Ideally, you will monitor the water-air meniscus in the outflow tube over a several hours and when flow stops you can assume equilibrium has been attained. For clay soils this can take days. The re-initiation of the flow you are mentioning may be due to the temporary cessation of outflow due to loss of hydraulic contact of the soil water with the water in the ceramic plate. The subsequent flow you are seeing may be due to the restoration of the lost hydraulic connection. To avoid this problem, it is important to fully saturated the sample at the beginning of the experiment by displacing the soil pore air by flushing with a highly soluble gas, i.e. CO2 or Helium, prior to saturation. This will ensure complete saturation of the soil sample. When you place the fully saturated sample on the ceramic plate it is important not to move the sample until equilibrium has been reached. Even a slight vibration from unscrewing the pressure chamber is enough to break the hydraulic contact especially when the soil becomes drier. Once a sample is removed for weighing you cannot put it back on the ceramic plate to get the next pressure step because of the loss of hydraulic connection of the soil pore water with the water within the pores in the ceramic plate! One way to ensure this is to start with 18 rings of soil samples on the plate and remove 3 samples, carefully without touching the other samples, for each pressure step. Using this protocol, you will get replicated data for 6 pressure steps.

I'm using HYDRUS 1D with estimated soil parameters, obtained from texture and bulk density, to simulate the matric head with time. After all, steady state means dh/dt=0

How is your experiment mounted? You could weigh the samples until you get no more change in weight (=no more water flowing out)... But there some experimental setups out there (for example using Decagon's/UMS Hyprop device) where you never will know if you reached equilibrium, so it's also dependant from the setup.

Yes I am using ceramic plates. The difficulty I have found with undisturbed cores is that if you do not remove the samples after water flow has ceased, flow may reinitiate after several days of continued pressure. Hence, true equilllibrium has not been reached. The alternative is to use disturbed samples, but obviously these do not reflect the structural component of the water retention curve as accurately.

Ideally, you will monitor the water-air meniscus in the outflow tube over a several hours and when flow stops you can assume equilibrium has been attained. For clay soils this can take days. The re-initiation of the flow you are mentioning may be due to the temporary cessation of outflow due to loss of hydraulic contact of the soil water with the water in the ceramic plate. The subsequent flow you are seeing may be due to the restoration of the lost hydraulic connection. To avoid this problem, it is important to fully saturated the sample at the beginning of the experiment by displacing the soil pore air by flushing with a highly soluble gas, i.e. CO2 or Helium, prior to saturation. This will ensure complete saturation of the soil sample. When you place the fully saturated sample on the ceramic plate it is important not to move the sample until equilibrium has been reached. Even a slight vibration from unscrewing the pressure chamber is enough to break the hydraulic contact especially when the soil becomes drier. Once a sample is removed for weighing you cannot put it back on the ceramic plate to get the next pressure step because of the loss of hydraulic connection of the soil pore water with the water within the pores in the ceramic plate! One way to ensure this is to start with 18 rings of soil samples on the plate and remove 3 samples, carefully without touching the other samples, for each pressure step. Using this protocol, you will get replicated data for 6 pressure steps.

Ramanathan sums it up perfectly. One other thing you should consider is that the pressure plate method (due to some of the reasons given earlier) never achieves true equilibrium and when compared with other methods - always yields higher water contents at any given potential. But if that is what is available, then waiting until no water comes out is your best bet of "equilibration". See below two very important papers on the subject:

Bittelli, M. and M. Flury. 2009. Errors in water retention curves determined with pressure plates. Soil Sci Soc Am J 73: 1453-1460.

Solone, R., M. Bittelli, F. Tomei and F. Morari. 2012. Errors in water retention curves determined with pressure plates: Effects on the soil water balance. J Hydrol 470: 65-74. doi:DOI 10.1016/j.jhydrol.2012.08.017.

Thanks everyone for your very helpful answers. I find it strange that pressure plates have traditionally been so popular given the large opportunity for error compared to other methods. I can certainly see why other devices such as the Hyprop device are becoming popular.

There are still problems associated to pressure plate regarding to volume change of the soil which affects void ratio and water content of the sample, however, the equilibrium can be achieved when water stop to flow out from the sample and this can be observed in the out flow unit ( burette).

