Skip to content

Planning drinking water cisterns

Reading time 6 minutes

Updated - December 10, 2025

The planning of drinking water cisterns involves several physical factors that need to be taken into account if the desired drinking water quality and pumping capacity are to be achieved with as little effort as possible.

Water analysis and limit values

First of all, a laboratory analysis of each water source must be prepared, by means of which the water values are compared with the applicable limit values in order to be able to take specific measures to comply with them if necessary:

  • Acrylamide 0.10 μg/l
  • Antimony 10 μg/l
  • Arsenic 10 μg/l
  • Benzene 1.0 μg/l
  • Benzo(a)pyrene 0.010 μg/l
  • Bisphenol A 2.5 μg/l
  • Boron 1.5 mg/l
  • Bromate 10 μg/l
  • Cadmium 5.0 μg/l
  • Chlorate 0.25 mg/l
  • Chlorite 0.25 mg/l
  • Chromium 25 μg/l
  • Copper 2.0 mg/l
  • Cyanide 50 μg/l
  • 1,2-Dichloroethane 3.0 μg/l
  • Epichlorohydrin 0.10 μg/l
  • Fluoride 1.5 mg/l
  • Haloacetic acids (HAAs) 60 μg/l
  • Lead 5 μg/l
  • Mercury 1.0 μg/l
  • Microcystin-LR 1.0 μg/l
  • Nickel 20 μg/l
  • Nitrate 50 mg/l
  • Nitrite 0.50 mg/l
  • Pesticides 0.10 μg/l
  • Total pesticide 0.50 μg/l
  • PFAS total 0.50 μg/l
  • Sum of PFAS 0.10 μg/l
  • Polycyclic aromatic hydrocarbons 0.10 μg/l
  • Selenium 20 μg/l
  • Tetrachloroethene and trichloroethene 10 μg/l
  • Total trihalomethanes 100 μg/l
  • Uranium 30 μg/l
  • Vinyl chloride 0.50 μg/l

Water hardness

The water hardness represents the content of calcium carbonate (CaCO₃) in mmol/l, ppm or mg/l (1 mmol/l = 1 ppm/l = 1 mg/l) Alkaline earth ions according to the international system of units SI (Système International d'Unités), outdated also in °dH (German hardness 1 °dH corresponds to 0.1783 mml/l).

  • soft water -> less than 8.4 °dH, corresponding to less than 1.5 mmoll
  • medium -> 8.4 ... 14 °dH, corresponding to 1.5 ... 2.5 mmol/l
  • Hard water -> greater than 14 °dH, corresponding to greater than 2.5 mmol/l

Water hardness is reflected, among other things, in limescale deposits in pipes, on surfaces wetted by water and in the consumption of more detergent the more calcareous, i.e. harder, the water is.

Water softening

Water softening can be carried out in two ways: 

  • Ion exchanger (here, water is passed through a resin filter saturated with sodium ions, whereby calcium and magnesium ions dissolved in the water are exchanged for the sodium ions of the resin.
    Depending on the flow rate, ion exchangers are saturated with calcium and magnesium ions sooner or later and must then be rinsed and regenerated with a highly concentrated common salt solution (NaCl). They can then absorb calcium and magnesium ions again and release sodium ions accordingly.
    Depending on the throughput, such systems require large quantities over the course of a year. EN 973 Type A certified, high-purity salt with a purity of over 99.5 %.
    The disadvantage is the enrichment of the water with sodium, which counteracts the idea of „low-sodium“ drinking water. Likewise, of course, the recurring costs for the regeneration salt and the associated maintenance costs.
  • Reverse osmosis, by pressing the water at high pressure through 0.00001 µm fine filter pores. The disadvantage, however, is that the resulting water no longer contains any minerals and therefore has to be re-mineralized in order to be usable by the human organism.
    The disadvantages are the regularly recurring costs for the osmosis membrane (filter cartridge) and the „water“ that is produced during reverse osmosis, which can amount to up to 50% of additional water consumption.
  • Distillation, which also results in water without any mineral content and is therefore counterproductive for human health.
    Another disadvantage is the high energy consumption.
  • Seed crystallization, in which a catalyst (filled with specially coated ceramic or polymer beads to which calcium and carbonate ions dock and crystallize) converts the calcite that causes limescale deposits into needle-shaped aragonite crystals that no longer adhere.
    The water still contains all minerals, including „lime“ in the form of aragonite crystals).
    This water is fully usable for the organism.
    Such devices have a service life of over ten years, depending on the flow rate and dimensioning. They are looped into the main supply line of the domestic water supply after the water meter.

