Water squandering

A typical human being needs some 2-5 L of water per day. Whereas the water consumption in the Netherlands per person is in the order of 120 to 140 L. A substantial delta of > 100 L. So the question that appeared in my cerebrum (just a fancy word for a part of the brain) was: “Where does al this water go to?”. To address this question I decided to monitor my (drinking) water consumption. Press here to see how I build a sensor to perform real-time measurements.

Toilet

So I started off simple. I installed the sensor and flushed my toilet. This is what I saw:

Figure 1: Cumulative water use after toilet use for three different toilet flushs

Some ~7 L of water were consumed in roughly 1 minute, so a flowrate of roughly 8 L per minute (0.12L per second). Moreover, flushing a toilet is quite reproducible, in terms of water consumption. I thought that this characteristic; the flowrate combined with a fixed total consumption would be representative for flushing the toilet. In a next step I could then distil my toilet use from the water consumption data. However, I soon realised that things are more complicated than I initially thought. For example, “clean” people will wash their hands immediately after pressing the flush-button on the toilet, see Figure 2. In other words, washing hands is performed in the same time span as when the water tank of the toilet is topped up. The integral (i.e. total water consumption) would be higher and the same would hold for the flowrate.

Figure 2: Cumulative water use after toilet use: Red; Toilet flush + hand wash, Blue; only Toilet flush

This resulted in the weird habit of first washing my hands before pressing the flush button of the toilet. Moreover, similar problems arises when someone decides to interrupt the flushing procedure or extending it, when additional water is required to get rid of their masterpiece.

Showering

In one attempt to clean myself (showering) I consumed more than 100 L in bit more than 800 seconds, i.e. 7.5 L per minute (0.125 L per second). Also for this application we can see that the slope is pretty steady.

Figure 3: Cumulative water use during showering

However, whereas the flowrate during toilet flushing is fixed (valve open/close action), showering can be performed for different flowrates as the operator can change the setting of the faucet. In one attempt you can see that intial flowrate was around 4.5 L per minute and later 12.5 L per minute.

Figure 4: Cumulative water use during showering

This water consumption data on showering also triggered me to estimate the energy consumption to provide the requested hot water. Let us consider the scenario that the final water temperature is 38 C and that water inlet temperature is 20 C, which might be pretty optimistic in winter. But hey, I am an optimist.

\[P = \phi_{v} \cdot \rho \cdot C_{p} \cdot \Delta T\]

Plugging in the values; $\rho$ = 1 kg / L, C_p of water is 4.2 kJ / kg K, $\Delta$T in our scenario is 18 and let us consider a water flowrate of 7.5 L per minute (0.125 L per second).

\[P = \phi_{v} \cdot \rho \cdot C_{p} \cdot \Delta T = 0.125 \ \frac{L}{s} \cdot 1 \ \frac{kg}{L} \cdot 4.2 \ \frac{kJ}{kg\ K} \cdot 18 \ K = 9.5 \ \frac{kJ}{s} = ~10 kW\]

I am pretty staggered by this number, but still think that my calculation is correct. I found some support in the fact that fully electric showers start from 7 kW and up which is a similar order of magnitude. A ten minute shower would require 6 MJ which is roughly equal to 0.2 m³ of natural gas.

\[Q = P \cdot t = 10 \ kW \cdot 600 \ s = 6 MJ\]

So yes, I should shower less long…

Washing Machine

I do possess one of the indestructable washing machines (Miele) which is already more than 20 years old I believe. I do think this a good thing, using a machine until its very very last day. However, this beast does consume quite a bit of water (>80 L) in one cycle, see Figure 5. This is probably not even the “washing program” which requires most water. New, fancy machines could trim the water consumption per cycle down to some 50 L, but currently I do not feel it is worth the investment.

