Modern substation control and communication have revolutionized the ability to troubleshoot system faults and mis-operations. Event fault records and documenting every input and output provide invaluable information to help answer the question, “What happened?”
However, there are hundreds of thousands of systems with legacy equipment. These systems can present seemingly baffling challenges that could necessitate several attempts (test and try) to work through various scenarios to resolve the issue.
Life in the electrical maintenance and acceptance testing business can throw strange situations your way. This is also true of my experience in the electrical testing business. These three stories occurred over the course of my career and might help you be more efficient at identifying and correcting issues without all the benefits of modern communication. These stories, focused on substation transformers, illustrate that regardless of the complexity of the situation and system, it is often the simplest of solutions that can be the hardest to find.
TWO TRIPS — ONE INPUT
Springtime in Tennessee can present some variable weather. Freezing cold mornings and bathing-suit temperatures in the afternoon are not uncommon. These temperature swings played an interesting role in the case of a mysterious transformer trip. Shortly after my third cup of coffee, the call came from the plant engineer. “Get over here now! T1 just tripped off.” Knowing little more than that we headed to the customer’s plant, where we found a flurry of activity and questions regarding why transformer (T1) had tripped off-line. Physical inspection of the system noted no protective relay targets and an 86T (transformer lockout relay) operation. No protective relay operations? That’s odd. Let’s dig deeper and review the HMI event records. As we reviewed the event log there was only one unrecognized item on the list: IN604.
This input (IN106) is from the SEL 2411 that was installed to capture and communicate transformer trips and alarms. It was odd to note that the input was high on the event list and a short time later low. The input came in and then cleared (Figure 1).
A review of the drawings (Figure 2) illustrates that IN604 is tied to two fault conditions associated with the load tap changer (LTC): Did someone grab the manual LTC crank handle (89C), or was there a pressure relief operation (63PR)?
Noting that the LTC handle is behind a bolted cover (Figure 3) and that no one had opened the cover, we turned our attention to testing the pressure relief device (PRD). Tests and inspection proved that the switch operated properly and wasn’t the source of the issue.
With the mindset of “better safe than sorry,” we did some basic tests on the transformer and the LTC. All results were comparable to the previous maintenance results. After much discussion, a decision was made to energize the transformer. It was energized without issue and was left to soak overnight.
The following morning, however, the transformer tripped again. Once again, the event log documented IN106 going high and then going low. There was no PRD operation and no one had messed with the bolted cover that housed the LTC manual crank handle.
All sorts of ideas began to fly about how the system must be mis-operating. To leave no stone unturned, we opened the LTC handle cover to inspect the switch. Repeatedly removing and replacing the handle proved that the switch was doing its job: Handle out, IN106 went high; handle in its holder, IN106 went low. More tests, more ideas, more head-scratching. No smoking gun. “Let’s turn it back on and let it soak overnight,” we decided.
The following morning, shortly after sunrise, another trip. Now I had enough information to see a pattern. What was causing a trip every morning shortly after sunrise? Then it dawned on me (pun intended) that perhaps the issue had something to do with the sun warming up the transformer from the overnight freezing temperatures. My theory was that there must be something wrong with the LTC handle or the PRD switch that only shows up in the transition from below-freezing to above-freezing.
Fortunately, the weather continued to cooperate and I had a chance to prove my theory. After all, the client wasn’t going to be confident in the solution unless we could prove the theory. Early the following morning, our team showed up before sunrise equipped with a hair dryer (not standard test equipment). We applied some heat to the LTC handle switch and noted that IN106 went high — but only for a moment and then it went low and stayed low. As the switch warmed, moisture created a moment that caused the trip input to go high. With the theory confirmed, we were able to replace the switch and confidently bring the transformer online and restore the plant to normal service.
BLAME IT ON THE RAIN
Another tricky example began with a call from a client that a transformer had tripped during a heavy rainstorm. A visual inspection found no protective differential or over-current relay targets, yet the 86T relay had been tripped. As in the previous example, you must investigate all the other protective device trips that could operate the 86T (Figure 4). The list of those items that could operate an 86T is typically short and includes devices like PRDs and a sudden pressure relay (SPR).
It should be noted that both devices provide targets. SPRs are used in conjunction with a seal-in relay (Figure 5). This device, typically located in the transformer control cabinet, provides a target and seals in the trip so that manual resetting is required.
PRDs will flip up a yellow flag if they lift, assuming the flag isn’t missing or broken. The PRD switch is a latching device that also requires a manual reset. The PRD isn’t as easy to access as it is typically located on the top of the transformer, and the switch is located under the bell housing of the device (Figure 6).
