Professional Documents
Culture Documents
FM Global Property Loss Prevention Data Sheets
FM Global Property Loss Prevention Data Sheets
Table of Contents
Page
©2001 Factory Mutual Insurance Company. All rights reserved. No part of this document may be reproduced,
stored in a retrieval system, or transmitted, in whole or in part, in any form or by any means, electronic, mechanical,
photocopying, recording, or otherwise, without written permission of Factory Mutual Insurance Company.
2-1 Internal Corrosion in Automatic Sprinkler Systems
Page 2 FM Global Property Loss Prevention Data Sheets
List of Figures
Fig. 1. Sample arrangement of wet sprinkler piping to prevent air accumulation in branchlines
and other high points of the system. ................................................................................................. 5
Fig. 2. Schematic development of MIC. ....................................................................................................... 7
Fig. 3. Corrosion at roll groove in a wet pipe system. ................................................................................. 7
Fig. 4. Pinhole leak site in cut-away section of pipe. ................................................................................... 8
Fig. 5. Pinhole leak site on outside surface. ................................................................................................ 9
Fig. 6. Localized nodule formation below water line. ................................................................................. 10
Fig. 7. Heavy tubercle formation in a 8 in. (200 mm) sprinkler pipe. ......................................................... 11
Fig. 8. MIC test kit and automatic delivery system – courtesy of Bio Industrial Technologies Inc. ........... 18
List of Tables
Table 1. Guideline for Retaining Pipe Based on Extent of Corrosion Damage ............................................ 4
1.0 SCOPE
This document addresses prevention and control of corrosion in automatic sprinkler system piping with
particular emphasis on Microbiologically Influenced Corrosion (MIC).
1.1 Changes
This is a new document.
2.1 Introduction
The following recommendations are intended to:
• Provide proper diagnosis of corrosion cases involving sprinkler systems;
• Attempt to keep existing sprinkler systems affected by corrosion free of piping obstructions;
• Minimize the exposure from possible water leakage losses resulting from MIC affected sprinkler systems;
• Provide advisory guidance on possible corrosion prevention measures that can be considered for any new
wet, dry and preaction sprinkler system. Recommendations for new systems should be considered based
on the occupancy and exposure present.
Locations wishing to pursue further corrosion mitigation than what is being recommended in sections 2.2
and 2.3 should follow the guidelines given in Appendix C of this document; these guidelines are advisory
in nature and provide a systematic approach to corrosion mitigation. While these advisory guidelines may
increase the chances of success in mitigating corrosion, and in particular MIC, they are not intended to rep-
resent a complete solution to the problem. Their primary goal is to ensure that the integrity of the sprinkler
system is not further compromised and to prevent loss of protection or further impairments to the sprin-
kler system.
2.2.1.2 Conduct metallurgical examination on a sample of the affected component to determine the type
and extent of corrosion mechanism involved.
2.2.1.3 Where the extent of corrosion damage to remaining pipe needs to be determined, use the FM Glo-
bal protocol for in-situ UT examination of piping given in section 3.3 below. Consider replacement of any
section of pipe exhibiting pit sites with remaining wall thickness less than shown in Table 1.
2.2.1.4 Further mitigation in the form or cleaning of the piping or treatment of the water is not presently being
recommended. However, in locations where these are being considered, follow the MIC Mitigation Guide-
lines in Appendix C.
2.3 Protection
to absorb pressure increases caused by thermal expansion. Set the relief valve to operate at pressures
not greater than 175 psi (12.1 bars) or 10 psi (0.7 bars) in excess of the maximum system pressure when
the maximum system pressure exceeds 165 psi (11.4 bars).
Fig. 1. Sample arrangement of wet sprinkler piping to prevent air accumulation in branchlines
and other high points of the system.
alternatives in cleaning of pipe and a better understanding of treatment options have made MIC mitigation
attempts possible. However, mitigation efforts in fire protection systems are still evolving and remain largely
empirical at present with no conclusive available field data on their long-term success. Conversely, severe
cases of MIC left unchecked in sprinkler systems can progress over time leading to an increased number of
leaks and possible obstruction of pipes and sprinkler heads.
Given the nature of some of the aerobic bacteria, MIC generally results in the growth of biofilms and nod-
ules within the piping. The biofilm produced will generally present itself as a black slime that, in certain cases,
can be flushed out. Nodules, on the other hand, are hard, well adhered to the pipe wall, and will not flush
out. Nodules and biofilm may result in increased friction loss, and affect the ability of the pipe to carry the nec-
essary flow. Figures 6 and 7 show typical nodules and tubercles formation in a sprinkler pipe sample.
In addition to the above, MIC also may affect sprinklers and valves, and block water motor gongs and alarm
lines and other components of a fire protection system.
Evidence of MIC was present in about 40% of all the cases involving sprinkler system components. The data
collected shows no indication that MIC is restricted to a particular geographical area, as MIC was evident
in components sampled from different geographical areas within North America and abroad. Also, there is no
evidence that MIC is restricted to certain types of pipes or sprinkler components. MIC was found in corro-
sion cases involving both black steel pipe and galvanized pipe of different pipe schedules as well as in cooper
piping. MIC also has been found in some cases involving corrosion of sprinkler heads, flexible steel hoses
and other sprinkler components.
case of corrosion in sprinkler pipes, the condition of the affected surface is more irregular, and usually
involves pitting. In sprinkler piping, corrosion pits are usually localized, and can sometimes be quite deep
in relation to their diameter. The evidence of pits is indicated by a numerical digital UT thickness gauge (if at
all) as instability of the numerical read out. Because the UT beam is being reflected irregularly by the shape
of a pit, a numerical digital thickness gauge does not receive a coherent reflection. In such cases, the read-
ing by an experienced operator of such a device might strongly suspect that a pit is being indicated. There
is no information obtainable regarding the size, shape, or depth of the pit, unless it is quite large or large
areas of the surface are affected.
