This article speaks to recent experiences from engineering and project managing the retrofit of a 300-ton chiller. It focuses on the philosophical issues for retrofitting anything in a mechanical or electrical system rather than speaking specifically about chiller projects. You might consider the principles and path described here for many projects and plans moving forward. It also addresses the realities of trying to do a project in today’s world — in the middle of a pandemic and through the ongoing supply chain issues we’re all facing. I’ve also included several “pearls of wisdom” I have learned going through a lot of painful oysters over the past 40 years.

The 22-year-old, failing, 300-ton chiller showcased in this article operated in an institutional setting. The building was initially designed to be a high school and was then converted to its current institutional purpose (multiuse — regional offices, storage, retail, and light processing). The chiller piping and air handlers were neglected as the building transitioned over the years. In this same scenario, older, failing equipment; challenging surroundings; equipment not in great shape — you can change this project from a chiller to one of the many boilers I have also been involved in retrofitting. Again, it’s the process more than it is the equipment.

1.     Set the Proper Expectations Immediately!

Theme: You need time for research, scoping, and preliminary engineering (probably 20% of the overall engineering effort is here!) — There are always expectations to set with someone who is paying the bill. The two most elemental, of course, are schedule and budget. Getting these right takes time, and scoping and research are the first things you should negotiate. It’s a walk down of the site, including collecting maintenance and repair records, utility bills, utility rate schedules, and the names of all contractors currently providing materials or services to these systems. In this case, it meant the water treatment vendor (Chemtreat); controls contractor on-site (Siemens); chiller service team (Carrier); existing maintenance personnel; and, of course, the maintenance supervisor. All of the information received had to be looked over, and all of these people had to be interviewed. This investigation and research revealed the existing chiller had failed vibration tests and would not make it into the next season. The previous building owners had not treated the recirculated loop's piping, and it was failing randomly. The cooling tower fan needed bearings, the chilled water pump needed to be rebuilt, and a few other things needed to be completed.

FIGURE 1: Delivery of a new 50-HP motor, pump seals, and bearings was going to take 16-18 weeks. There simply wasn’t enough time to complete this work before the cooling season. The lesson is to have spares on hand. Photos courtesy of Prescient Technical Services

At this stage of the project, preliminary engineering may require confirming sizes, weights, performance requirements, voltages, wire sizes, piping feeds, and other basic big-picture things. It might include schematic diagrams of how things will be piped and wired.

Forget What You Thought You Knew About Lead Times

In this case, with schedules for advertising for bids on a public job, scheduling approvals for board meeting dates, etc., choices for what kind of chiller to get and where to get it were limited. Lead times, in some cases, were 16-18 weeks. One vendor said privately, “When someone says 16-18 weeks, it means they have no idea when you will actually get your machine.”

Because of all of this uncertainty and limited months until spring, getting a stocked chiller that could immediately be purchased and shipped made for a very safe choice that was welcomed by all. It also, however, limited some of the technology choices.

2.     What are the New Rules (Codes and Standards)?

Before further digging into things, research had to be conducted to identify the new codes and standards applicable to all parts of this project. In this case, it meant reviewing the latest editions of piping standards (ASME B31.3), ASHRAE 15, ASHRAE 34, and the National Electric Code NFPA 70. These all needed to be read and reviewed. Several historic editions also had to be purchased to better understand what changed since the original installation. A basic understanding of this information arms you for starting any field investigation. You can’t assume that everything was done correctly the first time. Remember, engineering is a practice, like medicine and law. We’re all human, and everyone interprets things differently. So, please don’t assume that because it’s been operating for 20 years, it was all put in according to the applicable codes at the time. It’s a new ball game, and you are now on the hook for code compliance moving forward as the “new” engineer or project manager of record.

Complying with codes and standards helps ensure the project goes in safely and protects the owner and designer from legal liabilities down the road. Plaintiffs’ attorneys and the U.S. Occupational Safety and Health Administration (OSHA) always look for what recognized and generally accepted good engineering practices (RAGAGEP) were not followed if something goes wrong. If something were ever to happen, you had better prove you had a process, knew the latest applicable codes and standards, and followed them. Trust me, on the other side, there will be a team of guys like me looking at every detail and everything you did or did not do.

Pearl of Wisdom

If you grab a valve handle or hit a start or stop button and something goes wrong, the results could be devastating. I have seen it. Fight the temptation to do that. As an engineer, nothing good comes from even the most innocent act you might do regarding physically touching a client’s equipment.

Having reviewed codes, standards, and drawings, it’s now time to gather the flashlights and personal protective equipment (PPE) and go into the field. As engineers and designers, remember, don’t ever physically touch anyone’s equipment. Don’t turn things on or off, change valve handle positions, or remove panels covers. You can do that at home but not at a customer’s site. If anything happens, regardless if it was related to your actions, you own lots of liability. You always need to have someone from the customer’s staff to execute even the most minor innocent things — even wiping a nameplate on a motor to be read.

