1-805-484-2992
Quotes - Engineering - SalesVacuum Condensers
Condensing at Low Absolute Pressure Without Destroying the Vacuum
Under vacuum, pressure drop is not a secondary consideration — it is the constraint that decides whether the system reaches its target. We design vacuum condensers, ejector inter- and after-condensers around the vapor-side pressure budget from the outset.
Condensing under vacuum inverts the usual design priorities. At atmospheric pressure and above, a fraction of a psi of vapor-side pressure drop is a rounding error. At 20 torr it can be the difference between a process that works and one that does not, because the pressure the process actually sees is the condenser pressure plus everything lost getting the vapor there. Specific volume is enormous at low pressure, velocities are high for very modest mass flows, and nozzle and header sizing that looks generous on a normal exchanger becomes restrictive.
What this means in practice. Vapor-side pressure drop is treated as a design specification rather than a calculated output. That drives large nozzles, short vapor paths, generous tube pitch and often a different orientation than the same duty would use at pressure. It is also why simply adding surface area to a struggling vacuum condenser can make matters worse: more tubes means more path length and more resistance.
A condenser between ejector stages that removes motive steam and condensable load so the next stage handles a smaller volume. An aftercondenser sits at final discharge and condenses the last stage’s motive steam before venting.
Ejector systems and staging. Steam-jet ejector systems use condensers between stages to knock out motive steam and condensable load so the next stage handles a smaller volume. An intercondenser between stages and an aftercondenser at discharge is a standard arrangement, with cooling water flowing in series from the intercondenser inlet through to the aftercondenser discharge. Because these condensers set what each ejector stage has to move, their performance governs the achievable vacuum for the whole system. Multi-stage sets are commonly specified as thermally and mechanically guaranteed to ASME Code Section VIII Division 1 with a defined TEMA class and type, and a fouling allowance applied to both sides.
Noncondensables dominate. Under vacuum, air in-leakage matters more than anywhere else because the pressure differential driving it inward is at its maximum and the volumetric consequence of a given mass of air is at its largest. Gas accumulating in the cold end blankets surface and raises pressure, and the vacuum equipment then works harder to hold the same condition. Vent placement and gas-side rating deserve more attention here than in any other service.
Common duties. Vacuum dryers and tray dryers; edible oil and fatty acid deodorizers; vacuum crystallizers; vacuum distillation overheads; degassing and deaeration systems; steam-jet refrigeration; and the ejector or liquid ring vacuum sets serving all of them. Several of these carry a fouling or freezing risk that further constrains geometry — for instance a stream where water vapor can ice up a surface that is running cold enough to condense an organic.
Coolant temperature sets the floor. The lowest pressure achievable is bounded by the saturation pressure at the coldest surface temperature the coolant can produce. Chasing a deeper vacuum with a cooling tower loop in summer eventually becomes futile; chilled water or a refrigerated stage is the honest answer. It is better to establish that early than after installation.
Related pages: surface steam condensers, vent condensers and overhead condensers.
Low pressure design questions
Because the pressure lost in the condenser is subtracted directly from the process vacuum, and at 20 torr there is almost nothing to subtract from. The volumes involved explain the rest: a pound of water vapor at 20 torr occupies roughly 800 ft3, about 30 times its volume at atmospheric pressure, so 500 lb/hr of vapor is moving 110 ft3 per second through the inlet nozzle.
A 2 torr loss through nozzle, bundle and piping is 10 percent of the pressure budget, and on a deodorizer or a crystallizer that may be the whole margin between the specified product and an off-spec batch. We therefore fix the allowable vapor-side drop first and let the exchanger geometry follow it, which usually means an inlet nozzle two or three sizes larger than a pressure-rated unit of the same duty would carry.
Adding tubes to a vacuum condenser that is not holding pressure usually treats the wrong problem, and a short diagnostic before any hardware change saves the cost of finding that out. First compare the condensate outlet temperature with the saturation temperature at the measured shell pressure: if the condensate is within a few degrees of saturation, the surface is already keeping up and the limit is pressure drop or air.
Then read the ejector suction pressure against the condenser pressure; a gap of more than 1 or 2 torr means the connecting line, not the exchanger, is where the vacuum is being lost. Only a condensate outlet far below saturation with a vent running mostly steam points to a genuine surface shortage, and even then a second shell in parallel beats a longer bundle because it halves the velocity instead of lengthening the path.
The intercondenser is the one that decides the size of the downstream ejector stage, while the aftercondenser is there to recover motive steam and stop a plume, and that difference changes how each is sized. An intercondenser typically runs at 50 to 150 torr between the first and second stages, so it is still a vacuum exchanger with the same pressure-drop discipline as the main condenser, and its outlet temperature sets the suction load on the next stage: every 5 F of extra cooling at that point removes a measurable fraction of the steam the second stage must compress.
The aftercondenser discharges near atmospheric pressure, where pressure drop is cheap, so it is a small and simple unit. Both drain through barometric legs or condensate pumps, and a leg that is too short to seal against the intercondenser vacuum is one of the most common commissioning faults we see.
The coolant inlet temperature plus a realistic approach, converted to a saturation pressure, is the floor, and running the arithmetic at summer conditions is the honest way to find it. With tower water arriving at 85 F and a 10 F approach on the condensing surface, the coldest surface is 95 F, and water vapor at 95 F has a saturation pressure of about 42 torr. No ejector, no extra surface and no amount of vent attention gets the condenser below that.
Switch to chilled water at 45 F and the same 10 F approach lands at 55 F, or about 11 torr, which is a different class of process. For a deodorizer or a crystallizer specified at 20 torr or lower, that comparison usually settles the coolant question before the exchanger is even sketched, and it is worth running for the warmest week of the year rather than the annual average.
Ice forms whenever any part of the condensing surface runs below 32 F while water vapor is still present in the stream, and under vacuum that happens at lower coolant temperatures than most people expect. Water carried with an organic at 20 torr has a very low partial pressure, so it does not condense as liquid on a warm surface at all; it travels on to the coldest tubes, where it deposits directly as frost.
A refrigerated stage running on minus 10 F glycol to recover a light solvent will blind itself within hours on a feed carrying even 1 percent moisture. The usual answers are a warm first stage at 50 to 60 F that removes most of the water as liquid before the cold stage sees it, two cold units alternating through a defrost cycle, or a controlled coolant temperature held just above the frost point.
Yes, because the condenser and the vacuum equiptment are sized against each other, and a unit designed without knowing what sits downstream will be wrong in one direction or the other. A steam-jet ejector set wants the condenser to deliver its gas load at a defined pressure and temperature; the intercondenser outlet temperature decides the next stage's capacity, and a 5 F difference there can push the stage off its curve.
A liquid ring pump has a different limit: its seal liquid temperature sets the lowest suction pressure it can hold, so it needs the condenser to knock out condensables before they reach the pump and cavitate it. The stage pressures, the pump type and the margin the system was sized with all belong on the data sheet, so the condenser fits that budget rather than being specified in isolation.