The equilibrium depends on several factors such as soil type (e. g. Clay silt sand) and value of the applied suction . For coarse grained soil it takes several hours to get equilibrium while for clay takes several days also it depends on sample size.

I recommend you if you have coarse grained soil to use hanging column system which is suitable in range of suctions of 0 to 10 kPa.

A point about time of equilibrium in pressure plate is sandy soils samples can take even longer as the clayey soil samples as you have continuity and tortuosity aspects that are magnified as soil sample drains. As mentioned above ... the "equilibrium" is utopia here. The hight of the samples it is a very important aspect... you can easily simulated the travel time of water in drier samples and estimated the time to water arrived the pressure plate... it can takes months in samples of 5 cm height. If you are interested in drier points (> 800 kPa) reflect to use deformed samples in lower cylinder or psycrometry techniques (as WP4 from Decagon).

Hi - the general definition of equilibrium is that dS/dt is zero or very very small. however it is difficult to achieve due to dynamic effects. Some of our papers on the subjects may be helpful to you

1. Das, D.B., Gauldie, R., Mirzaei, M. (2007). Dynamic Effects for Two-Phase Flow in Porous Media: Fluid Property Effects. AIChE Journal, 53(10), 2505-2520, October

2. Mirzaei M. and Das, DB. Dynamic Effects in Capillary Pressure-Saturations Relationships for Two-Phase Flow in 3D Porous Media: Implications of Micro-Heterogeneities. Chem Eng Sci. 2007:62, 1927-1947.

3. Hanspal, N and Das, DB. Dynamic Effects on Capillary Pressure-Saturation Relationships for Two-Phase Porous Flow: Implications of Temperature. AIChE Journal, 2012: 58(6), 1951-1965, DOI: 10.1002/aic.12702.

4. Das, DB, Mirzaei, M. Dynamic effects in capillary pressure relationships for two-phase flow in porous media: Experiments and Numerical Analyses. AIChE Journal. 2012: 58, 12, 3891-3903, DOI: 10.1002/aic.13777.

5. Das, DB and Mirzaei, M. Experimental measurement of dynamic effect in capillary pressure relationship for two-phase flow in weakly layered porous media. AIChE Journal, 2013: DOI: 10.1002/aic.13925.

6. Mirzaei M and Das DB EXPERIMENTAL INVESTIGATION OF HYSTERETIC DYNAMIC EFFECT IN CAPILLARY PRESSURE-SATURATION RELATIONSHIP FOR TWO-PHASE FLOW IN POROUS MEDIA. AIChE Journal, 2013, DOI 10.1002/aic.14121.

7. Hanspal, N, B Allison, L Deka, DB Das. Artificial Neural Network (ANN) Modeling of Dynamic Effects on Two-phase Flow in Homogenous Porous Media. Journal of Hydroinformatics, 2013: 15(2), 540-554, doi:10.2166/hydro.2012.119

Hi Diganta,

Thanks very much. Im looking to get these articles and read them. Hopefully they can lighter the determination (estimation) of equilibrium - dS/dt ~ 0

Hi, In our case with relative compacted sandy soil we waited for around 3 weeks

Hi M. Rezaei - do you do any measurements whin the soil during this 3 weeks and if so what kind ? and, what kind of permeability do these sample have? I think in our case we reached equilibrium quite soon but ofcourse our sample was smaller in size and they were unconsolidated..

Hello, No we did not measure anything at pressure plate step. the average Ks was around 1 cm/h. Normally size is small in this step. in our department we determine WRC based on following article:

Cornelis, W.M., Khlosi, M., Hartmann, R., Van Meirvenne, M., De Vos, B., 2005. Comparison of unimodal analytical expressions for the soil-water retention curve. Soil Sci Soc Am J 69, 1902-1911.

we believe for sandy soil 2-3 weeks is enough (confirm that they reach to the equilibrium even they reach sooner) and for heavy soil more time is needed.

Hello Mr Rezaei, thanks. I have just had a look at the paper which I did not know of - very good work indeed. Some of our work (paper listed above) relate to the WRC before reaching flow equilibrium and they discuss the significance of dynamic WRC. I might be predicting too much here, but it would be interesting to see if these have any significance/validity in your case while you wait for equilibrium..