FIGURE 2: Corrosion under insulation (CUI) is an important issue to consider. Make sure samples are taken in several vulnerable areas.

3.     What Will it Take to Get Things Back to Original Design Conditions?

It's always easier to design new systems. There’s no mystery about the condition, or “fitness for use,” of the supporting systems. Remember, the equipment that is the project’s focus was part of a system. If it needs to be replaced, then there’s a good chance other parts of that supporting system are also near the end of their useful lives. It will never be just about a chiller, boiler, industrial oven, or furnace.

Sometimes there is so much wrong on a project and so many moving pieces that it’s hard to wrap your head around the oceans of choices and decisions to be made. In these cases, I’ve always found it helpful to try and first understand what it would take to get everything back to its original design and then go from there. Of course, it would not be intended that the original plan or configuration would be where things end up. However, it’s a helpful starting point in chaotic situations.

There were rust stains from water that had leaked from the piping all over the mechanical rooms. When asked about this, the maintenance staff indicated that piping systems had been randomly failing. When these pipe breaks occurred, the water coming out was dark, murky, and rust-colored.

I interviewed the water treatment vendor, who told me the recirculated water systems had not been chemically treated with oxygen scavenger for many years. This is despite seasonal dumping and refilling this system with fresh city water to avoid freezing coils in air handlers where outside air dampers were not functioning correctly. Every time a system is flushed and filled with fresh water, dissolved oxygen is introduced to the system, which loves to run around and create havoc.

Pearl of Wisdom

You can never go wrong by identifying the possible need for non-destructive testing (NDT) of customers’ systems. Please, don’t be ridiculous about it by calling out onerous, unrealistic requirements. However, being reasonably prudent never hurts. You can always ask for more testing if initial observations or tests indicate problems. They can always say no, and then you are protected because you will have documented the request. Why should you take responsibility for systems that have been there forever and may have had questionable maintenance?

These conditions also clogged some air handler coils and had destroyed at least three of 10 air handler three-way valves for coil temperature control. However, besides randomly failing piping throughout the building, the biggest issue was the possibility of fouling the new chiller’s heat exchanger within days of starting up.

In this case, it was clear that at least the following needed to be added to the project’s scope before a new chiller could be considered.

FIGURE 3: Plugs from ultrasonic thickness testing locations at elbows and areas of turbulence. Note the discoloration from the rust and sludge in the recirculated water.

a. The piping systems had to have integrity — There’s no way to guarantee this after years of mistreatment, but at least well-placed ultrasonic thickness testing will help us understand some of the overall conditions.
b. The piping systems had to be clean — This can only be made better and not perfect. It will mean renting a bag filter and its temporary installation and operation for several days. This was an expense no one expected. Several popular rental houses have these available with replaceable bags down to small numbers of microns.
c. The outside air dampers needed to be functional — These will need to be replaced, so winter operation does not include draining the system. Even though this reduces peak chiller capacity, the overall strategy might also include some glycol in the system.
d. Three-way valves needed to be replaced — You can’t spend lots of money on a new chiller system and not have temperature control. The executive offices paying the bills would never understand rampant discomfort after hundreds of thousands of dollars were spent on refurbishing the air conditioning systems.   

4.     What Functionality Is Required Today and Over the Life of the Equipment?

This is a complex, abstract, and high-level discussion, but it must be had. It forms the basis of almost every decision you must make moving forward except for one. Safety can never be compromised, which means compliance with codes and standards is not negotiable, but almost everything else is somewhat discretionary.

When I say “functionality,” I mean things like the overall size or capacity, the energy efficiency and carbon footprint, and how much technology should be brought into this facility. Let’s take these each one at a time.

  • Overall Capacity

Is “like for like” the best? How has the use of the facility changed? What about sparing philosophies? Are two 60% machines better? Is reliability a big issue? There’s a lot also going on today with outside air requirements. Suppose you’re doing a boiler or chiller replacement. In that case, you have to consider that more outside air in the future is likely in the cards, and having the extra boiler or chiller capacity is the first step to making that even possible. Who knows what the following virus variant might bring in recommendations for more outside air.

How does your capacity today meet or not meet the facility’s needs? Of course, you can do load modeling. Unfortunately, our client did not have the budget or time for this. In our case, the chiller was already somewhat oversized. How would I know this? On an 85°F day, the chiller was operating at less than 30% load. A review of the outside air dampers in all 10 air handlers showed that, in at least seven of them, seals were missing or were severely damaged, so this low load was even occurring with ventilation rates higher than intended.

