Natural Esters in Transformers: Which Mineral Oil Habits Still Apply?
The transformer industry has about 130 years of habits built around mineral oil, and roughly 30 years of experience with natural esters. Bruce Forsyth set those two figures side by side near the end of the FR3™ fluid expert panel discussion. They explain most of what came before.
Forsyth, Director of Global Application Engineering at Cargill Bioindustrial, moderated the APC Media panel on 29 October. With him were four regional application engineering leaders from his team: Revin Wang (China, Japan and South Korea), Naveen Jain (India and the Middle East), Jorge Gabiatti (South America and South Africa) and Ahmed Gamil (Europe). Over more than an hour they covered five subjects that users of FR3 fluid, Cargill’s soybean-based natural ester, keep raising: moisture, cold weather, oxidation, dissolved gas analysis and retrofilling.
At the heart of all five topics was the same question: when the liquid changes, which of those 130 years of habits still hold?
It is tempting to say most of them. The DGA gases, the paper and the tank are the same. And the reason for switching is easy to state: FR3 fluid is a less-flammable K-class liquid with a fire point of about 360 °C, against roughly 160 °C for mineral oil, according to Forsyth.
But the panel kept returning to differences that only show up in service, inside parts of the transformer most people never look at.
Water moves where nobody is watching
In a sealed transformer, moisture is shared between the cellulose insulation and the liquid, and the two drift toward the same relative saturation, Jain explained. Natural esters can hold around ten times more water than mineral oil before free water appears, depending on temperature, Wang said. So water migrates out of the paper and into the liquid, and the paper slowly dries.
That changes the liquid’s job description. It is no longer just a coolant and a dielectric. In an ester-filled unit it also behaves like a slow, continuous dryer for the insulation, and because water accelerates paper ageing, the panel tied that directly to longer insulation life. After a retrofill, Gamil cautioned, none of this happens overnight. The chemistry of relative saturation is covered in more depth in the effects of water on dissolved gas analysis.
Pour point is one number. Cold service is a system.
FR3 fluid’s pour point is typically −21 °C, Wang said, yet hundreds of units operate in Chinese regions where minimums reach about −35 °C. The explanation is thermal mass. Jain’s example: a small 15 kVA single-phase transformer at −25 °C needs more than three days of cold soaking before the fluid turns slushy.
“For me, the key point is don’t look only at the lowest ambient temperature,” said Gamil. Viscosity, cold-soak duration, pumps, tap changers and exposed piping tell you more than a single laboratory figure. Gabiatti added a nuance worth hearing in full: he finds the staged loading in IEEE C57.93 useful, but would rather steer a cold start by actual fluid temperature. For a more cautious view of cold starts, see the “dual nature” of dielectric esters in extreme cold.
Air does its damage in thin layers
Natural esters oxidise more readily than mineral oil. A few hours of exposure is negligible. Weeks or months of exposure are not. The real concern is the thin film, liquid spread over a large surface relative to its volume, which polymerises from a tacky residue into a hard, glossy layer that can strip paint when removed, Jain said. Ester-soaked active parts should not go into a conventional hot-air oven, Gabiatti said; vacuum or other oxygen-free drying is the route.
A DGA result is only as good as the sample
Mineral oil DGA limits cannot simply be carried over to esters, Jain said, pointing to IEEE C57.155 and IEC 63585 for ester-specific interpretation. A poor sample or an unsuitable lab method can mislead in either direction, Gabiatti warned: “Otherwise, we may end up chasing a problem that is not really there, or also missing a real problem.” Gamil’s sequence was short. Verify the result, check the trend, correlate, then act. An earlier webinar covers getting the most out of online DGA.
Retrofilling starts with “why”
Gamil’s first question for any retrofill is the motive: fire safety, sustainability, or simply more value from an existing asset. The technical trap is leftover mineral oil. Too much of it can drag the blend’s fire point below 300 °C, the K-class threshold, Jain warned. His rule of thumb: lean toward retrofilling if the unit is healthy and residual mineral oil stays below 5%. The panel’s discussion of how to weigh this option against purchasing a new transformer is worth hearing in full.
Jain answered the opening question in the closing round: “Practices, interpretation methods, design assumptions, and diagnostic criteria developed for mineral oils should not be automatically applied to natural ester-filled transformers.”
