A Case Study of R-22 Replacement Problems
When dealing with the coming “R-22 Replacement Reality”, there will be times when there is no “easy” path to take. The customer does not want to or cannot replace the equipment, yet R-22 is either far too expensive or simply not available. The customer may also be possibly looking to extend the life of their equipment, but the current equipment has some “issues” – whether from normal wear and tear, abuse, or perhaps most importantly they have inherent design flaws that are exaggerated when using a blend (which ALL R-22 replacements are). Most of these issues can be identified and easily addressed prior to converting the system to an R-22 Replacement refrigerant. Other issues will be harder to identify and will require technicians with a dynamic mindset and a ton of experience – or access to a technical support team with powerful tools, resources and community such as Bluon’s!
Bluon Energy working with a large local mechanical contractor encountered one of the “harder” scenarios this past winter while replacing R-22 and overhauling 24 large 120 Ton Split Systems (Trane RAUC124 Condensers and Bousquet HDGH-400 Air Handling Units). This project faced many foreseen and unforeseen challenges.
Many of these issues could have been preventable through better due diligence and planning prior to starting the project and by onsite technicians following newly established, and always growing, best practices surrounding replacing R-22 created by Bluon Energy and its in-house technicians.
However, some of the issues were unique to this specific situation and culminated in a “perfect storm” surrounding Replacing R-22. The silver lining in this project, along with future challenging projects, is the ability for Bluon to share these experiences and discovered solutions across a growing community of contractors and technicians engaged in tackling the R-22 Replacement Reality. As we will expand upon below, some of the challenges and their discovered solutions can be summarized as:
- Difficulty with evacuation and getting system under deep vacuum;
- Very cold weather exaggerated every issue with system evacuation
- Cold weather enabled oil to get trapped in system suction lines (low spots) which in turn trapped R-22 within the oil making it challenging to draw a deep vacuum
- Metering device replacements, without proper nitrogen flushing, allowed moisture contamination contributing to the vacuum challenge
- Minor leaks in the original liquid line valves caused major challenges as the leaks were aggravated by the evacuation process and required replacements before a system vacuum could be achieved
- Improper charging procedures;
- Newly installed Electronic Expansion Valves (EEV’s) were not properly closed prior to charging, causing flood backs on startup and failed compressors via locked rotors
- “Leaky” check-valves within 3D scroll compressors impact on large systems
- In large systems with large piping, blend refrigerants (which all R-22 replacements are) can create startup challenges due to off-cycle refrigerant migration & condensation
- Problem can be solved with the addition of pump-down circuity and/or discharge line check valves
- Faulty electrical breakers
- If breakers are routinely just “switched back on” by customer, any pump-down circuity is effectively made useless and migration based startup problems can re-occur
- Apparent “need” for more and more oil after charging
- Improper suction line piping causing low points (or gravity wells) can take days of runtime to effectively “fill” and cause the system oil needs to reach equilibrium
The first major challenge faced was properly baselining the system in the middle of a Colorado winter. This site was home to a large manufacturing facility that simply could not risk converting the systems during the summer heat when properly cooling the building was vital to manufacturing , but the cold temperatures severely hindered the project – from pre-inspect, to recovery, to evacuation, and finally system start-up. Despite the cold weather, the technicians went through all of the units prior to recovering the R-22 but the vast majority of them were not running due to the very cold ambient conditions. Eight compressors were found to be dead on arrival due to locked rotors, giving the technicians some initial suspicions of dealing with a challenging system. The remaining compressors were “bumped” to ensure they were not
locked, however this later proved to be completely insufficient at identifying the health of the system and individual compressors. Not being able to properly baseline the system prior to conversion was, no doubt, a large contributor to the multitude of issues experienced through the remainder of the project.
A third-party recovery company was used with industrial recovery equipment – but even utilizing the heavy equipment the recovery took much longer than anticipated and it was particularly difficult to pull the systems into a vacuum. Several weeks after the recovery portion of the project was complete and the weather had warmed a bit, several systems were found to have some positive pressure of refrigerant still in the system – presumably from refrigerant that was trapped in the oil throughout the system but was not pulled out during the recovery due to the cold temperatures and inherent piping of the systems creating gravity wells of oil throughout.
