HomeFeaturesExamplesTutorialsHelpTheoryVerificationPricingContact
Theory manual

26. Input checks and safety checks

A calculation that succeeds is not necessarily a sound design. Inputs can be valid numbers and still describe something unintended, such as a peak that is never applied or a time series that ends after half a year, and a design that meets its limits on the mean fluid temperature can still freeze at the inlet. BoreTherm therefore checks the inputs before every calculation and the results after it. Errors stop the calculation and name the input; warnings accompany a valid result and appear in the result list, on the command line and in the report. This chapter lists the checks, gives the freezing check and the load coverage in detail, and explains what the warnings mean for the design.

26.1 One set of input rules

The same rules run before a calculation, before a sizing, when a project is validated without calculating, and on the command line, and they use the same messages, so the texts agree wherever they appear. When a project file is read, keys the program does not read and physical inputs that are missing and filled with their defaults (ground conductivity and heat capacity, surface temperature, geothermal heat flux, borehole length and radius, flow rate, fluid, the monthly energies, the design limits) are reported as well.

Table 51. Input rules applied before every calculation.
InputRuleSeverity
Monthly loadsnon-negative and finite; the first bad month per column is namederror
Peak durationsbetween 0 and 730 h (one month)error
Hourly loadsnon-negative and finite, with the hour named; all columns of equal length; 8760 values or a whole number of years (8784: leap-year hint)error
Time seriesreadable, ordered times, finite valueserror
Seasonal performance factorsheating and DHW greater than 1; cooling positiveerror
Ground loads after conversionfinite (an input such as 10307 MWh per month is not)error
Borehole tiltfinite and at most 85°error
Borehole axesboreholes that intersect or coincideerror
Design limitsminimum mean fluid temperature below the maximum; an inlet limit, when on, finiteerror
Sizingminimum length positive and below the maximum; tolerance positive; at least two iterationserror
Monthly peaksa peak power with a duration of 0 h; a peak that does not exceed the month's mean ground load (Section 26.6)warning
Time-series columnscolumns the chosen borehole model does not apply (Section 26.7)warning
Load coverageloads that do not cover the design period (Section 26.5)warning

A seasonal performance factor of exactly 1 for heating or DHW is refused because it means that the heat pump takes no heat from the ground; loads that are already ground loads are entered as direct ground loads. A non-finite result (the effective borehole resistance, a monthly or hourly mean, inlet or outlet temperature) is reported as a numerical failure, with the quantity and the month or hour, and never presented as a design.

26.2 Plausibility warnings of the design

After a calculation the result is checked against the design limits and the usual ranges of borehole design. None of these warnings stops the calculation.

Table 52. Warnings checked after every calculation.
WarningConditionMeaning for the design
LimitsTf,min below or Tf,max above the design limitthe design does not meet its limits: size the field, or change the design
Near freezing (mean)Tf,min within 1 K of the freezing point of the fluid, or below itthe limit itself is too close to freezing: raise it or use a fluid with a lower freezing point
Freezing risk (inlet / outlet)lowest inlet or outlet temperature less than 1 K above the freezing point (Section 26.3)the mean meets the limit but the coldest fluid does not have the margin
Laminar flowReynolds number in the pipes below 2300 (Section 26.4)poor heat transfer: a higher borehole resistance and a longer field
Length rangeborehole length outside 20–300 m (energy piles excepted)outside the usual range of borehole design; check the inputs
Spacing ratiosmallest spacing divided by the length, B/H, outside 0.02–0.5unusually dense or sparse field; check the layout
Close axestwo borehole axes closer than the warning distance (default 1 m)check the drilling tolerances; the calculation treats boreholes as lines
Site planboreholes outside the drilling areas, in no-go zones or closer than the setbacksthe layout does not fit the site

26.3 Freezing check of the heat carrier fluid

The design limits apply to the mean fluid temperature Tf, the mean of the inlet and outlet temperature of the field. In heating the fluid entering the field is colder than Tf by half the temperature change across the field:

Tin=Tf+Q2ṁcp,Tout=Tf−Q2ṁcp
(26.1)

with Q the heat rate into the ground (positive for injection, negative for extraction) and ṁcp the heat capacity flow of the whole field. The lowest temperature in the circuit is therefore the inlet in heating, and it is lower the higher the specific load and the lower the flow. The check evaluates