In the Q&A, an attendee described tap-changer makers recommending a 115 kV-rated no-load tap changer for a 72 kV transformer, a margin born of uncertainty. Manufacturers trust mineral oil because they have built on it for generations, and that confidence has not yet transferred to esters. “I believe that over time, that will change,” he said.
The distance between 130 years and 30 is shrinking one installed unit at a time. How quickly it closes, and how much of the industry’s conservative approach changes along the way, depends on what the industry chooses to measure in the meantime.
Who should watch, and why
- Utility asset managers and maintenance engineers running ester-filled units, or planning to, who need sampling, monitoring and cold-start practice that fits the fluid.
- Transformer designers and OEM engineers weighing design margins, accessories and drying methods for natural esters.
- Specifiers and project planners choosing an insulating liquid for fire-sensitive sites, cold climates or new voltage classes.
- DGA and condition-monitoring specialists who interpret results from ester-filled transformers and need to know where mineral oil limits stop applying.
- Owners considering a retrofill, from the first question (why?) through to the new baseline after refilling.
Why it’s worth the hour
- Five engineers, five regions. The Cargill Bioindustrial application engineering team brings field experience from North America, Europe, China, India and the Middle East, and South America and South Africa, including cold-climate installations and a 30-plus unit retrofill programme.
- Practical rules you can apply on Monday. Examples include when to treat a start-up as a cold start, why hot-air ovens are off limits for ester-soaked parts, and how far to dry the fluid without degrading it.
- The standards behind the advice. The panel refers to IEEE C57.155, IEC 63585, IEEE C57.93 and the forthcoming IEC TR 63646 on retrofilling.
- An audience Q&A that goes past the script. Attendees asked about furan analysis in ester-filled units, silicone contamination, moisture limits for processing, tap-changer dielectric margins and operation down to −50 °C. Most of these questions were not part of the prepared topics.
- Room for disagreement. One panellist prefers steering cold starts by fluid temperature rather than fixed loading steps, and the reasoning is worth hearing in full.
Resources and Next Steps
Download these Cargill resources:
Meet the Expert Panel
All five panelists work in Cargill Bioindustrial’s transformer fluids application engineering team.
Panel bios:
Director of Global Application Engineering, Cargill Bioindustrial – Transformer Fluids
Bruce Forsyth leads Cargill’s global Application Engineering team supporting transformer fluid applications. He holds degrees in Physics, Mathematics, and Electrical Engineering and is a Registered Professional Engineer and Life Senior Member of IEEE. With more than 35 years of experience in the IEEE Transformers Committee, Bruce has extensive expertise in transformer design, forensic analysis, and the application of natural and synthetic ester fluids.
Regional Application Engineering Leader – Europe, Cargill Bioindustrial
Ahmed Gamil brings more than 15 years of experience in transformer technology, specializing in dielectric performance, thermal behavior, insulation systems, and transformer asset management. He holds an MSc and PhD focused on transformer dielectric and thermal performance and actively contributes to IEC and CIGRE working groups. His current work focuses on sustainable transformer technologies and the application of natural ester fluids.
Regional Application Engineering Leader – China, Japan & South Korea, Cargill Bioindustrial
Revin Wang has 17 years of experience in transformer design and dielectric liquid applications. He was involved in the development of Cargill’s FR3™ Fluid facility in China and contributed to the realization of the world’s first natural ester-immersed 750 kV power transformer. Revin is also active in IEC and CIGRE activities and focuses on expanding the application of ester liquids in transformers.
Regional Application Engineering Leader – South America & South Africa, Cargill Bioindustrial
Jorge Gabiatti has more than 20 years of experience in transformer design and technology, covering power transformers, shunt reactors, and specialized transformer solutions. An Electrical Engineer, he actively contributes to international industry and standardization initiatives related to transformers, alternative insulating liquids, bushings, tap changers, and fire safety. His current focus is the application of natural and synthetic ester fluids in power and distribution transformers.
Naveen Jain
Regional Application Engineering Leader – India & Middle East, Cargill Bioindustrial
Naveen Jain brings more than 22 years of experience in the electrical power industry, with a background in Electrical Engineering. He is an active contributor to BIS, IEEMA, and CBIP initiatives and has participated in national standardization committees for more than a decade. His current focus is supporting the application of natural and synthetic ester liquids in power and distribution transformers across India and the Middle East.