Prior to project start, it was decided to utilize Carel Electronic Expansion Valves (EEV’s) to increase system efficiency while also addressing low airflow and liquid flood-back concerns. While brazing in the EEV’s, the technicians did not flow nitrogen through the system to prevent contamination which likely introduced moisture into the system. This, along with the residual refrigerant and subfreezing temperatures, appeared to substantially affect the evacuation times of these systems increasing it from hours to weeks.
The technicians tried repeatedly changing vacuum pump oil, breaking the vacuum with nitrogen, pressure testing the systems with nitrogen, conducting thorough leak checks, and conducting vacuum decay tests which all indicated the systems were tight but some of the initial oil changes did show a “milky” site glass – a strong indication of moisture in the system. It was then decided to apply 50,000 BTU electric heaters and heat tape to the condensers to help boil off any residual refrigerant still trapped in the oil. While this helped the evacuation process somewhat it did not solve the problem.
After several weeks of evacuating the systems, specific liquid line ball valves were found to have very minor leaks, which were causing significant “leaking” when the system was under deep vacuum (less than 5000 microns). These ball valves were replaced, and the evacuation times were reduced to a day or two (still very slow vs. industry norm due to the cold ambient conditions). The leaky ball valves demonstrated the importance of several
considerations when replacing R-22 in existing equipment. One key consideration is that many elastomer seals exposed to R-22 tend to expand and harden over time. More importantly, when the R-22 is evacuated, these seals tend to shrink and/or become brittle leading to substantial leaking particularly when under a vacuum. Leaking in this case is typically a culmination of age, normal wear and tear, and the pre-described chemical reaction of the seals with R-22, but in practice it is the direct result of simply removing the R-22 from the system which causes the seals to dry out. It also demonstrates the importance of a proper assessment as to which seals are most susceptible to leaking (the high side of the system being more prone to leaks), which are critical in nature (must recover refrigerant to replace), and which seals are the best candidates for preemptive replacement.
Once the systems were properly evacuated, the technicians began to charge the systems, through the liquid line service port, to the starting point charge of 80% of nameplate or amount of refrigerant recovered, as per the Bluon R-458A instructions. The nameplate charge of these systems was approximately 160 lbs. of R-22 (80% charge = 120 lbs.). Liquid migration was a concern during the cold weather startup, so precautions were strictly adhered to and technicians ensured the crankcase heaters were operating for at least 8 hours prior to system startup. Pressure from the end-user/customer did however lead to one system being started with very low return air and low ambient conditions (55-60 F OAT). The system was operated for several hours before shutting down for the evening. The following morning, the technician working on the specific system went to start-up the system and immediately had one compressor on each circuit trip the breaker due to a locked rotor. What caused the failure of these compressors wasn’t initially fully understood, but it was attributed to being outside of OEM parameters recommended for startup and the fact that the hot gas bypass valves were not operational. Due to these circumstances, it was agreed that the remaining units would not be started unless there was adequate ambient temperatures and minimum 70 F return air until the hot gas bypass valves could be put back into operation.
The next unit startup was data-logged to ensure startup was conducted properly (ambient conditions, return air, superheat, etc.). The system was brought online and ran for approximately two hours and was tuned as best as possible before being shut down for the day. The customer’s in-house technicians put the system offline and not available for mechanical cooling. The following day when technicians returned to site, the system was found to have a lead compressor with a locked rotor. When the data logger was reviewed, it was found the system came online in the middle of the night with low return air and low ambient conditions (approximately 50-55 F) and had a clear electrical spike in amperage indicative of the compressor failure. This revelation caused us to question the intention of the sequence of operation. Either the intended sequence of operation allowed the compressors to start during undesirable conditions or one or more of the controls did not operate properly. To this day the sequence of operation remains unknown which would help us understand what actually occurred and/or to resolve subsequent similar control anomalies.