Seed crystallization technologies

The water hardness does not change with any of the following technologies. However, limescale deposits are greatly reduced, as the CaCO3 is already bound in stable fine micro- or very fine nanocrystals.

TAC

Template Assisted Crystallization - The carrier of the crystallization nucleus is a solid medium, e.g. resin or granules. Calcium carbonate crystals form on these when the water flows through them, which immediately detach from the medium and are carried away with the water.
Ideal for low flow rates, as the formation of microcrystals takes longer than nanocrystals in the NAC process.

NAC

Nucleation Assisted Crystallization - is basically identical to the TAC process, but here nano- instead of microcrystals are produced. This means that the number of particles is larger and the crystals smaller.
This is advantageous if a high flow rate is required, as the contact time with the medium is therefore shorter and crystal formation is faster than with the TAC process due to the system.

MAC

Media Assisted Crystallization - combines both processes, TAC and NAC, under one umbrella term. The term MAC is usually used when proprietary technologies are used that differ from the usual TAC/NAC process technologies.

Supplier of systems with seed crystallization

As always, it is important to separate the wheat from the chaff, i.e. which manufacturer not only claims, but also proves the function of its products with corresponding certificates from independent test centers and / or studies.

As mentioned above, the contact time of the water flowing through is functionally significant, which is why the data sheets of the individual products must be evaluated separately with regard to this criterion and the overall system must be adapted accordingly.

The following manufacturers stand out positively here:

Yield of the source

In the case of an existing or planned well drilling, the parameters of flow rate and extraction rate, in addition to other parameters, are elementary in the calculation of the quantity of groundwater to be extracted in the time unit. The explanation, including the calculation, can be found very clearly on this website.

The withdrawal of water from a river is often subject to approval and is limited in terms of quantity.

Selecting the feed pump

A deep well pump is used to pump groundwater from depths of between 8 and 90 meters. It should be noted that the extraction head and pump depth (suction depth) add up to the total delivery head.

ApplicationExample:

Pump depth (suction strainer) 20 m + highest extraction point 30 m = 50 m is the total delivery head of the deep well pump.

However, in addition to the pure height difference, friction losses in the pipe routing (roughness of the pipe, fittings, etc. = dynamic head) must also be taken into account. These must be added accordingly when selecting the pump.

Assuming a pump capacity of 3,000 l/h (Q = 3/3600 m3/s = 0.0008333333333333333 m3/s), using e.g. DN65 (D = 0.0752) PE/HDPE pipe with a roughness according to the data sheet of ε = 1.5 µm = 1.5 × 10-⁶ m, the kinematic viscosity of water of ≈ 1-10-6 m²/s, and a density ρ = 1000 kg/m3 and flow velocity of g = 9.81 m/s2, as well as consideration of the equivalents of fittings, etc. by the calculated 1.2 times the pipe length of 200 m, corresponding to 240 m and a height difference to be overcome of Hs 50 m (static head) at an average acceleration due to gravity of g = 9.81 m/s2 (as a constant), the following calculation results:

