Frequently asked questions

GHE Analysis sizes the ground loop and simulates the entering fluid temperature hour by hour, over a design horizon of one to fifty years and up to 625 boreholes. The g-function of the field you draw is recomputed automatically, rather than looked up from a catalogue of predefined layouts.

GHE Analysis integrates up to 625 boreholes directly, at depths up to 500 m, in regular or fully irregular layouts — larger fields, up to 3,000 boreholes, are available on request. The g-function is computed for the field as drawn, never looked up from a catalogue of predefined configurations.

Yes. Moving a group of boreholes recomputes the field's g-function, the fluid temperature simulation and every chart — together, before you let go, not as a staged sequence you wait through.

Yes, and throughout the entire simulation, not as an add-on: regional flow, set by direction and Darcy flux, acts on every borehole for the full design horizon. Among geothermal design tools, this native integration is unique to GHE Analysis.

Yes — water-to-air, water-to-water and hybrid systems are all simulated hour by hour, with geocooling and domestic hot water integrated rather than left as manual adjustments, and heat pump COP and capacity interpolated from the entering fluid temperature the loop actually delivers, not read off a nameplate curve. A fluid cooler or auxiliary heater is dispatched alongside the heat pumps rather than sized by a rule-of-thumb margin, so you see exactly how much field the hybrid strategy buys back.

No. The interface is built for a fast, guided first session, not a course of video tutorials before you can start a design. Context-sensitive help is one click away on every panel, and the full manual is available online or as a PDF whenever you need more depth.

Yes. The application ships with 137 verification scenarios covering every stage of the calculation — borehole resistance, g-functions, sizing and fluid temperature — each checked against independent published methods and reference tools. Every scenario runs from the Examples and verification menu, so the comparison isn't just a claim.

Yes. A comma-separated file (.csv) is generated automatically every time a project is saved, written beside the project file in the same folder. It carries the full hourly results — no export step, no extra dialog, and nothing to remember to do before closing the project.

No — flow is modeled as a uniform Darcy flux through a homogeneous medium, the same simplification every geothermal design tool of this kind makes, so layer thickness is not a required input. Site stratigraphy and depth-varying flow are not modeled today.

Yes, when the same field alternates between charging and discharging heat over the year. What it does not model is a field split into permanently separate hot and cold zones operated at the same time — the field is simulated as a single thermal system.

There is no limitation to work around: heat pump performance is entered directly, as a table of capacity and COP against entering fluid temperature. Paste it from a spreadsheet or import a CSV file, and GHE Analysis fits a smooth curve through the points — any make or model works the same way.

No — Rb* is recomputed by the multipole method at every temperature and flow rate the simulation reaches, never entered once as a constant. Grout and pipe conductivity, spacing and the fluid's own properties all shift Rb* as conditions change over the design horizon.

Three fast methods combine: a fast, parallel, successive-flux-estimation solve for the g-function; a rapid spectral (FFT) convolution of the load history instead of a step-by-step sum; and a fast interpolated solver for heat pump COP and capacity at each hour's fluid temperature. Together they return the full fifty-year horizon in under 50 ms for most cases.

Yes. A part-load degradation coefficient (Cd) is applied to the heat pump's COP whenever it cycles below full capacity, rather than assuming full-load efficiency at every operating point. Ignoring this effect is a common source of over-optimistic seasonal performance estimates.

Yes. A georeferenced background image — an aerial photo, a site survey, a property plan — can be imported and used as the canvas for the borehole layout, so boreholes are drawn against real site constraints instead of arbitrary coordinates.

Both. Circulator power is folded into the seasonal COP like any other parasitic load, but the flow rate it implies also feeds back into the Reynolds number and the borehole's convective resistance — an oversized circulator or excess antifreeze can quietly erode the efficiency a well-designed field was meant to deliver.

Several. Distributed water-to-air heat pumps per zone, distributed water-to-water units serving zone hydronics, and a central water-to-water plant feeding a building loop are all supported, alongside geocooling variants — the schematic redraws itself to match whichever configuration you choose.

TRT Analysis reads a thermal response test record and returns the ground conductivity, the borehole resistance and the undisturbed ground temperature — including tests interrupted by a power outage or a stopped pump, which the standard line-source method cannot use.

Yes. The short-term g-function model is convolved with the heat-injection history exactly as it was measured, so it follows every ramp, pause and shut-off in the record instead of assuming a constant injection throughout.

Yes, when the test allows it. The first-order approximation and the short-term g-function use different data through different regressions, so running both on the same test and comparing is a built-in consistency check, not an extra step.

Yes. Six sliders calibrate the short-term g-function directly against your measured record, and the residual statistics update with every move — so you see in real time which parameter the data actually pins down and which one is only guessed.

No. The interface is built for a fast, guided first session, not a course of video tutorials before you can start an interpretation. Context-sensitive help is one click away on every panel, and the full manual is available online or as a PDF whenever you need more depth.

Three methods are available: a manual entry, a reading from the test's own circulation phase before heat injection starts, or a logged vertical temperature profile averaged over depth. The method best supported by your data drives the rest of the interpretation.

Both an arithmetic mean and a p-linear average of the fluid temperature are available, and the two are not interchangeable: the p-linear average weights the temperature correctly along its full range, which a simple average does not. Which one matches your test's conditions is part of an accurate interpretation.

It is a neural network pre-trained on three-dimensional transient finite-element simulations, matching that reference data to within 0.0001°C — not fitted to field tests after the fact. Interactive calibration against your own record then refines it, with the residual error shown live as you adjust.

