1-805-484-2992
Quotes - Engineering - SalesCooling insulating oil without risking it
A transformer or a rectifier cooled by forced oil moves its losses out of the windings into the oil and then needs somewhere to put the heat. A power transformer in the OFWF class sends the oil through a water-cooled exchanger; an outdoor unit in the OFAF class blows air over an oil coil; a large rectifier or an induction heating supply does the same at a smaller scale. The oil leaves the tank between 140 and 195°F and needs to come back 20–40°F cooler. Mineral insulating oil and the natural ester oils are chemically mild toward stainless, so the oil side of the cooler is rarely the problem.
The problem is the other fluid. A transformer oil cooler is the one coil where a tube leak is more than a nuisance: water in the oil collapses its dielectric strength and a few hundred ppm can take the transformer out of service. The water side has to be a grade the water will not pit, and the cooler arranged so that a failed tube leaks oil outward rather than water inward.
Insulating oil is kind to any stainless, so 316L is chosen for the water side where tower or plant water carries chlorides; 304L serves oil-to-air coils and clean closed-loop water. Material & Construction explains how the grade is chosen.
The oil side
- Viscosity about three times higher at 100°F than at 200°F; the oil side runs laminar almost always
- Oil film controls the overall coefficient; sizing is done on the cold-start oil temperature, not the running one
- Oil pressure kept above water pressure by pump placement, so a leak goes oil-to-water and shows in the water
- Low pressure drop on the oil side, a few psi, to match the transformer's own oil pump
Oil-to-water
- Water inside the tube, oil outside in a shell or a tank; 316L tube on tower, river or well water
- Tower water can reach 600–900 ppm chloride from a 150 ppm makeup, past the 200 ppm typical for 304
- Double-wall tube with a vented gap where the plant cannot tolerate any oil-water contact
- Water velocity held at 3–6 ft/s to control fouling; a cooler fed from a pond silts up and loses a third of its duty
Oil-to-air
- Stainless tube with stainless or aluminum fins, fan-driven, mounted on the transformer frame or stand-alone
- Fins at 6–10 per inch for outdoor dust; tighter spacing clogs and a clogged air cooler runs the oil hot in July
- 304L tube and fins for inland sites; 316L for coastal and chemical-plant air with chloride in it
Talk to an Application SpecialistSize the oil cooler for cold-start oil and worst-case chloride water: the first sets the surface, the second sets the grade.
Oil side laminar, water side 316L, leaks outward.
Oil cooler design notes
Why the oil film sets the size
Insulating oil has roughly a hundredth the thermal conductivity of stainless and a viscosity that rises steeply as it cools, so the oil-side film is the resistance that matters. At running temperature the flow may just reach transition; at a 40°F cold start it is deeply laminar and the coefficient is a fraction of the warm value. A cooler sized on hot-oil properties looks fine on the data sheet and then runs the transformer warm every winter morning until the oil thins. We size at the lowest oil inlet the site will see and accept a cooler that is oversized on a summer afternoon. Turbulators raise the oil coefficient but add pressure drop, and the transformer's oil pump usually has only a few psi to give, so we check the pump curve first.
Keeping water out of the oil
Three things keep water out of the oil. First, the water-side grade: 316L tube on any open-loop water, because tower water at 600–900 ppm chloride and well water with iron and chloride can pit 304 within a season or two, and the pinhole is on the pressurized water side. Second, pressure: the oil pump is placed so the oil in the cooler sits above the water pressure, and a leak then shows as oil sheen in the cooling water. Third, where the plant cannot accept even that, a double-wall tube with the gap vented to atmosphere, so a failure of either wall shows at the vent before the fluids meet. A moisture monitor on the oil return is cheap insurance on any of the three.
Oil-to-air for outdoor units
Where there is no cooling water, the oil cooler is a finned tube bank with fans, and the air-side film is now the controlling one. Stainless tube with aluminum fins is the economical build for an inland substation; stainless fins on 316L tube are specified for a coastal site or a plant where the air carries chloride or sulfur compounds that eat aluminum. Fin spacing is a trade between surface and fouling; 6–10 fins per inch stays cleanable with a hose. The bank is sized at the design ambient for the site, because an air cooler sized on an average ambient is 20% short on the day the transformer is at full load.
The mistake: tower water through 304 tube
The transformer oil cooler failures we replace share one history. A 304 cooler was bought for a closed-loop duty, or because the water was tested once at 150 ppm chloride and nobody allowed for cycles of concentration, and it was piped to the plant's cooling tower. Two summers later a pit went through on the water side, water went into the oil because nobody had set the pressures, and the dielectric test failed. The replacement is 316L tube, the oil pump moved to the inlet side of the cooler, a moisture monitor on the oil return, and tower water sampled for chloride at the high-cycle condition, not at makeup.
Transformer oil cooler questions
Yes. Natural ester and synthetic ester oils are no harder on stainless than mineral oil, and the cooler is built the same way. They are more viscous at a given temperature, especially below about 70°F, so an ester-filled transformer gets a larger oil-side surface and we ask for the oil's viscosity at the lowest site temperature before sizing.
Pressure and, where specified, a second wall. We arrange the oil pump so the oil in the cooler is at a higher pressure than the water, so a pinhole leaks oil out into the water and shows up as sheen rather than as moisture in the transformer. For plants that will not accept any contact, the tube is double-wall with the gap between the walls vented to atmosphere; a failure of either wall shows at the vent first.
With 316L tube and a water velocity of 3–6 ft/s, yes for most towers. Send a water analysis taken at the concentrated condition the tower actually runs, not at makeup, because 150 ppm chloride in makeup can become 600–900 ppm in the loop at normal cycles. Chlorides above about 1,000 ppm, or tower water that runs warm, push the water side to 2205 duplex or titanium tube.
A few psi at running temperature if we are allowed to size it that way, which is why we ask for the transformer oil pump curve with the inquiry. Oil pressure drop rises sharply as the oil cools, so the cold-start figure can be several times the running one. If the pump cannot cover it, we add surface rather than fitting turbulators the pump cannot push through.
Related
Heating & Cooling Coils
Welded Coils
Fully welded coil assemblies in 304L or 316L, with welded headers, support frames and sloped runs that drain, for steam, water and glycol in tanks, ducts and flues.
+ Learn More
Heating & Cooling Coils
Ammonia Coils
Evaporator coils for ammonia in brine, glycol, ice-bank and process tanks, built entirely in stainless with no copper-bearing part on the circuit.
+ Learn More
Heating & Cooling Coils
Custom Coils
Helical and serpentine coils made to your tank drawing in 304L or 316L, with the tube size, pitch and nozzle layout chosen for the duty and the agitator.
+ Learn More