  • Cross section DN65
    A = ( π ⋅ D2 ) : 4 = ( π ⋅ 0,0752 2 ) : 4 = 0,004417865 m2
  • Flow velocity
    v = Q : A = 0.0008333333333333333 m3/s : 0,004417865 m2 = 0.1886280807 m/s
  • Reynolds number (characteristic number, low = laminar, high = turbulent flow)
    Re = υD : v = ( 0.1886280807015056 ⋅ 0.0752 ) : ( 1⋅10-6 m2/s ) = 14147.10605261292 m2/s
  • Pipe friction coefficient (Swamee-Jain)
    f = 0.25 : [ log10 ( (ε : ( 3.7 ⋅ D )) + ( 5,74 : Re0,9 ) ) ]2
    f = 0.25 : [ log10 ( 5,405405405405405 × 10-6 + 0,000728728 ) ]2
    f = 0.25 : [ log10 ( 0,0007341334054054 ) ]2
    f = 0,25 : [ -3,134490 ]2
    f = 0,25 : 9,825866 = 0,028256663933258565
  • Pipe friction loss (Darcy-Weisbach)
    H = f ⋅ ( L : D ) ⋅ ( v2 : 2g )
    H = f ⋅ ( 240 : 0,075 ) ⋅ ( 0,18862808070150562 : 2 ⋅ 9,81 )
    H = f ⋅ ( 3,200 ) ⋅ ( 0.001813634 m )
    H = 0,028256663933258565 ⋅ 5,803629 = ≈ 0,1639776104 m
  • Equivalent fitting loss
    (identical Fommel, instead of L (pipe length) Leq (set with 10 m)
    Heq = f ⋅ ( Leq : D ) ⋅ ( v2 : 2g ) = ≈0.0068324004 m
  • Total head
    Hdead = Hs + h + heq
    Hdead = 50.0 m + 0.1639776104 m + 0.0068324004 m = ≈50.1708100 m
  • Pressure at the pump outlet
    p = ρgHdead
    p = 1000⋅9.81⋅50.1708100108 = 492175.6462064206 Pa
    p = 492175,6462064206 Pa : 10,000 = 4.921756462064206 bar
  • Hydraulic pump capacity
    Ph = ρgQHdead
    Ph = 1000 ⋅ 9,81 ⋅ 0,0008333333333333333 ⋅ 50,1708100108 = 410,1463718386839 W
  • Electrical motor power Pmotor with efficiency η = 0.65
    Pmotor= Ph : η
    Pmotor= 410,1463718386839 : 0,65 = 630,9944182133598 W

Nominally, a pump with around 630 W provides the required power. In practice, with an 80 percent increase as a safety reserve, this is assumed to be around 1.1 kW.

Valves

If each cistern in the network is to be individually separable, which makes sense in the event of maintenance work or leaks, a shut-off and non-return valve as well as a flush or drain valve are required for each cistern.

Motorized valves should have the option of manual (emergency) actuation.

All components must be designed in accordance with WRAS/DVGW for drinking water requirements (EPDM seat and NBR diaphragm). Mounting flanges for the connection of actuators must be designed in accordance with ISO 5211, motor flanges in accordance with ISO5211 F05/F07.

The leakage rate should be in accordance with Class VI which means absolute (bubble-free) tightness, i.e. a leakage rate of zero. All valves with a (PTFE) / EPDM seat meet this requirement.
Metal seats only reach class IV: a leakage rate of 10 ml/min under test pressure is permitted.

Whether 230 V AC or 24 V DC (battery) operation of the actuators is decided by the requirement for automated operability even if the public power supply fails.
All electrical components in systems exposed to the weather must comply with IP65 (dust-tight, protection against water jets), better IP67 (dust-tight, protection against short-term immersion).

Motorized valves must have limit switches, ideally a control via 0 ... 10 V, resp. 4 ... 20 mA if positions other than open / closed are to be controlled.

Level measurement

Ultrasonic and pressure sensors are suitable for level monitoring. While pressure sensors, positioned at the bottom of the cistern, are constantly exposed to water, ultrasonic measurement is contactless: the sensor is mounted on or under the cistern lid and is therefore quickly accessible.

Industrial ultrasonic sensors output a current of 4 ... 20 mA, which generates a measured value-dependent voltage via a calibrated IU converter (current to voltage), which is evaluated by a microcontroller and displayed as a measured value, converted into liters, cubic meters or percent.

However, the prices for ultrasonic sensors are rising in line with the increasing range, reaching into the four-digit euro range. Sensors with a measuring distance of up to 2.2 m are around the 200 Euro, which limits the cistern depth to around 2 m if the budget is not to be overstretched.

Leave a Reply

Your email address will not be published. Required fields are marked *

en_USEnglish