Read the method, check the equations, and run it yourself before you rely on a number.
Your situation
A screening result is only as useful as your ability to reproduce it. The relations behind each constraint are published with their inputs and a worked example, so you can recompute a result by hand and see where you disagree.
The engine takes inputs a client rarely has complete. What matters to an engineer is that every gap is named, every default is labelled an assumption, and the result carries the weakest evidence class that went into it, so you know what to measure.
The engine runs with no dependencies on the standard library, so it can be reviewed and run in an air-gapped environment, and exposes the same solve as an API and an MCP endpoint for tooling you already have.
The decision this informs
Whether the method is sound enough to use as a screening tool, where it disagrees with your own calculation, and what to measure to close the gap.
THE FIRST QUESTIONCan I reproduce this number, and which of its inputs are assumptions?
What you probably already have
- Your own single-lines, ratings and calculations to compare against
- The hall's measured data, if there is any, which the engine accepts as client-supplied (E5) input
The constraints most likely to decide your answer
- Busway ampacity
A pooled-run model and a run-by-run model give different rack counts; the difference is documented.
A 400 A busway run cannot feed three 130 kW racks, whatever the hall total says.
- Hydraulic flow capacity
A flow limit compared against the loop's demand is a check you can repeat from the published relation.
A flow figure without a head figure is not a duty point.
- Achievable facility water temperature
A temperature approach is where assumed and measured inputs diverge most.
Legacy plant makes 6 °C water beautifully. Liquid cooling would rather have 30 °C.
The route through what can be bought
It starts free and each step credits or informs the next; none is required. Prices are read from the live catalogue.
- Free capacity checkFree, no card
See what the engine asks for and what it returns, with nothing to sign.
- API — Triage€900 per month
Run the engine from your own tooling with a metered key, screening-mode output only.
- Density Screen€4,500
5 working days from a complete intake
A reviewed document for one hall, to put in front of a colleague or a client.
What to ask us before you rely on the answer
- Which review findings are still open, and where is the calibration record?
- Which models are simulations and which are closed-form relations?
- Where do your published limitations say the model is weakest?
What it is not
- The calibration ledger is empty today: the models have not been checked against measured data from a real hall.
- Output is screening-mode. It carries no named signatory and no professional indemnity.
Time to Power sells no equipment, holds no inventory and owns no megawatts. A figure is modelled unless its evidence class says otherwise, and a date that rests on an assumed lead time is labelled as one.