Roof manufacturers recommend that the subcontractor perform test welds before welding actual field seams. Test welds consist of the following steps:
A good weld will be demonstrated by the exposure of a solid 1.5- to 2-in.-wide (38- to 50-mm) area of reinforcing scrim, as shown in the scrim on the right side of Fig. 14. It is important for the designated welder operator to perform these test welds each time they start up the equipment and after the equipment has achieved the operating temperature. The time of day, direct sunlight versus indirect sunlight (shade), high winds, or cloud cover versus sunshine can all affect the temperature, speed, and overall weld quality.
Hand-welding seams and flashing details are at least as important as, and often more important than, robotic seam welding. The hand-welding-detail roof technicians must be well trained and experienced in the use of a hand welder. Hand welding requires a great amount of patience and skill. It cannot be rushed, as any attempt at expediting the process can result in poor welds, which can permit moisture intrusion into the roof assembly. Figure 14 provides an example of a poor (cold) weld. A cold weld may visually appear to be a good weld/splice, but minimal pressure from a seam probe or wind can cause a cold weld to open and fail.
The detail technicians should also perform test welds to determine optimum welder temperature, the proper speed at which to move along while welding, as well as the appropriate pressure to exert with the 2-in. (50-mm) handheld seam roller. Figure 15 shows a hand-welding example. Typical hand-welding equipment can be seen in Fig. 16.
All welded seams—whether completed with an automatic welder or hand welding—must be probed by the end of the day, every day. The welded seam must be allowed to cool before probing. Probing is accomplished with a roof manufacturer-supplied seam probe (Fig. 17) or a common cotter pin puller tool. The tip of the tool is placed along the splice edge, and light pressure is applied against the splice while the tool is pulled along the length of the splice (Fig. 18). Any defective (cold) splices or wrinkles will open up with minimal pressure from the probe. All deficiencies should be properly cleaned and repaired per the roof manufacturer’s specifications.
Gary Gilmore, RRO, REWO, CIT Level I, is director of the Roof Consultant Group, LB Pie Consulting & Engineering, in Texas, where he is responsible for overseeing and executing roofing and building enclosure assessments, infrared scanning, design, contract document review, quality assurance observations, and field performance testing services. Gilmore has extensive experience working with owners, architects, general contractors, and trade contractors, assisting them in selecting and installing roofing and facade systems that are appropriate for their specific project needs with regard to building code and energy code requirements, building type and occupancy, and cost constraints. He has direct experience in field installation of roofing and cladding systems obtained through his early career on the contractor and manufacturer representative side of the industry.
This is the first article in a several-part series about thermoplastic roofing systems.
REFERENCES
1. American Society of Civil Engineers (ASCE). 2016. Minimum Design Loads and Associated Criteria for Buildings and Other Structures. ASCE 7-16. Reston, VA: ASCE.
2. FM Global. 2021. Wind Design. Property Loss Prevention Data Sheets 1-28. Reston, VA: Factory Mutual Insurance Company.
3. ASTM International. 2019. Standard Specification for Thermoplastic Polyolefin Based Sheet Roofing. ASTM D6878/D6878M-19. West Conshohocken, PA: ASTM International. doi: 10.1520/D6878_D6878M-19.
4. ASTM International. 2021. Standard Specification for Poly(Vinyl Chloride) Sheet Roofing. ASTM D4434/D4434M-21. West Conshohocken, PA: ASTM International. doi: 10.1520/D4434_D4434M-21.
5. ASTM International. 2015. Standard Specification for Ketone Ethylene Ester Based Sheet Roofing. ASTM D6754/D6754M-15. West Conshohocken, PA: ASTM International. doi: 10.1520/D6754_D6754M-15.
6. Cool Roof Rating Council (CRRC). 2021. Product Rating Program Model. CRRC-1. Portland, OR: CCRC. https://coolroofs.org/documents/CRRC-1_Program_Manual.pdf.
7. CCRC. 2021. Standard Test Methods for Determining Radiative Properties of Materials. ANSI/CRRC S100. Portland, OR: CCRC. https://coolroofs.org/documents/ANSI-CRRC_S100-2021_Final.pdf.
8. Energy Star. n.d. “Energy Star Product Finder.” Accessed September 16, 2021. https://www.energystar.gov/productfinder/product.
9. LEED. https://www.usgbc.org/leed.
Please address reader comments to chamaker@iibec.org, including “Letter to Editor” in the subject line, or IIBEC, IIBEC Interface, 434 Fayetteville St., Suite 2400, Raleigh, NC 27601.