Skip to main content
LibertyCES — Engineering Line Direct: Call James Riggins at (559) 395-5500 — 30+ Years Spec Experience. No sales queue. Real engineering answers.
Guides / Semiconductor & High-Purity

Duplex/Standby Chemical Dosing Architecture for Semiconductor Facilities

Two pumps on one skid isn't automatically redundancy. The failover logic is.

Read This First

Same boundary as the rest of this cluster: facilities-side dosing architecture, never wafer-contact UHP chemical delivery. See the flagship guide for the full explanation.

A duplex skid with a standby pump that never gets a start signal isn't redundant — it's a spare that happens to be already plumbed in. Two pumps mounted on the same frame is real hardware; a system that detects a duty-pump failure, starts the standby, confirms it's actually delivering, and tells someone what happened is engineering. Blue-White's own literature is honest about which one it ships — worth being equally honest about it here.

The Real Sequence

Five Steps, Not One Switch

1

Duty pump runs, standby idle

Both pumps are mounted, piped, and valved on the same skid — the standby unit isn't a spare sitting in a closet, it's already plumbed in and ready.

2

Failure signal fires

A real Blue-White signal trips — TFD (tube rupture), DFD (diaphragm failure), or MS6 (measured flow drops below its configured low-flow trigger). This is the same failure-detection hardware covered in the flow verification guide, not a separate system.

3

Standby pump starts, duty pump isolated

The control panel — not the pump itself — makes this call: stop the failed duty pump, start the standby, and isolate the failed unit so it can be serviced without shutting down the dosing point.

4

Flow re-verified on the new duty pump

MS6 (or an equivalent flow signal) confirms the standby pump is actually delivering chemical — the same command-vs-delivery discipline applies to the pump that just took over, not just the one that failed.

5

Alarm to SCADA

A failover event gets reported, not silently absorbed. If nobody's told a pump failed and the skid quietly switched to its last line of defense, the next failure has no backup left.

What's Real vs. What's Engineered

What Blue-White Ships, What LibertyCES Designs

Blue-White's real product line confirms the duplex/triplex skid hardware: CFPS-2, CFCS-2, and CFWS-2 each mount two pumps on one factory-assembled frame; CFPS-3 mounts three. Distributor literature describes the duplex configuration as usable either for simultaneous dual-chemical injection or with one pump held in standby. What isn't documented as an off-the-shelf Blue-White feature is the automatic failover control logic itself — checked directly against Blue-White's own CFPS-2 product page, which describes the hardware and compatible pumps but no lead-lag alternation or auto-failover controller spec. That gap is exactly where LibertyCES's control-engineering work sits: wiring the skid's real TFD/DFD/MS6 failure signals into a control panel that actually executes the five-step sequence above.

FAQ

Frequently Asked Questions

Does Blue-White's duplex skid include automatic lead-lag pump alternation as a packaged feature?

Blue-White's own product literature confirms the duplex skid hardware itself — CFPS-2, CFCS-2, and CFWS-2 all mount two pumps on one factory-assembled frame, and distributor material describes the configuration as usable either for simultaneous dual-chemical injection or with one pump held in standby. But the specific automatic failover control logic — detect a duty-pump fault, start the standby, verify its flow, alarm to SCADA — is not documented as an off-the-shelf Blue-White controller feature. That's control-system engineering built around the skid's real hardware and the pumps' real failure-detection outputs, which is exactly the kind of work LibertyCES does.

What real signals can actually trigger a failover?

The same three covered in the flow verification guide: TFD (Tube Failure Detection) on a peristaltic pump, DFD (Diaphragm Failure Detection) on a diaphragm pump, and a low/no-flow trigger from an MS6 flow meter. All three are real, patented or measured hardware signals — not assumptions about what "should" be happening.

Why does the standby pump need its own flow verification after a failover?

Because "the standby started" and "the standby is delivering chemical" are two different claims — the same distinction that matters for the original duty pump. A failover that swaps to a pump with its own undetected problem (lost prime, closed valve, empty day tank) looks like a fixed system while doing nothing.

Is a duplex skid the same thing as a triplex skid?

No — duplex mounts two pumps, triplex (Blue-White's CFPS-3) mounts three. A triplex configuration can support a true duty/standby-with-second-standby arrangement or split duty across two active pumps with one standby, depending on how the process is specified.

Related
Chemical Dosing Systems for Semiconductor Fab Facilities →Flow Verification & Failure Detection — the real signals this page's failover logic uses →Chemical Feed Skid Selection — which family actually supports duplex →Pre-Assembled Chemical Skids for Semiconductor Support Systems →Blue-White Manufacturer Hub →Facilities vs. Wafer-Contact Boundary →
Spec Review

Get real redundancy, not just a second pump

Ready to buy or need a fast answer? Call, text, or email James directly — (559) 395-5500 · [email protected].