Chillers & Boilers in the Same Room: A Cautionary Tale.
1. There are two existing boiler flues that serve two boilers in the basement of the building, which are routed from the basement through three stories within a vertical chase and penetrate the roof.
2. The flues are listed Category IV double-wall special gas vents intended for use with high-efficiency condensing boilers, factory-built by a reputable manufacturer.
3. The flues have suffered acute deterioration at a much faster rate than typically expected. The boilers and flues were new in 2013 and the flues failed dramatically (Photo 1) in less than three years.
4. Condensation has been observed with water dripping from the flues into plastic buckets.
5. The original consulting engineer (not the firm that employs this author) and flue manufacturer were both unable to explain the premature flue failure.
This writer visited the facility to observe the existing mechanical room and roof areas, record data, and interview the maintenance personnel. The visit occurred during normal weekday business hours, allowing observation of the equipment and building functions operating in routine, normal status mode.
The existing boilers are high-efficiency condensing type with a forced combustion air fan. The boilers appeared to be in good condition, at least externally.
One significant problem we noted with the boilers is that their top covers have been removed and set aside, apparently more-or-less permanently; and the boilers continued to operate with their top covers off (Photo 2). This unintentionally created a violation of Standard 15 (5) and is considered the major contributing factor to the premature deterioration of boiler flues, as explained in the next section of this column.
In fact, when I first walked into the boiler room, I thought it odd that the boilers would lack a top cover, exposing its internal working parts to the dirt and dust of a mechanical room. Seems like a poor design,
I thought to myself. Next, I looked for anything I could use as a makeshift table for my laptop computer, notepad, and drawings.
Up against the far wall of the mechanical room, I found a pair of large folded metal assemblies of about the right size and height for my "table" and began to set my belongings down on them. Then, it occurred to me; my makeshift tables were the boiler top covers (Photo 3).
The boiler flues are factory-built double-wall chimneys, with an inner wall constructed of a superferritic grade of stainless steel specifically chosen for its superior corrosion resistance against condensing boiler flue gasses. The outer wall is constructed of aluminized steel. The joints between segments were assembled with a field-applied sealant recommended by the manufacturer.
By external observation, the boiler flues were clearly in a failed condition. Many examples of acute corrosion and leaking flue gas condensate were found. In many places it appears the sealant between joints has failed, allowing flue gas condensate to drip out of the flue onto nearby pipes or the floor surface below. Decay and outright loss of product is especially prevalent on the roof and on the galvanized steel curb cap at the roof penetration (Photo 4). Finally, a few instrument test ports have been drilled into the flue for flue gas measurements, and the areas around those test ports are showing accelerated deterioration (Photo 5).
The boiler flues appear to be beyond repair and will require a full replacement. The walls and joints of the flues have been breached, creating a potentially hazardous situation if not repaired via replacement.
Each flue follows a similar path of approximately 72 ft [22 m], about equally divided between horizontal runs and a vertical run up through the building. Additionally, each has five 90[degrees] changes of direction and a stack cap termination on the roof.
In addition to the two boiler flues, there is a third duct which is the combustion air duct, whose purpose is to allow dedicated outdoor air to be delivered to the boiler combustion chamber. Without it, the boilers would use air from the boiler room for combustion, but instead, these boilers were intended to use outdoor air. The combustion air duct is not showing signs of deterioration and does not require replacement.
Because the mechanical room houses chillers, Standard 15 (6) requires ventilation of the room via exhaust to the outdoors. An exhaust fan was found on the roof and serves to continuously ventilate via exhaust the boiler/mechanical room in the lower level. It is a spun aluminum, downblast centrifugal roof exhauster; also new in 2013. This exhaust fan is mentioned because its location on the roof is very near the combustion air intakes and exhaust flues and, therefore, could be playing some part in disturbing the airflow pattern of the flues; and because exhaust in a boiler room is frowned upon because it complicates combustion air draft to the boiler. The exhaust fan appears to be in very good condition, with no sign of the type of deterioration seen on the boiler flues.
Why Did The Boiler Flues Deteriorate so Prematurely?
The boilers have been operated for a long period of time with their top covers removed and set aside.
