Why Adding More Effects Stops Paying Off: The Steam Economy of Multiple-Effect Evaporation
Every evaporation plant pays an energy bill twice unless it plans carefully. Water boiled off in a vessel leaves as secondary vapour, and that vapour still carries most of its latent heat when it exits the separation chamber. Capturing that heat instead of venting it is the single biggest lever on the steam bill, and the machinery for doing it is the multiple-effect train.
When secondary vapour can do a second shift
The vapour earns its keep again if it is piped into the heating chamber of another evaporator whose operating pressure, and therefore whose boiling point, sits below those of the original vessel. Under that condition the incoming vapour still performs a heating duty: it condenses on the far side of the heating surface while fresh solution boils on the near side. Chain several such vessels together and the arrangement becomes multiple-effect evaporation, with each evaporator referred to as an effect. Live steam feeds the first effect; the vapour from each effect heats the following one.
The purpose is straightforward: reduce the consumption of heating steam. In theory 1 kg of heating steam evaporates about 1 kg of water. Practice is less kind. Heat losses intrude, and the latent heat of vaporisation in the separation chamber exceeds the latent heat of condensation in the heating chamber, so evaporating 1 kg of water actually consumes more than 1 kg of steam.
The steam economy curve and its flat tail
Engineers compress the trade-off into the steam economy U = W/D, the water evaporated per kilogram of heating steam. Accumulated experience assigns it values of 0.91 for a single effect, 1.76 for a double effect, 2.5 for a triple effect, 3.33 for a quadruple effect and 3.71 for a quintuple effect. Read as increments, the curve tells its own story: going from one effect to two saves roughly 50% of the heating steam, while the step from four effects to five returns just 10%.
The flattening has structural causes. Each added effect consumes part of the available temperature difference for heat transfer, production intensity falls sharply as the train lengthens, and equipment cost multiplies with every vessel. Beyond a certain number of effects, the steam saved by one more stage no longer justifies the capital it demands. Very long trains therefore appear only where fuel is dear and steam is the dominant cost item.
How pressure cascades down the train
Inside a multiple-effect train the plumbing follows a fixed logic. Live steam enters the first effect; the secondary vapour it releases serves as heating steam for the second effect, whose heating chamber simultaneously acts as the condenser of the first. The pattern repeats along the line until the last effect, whose vapour passes to a condenser, is cooled by water and leaves as condensate. Because operating pressure steps down from effect to effect, the final effect of a production-scale train is connected to a vacuum system. Heating-steam temperature and the boiling point of the solution both fall in sequence, while the concentration of the finished liquor rises from stage to stage.
Choosing between forward and backward feed
To spend the available temperature difference efficiently, plants pick one of three flow arrangements based on how the liquor behaves.
Forward feed sends the liquor from effect to effect under its own pressure differential, so no interstage pump is needed. Because the boiling point upstream is higher than downstream, the liquor flashes on entering each next effect, boiling off part of its water as bonus vapour. Operation stays simple and easy to hold stable. The weakness shows up in the heat-transfer coefficient: downstream effects run at higher concentration and lower boiling point, viscosity climbs, and heat transfer deteriorates stage by stage.
Backward feed pumps the liquor against the vapour flow. Here the effects of concentration and of temperature on viscosity largely cancel each other, so heat-transfer conditions, and the coefficients that follow from them, stay close to uniform across every effect. The arrangement pays for this with pumping power and with the loss of flashing on admission, which confines it to liquors whose viscosity varies strongly with temperature.
Closing the design: the trial-and-error loop
Three quantities define a finished design: the water evaporated in each effect, the heating-steam consumption and the heat-transfer area. Many effects mean many unknowns, and the arithmetic quickly outgrows a single-effect calculation; electronic computers have handled the routine since 2015. The basis never changes, being the material balance, the heat balance and the heat-transfer rate equation, but unknown parameters force a trial-and-error procedure:
- Determine the total evaporation rate from the material balance.
- Estimate each effect's evaporation rate from experience and back out the concentration in every effect. For forward feed, where flashing occurs, the ratios may be assumed: for two effects W1 : W2 = 1 : 1.1, and for three effects W1 : W2 : W3 = 1 : 1.1 : 1.2.
- Assume the operating pressure of each effect so the pressure drop divides evenly between adjacent effects, then read off the boiling point. Here p1 denotes the pressure of the heating steam in Pa and pe the pressure in the condenser.
- Apply the heat balance to fix the heating-steam quantity and the water evaporated per effect.
- Distribute the available temperature difference on the principle of equal heat-transfer area, then compute each area from the heat-transfer rate equation.
- Check the areas. If they disagree, redistribute the temperature difference and recalculate until the areas match, or come close enough.
No single step is difficult. The craft lies in the loop: iterating until every effect ends up with the same area is what turns a stack of equations into a buildable evaporator train.