So how could an unflagged PRD or SPR operate the 86T relay, and what’s the weather got to do with it? Simply put, damaged wiring insulation and/or device connectors (Figure 7) combined with water intrusion can create a sufficient low-resistance path to complete the trip circuit. This failure mode is almost always related to the PRD since it operates on one normally open contact going closed. In comparison, the SPR has two contacts (one open and one closed), and both sets must change state to trigger an operation in the seal-in unit.
IT TRIPS WHEN WE…
In this final story, a transformer (let’s call it T3) tripped on an 86T relay operation that opened the SF6 breaker ahead of the transformer and tripped the secondary main in a separate e- house. Keeping with the theme, the strange part…no targets to help explain what caused the operation. A review of the drawings showed several items that could operate the 86T relay: transformer differentials, low gas pressure in the SF6 breaker, and the sudden pressure relay (see Figure 4). We verified all the trips, and they all appeared to be operating and targeting properly.
We tested the transformer, and all results including a DGA came back acceptable. With no smoking gun, the decision was made to turn the transformer back on and load it, which was all done successfully and without any drama. The transformer serviced the plant fine for about 18 hours. Then, without warning, it tripped by the 86T relay with no targets on the protective devices. What’s going on here? It’s not the SF6 gas breaker low-pressure trip or sudden pressure relay. It must be the HU-model relays, but there were no targets. So we tested the relays and verified that the targets operated. Everything on the relays worked perfectly.
We energized the transformer and loaded it again without any issues. The transformer carried the plant for about 18 hours and then, once again, the 86T rolled with NO targets!
The question now was what changed in the plant. Fortunately, one of the engineers spoke up and shared that both of the previous two trips occurred when they were attempting to start a large HP motor. Ahh, now we are getting somewhere. The transformer system was fine when lightly loaded. Only when a full load was being added did the problem reveal itself. Many readers know that it takes a minimum load (usually 30%) before a transformer differential will operate.
Now that the relays have tested fine, what’s the problem? With today’s modern relays, troubleshooting a differential trip is so easy a caveman could do it. Older systems with an electro-mechanical differential relay like HUs represented a greater challenge. Our plan to identify the issue required us to energize and load the transformer with the differential relay trips pulled/opened.
Want to measure current magnitude or compare your phase angles on a system with HU relays? You need a current thief and a phase angle meter. Fortunately, we had the necessary equipment. Start with measuring the primary and secondary current into the relay. In most cases, you must do the math to compare the primary current magnitude to that of the secondary. In our case… there was NO secondary current. Now it wasn’t zero, but it wasn’t anywhere close to the amount we were expecting. With this new piece of information, our investigation turned to the secondary main current transformers (CTs).
When we walked into the e-house we could hear the audible groan of CTs that are in saturation because of an open CT secondary circuit (more specifically in this case, an extremely high burden/resistance). After making the system electrically safe, the CT secondaries to the remote e-house were tested, and it was confirmed that they had failed in the underground conduit. After pulling and landing new CT secondary wires, the transformer was successfully returned to service.
Now I hope you are asking yourself, “So why didn’t the differential relays target?” It was a legitimate differential fault because of the lack of secondary current input. As it turns out, this issue goes all the way back to improper and incomplete commissioning of the system when installed a decade earlier. The relay targets were set to the 2.0 amp tap (Figure 8), which is an issue in this application because the current necessary to operate the 86T relay is much less than 2 amps. For that reason, no targets dropped when the HU relays operated the 86T relay.
After adjusting the relay targets to 0.2 amps and testing the system, we verified that if the HU relay operated and tripped the 86T relay, the targets would operate on all three relays.
Transformer trips are a common problem faced by engineers and technicians in the field. They can be caused by various factors, including weather conditions, faulty wiring, damaged insulation, and malfunctioning protective devices. In many cases, identifying the root cause of the problem can be a challenging task that requires a thorough investigation of all possible scenarios. As demonstrated by these examples, the process of troubleshooting transformer trips involves a combination of technical expertise, attention to detail, and persistence (lots of persistence).
Mose Ramieh is Vice President, Business Development at CBS Field Services. A former Navy man, Texas Longhorn, Vlogger, CrossFit enthusiast, and slow-cigar-smoking champion, Mose has been in the electrical testing industry for 24 years. He is a Level IV NETA Technician with an eye for simplicity and utilizing the KISS principle in the execution of acceptance and maintenance testing. Over the years, he has held positions at four companies ranging from field service technician, operations, sales, business development, and company owner. To this day, he claims he is on call 24/7/365 to assist anyone with an electrical challenge. That includes you, so be sure to connect with him on the socials.