Evaporator
CASE STUDY - IMPROVING EVAPORATION EFFICIENCY
CASE STUDY: IMPROVING EVAPORATION EFFICIENCY
A producer of high-quality malt extract syrups used mainly in the brewing and related industries was faced with rapidly increasing energy costs. The process was batch-wise preparation of a 13% solids wort from malted barley followed by evaporation in a triple effect evaporator set followed by concentration to 75 - 80% solids in a finishing evaporator vessel heated by live steam.
Scope of works
- Evaluate options available for improving efficiency of an evaporator set
- Evaluate the cost-effectiveness of the options
- Produce a constructability program to enable modifications required to be carried out with minimal interruption of the production process
- Compile and present a full report to the Owner's management team
- Monitor progress of modifications
Considerations
Many industrial processes such as desalination, treatment of waste streams, concentration of solutions and spray drying involve evaporation. In most cases, the source of heat is live steam which is an expensive resource due to both the cost of the energy required to produce steam and of the high purity boiler feedwater.
Steam usage is reduced by multiple effect evaporation in inverse proportion to the number of effects – a triple effect evaporator requires approximately one third of the live steam which would be needed for simple evaporation. The process fluid is evaporated in a cascade of vessels over a pressure differential typically between 100 kPa of steam pressure and 80 kPa of vacuum final vapour pressure corresponding to a temperature differential of 75° C. The vapour produced in each of the effects is used to heat the following effect. Efficiency of steam usage can be further improved by employing mechanical or steam ejector recompression of one or more of the vapours. The most common evaporator configuration is for the process fluid to feed into the first, highest pressure and temperature effect and the concentrated product to discharge from the final, lowest temperature effect. However other configurations are possible, for instance, the process fluid can be pumped in the opposite direction which has the advantage that viscosity is reduced and heat transfer improved when the more concentrated and viscous product is evaporated at the highest temperature. Evaporator vessels may transfer heat by natural circulation in a tube heat exchanger within the vessel body or heat transfer can be external, for instance in a plate and frame heat exchanger with a circulating pump which can also provide the pressure required to reverse feed the process fluid.
As the process fluid increases in concentration, vapours produced in the evaporation process become increasingly contaminated with entrained process fluid which adversely affects the quality of the condensate. Non-condensing gasses need to be removed by venting either to atmosphere or to a vacuum for vapours produced at negative pressure. Wherever process geography and economics permit, process heating should always be done by vapour at the lowest possible temperature.
Condensate from the first effect and any other processes using live steam is returned to the boiler as feedwater. However, some of this feedwater has to be blown down from the boiler to maintain a satisfactory level of dissolved solids. There is therefore a deficiency of feedwater which has to be made up with additional purified water. The cheapest source of purified water is vapour condensed in the second effect provided that the separation of vapour and entrained process fluid from the first effect is sufficient to produce usable feedwater. The conductivity of this vapour condensate needs to be carefully monitored, especially during upset conditions, so that the boiler feed water always meets the required specification. Condensate from other effects can be used efficiently to provide preheating of process fluids. A useful amount of energy can be recovered by flashing higher temperature condensates to provide additional lower temperature vapours.
The properties of the process fluid are important when considering the best type of evaporation configuration. In the case of sugar solutions, factors such as boiling point elevation and Maillard condensation reactions must be taken into consideration when designing evaporator layouts. Desalination and the treatment of waste with low dissolved solids having little or no value are applications where vapour recompression becomes an attractive option. Vapour recompression adds sensible heat to the vapour and requires substantially less energy than evaporation. Recompression can also be applied to more sensitive process fluids often in combination with multiple effect evaporation. In the case where the process fluid is temperature sensitive, the available temperature difference may be considerably reduced; possibly limiting the number of effects to 2 or even a single effect operating under high vacuum where vapour recompression can significantly improve process economy.
Implementation
Work commenced with site visits to gather data and evaluate the design constraints. Plant layout and associated geographical factors featured importantly in the evaluation of possible improvements. An evaporator survey was carried out to establish the operating parameters and the performances of the existing effects and the finisher. A major constraint was the continuity of production and therefore modifications had to be carried out in stages to take place over weekends or during planned maintenance outages.
Floor space was severely constricted as was the available height overhead. Foaming of dissolved gasses was a problem in the first effect due to the large temperature difference between the wort and the vapour produced in the first effect and close control of the finisher temperature was required to achieve the specified product colour. The three effects and the finisher were similarly configured with heating carried out in a plate heat exchanger with a circulating pump and the superheated wort fed to a separator vessel where the vapours V1 and V2 passed through an entrainment separator to heat the following effect and the concentrated worts W1, W2 and W3 fed forward to the circulating pump of the following heat exchanger. The final vapour V3 and the vapour from the finisher were condensed under vacuum. The condensates C0 were returned as boiler feed water and the condensates C1 and C2 were used to preheat the process. The configuration which existed is shown in the schematic diagram below.
After evaluating the possible options for improving efficiency the resultant final configuration is shown in the schematic diagram above and the modifications are described below.
The temperature in the second effect was better matched to that of the feeding wort W0 and the first modification was to introduce W0 to the second effect and back feed wort W1 to the first effect and feed wort W2 to the third effect. This reduced the foaming problem significantly and there proved to be sufficient head to prevent cavitation in the circulating pump of the first effect. Increasing the plate area in the first effect reduced the temperature difference across the effect while still giving reasonable vapour velocity consistent with good separation. The next step was to add an additional parallel separation vessel to the first effect so that the planned increase in vapour V1 quantity could be accommodated. Following this, the thermo-compressor was installed to recompress part of the first vapour V1. This potentially could have adversely affected the quality of the condensate fed back to the boiler and, as a precaution, conductivity monitoring with a dump facility was added to the boiler feed water return system. In the event, condensate C0 quality remained acceptable.
Because of the particular plant layout, it proved better to reconfigure the finisher as the second effect. This required careful prefabrication of connecting pipework modifications so that the operation could be carried out over a weekend. Simultaneously, the plate area of the finisher was increased to improve heat transfer. The next modification was to increase the plate area on the third effect and finally to modify the evaporator set so that the finisher became the second effect and second and third effects became the third and fourth effects. This step required the longest shutdown period of one week; however supply to customers was not affected as stocks were built up to take account of the planned downtime. It was found after trial that a small increase of plate area was required on the new third effect to achieve the desired throughput and temperature profile.
The net results of the modifications were the halving of steam consumption, a drastic reduction in the amount of boiler feedwater make up required, the virtual elimination of foaming problems and improved product colour control due to the reduced temperature differential over the final stage of concentration to 79% solids. In parallel, a number of modifications were made to the condensate system to recover heat from condensates C1, C2 and C3. This further reduced steam consumption.
The project took several months to implement but was completed according to schedule and with minimum shutdown time. The calculated payback period was just over one year but in practice this was reduced by reusing and modifying available spare and redundant equipment.
Similar energy recovery and utilities studies can be carried out to optimise heat usage for a variety of processes. For further details contact This email address is being protected from spambots. You need JavaScript enabled to view it.