Ask the supplier which of the two items is the unit being bought. Two plate-heat-exchanger documents arrived in the same supplier intake, both dated 15 September 2026, both carrying the same model designation, and each carrying a different item identifier. The datasheet lists 67.3 m³/h secondary and 100.1 m³/h primary flow at pressure drops of 69.2 and 81.1 kPa. The dimensional drawing lists 52.2 and 77.6 m³/h at 42.76 and 50.30 kPa [MODELED]. The reviewer graded the finding class A and typed the evidence as a confirmed document difference rather than an inference, and left it open. Nothing in the package links the two identifiers.
Both Files Are Same-Day Peers, Not Successive Revisions
The intake manifest is worth noting because it removes the ordinary explanations. Both documents are separate files in one received package, each with its own digest and byte count, logged on one receipt date. They are not successive revisions separated by weeks. Neither supersedes the other on its face.
Same day, same model name, same delivery, two item identifiers, two duty points. The reviewer's required response is correspondingly narrow: a written statement from the manufacturer relating the two identifiers, and a single unified configuration, drawing, duty point, interface and certification basis.
The Drawing Point Is About 700 kW, Not the 900 kW Rating
The flows differ by enough to matter. Using the fluid properties listed with the datasheet, the reviewer back-calculates the drawing's temperature and flow point at roughly 700 kW, against the 900 kW [TARGET] selection the datasheet is built around [MODELED]. The pressure drops move with the flows, as they must, and it is the flows that carry the duty.
The reviewer also checks the datasheet against itself. Its listed terminal temperatures and flows reproduce about 904.97 kW on the hot side and 897.79 kW on the cold side, inside 0.6 % of the 900 kW rating [MODELED] — internally consistent, and therefore not evidence of an arithmetic error anywhere. The package's constant conductance figure is likewise the rated duty divided by the log mean of the stated end differences, which is a restatement of the rating rather than an independent test across operating points.
The 700 kW Point Looks Like the Normal Operating Case
Here the article departs from the review record, and the reader should treat the next two sentences as author judgement rather than a reviewer finding.
The same package's duty table gives the unit's long-term normal external liquid heat as 697.5 kW, rising to 709.405 kW at the exchanger once secondary pump heat is added, with a newly introduced long-term peak of 775 kW [MODELED]. A drawing point near 700 kW therefore sits on the normal operating case, not somewhere arbitrary, which makes the innocent reading — a rating document and an operating-point document, both true — the most likely one.
It is also the reading that does the most damage if it is wrong, because it is the one a buyer adopts without asking. Nothing in the two documents says which is which.
The Document Gap Does Not Prove a 700 kW Hardware Limit
The disposition recorded against the finding is careful in a way worth preserving. Estimating from the datasheet's properties, the drawing's listed point corresponds to about 700 kW and not 900 kW; this may be a different selection point on the same plate type; it is not sufficient to conclude that the physical unit can only achieve 700 kW; and the supply record is not closed.
That is four separate statements, and only the last is a conclusion. The reviewer treats a document discrepancy as a document problem, which is why the required response is a letter rather than a re-test.
The same restraint appears in the capacity-margin table, which reports 26.867 % relative margin against 900 kW at the normal case and 14.372 % at the new peak [MODELED], and then notes that a normal case plus 20 % needs 851.286 kW, which 900 covers, while the peak plus 20 % needs 944.286 kW, which it does not. The margin question is live independently of which item identifier is supplied.
Limits and open items
Held: which item identifier is the supplied configuration, whether the two are the same physical assembly, and what duty is guaranteed for the item actually delivered. [HOLD]
Not established: any conclusion about the hardware's real capability, any certification status for either identifier, and any statement that either document is wrong. Nothing here reflects Engineer of Record review, authority approval, equipment-supplier confirmation, commissioning acceptance, certification or procurement release. No vendor, model designation or item identifier is reproduced.
The condition that changes the conclusion is a single written manufacturer statement tying the identifiers to one configuration and offering a guarantee point at the rated duty for the item actually supplied. If it arrives and confirms one assembly with two selection points, this stops being a finding and becomes a filing note. If it arrives and names two assemblies, the interface, dimensional and certification work done against the wrong one has to be redone — which is why the response is owed before, not after, the equipment is committed.
© 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., “Same-Day Plate Heat Exchanger Datasheet and Drawing Disagree,” https://www.kkdatasvc.com/lab/direct-liquid-cooling-and-cdu/same-day-phe-datasheet-and-drawing-two-different-items/.
Discuss This Constraint. 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.