No. Standard, pulsed and interrupted tests are all interpreted the same way, from the measured heat-injection history rather than an assumed constant rate — heating and recovery phases are read together, so a test that was paused or restarted is not discarded.

Yes. The Reynolds number and flow regime are flagged for every phase of the test, because a line-source interpretation assumes turbulent flow — outside that regime, it is the effective borehole resistance (Rb*) that is most affected, more than the ground conductivity itself.

The critical time is drawn directly on the regression chart and updates automatically as the analysis progresses, with nothing to recompute by hand. It marks where the infinite line-source assumption starts to hold, and choosing the fit window around it is the single factor that affects the result most.

Yes. The module ships with 43 verification scenarios and 294 individual checks, most analytical or built from data shaped so the expected result is hand-computable, plus a sandbox test result checked against a published field experiment. Every scenario runs from the Examples and verification menu.

GLN Analysis solves the full hydraulic system, not just the ground loop: the mechanical room and the building's own distribution network are drawn and solved the same way — flow, velocity, friction and pump head in every pipe, plus the spread of flow between borehole loops that says whether the field is balanced.

Instantly. Draw the system as a piping and instrumentation diagram — several independent networks in one drawing, if the project needs it — and the full solution comes back before you've moved on: flow, velocity, friction and pump head in every pipe and network, even on a five-hundred-borehole field.

Yes. Its accuracy is checked against EPANET, an independent solver we did not write, across dozens of reference networks — not tuned to agree with itself, but tested against a solver with no reason to.

Yes — for example, whether the antifreeze concentration chosen for thermal protection has moved the pump's operating point somewhere the circulator cannot follow, a question a purely thermal analysis never raises.

No. The interface is built for a fast, guided first session, not a course of video tutorials before you can start a design. Context-sensitive help is one click away on every panel, and the full manual is available online or as a PDF whenever you need more depth.

Yes, truly closing. The solver arms a check valve from the network's topology and shuts it the moment flow would reverse, rerouting through any parallel branch — it does not apply a fixed resistance in both directions the way a simplified model would.

Yes. Each network in the same drawing can carry its own fluid, flow rate and temperature — a heating loop and a cooling loop, for example, solved side by side rather than forced to share one set of conditions.

Yes. Every pipe is checked against a velocity window and a friction-gradient window as soon as the network solves, and the result table flags any pipe outside either range — no separate check to run, and nothing to look up in a chart.

Yes — a 120-borehole reverse-return field with 362 nodes and 480 pipes has been solved and checked against EPANET, an independent solver, to within 0.042% on flow. The same sparse solver that handles a five-pipe network scales to a full campus field.

No — extra antifreeze buys no added safety once the freeze point sits a few degrees below the coldest fluid temperature expected. It thickens the fluid, raises the pump head, lowers the Reynolds number and can push a marginal leg into laminar flow — the size the design actually calls for, not more, is the safer choice.

The curve is completed automatically — extrapolated to the shutoff head at zero flow if the first point isn't there, and extended to the runout flow where the head reaches zero. Without it, a truncated curve could let the solver push the operating point onto a non-physical extension and return an absurd flow.

Yes — they are different quantities that multiply together, not one combined number. Typical hydraulic efficiency runs 25 to 45% for a small wet-rotor circulator and 60 to 80% for a well-sized centrifugal pump; setting it to 100% would ignore real pump losses and understate the power the system actually needs.

Not automatically. A variable-speed pump can be solved in constant-head or constant-flow control, but the curve itself is not rescaled by speed — a genuine variable-speed analysis needs the already-scaled curve entered as its own pump slot.

Yes — Windows 10 or later and macOS 14 or later, as a desktop application. Nothing runs in the cloud: projects stay on your workstation, and the simulation uses the cores of the machine you already have rather than a metered service.

Yes. The application ships with executable verification scenarios that reproduce published benchmarks on your own machine, and every model is cited on its module page. The hydraulic solver's own accuracy is checked against EPANET, a second solver that was not written by us.

Engineers and researchers run it in eleven countries, from single houses to campus-scale systems, and students are taught on it every year through the free academic licence. It is actively developed, too — new modules and verification scenarios continue to ship, not a research tool left on a shelf.

Not yet — every module works and is reported in SI units today. Support for imperial units is something we could add if enough people ask for it; write to support@gheanalysis.com if that would help your workflow.

It depends on how big the projects you design are — the number of boreholes, pipes and how much detail you feed the model. All three modules are included from the Intermediate tier up; the Small tier covers GHE and GLN only, sized for a residential loop with no thermal response test to interpret.

Move up a tier whenever you need to, and we make it fair for the time remaining on your current one — you are never stuck mid-project.

Per seat, per year, for Small, Intermediate and Large — priced per person, not per company. Large is also available for the whole team, with unlimited seats, and as a monthly, no-commitment plan instead of the annual one — write to support@gheanalysis.com to arrange either.

Yes. The demo needs no account, no email address and no credit card: download it and it runs. It exposes every feature of the application with limited input ranges, so you can follow a complete workflow through all three modules before deciding anything.

You have thirty days from the date of purchase to change your mind and be refunded, no conditions beyond that window.

Yes, for academic research and classroom use, at the same ceilings as our largest tier. Write to us from your institutional address and tell us what you are working on.

Your projects stay readable, free of charge, in consultation mode: Analysis Suite opens any project file without a licence, whatever its size, recomputes it and draws every chart, so it can still be read and printed years later, by you or by anyone else.

Write to support@gheanalysis.com and ask. Several features already in the software started as a customer's question, including some answered on this page — there's no guarantee every request becomes a feature, but every one is read by the people who write the code, not routed through a support queue.