A controls package written to one number will either hold 35 per cent of rated branch flow during filling, which is outside the criterion that justifies the number, or hold a higher figure permanently, which discards the part-load band the parallel-running strategy depends on. The control sequence carries one written lower bound on cabinet branch flow, 35 per cent of rated, and the reason recorded for it is turbulence rather than temperature [MODELED]. That rule and the threshold behind it are already on the record and are taken here as stated. What is not on the record is that the same percentage describes two different flow regimes, because the viscosity the criterion depends on moves with the fluid temperature and the written figure does not.
A Reynolds Floor Is Temperature-Dependent, Not a Percentage
The criterion is a Reynolds number, and Reynolds number is velocity times bore divided by kinematic viscosity. For the 25 per cent glycol mix the chapter tabulates, that viscosity is 1.14 square micrometres a second at the 40 degree design mean and 2.88 at 10 degrees, a factor of 2.53 [TARGET]. So the same branch that sits at 60,000 at rated flow and design temperature sits at 23,800 at rated flow and fill temperature. Applying the one written percentage to both gives about 21,000 in the first case and about 8,300 in the second [MODELED]. Those are not the same flow regime, and the second is below the transitional threshold the rule exists to stay clear of.
Exposure to that threshold is local, which is the other half of the finding. On the same property basis the zone ring at its design velocity sits at 261,700 and the distribution-unit connection at 262,100, and both are still above 103,000 at fill temperature [MODELED]. Those runs are two orders of magnitude clear of transition and would stay clear at any flow the plant can physically reach. Only the cabinet branch, which is an order of magnitude smaller in Reynolds number at full flow because its bore is a third of the connection's, comes anywhere near the threshold. A single plant-wide minimum-flow rule is therefore solving a problem that exists on one pipe class.
Below the line the chapter does not say to pinch valves. It says to close the branch and use bypass. Diagnosing the same low-flow state as a capacity miss is a different error, recorded on the heat-rejection side.
Fill Temperature Compresses the Branch Reynolds Margin
The chapter also prints a cold-start figure a few lines from the rule and never sets the two against each other. That pairing, and the viscosity shift that produces it, are recorded elsewhere in this series and are not reworked here. What is not recorded elsewhere is the shape of the ladder at each temperature. At the design mean the branch reaches the criterion at about a third of rated flow and stays clear of transition for a long way below that; at fill temperature the same ladder is compressed by the viscosity ratio, and the one written setting lands inside the compressed part of it rather than above it.
A supplier review of a different packet records the same family of gap: room-temperature viscosity does not cover a cold start. The parametric engine that expands the hydraulic formula across a grid does not size this pipe and does not replace the branch Reynolds table.
One Min-Flow Setpoint Cannot Cover Fill and Part-Load
What follows is a setpoint table rather than a setpoint. A controls package written to one number will either hold 35 per cent during filling and early commissioning, which is outside the criterion that justifies the number, or hold a higher figure permanently, which discards the part-load operating range that the parallel-running strategy depends on for its pump energy saving. Writing the bound as a Reynolds target with a temperature-indexed flow floor costs one extra column in the point list and removes the choice.
A different inhibitor package or a different glycol concentration moves the viscosity, and every Reynolds number above moves with it, including the one the written percentage produces at fill. The chapter grades the property set as general correlation data, with the purchased fluid's own sheet not obtained, so the table carrying all three figures is unconfirmed and the formulation behind it is not fixed.
Rated branch flow is not settled either, so the denominator moves as well. It is derived from a cabinet heat load and a temperature difference, both of which are open items against the equipment supplier, as is the manifold and cold-plate pressure drop that would fix the branch characteristic in the first place. Until that characteristic arrives, the written percentage has no supplier floor under it.
Limits and open items
Confirmed: at full load the tabulated runs sit far above transition; the risk the written percentage is addressing is part-load flow on the cabinet branch. [FACT]
Property table, unconfirmed: kinematic viscosity 1.14 square micrometres a second at the 40 degree design mean and 2.88 at 10 degrees, a factor of 2.53. [TARGET]
Written bound and the two regimes it produces: 35 per cent of rated; 60,000 at rated flow and design temperature, 23,800 at rated flow and fill; about 21,000 and about 8,300 at the written percentage; zone ring 261,700 and connection 262,100, both still above 103,000 at fill. [MODELED]
Open: a temperature-indexed column in the point list; the purchased-fluid sheet; the supplier min-flow, rated-flow and max-flow band. [HOLD]
Nothing here has been reviewed or sealed by an Engineer of Record. Nothing describes how any installed system behaved.
Add a Temperature Column Before Issuing the Sequence
None of this makes the 35 per cent wrong. It makes it a number that cannot be read without its temperature, and a control sequence has nowhere to write a temperature next to a percentage unless someone adds the column before the sequence is issued.
© 2026 K&K Data Service Inc. All rights reserved. Reproduction or republication is permitted only with clear attribution to K&K Data Service Inc. and a working hyperlink to the canonical URL of this article. Excerpts must preserve the technical context, maturity labels, assumptions, and limitations. No excerpt may imply project approval, field validation, certification, or endorsement that the original article does not state.
Source: K&K Data Service Inc., “One Cabinet Min-Flow Percentage Describes Two Reynolds Regimes,” https://www.kkdatasvc.com/lab/hydraulic-calculation-detail/one-written-percentage-is-two-reynolds-numbers/.
Request the Public Calculation Note. If your project record shows a different result, or the same failure domain under another name, we want to see it. Email inquiry@kkdatasvc.com or use the contact page.