Deliberate Actions

An Unknown Threat Unexpectedly Revealed Through Ultrasonic Investigation

Preface

A high rise 54 story office building changes owners.  When the new chief engineer takes over, he notices what seem to be abnormally high rust deposits at the package air condition units served by a central condenser water system.  Soon after, CorrView International, LLC is hired to perform an ultrasonic investigation of the condenser water system.  The building is approximately 20 years old and has been well maintained.

Investigation

Our investigation shows low to moderate corrosion activity of near 1-3 MPY throughout all areas tested.  Like many more modern buildings, multiple futures have been installed at each floor but have never been used.  For some floors we identify one set of larger 4 in. distribution lines serving the floor A/C unit, and then 3-4 additional dead ended lines ranging in size of 2 in. to 3 in.  In some examples a small copper by-pass line has been installed across the dead ends in the hopes of mitigating damage caused by no water flow.  Most, however, are valved off and capped.  In reality, such efforts are worthless and do not slow the corrosion rate.

Investigation immediately discovers a significant threat at the capped future lines.  In addition to higher corrosion and pitting activity, brass valves end each line and have had a noticeable galvanic impact against the steel pipe.  We identify and photograph multiple locations showing rust encrustation and bluish green discoloration around the valve side of the future which defines galvanic activity.  Some wall thickness measurements are not only below minimum standards, but near the limit of the thread cut itself.

We recommend the identification and replacement of all threaded pipe throughout the entire building, and highly recommend using either dielectric fittings or heavier schedule 80 black steel pipe.

During our investigation we are advised to the existence of an above ceiling located 4 in. distribution line running the entire length of the third sub-cellar where the building offices are located.  Testing in this area identifies the most severe pitting of our entire investigation and shows that the pipe is very near failure.  Thickness values in many areas are near 0.115 in. or less, and with the thread cut for 4 in. pipe at 0.105 in., very little wall thickness remains for pipe operating under a pressure of near 400 PSI.

We bring this information to the attention of the chief engineer and further warn him that while performing our testing at the very bottom of the main risers where the 4 in. distribution pipe originated, there were no isolation valves.  The chief engineer expresses shock at our statement, and upon inspecting the riser pipe personally, finds that the very heavily damaged 4 in. pipe has been simply welded into the side of the main risers.  Any failure of this 4 in. pipe will drain both risers and cooling towers into the basement of the building.

Within a few days, the engineer performs an emergency installation of new 4 in. isolation valves at the bottom of the risers by freezing the 8 in. lines.  This provides some safety margin while preparations are made to replace hundreds of feet of the 4 in. pipe routing through multiple offices.   Plans are also underway to shutdown the risers and replace some of the threaded pipe.

Conclusion

OLYMPUS DIGITAL CAMERA

Upon cutting apart the 4 in. sub basement distribution piping, engineers find multiple lengths of 1 x 4 wood studs inserted into both supply and return lines.  For whatever reason, and likely due to some form of dispute with the contractor or between employees and their firm, steam fitters deliberately inserted lengths of wood into the pipe before threading it into place.  This obstruction then slowed the flow of water down to explain the long term balancing and operation problems the engineers had not been able to correct.

Slow flows over the prior 20 years had led to the settlement of massive amounts of rust and particulates creating a severe under deposit condition resulting in the wall loss measured ultrasonically.  Some lengths of 4 in. pipe fracture at their threads and separate when handled.

Months later, when the dead ended futures are replaced, engineers find that many simply snap off from the threadolets with only minor handling, requiring a re-cutting of the threads.  Our final written report from ultrasonic testing had suggested this may occur, and we had recommended having thread cutting dies available in the different size pipe they were planning to replace.