Two Records Ban Double-Counting 642 kW Recharge, Then Disagree What It Is

If the paste-ready sentence is applied tonight, which 642 kW does it insert into four calculations? Taking 233 kW away from 642 kW to reach 409 kW is a double deduction, and the shared project memory and the errata chapter both forbid the result [TARGET]. What 642 kW is, the two documents do not agree on. Read as the memory writes it, 642 kW is the headroom remaining against the facility input limit at the charging peak, and the error fabricates a shortfall. Read as the errata writes it, 642 kW is a branch load that already contains the charging term, and the error understates a load. Consequence points in opposite directions under the two readings.

Shared Memory Treats 642 kW as Remaining Headroom

Steady-state facility power is given in the memory as 14,125 kW and the charging peak as 14,358 kW, with the margins against the 15 MW input limit stated as 875 kW and 642 kW [TARGET]. All three subtractions close exactly: 15,000 less 14,358 is 642, 15,000 less 14,125 is 875, and the difference between the two power figures is 233 kW [MODELED], which is the charging increment itself. That last identity is what makes the second deduction a double one, because the 233 kW is already inside the 14,358 kW that produced the 642 kW.

A neighbouring line in the same memory records a different steady figure: 14,391 kW of facility power, leaving 609 kW to the 15 MW cap [TARGET]. The heat-rejection chapter states that 609 kW is to be compared with 875 kW, not with 642 kW, and that the move from 609 to 875 is a re-itemisation of the column, not a design that got better.

Errata Replacement Text Still Calls 642 kW a Load

Reading 642 kW as a branch load that already includes the charging component, the errata chapter concludes that 409 kW understates that branch by 233 kW. It computes the understatement as 36.3 per cent of 642 kW [MODELED]. Paste-ready replacement text is then issued, instructing that facility-side, generator-side, heat-rejection and efficiency-ratio calculations all take 642 kW, and four consequences of not doing so are listed [TARGET]:

• branch current and protection settings;
• generator capacity and the derating check;
• the matching heat to be rejected;
• the numerator of the efficiency ratio.

Under the margin reading that instruction inserts a headroom figure where a load belongs, in four calculations at once. Under the load reading it is correct. The ratio survives both readings, because 233 divided by 642 is 36.3 per cent either way, and the ratio is the only thing that does.

The errata chapter is explicit about why it cannot settle the question alone. It opened no file outside its disposition matrix. It grades the exact wording of its own short quotations as unconfirmed. It instructs that where the anchor wording differs, the target document governs and the replacement is adjusted to the same meaning.

Heat Rejection Rewrote 642 kW as Remaining Headroom

When the heat-rejection chapter later executed that errata, it did not paste the branch-load sentence. It rewrote the row as two remaining-headroom readings, 875 kW steady and 642 kW with charge already in, and forbade chaining them. Downstream files that still show 409 kW are told to correct to that pair and to name which reading they took. The candidate text and the executed text therefore still disagree about what 642 kW is, after the correction has been applied.

The overall controller, named in the entry as its responsible party and named again in its closure evidence as the signatory, can settle it in one line by stating which of the two 642 kW is. Until that line exists the paste-ready text cannot be applied, because the same sentence is right under one reading and wrong under the other.

Charging Input Also Rose; the Two Corrections Accumulate

A third figure in the same chain moved while this one was being argued. Charging power in the third load column was corrected from 535 kW taken at the battery side to 581.8 kW at the charger input, an increase of 46.8 kW [MODELED] at an implied channel efficiency of about 0.92, which is graded as an assumption by the errata, awaiting nameplate data and a state-of-charge recovery curve. Both corrections point the same way, the same chapter notes, and therefore they accumulate rather than cancel. A third errata in the same batch points the other way and is forbidden from being netted against these two.

One consequence is already visible upstream. The heat-rejection chapter's interlock failure case, the one where the control system is down while the generators start, now writes the charging term as a range from 233 to 581.8 kW [MODELED] instead of a single figure. So the quantity the double deduction removed twice no longer has one value to remove, and any later correction of the deduction inherits a range where it used to inherit a number.

Both readings accept one legitimate meaning for the banned figure, the branch taken without battery charging. No document in the read set says where that meaning may still be used, and none gives that residual interpretation an owner.

Limits and open items

Confirmed: both documents forbid 409 kW as a load and as a remaining margin. [FACT]

Column figures, not a freeze: 14,125 kW steady and 14,358 kW charging peak; 875 kW and 642 kW remaining against 15 MW; the neighbouring 14,391 kW / 609 kW steady pair. [TARGET]

Study values: 233 kW as the charging increment already inside 642 kW; 36.3 per cent; 535 kW to 581.8 kW at about 0.92; the 233 to 581.8 kW range on the interlock-failure row. [MODELED]

Open: a one-line ruling on which of the two 642 kW is; an owner for the residual "without charging" reading of 409 kW; nameplate efficiency and a state-of-charge recovery curve behind 581.8 kW. [HOLD]

Nothing here has been reviewed or sealed by an Engineer of Record. Nothing describes how any installed system behaved.

Name Whether 642 kW Is Headroom or Branch Load

The useful test is not whether 409 kW is banned. Write the sentence you would paste into the four calculations, and say whether the 642 kW in it is remaining headroom or a branch load. If that sentence cannot be written, the controller's line is still the missing product.

© 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., “Two Records Ban Double-Counting 642 kW Recharge, Then Disagree What It Is,” https://www.kkdatasvc.com/lab/errata-and-open-questions/both-documents-ban-the-result-and-disagree-on-what-642-kw-is/.