Moving air and liquid through the same block

Why a shared manifold has to handle two fluids with opposite behaviour.

A manifold that carries both air and liquid has to satisfy two contradictory sets of design rules at once, because liquid is incompressible and wets whatever it touches while air is compressible and generally does not, and a channel geometry chosen purely to suit one of those fluids will quietly work against the other every time it is used.

Radiators and their bleed valves

A home heating system's water pipes and radiators are plumbed as one continuous network, yet every radiator still needs its own small bleed valve at the top, because air that gets into the system behaves so differently from the water that it needs a separate mechanism. Air trapped at the top of a radiator sits there as a stubborn pocket, keeping hot water out of the top of the radiator and leaving the room colder than the boiler's output should allow. Pushing harder with the same pump does not shift it. The fix is a dedicated valve built to let trapped air out without losing the water around it. A shared manifold is built around the same acknowledgement: a network carrying two fluids with such different behaviour needs deliberate, separate provision for each.

On a drawing, a liquid channel and an air channel look like the same feature, a round or rectangular groove cut through solid plastic. A liquid channel has to worry about surface tension pulling at every corner, a meniscus catching on a rough wall, and a bubble lodging in any pocket the flow does not sweep clean. An air channel has none of those concerns, since air does not wet a surface or leave a retained film behind. Its problem is liquid getting in, whether condensed moisture, a splash from a nearby junction or a seal that weeps slightly, where it becomes a contaminant in a path never designed to carry it.

One sharp corner, two different verdicts

A sharp internal corner in a liquid channel is a place surface tension can pin a bubble or leave a stagnant pocket of retained fluid, and manifold designers spend real effort rounding or eliminating such corners. The identical corner in a purely air-carrying channel causes essentially none of that trouble. A rule copied wholesale from the liquid side onto the air side, rounding every corner regardless of what runs through it, spends manufacturing effort on a problem that was never present. A rule copied the other way, leaving a liquid channel's corners sharp because the equivalent air channel tolerated them, invites the bubble-trapping and dead-volume problems the rest of this set keeps returning to.

Where the two fluids push on each other

Starting from atmospheric pressure, air is roughly twenty thousand times easier to squeeze than water, and that gap underlies every difference described here. An actuator driving a liquid transmits force almost instantly and precisely, while the same actuator driving air through a comparable path has springy compliance built in before anything downstream moves. A junction where the two fluids meet, such as a pneumatically actuated valve pinching a liquid line closed, has to be engineered around that mismatch; treating it as a simple mechanical joint ignores the delay and give on the air side. The parts of the drawing where the two networks come close, a shared wall between an air channel and a liquid channel or a junction where pneumatic pressure drives a liquid valve, deserve more scrutiny than either network would need alone.

Air as a moving wall

Some channels carry both fluids on purpose. A design can push a train of separate liquid segments down a single channel with a small plug of air between each one, using the air as a moving wall that stops one sample from mixing with the next while both travel down the same path. Here the sharp behavioural split is exactly what the designer wants, since the air plug's refusal to mix with the liquid on either side keeps each segment isolated, and the channel geometry is chosen to preserve that difference.

A manifold serving both fluids is, in effect, two design problems solved inside one physical block. Applying the same channel rules everywhere is the fastest way to end up with an air path that traps condensation like an unbled radiator, or a liquid path where a design choice borrowed from the air side lets a bubble take hold. A drawing review worth trusting asks, at every channel, which of the two fluids it carries, and checks each channel against that fluid's rules.