A hydraulic power unit with a 55 kW pump, fitted with a plate heat exchanger sized by the manufacturer’s catalog. The oil ran at 72°C. The spec said the exchanger should hold it at 55°C with 30°C incoming water. The customer installed it and watched the thermometer creep up to 72°C. They called the exchanger vendor. The vendor said “water flow is too low.” It wasn’t — the water flow meter showed 25 L/min, exactly the catalog value. The problem was something else entirely.
The catalog assumes clean water and clean plates
Plate heat exchanger performance curves are generated with clean water, clean plates, and a temperature difference measured at the inlet. In the real world, the water has some mineral content. After three months, the plates scale up with calcium. The heat transfer coefficient drops by 30%. The exchanger that was sized for 55°C now holds 65°C. After a year, it holds 75°C. The customer thinks the exchanger was undersized. It wasn’t — it just got dirty.
The second assumption the catalog makes: the oil side is clean. Hydraulic oil that runs at 60-70°C for a year forms a varnish film on the plates. The varnish is an insulator. Heat transfer drops another 15%. Combined with water scaling, the exchanger is now at 40% of its rated performance. The customer sees 72°C and blames the initial sizing.
What actually happened on the 55 kW unit
We pulled the plates and inspected them. The water side had a thin layer of calcium scale. The oil side had a dark varnish film. We soaked the plates in a 5% citric acid solution for 4 hours. The scale dissolved. The varnish came off with a nylon brush. Reassembled, the oil dropped to 58°C. Back to spec.
But that was a one-time fix. The customer needed a permanent solution. We upsized the exchanger by 30%. The catalog rating for the new exchanger was 55°C clean. After scaling and varnish (which will happen within a year), it holds 65°C. We also installed a 5 μm oil filter upstream of the exchanger. It catches the varnish particles before they reach the plates. And we added a manual flushing connection so maintenance can circulate acid through the water side every six months without disassembly.
The sizing rule that actually works
Size the heat exchanger for 70% of the catalog rating. If the load is 55 kW and the exchanger curve says unit X holds 55 kW at 30°C ΔT, pick the unit that holds 75 kW at that ΔT. The extra 35% margin covers scaling, varnish, and the slow fouling that happens over a year of service. The initial cost difference is small — a 75 kW exchanger costs about 15% more than a 55 kW one. The alternative (overheating oil, shortened seal life, carbon buildup, and 100°C thermal trips) costs more in downtime.
Also, size the water flow. The catalog shows 25 L/min for the 55 kW unit. Run it at 35 L/min. The extra flow increases the turbulence on the water side, which actually improves heat transfer. The water pump is slightly bigger. The electricity to run it is negligible. The oil runs 5°C cooler.
The thermometer that lies
Most hydraulic systems read oil temperature at the reservoir wall. The oil near the wall is cooler than the oil in the middle. The wall thermometer might show 55°C while the oil at the pump suction is 65°C. Install the temperature probe in the return line, not the reservoir wall. That’s the actual oil temperature going through the system. The wall reading is marketing. The return line reading is the truth.
Plate heat exchangers foul. Size at 70% of catalog rating and you won’t be the engineer debugging overheated oil in year two. Clean the plates annually. Put the temp probe in the return line, not the reservoir wall. The extra margin costs less than a single weekend troubleshooting call.