The other issue to consider is that there isn’t much first-cost difference, for example, between a 250- and a 300-ton chiller. There was also no considerable difference in part-load operating efficiency between the two sizes.

  • Energy Efficiency and Carbon Footprint

If the owner tells you he or she is selling the building within five years, is it best to install something that pays back in 10-15 years from energy savings? One can argue that lower energy costs make the building more valuable to a prospective buyer, but sometimes that’s a complex argument to make to folks without much of a budget right now who are focused on getting out of the building.

In my analysis of energy cost scenarios for this client, it wasn’t easy to get every vendor to provide comparable efficiency factors. The comparison is even more difficult when you don’t have great information about a building’s load profiles. In our case, a variable-frequency screw machine was the best option. There were more efficient technologies, like magnetic-bearing machines, but we couldn’t justify the added first costs given the number of years this client intended to own the building.

FIGURE 4: Outside air dampers with missing seals.

Remember, when considering energy costs, you cannot use average cost per kilowatt hour for your calculations. The kilowatt hours you’re saving are the incremental ones and not the average ones when it comes to utility rate schedules. They are likely the lowest-cost ones. You are likely to be removing some demand, so those savings can be counted, but, again, make sure you do not have the time-of-day rates and/or some form of ratcheting going on in the rate schedule that you need to take into account.

What’s the Right Level of Technology for This (Facility, Plant, Site, Workforce)?

Be careful here. Engineers like to let everyone know how smart they are. Sometimes this is done in very subtle ways, and we don’t even realize we’re doing it. Consider when we speak to people for five minutes, and 30 seconds in, their eyes are glazed over because they have no idea what we are talking about. It’s hard to remember that we do this stuff all day. Sometimes, you have to put away some of the blinking lights and software and think through things like where will they get service? Can they fix this stuff themselves? Is this now going to be a unique piece of equipment for this facility? Will someone need training?

Pearl of Wisdom

The first time production goes down because the wrong person is not comfortable in his or her space, you will stop being thought of as a genius for the thousands of dollars a year you saved them with your brilliance that no one can really tangibly see or feel anyway. Think of reliability, functionality, spares, service people, spare parts, etc.

Did We Forget Anything?

There are many other things I wish I could have reminded you about to help you avoid problems, including:

  1. There is a possible need for more changes to fusing and or electrical switchgear protection if you end up with a smaller or more efficient chiller.
  2. Changing out wetted components (temperature sensors, flow switches, and differential pressure switches).
  3. Human factor considerations, like space in the room and clearances for maintenance.
  4. Review the maximum operating ambient for the new chiller’s control systems, and make sure you can maintain machine room temperatures (i.e., maybe it’s time to add more ventilation).
  5. What will you do with the old refrigerant? Sell it, clean it up, and store it for your use for other equipment you might have?
  6. The cooling tower likely needs to be inspected, and serious work needs to be done on fill, fans, motors, gearboxes, and water distribution systems.
  7. Noise levels that can be tolerated.
  8. Physical size and doorways, access areas for rigging.
  9. Refrigerant sensors.
  10. Vibration isolators (pads or springs?).
  11. Leakage pathways from the room to occupied areas (i.e., tightness of the room).
  12. Fire protection systems (existing heads, new sprinkler heads, locations).
  13. Emergency eyewashes for chemicals handled for water treatment.
  14. Air venting from the piping systems (is it adequate today?).
  15. Piping insulation (is it all in decent shape), uninsulated, makes for lots of corrosion.
  16. Make sure you buy start-up and training.

What Else Can I Do to Enhance My Chances for Success?

Get some basic project management infrastructure in place, such as task lists and Gantt charts. There are plenty of inexpensive and straightforward software choices out there today for accomplishing this. Remember, though, that it’s not about the software as much as knowing all the tasks, sequences, and timing. Plan on some startup time and unexpected troubleshooting as well.

Verify that all safety issues are addressed. These include a robust lockout tag-out process, PPE rules, emergency plans, refrigerant handling safety, rigging and lifting safety, and hot work welding and pipe cutting.

Most importantly, communicate a lot regularly. Nothing replaces job meetings at the site on at least a weekly basis. Make sure there’s an agenda and someone’s documenting the discussions occurring at every meeting. I recently attended a professional development course that discussed what can happen if you’re not careful regarding meeting minutes. There can be tremendous legal liability created with meeting minutes. Try to be the person who takes the lead and is in charge of them. Review minutes carefully and state that if someone disagrees, you need their comments in writing within a specific time frame, like 48 hours after their issuance.

Hopefully, there’s something here that puts you on a track for success and a project that is as pain-free as possible.

Lead photo courtesy of Prescient Technical Services