When the volume of the vacuum line buffer tank is insufficient, the pressure pulses caused by the vacuum pump’s periodic evacuation cannot be effectively dampened; they are transmitted in the opposite direction along the pipeline into the evaporation chamber, directly disturbing the gas–liquid interface and having a significant negative impact on the separation efficiency of the demister. This disturbance is not a constant negative pressure deviation, but rather a periodic oscillation in pressure; in many cases, this issue is mistakenly attributed to an inadequately sized demister or excessive foaming.

The gas–liquid interface within the evaporation chamber is inherently in a state of negative pressure equilibrium. If the buffer tank is too small, the instantaneous suction pulses generated by the vacuum pump’s piston or rotary vane will cause the pressure within the chamber to fluctuate slightly and repeatedly. At the moment the pressure drops, the gas dissolved in the liquid rapidly escapes, generating a large number of microbubbles in a short time. These bubbles converge and rise, lifting the foam layer, causing the height of the foam layer to fluctuate periodically. When the pressure rebounds, some of the bubbles burst rapidly, spraying a large number of fine liquid droplets. The particle size of these newly formed mist droplets often falls within the range that is most difficult for demisters to capture, exceeding the design capacity of wire mesh or baffle-type demisters. Droplets that can be effectively intercepted by a demister under stable negative pressure conditions may penetrate the demister when subjected to pressure pulsations, resulting in continuous carryover.

Pressure fluctuations also alter the upward velocity of the vapour phase within the evaporation chamber; the vapour flow velocity fluctuates in synchronisation with the extraction pulses. All demisters have a corresponding optimal operating velocity range: when the velocity is too high, droplets cannot settle by inertia and pass directly through the demister elements; when the velocity is too low, the liquid film already captured and adhering to the wire mesh cannot flow back down smoothly under the influence of gravity and will accumulate and thicken on the surface of the mesh; when the next instantaneous high velocity occurs, the accumulated liquid film will be torn apart again, forming secondary entrainment—that is, the demister itself generates secondary mist. This is a very typical failure mode caused by an insufficient buffer tank, and differs from the entrainment mechanism caused by excessive foaming in the feedstock itself.

Reverse impacts from pulsating pressure also disrupt the internal gas flow distribution within the chamber. The gas flow, which originally passed uniformly upwards through the demister cross-section, develops localised vortices and flow deviations. In some areas, the gas velocity far exceeds the design value, forming flow channels through which droplets pass directly through the demister elements, whilst in other areas, the velocity is too low for the accumulated liquid to drain back. Over the long term, periodic pressure oscillations cause the solid deposit layer retained on the wire mesh demister to be repeatedly subjected to airflow impacts, leading to cracking and flaking of the deposit layer. The tiny solid particles that flake off enter the downstream condensation system along with the mist droplets; simultaneously, once the deposit layer has flaked off, the local gaps in the wire mesh widen, further reducing the capture efficiency.

There is also an easily overlooked knock-on effect: pressure pulses are transmitted to the vacuum monitoring points, causing the vacuum reading to fluctuate continuously. The automatic control system may misinterpret this as insufficient vacuum, leading it to continuously increase heating power or feed rate. This further increases the evaporation rate and foam generation, creating a vicious cycle that exacerbates the product carryover problem. Although this phenomenon resembles vacuum drift caused by the accumulation of non-condensable gases on the condensation side, the root causes are entirely different. During troubleshooting, one can stabilise the evaporation load and observe whether the frequency of vacuum fluctuations synchronises with the operating frequency of the vacuum pump to distinguish between the two.

The intensity of disturbances caused by an undersized buffer tank is related to the type of vacuum pump: rotary vane vacuum pumps exhibit relatively milder pulse amplitudes, whilst the periodic suction pulsations of liquid-ring vacuum pumps are more pronounced; the disturbance transmission effect is stronger when the buffer volume is insufficient. Simply increasing the buffer tank volume can attenuate pressure pulses, but the position of the buffer tank is also critical. Only if the buffer tank is installed immediately adjacent to the vacuum pump can it effectively stabilise the pressure; if installed on the side close to the evaporator chamber, it can only accommodate a small amount of vapour, and its ability to dampen pulses will be significantly reduced.

There are also boundary conditions to consider: if the system itself has very low foam levels and a minimal mist load, the efficiency loss caused by minor pressure pulsations may be negligible; however, once the feedstock contains components that readily stabilise foam—such as colloids or silicates—the foam lift and secondary entrainment induced by pressure pulsations will rapidly become pronounced, and the problem of carryover will continue to worsen.