The need for measuring a liquid’s viscosity in the process line has grown over the years. Although laboratory measurements can give much more insight into a liquid’s behavior, it is the lab time required and the fact that just single measurement spots are taken in the lab which makes real-time process control interesting. Together with the demand for measuring process viscosity in different applications a vast number of different methods have arisen.
Viscosity control with process viscometers
Process viscometry
How do process viscometers work?
Viscosity measurement and the methods to perform it are historically grown and very much dependent on the kind of liquid analyzed and the application in general. For this reason, a vast number of different viscosity measurement methods is available, both for the laboratory and at-line. Some of the simple viscosity measurement methods used in the laboratory (compare How to measure viscosity) were easily transferred to process environments and therefore the measured values are comparable if temperature differences from process to lab are taken into account.
Over the years methods in the laboratory have become more sophisticated (see Basics of viscometry and Basics of rheology) making it possible to gain better and more information about the liquids measured. From simply measuring a single viscosity value, investigations such as flow curves, yield points, or the analysis of viscous and elastic properties were developed. Consequently, a gap between lab and process methods arose.
In addition, it became necessary to measure viscosity under harsh process conditions, which led to the development of process viscometers which look different from lab viscometers.
Today, most viscometers used measure a single point of the liquid’s viscosity, picking up relative changes in the process which can be correlated to lab values. The major advantage of a process measurement is that viscosity values are obtained over 24 hours, 7 days a week, enabling constant process monitoring or controlling. Lab values additionally complement the measurement to provide a “complete picture” of a liquid’s quality, but their frequency can be kept to a minimum – for example to once a day.
Why is it important to measure viscosity in the process?
In many industrial processes, ranging from the chemical and manufacturing industries to medicine and food processing, viscosity is an important parameter for intermediate and final product characterization. Traditionally, samples would be taken one by one out of the process and examined with a laboratory viscometer, typically including a range of shear rates as most of the fluids in industry are non-Newtonian. However, there are some difficulties when it comes to comparing laboratory and process methods. For example, viscosity can be directly affected by temperature, shear rate, and other variables that can be very different during a laboratory measurement from what they are in the production line. Since the rheological properties of many samples are closely related to the flow conditions in the process, the measurements in the laboratory may not reflect the true conditions very well.[3]
In contrast to laboratory (off-line) determination of viscosity, process viscometers gather inline information on viscosity directly within the process and in real time. There are several different inline viscometers available, implementing measurement principles ranging from capillary differential pressure, time of falling or oscillating elements to torsional oscillations damping and rotation rate measurement of a solid shape upon application of a known force or torque.[4]
It is common knowledge that, except for the rotation methods, most of the viscometers lack a well-defined shear field and/or operate at very high shear rates. Due to that, measurements of non-Newtonian industrial fluids, which in most cases exhibit pseudoplastic behavior, can be very difficult. On the other hand, standard rotation methods often allow only a bypass installation or do not allow for a good exchange of fluid in the measuring field.[2]
Measuring the process viscosity inline has many advantages, such as
- saving time and lab costs and speeding up the process as samples do not have to be taken to a lab viscometer, which would need proper and accurate handling for a reliable viscosity value,
- having a reliable value for process control and being able to change product parameters automatically,
- and providing continuous viscosity monitoring 24/7 and thus being prepared for any unforeseen changes in the process
Some of the terms used in connection with process viscometry have to be defined. A “process viscometer” is a unit used for measuring the viscosity directly installed in the process line. The term “inline” is used to describe the situation in which all the liquid passes the process viscometer; whereas “online” means that the viscometer will be installed in a bypass and therefore just a part of the liquid passes the viscometer.
“At-line” and “off-line” are commonly used for lab viscometers meaning either a lab unit used “at-line” directly at the process line or “off-line” in the lab.
With process viscometers we commonly use the term “inline” for convenience. However, the place and manner of installation have to be chosen carefully, either inline or in bypass, bearing in mind the effects of the flow rate, temperature, and type of viscometer.[1]
How do process viscometers measure?
Many inline viscometers employ different measuring methods in terms of the way they measure viscosity or how they ensure that liquid is replenished in the measuring region. Accordingly, this leads to some degree of compromise when comparing measured viscosity values from a process sensor to those obtained from lab equipment.[1]
The capillary type
In the lab, measuring the viscosity using a capillary will mainly involve filling a tube-shaped capillary with liquid and measuring the time needed to empty this tube under temperature-controlled conditions. The main force driving the liquid’s flow will be gravity. Just a few lab capillary types will use a constant pressure instead of gravity. The process capillary type will use mainly a constant flow through a well-defined tube maintained by either a pump or pressure valve.
The pressure drop across the capillary is measured with a differential pressure transducer, which is connected to the inlet and outlet sides of the capillary. The differential pressure transmitter output is a linear indication of the process viscosity and is used for indicating, recording, or controlling the process.[4] The temperature must be stable across the capillary to avoid a temperature influence on the viscosity readings and the flow has to be measured highly accurately.
The falling elements type
Falling- or rolling-ball methods are very well established in the laboratory. As the respective falling element moves in the liquid the time it needs to pass a defined length is measured. This can be applied to process using a cylindrical piston moving within the liquid. The speed of movement is then related to the viscosity.