The plot also shows that the water consumption profile is much more complicated than that for flushing a toilet or showering. During the “big” steps the machine takes in more than 10 L (typically ~13 L) water with a flowrate of around 12 L per minute. Which both appear significantly more than a toilet flush (~7 L total, 7 Liter per minute). On first hand, this difference appears to be sufficient to pinpoint the water use either to the washing machine or flushing a toilet. Moreover, washing programs are probably perfectly reproducible in terms of water consumption. In other words, the intervals between water intake in Figure 5 are always the same. The chart also gives an indication for the time that my washing machine needs some for one cycle; 100 minutes (1:40). Yes, probably the acutal run time is a bit longer as it needs to do centrifugation and other stuff. To clarify, I had no clue how long the washing machine takes to finish its job.

Figure 5: Cumulative water use during washing machine cycle

Dish Washer

A modern dishwasher is suppose to consume only a bit more than 10 L per cycle. Yes, I did run my dishwasher (it is one of my favorite devices) and found it to do a bit more, between 14 and 16.5 L. The dishwasher runs appear quite reproducible, except for the region around 9000 seconds. Possibly, water consumption around this moment in time does not belong to the dishwasher but to something else. To be investigated…

Figure 5: Cumulative water use during dish washer cycle

I typically do not run the dishwasher in ECO-mode, the mode to save the world, because in the past it resulted in poor cleaning.

Figure 6: Cumulative water use during dish washer cycle for different operating modes

The ECO-mode (9 L) is indeed saving the world by some 5 L compared to the regular mode (14 L). I then decided to take it one step further and looked up the numbers provided by the manufacturer of the dishwasher in their manual. The numbers indicated by the dishwasher vendor are displayed on the left of the table below. On the right side of the table you can see the numbers I measured. The vendor is not lying as the numbers match pretty well. Water consumption measurement does not define the duration, but it defines a minimum. In other words, after water is consumed, the machine might still do some stuff. So the measured value should be smaller than the value in the manual, which is the case for all scenarios.

The most important thing that I learned here is that I should try to avoid the “Regular” mode as much as possible, to save on water and power.

Table 1: Dishwasher water consumption for various operating modes. Numbers on in teft side of the table are according to the manufacturer. Values determined by my measurements are displayed on the right

	According to manual	According to manual	According to manual	Measured	Measured
Mode	Duration (min)	Water consumption (L)	Power consumption (kWh)	Duration* (min)	Water consumption (L)
Regular	144-154	12.5-14.5	1.3-1.5	100	14-16.5
ECO	225	10.2	0.932	100	9
Quick	30	10	0.9	26	9.5

Conclusion

I installed a sensor to read out my water sensor in real-time. From the obtained data I attempted to trace back the use of specific water consuming applications such as toilet flushing or showering. I have summarized my findings in the table below. In order to reduce my drinking water consumption I should either change my habits, i.e. shower less long, or use a different water source (e.g. rainwater) to flush my toilet. In addition, I could invest in a new washing machine, which is suppose to cut the water consumption in half. The dishwasher does not require our focus, it is rarely used in comparison to the toilets and consumes just a bit more than a toilet flush.

To paint a more accurate picture I could of course decide to install a ton of other sensors. However, I generally dislike such an approach; it is much harder/time consuming to implement, it is harder to adopt by others and it will likely not pass the WAF (wife acceptance factor). Though, I am already collecting data from my combi-boiler and electricity consumption. In particular reading whether the combi-boiler is in “tap water heating mode” would be very useful to differentiate “showering” from the other water consuming activities. Similarly, the use of the dish washer and washing machine is associated with substantial electricity consumption and combining data for water and electricity consumption will provide a more deceisive answer on what “application” is consuming my precious water.

So, enough to do…

Table 2: Water consumption per cycle of various applications

Application	Flowrate (L/min)	Total consumption (L)	Time duration (min)	Remark	Future strategy

Toilet flush	~8	8	~1	Interruption of the flush button makes tracking difficult
Showering	Variable, 5-13	Variable, >30	5-15		Check when heating system is in hot water mode
Washing Machine	~12	80	100	Complicated and long run time, easily intervered by other applications	Check electricity consumption data
Dish Washer		9-16.5	at least 26-100		Check electricity consumption data
Others	?	?	?	?	?