The result gives the lowest inlet and outlet temperatures with their times, the freezing point Tfr of the fluid at its property temperature (for aqueous solutions from the property fits of Melinder 2010), and the margin

ΔTfr=min(Tin,min,Tout,min)−Tfr
(26.2)

A margin below 1 K, the same margin as the warning on the mean fluid temperature, gives a freezing-risk warning with the end of the field (inlet or outlet), the value, the year and the month. Two examples show the size of the effect:

The check warns; it does not fail the design. The sizing criterion stays the mean fluid temperature, as in the classic monthly method. For designs where the inlet temperature matters (water as the heat carrier, energy piles, strict antifreeze rules), an optional minimum inlet temperature limit can be switched on with the design limits. It then takes part in the limits check and in sizing like the limit on the mean; the sizing violation becomes

v=max(Tf,min,lim−Tf,min,Tf,max−Tf,max,lim,Tin,min,lim−Tin,min)
(26.3)

over the whole period or the last year only. The lowest inlet temperature rises with the borehole length like Tf, so the length search is unchanged. Remedies for a freezing risk are an antifreeze mixture or a lower freezing point, a higher flow (a smaller difference between inlet and mean), or a longer field.

26.4 Laminar flow

The convective resistance between the fluid and the pipe wall depends on the Reynolds number Re of the flow in the pipes. In turbulent flow (Re≥4000) the Nusselt number follows Gnielinski (1975); in laminar flow (Re≤2300) it is the constant 3.66; between the two it is interpolated linearly in Re (Gnielinski 2013). The laminar Nusselt number is an order of magnitude below the turbulent values at typical flows, so the fluid-to-pipe resistance and with it the effective borehole resistance Rb* rise markedly when the flow becomes laminar. A Reynolds number below 2300 gives the warning "Laminar flow in the pipes: heat transfer is poor", with the value.

Laminar flow is not an error: the calculation includes its effect on Rb*, and the field length that results is correct for the flow entered. The warning says that the same field would need less length at a higher flow, at the price of more pumping energy, or with smaller pipes. Viscous antifreeze mixtures at low temperatures move the transition to higher flows. With a mixed field the flow counts as laminar if it is laminar in any borehole. The dynamic borehole model can evaluate the fluid properties at the current mean fluid temperature, which captures the change to laminar flow in cold periods.

26.5 Load coverage

A load input covers a certain time; what happens after its end is a policy, and the result states it.

Table 53. Load coverage and the policy after the end of the supplied loads.
Load inputSuppliedAfter the end
Monthly tableone yearrepeated every year (as in the classic monthly method)
Monthly loads per yearthe years of the tablethe last year repeated, with a warning
Hourly8760 or 8760 × years values8760 values repeated every year; other counts are errors (8784: hint to remove 29 February)
Time seriesrows × time stepexactly one year (8760 h within half a step) repeats; a longer series is cut; a shorter one is followed by no load

A time series that ends before the design period gives a warning with the hours covered and the hours of the period: after its end there is no load (the dynamic model also stops the pump), so the later years show the recovery of the ground, not a design case. A series of 8784 hours (a leap year) does not repeat and gets its own warning; removing 29 February makes it a repeating year. For a design, a series should cover exactly one year or the whole period.

26.6 Monthly peaks that are not applied

The monthly method applies the peak of a month only when its duration is positive and its ground load is more extreme than the month's net mean ground load in its direction: the peak extraction must exceed the mean extraction for heating, the peak injection the mean injection for cooling, both on the ground side (building powers converted with the seasonal performance factors). Two warnings point out peaks that were entered but do not act:

Absent peaks (0 kW) give no warning.

26.7 Time-series columns that are not applied

A time series can carry, besides the load, the flow, the flow direction and the inlet temperature. The steady and short-time models use the hour means of the load and the project's nominal flow and do not apply the other columns; the dynamic model applies them only when its flow, direction and inlet settings ask for them. In both cases a warning names the columns that are read but not used, so that a series imported for its flow data is not mistaken for a flow-controlled calculation.

26.8 What the warnings mean for the design

The warnings fall into three groups.

Every warning stays with the result in the report. A design passed on to others should state the warnings that remain and why they are accepted.