Both methods, in theory, measure absolute viscosity values in Newtonian liquids and therefore are mainly used for analyzing those liquids. When measuring non-Newtonian liquids the single viscosity value obtained can only yield an equivalent or average Newtonian viscosity.
The oscillation type
The oscillating type has no complementary method in the lab but some advantages when installed in process. All oscillating type viscometers are simple devices, easily placed in the flow line, and easy to clean. However, the measurement is dependent on the flow rate past the vibrating sensor, and the value obtained is at best an oscillatory parameter which, for the materials usually being measured, cannot necessarily be correlated with off-line measurements.[1]
The rotation type
A rotation type using a standardized cylinder-cup system and measuring the torque needed at a specific speed will fit best to laboratory methods giving the most sophisticated values. However, there are still limitations when using such a type in process. In general rotational types will not ensure a proper liquid exchange at the sensor and will not provide a sensor’s long-term stability. Therefore those specific types will only be used with compromises in design differing from the standardized setup and measurement. However, rotational types will be closest to what nowadays is measured in the lab.
Whether or not a certain compromise associated with any instrument is acceptable depends on the nature of the liquids being measured. Thixotropic, shear-thinning liquids are very difficult to measure even in the best inline viscometer under ideal laboratory circumstances. To measure such liquids inline is extremely difficult. While most instruments are calibrated to give an equivalent Newtonian viscosity, this is not always very useful for difficult materials. If on the other hand the liquids of interest are easy to measure, being virtually time-independent and only slightly non-Newtonian, then almost any method will suffice, since a reliable correlation can be established between inline and off-line measurements.[1]
The hydrodynamic process viscometer
The hydrodynamic process viscometer is based on a similar hydrodynamic effect which makes slide bearings work. In such bearings, the relative motion between two surfaces induces shear stress in a lubricant and leads to the formation of a lubricant film separating the sliding surfaces. The pressure necessary to carry the load can only develop if the film is wedge-shaped, so that the variable surface will be slightly inclined. The pressure distribution in the wedge-shaped film depends on the surface velocity and the dynamic viscosity of the lubricant.
For the process viscometer shown in Figure 1 this viscosity dependence is utilized to measure the viscosity given a fixed geometrical configuration of the sliding surfaces. The functional principle is shown in Figure 1. A rotor and a static outer surface define the wedge-shaped gap of the viscometer configuration. The rotor can be powered directly from a rigid shaft or using a magnetic coupling. The fluid under test is conveyed into the gap by the action of the rotor. The opposite static outer surface is rigidly fixed on the entry side and open-ended at the outlet. Due to the pressure rise in the gap, the outer surface is slightly displaced, acting as a spring. The displacement between the fixed and flexible parts of the spring element is proportional to the fluid’s viscosity and measured with an inductive sensor. The sensor coil is mounted on the fixed part of the spring while the counter piece is attached to the flexible part.
Figure 1: Partially open tube with a rotating cylindrical shaft
Liquid surrounding the system is drawn into the narrowing gap by shear forces. The liquid volume passing through the gap is constant and as a consequence the pressure changes. The pressure first increases as the gap gets tighter, then reaches a maximum inside the tube, then decreases again. This pressure is a measure of the liquid’s viscosity, being proportional to the dynamic viscosity, shear velocity, and a geometry factor:
Equation 1: Pressure as a function of viscosity, speed, and gap height
Besides the mechanical components outlined, the viscometer incorporates an electric drive with closed-loop control to achieve stable flow conditions and a stable shear rate in the wedge-shaped gap. The excess pressure is zero at the inlet and outlet and reaches a maximum after about two thirds of the gap length. The spring structure leads to a weighted integration of the pressure distribution, which is measured as displacement. An inductive displacement sensor measures the deflection of the spring, offering the advantage of being independent of the process fluid and decoupled from the rotating motion and hence not influenced by bearing friction changes. A temperature sensor immersed in the process liquid is used to compensate for temperature variations of the sensor front-end and to provide process temperature information.
Figure 2: Measuring principle
To avoid turbulent flow and centrifugal forces the gap shouldn’t be too wide and the rotating speed shouldn’t be too high.
Figure 3: Measuring head and its parts
Conclusion
Measuring the viscosity in the laboratory is very well established and also standardized in terms of equipment and methods to provide a proper measurement. Bringing viscosity measurement into process leads to the need for different equipment and also methods. The main challenge is to correlate both gathered values – from lab and process – properly.
References
[1] Barnes, H.A. (2000). A Handbook of Elementary Rheology. Aberystwyth: University of Wales
[2] Zavrsnik, M. and Joseph-Strasser, M. (2013). Inline viscometery for non-Newtonian viscosity characterization. Graz: AMA Conferences 2013
[3] Koseli, V. and Zeybek S. and Uludag Y.(2006). Online viscosity measurement of complex solutions using ultrasound Doppler velocimetry
[4] Liptak, B. G. and Venczel, K. (2014). Instrument and Automation Engineers' Handbook: Process Measurement and Analysis. 5th Edition. New York: CRC Press, chapter 8.64