A mechanic working in the area explained that the top covers, which need to be removed occasionally for maintenance, are unwieldy and difficult to maneuver, remove, and reinstall; therefore they were removed once or twice and then not re-installed. The mechanic who did this did so innocently and had no understanding of the ramifications. However, there is a decal on each boiler explicitly stating that the top covers must be in place while the boilers operate (Photo 6). Operation without the top covers went on for months, if not years, and, in the judgement of this author, led to the dominant reason why the boiler flues have prematurely deteriorated. The operation of boilers without their top covers potentially exposes the combustion chamber of the boiler to refrigerant leaked from any of three chillers in the same room.
Both the International Mechanical Code (7) and ASHRAE Standard 155 state that fuel-burning appliances and equipment having open flames that use combustion air from the machinery room shall not be installed in the same room with refrigerant-based equipment including chillers. As a former 12-year member of the SSPC 15 committee, I learned that this prohibition is not due to any specific flammability concerns related to the chiller refrigerant itself. Most refrigerants used for comfort air-conditioning purposes in commercial buildings are non-flammable, including the R-134a used in this case.
Rather, the prohibition exists because the by-products of exposure of common refrigerants to a flame can be exceptionally corrosive to boiler components and boiler flues downstream of the flame.
There are two exceptions, either of which allow chillers and boilers to be in the same room: if the boiler uses combustion air ducted from outside the machinery room and if the boiler is sealed in such a manner to prevent any refrigerant leakage from entering the combustion chamber; or if the combustion process is automatically shut down on detection of refrigerant. In this installation, there was no automatic shutdown, but the combustion air is ducted to the boilers from the outdoors, so the original design engineer did provide a design that complied with one of the exceptions in the code. However, once the boiler top covers are removed, the ducted combustion air is defeated.
These particular boilers have a combustion air duct inlet, then an unducted gap inside the boiler housing, and then a forced air fan that pushes air into the boiler's combustion chamber. When the lid is in place, the gap between the combustion air inlet and forced air fan is enclosed by the boiler housing itself. The lid forms a "duct" between the inlet and the boiler's internal fan (Photo 2). When the lid is off, the gap is open to the mechanical room. Therefore, the forced air fan merely draws air from the mechanical room itself, and the combustion air duct from the outdoors becomes useless.
Because the three medium-pressure chillers in the same room cannot be reasonably expected to be perfectly 100% leak-proof, small amounts of leaked refrigerant can be expected to eventually reach the combustion chamber of the boilers. Worse still would be if any of the chillers suffered a significant rupture or failure that released a large quantity of refrigerant.
In fact, that is exactly what happened, as related to me by the mechanic. A few years ago, one of the heat exchangers of one chiller cracked and released a large amount of refrigerant, approaching a full charge, into the mechanical room.
The Material Safety Data Sheet (MSDS) for refrigerant R-134a (8) states that by-products of exposure to open flame can be hydrogen fluoride, carbonyl fluoride, and hydrofluoric acid-chemicals that are exceptionally corrosive even to stainless steels and some butyl or silicone sealants and gaskets. In extreme cases, even phosgene gas can be formed. This is why Standard 15 prohibits open flames in rooms containing chillers or other refrigerant-containing products.
So what happened? Although I cannot prove it, my conclusion is that as the chiller refrigerant rupture dumped a large refrigerant charge into the mechanical room, some of that refrigerant was drawn into the boiler because the top cover was off. The refrigerant was exposed to the flame, breaking that refrigerant down into exceptionally corrosive acids that the boiler flues are not designed to resist. The joints and seals of the flues were likely the first to be breached, allowing some flue exhaust into the outer chamber of the double-wall flue; the outer chamber is less corrosion-resistant because it was not designed for condensing boiler flue gasses. There is no way to repair this situation other than to replace the flues with new, and never again operate the boilers with the top covers removed.
Bolstering that conclusion, one flue is considerably more failed than the other. Only one boiler of the two provided is needed to meet load most of the time, and the one boiler that was operating the day of the chiller rupture is the one whose flue is in significantly worse condition.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||COLUMN ENGINEER'S NOTEBOOK|
|Author:||Duda, Stephen W.|
|Date:||Sep 1, 2017|
|Previous Article:||Preservation On Location.|
|Next Article:||It's All Relative.|