3. Remove outside paint from the pipe prior to UT scanning.
4. Scan the entire circumference of the pipe in search for pits, with particular focus on the bottom of the
pipe between three o’clock and nine o’clock position.
5. Use any of the following modes of UT flaw detector display: A SCAN, B SCAN or C SCAN.
Ultrasonic examination instruments with cathode ray tubes, and more recently developed digital screens,
can display the information from a reflector in a variety of ways.
Modes of UT Flaw Detector Display:
• A SCAN
A SCAN displays are the most commonly used with portable UT flaw detectors. Ultrasonic reflectors are dis-
played on the rectangular screen as base line deflection spikes. The horizontal axis represents material
thickness, and the vertical axis represents the amplitude of the reflected energy. Other controls and options
enable an operator to precisely locate and measure small flaws.
• B SCAN
B SCAN displays are becoming very popular for monitoring thickness of tank and pressure vessel walls.
The rectangular screen displays a profile view of a wall’s thickness, and the shape and depth below the sur-
face of a flaw. In tank and vessel examinations, ultrasonic reflection data from all points scanned are recorded
and usually stored electronically.
• C SCAN
C SCAN displays data in a plan view. Images of pits would be displayed as round shadows on a flat sur-
face, similar to a radiograph.
Any one of these flaw detector displays would be sufficient for sprinkler pipe thickness survey/pit detection
examinations.
6. While using a flaw detector system to scan a section of piping, continuously monitor its relative thick-
ness and search for pits. Once pits are detected, stop and measure the pit depth, diameter and the
concentration of pits in a given area. Mark the location of the pit in the pipe with an indelible marker or paint.
7. Prepare a final report with the results of the survey indicating regions of the pipe scanned; location and
depth of the pits in the sprinkler plan.
3.5.2 Pinhole Leaks From Sprinkler System Over Data Processing Center
Pinhole leaks developed in a schedule 10 sprinkler piping system protecting a data processing center. The
leaks damaged records and computer equipment. This location was not insured with FM Global.
4.0 REFERENCES
4.1 FM Global
Data Sheet 2-8N, Installation of Sprinkler Systems.
Data Sheet 2-81, Fire Safety Inspections and Sprinkler System Maintenance.
4.2 NFPA
NFPA 13, Installation of Sprinkler Systems.
NFPA 25, Standard for Inspection, Testing and Maintenance of Water-based Fire Protection Systems.
where the water supply is known to be contaminated with MIC-causing bacteria and microbes. The second
group includes those activities to help mitigate or control MIC in existing sprinkler systems that are sus-
pect or known to have MIC-related problems.
10. Plan should identify what measures should be taken when system is flooded with untreated water after
a fire or accidental trip.
11. Records detailing the treatment agents, concentration, injection process for the initial treatment as well
as for all periodic monitoring should be kept on file and on a summary record sheet displayed on the riser
of each system undergoing water treatment.
C.2.4 STEP 4 – Installation of clean pipe and care during system acceptance
The last step in MIC prevention for new sprinkler systems is to ensure that all piping installed is disinfected,
and that proper hydrostatic test procedures are used. This step should be taken to minimize the possibility
that bacteria or microbes are present in the new piping to be installed or are introduced into the system acci-
dentally during hydrostatic test or other acceptance procedure.
Regardless of the condition of the water supply, all piping to be installed should comply with recommenda-
tion 2.3.1.1 above. Piping that is rusted or weathered should not be installed.
Where the water supply is treated, all hydrostatic tests and other acceptance test procedures should be done
using treated water in accordance with the guidelines established in Step 3 above.
C.3.1 STEP 1 – Diagnosis of the corrosion and of the condition of the piping
The first step involves determining the nature of corrosion in the piping and conducting a corrosion dam-
age assessment of the piping to determine extent of corrosion damage and feasibility of further treatment.
Diagnosis of the corrosion should be conducted per recommendation 2.2.1 above. Preferably, piping should
be sent in for metallurgical examination in the ‘‘as removed’’ condition, without any cleaning or removal of
internal deposits or substances, so that a metallurgical lab can analyze the nature of the residuals found in
the piping.
Assessment of the condition of remaining piping should be conducted per recommendation 2.2.2 above.
Any sections of piping that are damaged or that do not meet the criteria given in Table 1 should be replaced
prior to cleaning. Remaining piping should then be cleaned as outlined in Step 3 below.
5. Cleaning agents should circulate throughout the system to promote complete cleaning of all piping, includ-
ing all branchlines, pipe drops and sprig-ups, all risers, feedmains and crossmains.
6. After cleaning, metallurgical examination should be conducted on at least three more samples of piping
removed from the riser, mid-branchlines and inspector test connection. This examination should determine the
extent and uniformity of cleaning performed.
7. Chemical cleaning solutions should be NSF approved.
8. Cleaning agents should not result in damage to any of the metals present in sprinkler piping or damage
to natural or synthetic rubber, silicone, or any other elastomeric products used in gaskets, valve seats, o-rings
and other nonmetallic parts in a sprinkler system.
9. No residual cleaning agents or solutions should be left in the system after cleaning.
10. Any cleaning should be immediately followed by a treatment of the water. The treatment plan should
be in accordance with FM Global Guidelines for Water Treatment for Sprinkler systems, given above.
prescribed dosage of treatment solution to the system automatically upon flow in the system. The cost of
these delivery systems varies but should be in the range of US $3,500 for well-equipped delivery systems
uninstalled.
Fig. 8. MIC test kit and automatic delivery system – courtesy of Bio Industrial Technologies Inc.