Peak-Flow
Gut-Check

This tool estimates your operation's real peak water demand, the instantaneous flow rate during the moment several stations call for water at once, and compares it to a plain-language risk read on whether that peak is likely to cause a pressure drop severe enough to trip a low-water-pressure shutdown on connected equipment. Most filtration and plumbing gets sized to average demand. The peak is a different, much higher number, and it is exactly when undersized filtration chokes.

Foodservice operators, multi-unit operations teams, and plumbing contractors who want a fast read on whether their peak demand is a real risk before it becomes a service call.

• Quick Path, pick your operation type and enter a guest or seat count. Fastest way to get a directional number.
• Detailed Path, enter counts for each water-using machine on site (ice machines, coffee brewers, dishmachines, and so on). More precise, because it is built from your actual equipment, not a typical mix for your operation type.

Filtration, plumbing, and water service are usually specified against average or nameplate-style demand, a comfortable, steady number that looks fine on paper. But water use in a foodservice operation is not steady. It spikes hard during a rush: the ice machine refills, the dishmachine cycles, several prep and hand sinks run at once, the fountain and the coffee brewer are both drawing, all inside the same few minutes. That instantaneous total is the peak, and it is routinely several times higher than the average the system was actually sized for.

This tool's diversity model was checked against eight days of ten-second-interval flow metering at a large Aquamor customer site. The result: for casual dining at that site's scale, the model reproduces the measured baseline, peak, and multiplier within about 1 percent. An earlier version of this tool under-counted the peak by a factor of roughly 1.7 to 2.5. It applied two different discounts for the same physical fact (that not every machine draws water at the same instant), which is a double-count, not a conservative estimate. That defect is now fixed at the level of the calculation itself, not papered over with a bigger safety-factor multiplier, which is why this tool no longer applies a separate design margin on top of the diversified estimate; the diversified estimate is now the design peak. Casual dining is the one operation type in this tool with real measured ground truth behind it. The other nine formats run the same validated method, extrapolated to a site profile that has not itself been metered, see "How confident is this, really" below.

If you simply added up the rated peak flow of every machine on site, you'd get an absurd number, every ice machine, dishmachine, and sink calling for water at the exact same instant almost never happens in practice. The diversity factor corrects for that: it keeps your single highest-demand machine at its full rated flow (because SOMETHING is always drawing when a rush hits), then discounts the combined demand of everything else by a factor that grows more conservative as your equipment count grows. The result is a realistic system peak instead of a meaningless worst-case sum. Aquamor cross-checked the underlying equipment figures against a published foodservice fixture-flow sum for a small restaurant, before diversity was applied, and found close agreement, a real, if partial, external confirmation of the numbers this tool is built on.

Not everything on site runs through the filter. Three-compartment sinks and mop/service sinks are usually plumbed straight from the incoming line, not through the cartridge. This tool's headline number is the filtered-branch peak, what your filtration actually has to pass, with the whole-site peak (including those unfiltered fixtures) shown as secondary context. Size your filter against the headline number, not the whole-site figure.

Measured flow data shows something operators often say anecdotally: the highest sustained flow of the day frequently happens at closing, not during the dinner rush itself. At the validated site, the hour ending at 10 PM carried the highest average flow of any hour, on every meter measured, with late service and end-of-day cleanup overlapping. This tool does not model that as a separate number to check, a correctly calibrated single peak already reflects it, and adding a second closing-period calculation on top would double-count the same event. What it means in practice: if your pressure problems show up at closing rather than at dinner, that is consistent with where the real peak usually sits, not a sign that something different is happening.

Filtration and plumbing sized to average demand can look completely adequate for most of the day. It's the brief rush burst when everything calls for water at once that exposes an undersized system. Pressure sags, filtration chokes, and connected equipment that depends on stable incoming pressure can trip a low-water-pressure shutdown right in the middle of the rush. That failure mode almost never shows up outside the peak window, which is exactly why so few operators ever catch it before it costs them a shift.

A few equipment categories have a hard minimum incoming pressure to run correctly at all, independent of flow rate, boiler-type steamers are the most demanding at roughly psi, with espresso machines close behind at around psi. If your site is near one of those floors, that single piece of equipment can be your real limiting factor even when the flow-rate math looks fine. This tool flags it when it applies.

Both paths run through the identical calculation: Quick Path simply looks up a typical equipment mix for your operation type and size first, then feeds it through the same engine Detailed Path uses directly. That means the two paths cannot disagree with each other for a typical mix; the only way they diverge is if your site's real equipment differs from what's typical for its type and size, in which case Detailed Path is the more accurate read because it reflects your actual machines rather than an estimate.

Be plain-eyed about where these numbers come from. Casual dining is the one operation type in this tool checked against real field measurement, and the tool marks that result "Field-validated" for exactly that reason. Every other format runs the identical, now-validated calculation method, but extrapolated to a site profile that has not itself been metered, the tool marks those results "Engineering estimate." A smaller subset of results, where the calculated peak runs many times higher than the operation's baseline, most often small, low-headcount formats, carries a further "low confidence" flag: Aquamor's own engineering review treats those multipliers with real skepticism pending measurement at a second, smaller site, and you should weight them accordingly. None of this makes an estimate result useless. It means treat it as a well-reasoned engineering estimate, not a site measurement, and lean more heavily on Detailed Path with your own equipment counts when the stakes justify it.

This is a gut-check, not a full engineering study. It does not know your incoming service size, your static pressure, your pipe run lengths, or your specific filtration model's rated flow, all of which determine whether a given peak actually trips a shutdown at your site. Treat the risk verdict as a strong directional signal for whether it's worth a closer look, not a final engineering determination.