At this point, the project is already more than month behind schedule, with three failed compressors and none of the systems commissioned. The customer was seriously concerned about the progress of the project, the failed compressors, and the quickly approaching summer and, as a result, was putting heavy pressure on the contractor and Bluon to get their systems up and running. The technicians were rushing to get systems charged and online and, as a result, they charged one system with the EEV completely opened into the liquid line, through the open valve, and into the compressor suction line. The system was brought online to be commissioned and immediately had two compressor failures (locked rotor) due to the liquid flood back of the refrigerant charge that was just placed directly in the suction line. Subsequently, specific charging and startup instructions were created by Bluon to correct any improper charging procedures. The contractor agreed with the Bluon startup procedures and followed them accordingly. However, immediately at startup of another system there was another compressor failure, locked rotor again! Thoroughly inspecting the unit found the suction and discharge lines had equalized across the compressor, due to leaky check valves, and that enough refrigerant had migrated into the large suction line piping to cause the compressor to flood at startup. This issue is the result of today’s refrigerant blends (with their multiple boiling points) exacerbating a poor refrigerant management design inherent to the system. This issue, which will no doubt become wide spread in the coming years, is now known to the Bluon community.
By now the end-user/customer was on the verge of pulling the plug on the whole project. Something had to be done to ensure the safety of the equipment or there was a risk of shutting down their business which was a sophisticated manufacturing facility. It was decided between Bluon and the contractor to implement a continuous pump down circuit with discharge line check valves in the systems to ensure no liquid refrigerant was migrating to the compressors at startup. The continuous pump down logic involved a couple relays, some minor wiring and the check valves were to be installed first on the compressors that needed to be replaced and then in all the systems. Once the pump down logic was in place 19 of 32 circuits were released for mechanical cooling. However, it was noticed that several of the circuits were pumping down so frequently, due to the condition of the compressor’s internal check valves, they were taken off line until check valves could be installed.
With the pump down units now providing cooling to the customer’s manufacturing facility, Bluon’s technicians focused on commissioning systems and dialing them in as quickly as possible while the contractor focused on replacing the down compressors. During commissioning six additional compressors were found to be either running way over RLA, way under RLA, or various other issues considered dead on arrival and were replaced rather than risk compromising manufacturing. These compressors were more than likely in this condition prior to the project (not identified due to the inability to properly baseline the systems).
With the systems now running, several of the units were tripping the main breaker due to what was learned to be faulty breakers. This was a serious issue as it compromised the manufacturing processes as well as negated any protection afforded by the pump down logic. Great effort was taken to try to identify any electric current draw issues with the systems that could cause the breakers to trip and in many hours of data logging and physically observing systems run, the amperages were all within limits and notably lower than the units had been previously running on R-22. The cause was clearly not the compressor’s amp draw. The customer looked into having an electrician assess their circuit load to determine if the circuit was under-designed or something else, but no definitive cause was determined. Breakers tripping and being
automatically reset by the in-house technicians caused numerous additional compressor failures due to the pump down logic and action being effectively disabled upon startup.
The last major hurdle the project faced was several additional compressors lost from what was suspected as a loss of oil. Upon closer examination, many of the units had long suction piping sloped away from the compressors which is against industry standard practice and more importantly was creating several traps, or gravity wells, in the refrigerant lines which was causing the oil to collect in pools throughout the long piping. This had the effect of needing to add oil over several days until the various gravity wells were “filled” and the system’s piping reached a kind of oil equilibrium. Lastly, in the haste of getting systems online as quickly as possible the hot gas bypass valves were ignored and were not dialed in to the lower pressure of Bluon TdX 20, which left the question open of whether the systems were losing ability to carry oil under low load conditions due to resulting low refrigerant velocity.
Despite the many hurdles faced by Bluon and the large local contractor, the positive side of this story is that many VERY hard lessons were learned and protocols developed. Those lessons and protocols were subsequently implemented on a project with very similar systems resulting in an almost seamless conversion of the very same units.
Though the results of this project are not the type that Bluon or any contractor would ever want to repeat or advertise, it demonstrates the level of commitment Bluon has to its contractor partners, its customers, and to the industry in facing the R-22 phase-out and related
challenges with retrofitting the industry’s most difficult systems.
It is also our commitment to provide clear and concise documentation and protocols to the technicians so that they do not have to experience ALL of the hard lessons. It is Bluon’s mission to give the technicians and contractor community access to an ever-growing database of equipment, best practices and real technical support to give them actionable solutions, tools and resources to solve the issues they may already be facing.