Which device fits the best to my application?

If you want to perform accurate velocity or turbidity profile measurements on laboratory setup or industrial pipes, the UB-Lab is designed for you. Because it can control a wide variety of external transducers, it fits many purposes: flow visualisation in opaque liquids, flow metering in complex geometry, food quality control, concentration measurement, etc.

If you want to study and to monitor open channel, sewer flows or rivers, the UB-Flow is made for you. With its hydrodynamic and compact design, the UB-Flow is a fully integrated probe adapted to harsh environments. If the channel it too small, we will recommend to use the UB-Lab.

How to know the maximal measurable velocity for my setup? What is the Range-velocity ambiguity?

The velocity measurement by coherent pulsed Doppler technique gives an excellent spatial and temporal resolution combined with a very good accuracy, thus in a given limit of velocity range.

The repetition of the ultrasonic pulses gives a precise measurement of the Doppler phase in a small cell (or volume of fluid). Nevertheless, this approach induces a limitation of the velocity range for a given exploration depth. Thus, the pulse repetition period defines on one hand the exploration depth (all the echoes will have to come back from the medium before the next pulse is sent) and on the other hand the interval for the determination of the Doppler shift (between -π et +π and proportional to the velocity). When the interval between two successive pulses is too long, the measurement suffers from a phase jump inducing an ambiguity. On a frequency point of view, this is equivalent to overstep the limit given by the Nyquist-Shannon theorem.

This phase jump results in a velocity jump that can be observed on a profile with a velocity gradient. The figure above shows the measurement on a profile with velocity increasing with the depth. The red lines shows the velocity limits of the range; one can see that each velocity outside this range suffers from a jump that brings it inside the interval.

Thus the instantaneous velocities (in green) are reduced in negative values when they overstep the highest limit of the range.

The velocity range for the projected velocities (as measured by the device) is given by:

with: PRF the Pulse Repetition Frequency, c the sound speed (about 1480m/s in water), f0 the emitting frequency.

Finally, it is the pulse repetition frequency (PRF), defined in the setup, that will act on one side on the maximal exploration depth and on the other side on the maximal velocity. This limit is expressed by:

with : Rv the velocity range along the flow axis, Ry the exploration depth (orthogonal to the main flow direction, the pipe diameter for example), β the angle between the beam axis and the velocity vector. In monostatic mode, the velocities measured are those projected on the beam axis of the transducer used to measure. This is why, because of the geometrical uncertainty and mainly because of the turbulences, the angle between the beam axis and the velocity vector can not be too near to 90°.

What is acoustic turbidity? What is the difference with optical turbidity?

The turbidity (the cloudiness of water) is due to the presence of suspended matter or sediments composed by organic and inorganic particles, flocs or vesicles.

This physical effect was initially linked with the optical turbidity which depends on the colour, the size and the shape of the suspended sediments. Optical turbidity can, after adequate calibration, be linearly linked to the suspended sediments concentration. But OBS (Optical Backscattering Sensors) don’t see large particles because their sensitivity decreases inversely as the particles size increases.

However, optical turbidity is a point measurement which might not be representative of the entire depth of the water column. Another major drawback is its sensitivity to biofouling which leads to a weak signal.

The acoustic turbidity measurement is based on the capacity of the particles to diffuse the acoustic wave. It is determined by the ratio of energy received by the transducer (backscattered wave) over the emitted energy, corrected with the dispersion of the acoustic wave. Based on the backscattered amplitude measurement, it corrects all the effects of the transducer and of the electronics to bring a value that only depends on the medium and the suspended particles.

The acoustic turbidity allows to monitor the suspended sediment concentration (SSC) and thus the quality of water. As for OBS there is a link between the particle size and the sensitivity of the ABS (Acoustic Backscattering Sensors). The sensitivity of Ubertone’s profilers goes from 10 to 1000 microns (depending on the emitting frequency).

The acoustic turbidity is measured along the acoustic beam and can potentially be inverted in order to give a concentration profile. In a homogeneous suspension where concentration and particle size can be considered uniform, the theoretical relation between turbidity and sediment concentration at a given emitting frequency is given by:

where β and α are the backscattering and attenuation coefficients of the particles, Cv the volumetric concentration and αw the attenuation coefficient of the fluid. Thus in a homogeneous suspension, the logarithm of the turbidity is a straight line.

How to choose my seeding particles?

For detailed informations about the influence of a particle’s characteristics on the recorded RMS (root-mean-squared) voltage of the backscattered signal, you can have a look on this article and this thesis:

  • Schmitt (2016): Suspended Sediment characterization by Multifrequency Acoustic, Philippe Schmitt, Anne Pallarès,Stéphane Fischer and Marcus Vinicius de Assis, 2016, ISUD10.
  • Bricault (2006): Retrodiffusion acoustique par une suspension en milieu turbulent: application à la mesure de profils de concentration pour l'étude de processus hydrosédimentaires, Mickaël BRICAULT, 2006, INP Grenoble.

Usually, the natural particle contamination of water is sufficient for UVP Monitor measurement. But if the medium is too clean or if you want to improve your measurements, you can introduce tracing particles in you flow. These particles can be from a little bit mud in hydraulic models for example. But there are also artificial particles you can use: glass spheres or polyamide particles. If you are lost, you can have a look on the particles selected by Ubertone.

Basically, when you choose your particles, you will be able to choose their size, density and shape. And usually, you want to have influence on the size, on the sedimentation velocity and on the acoustic impedance of the particles.

The particles should have a low Stokes number to follow the flow, ...

Stokes’ formula gives:

where µ is the dynamic viscosity of the medium, r is the particle radius, g the acceleration of gravity and

the mass density difference between the particle and the medium.
Small size and a mass density as close to the medium’s as it is possible lead to a small sedimentation velocity.

... they should have an acoustic impedance that allows to have a good amplitude of signal ...

Le acoustic impedance of a matter is given by

where c is the sound velocity in this matter and the mass density of the same matter. The amplitude of the signal scattered by the particle depends on Zpthe acoustic impedance of the particle matter and on Zmthe one of the medium:

and if the medium is water,

where d is the particle matter density.
This means it is actually the difference of mass density and of sound speed between the medium and the particle that leads to more signal.

… and they should be big enough to be seen by the ultrasounds.

We know that the particles should be equal or smaller than the wavelength of the ultrasound wave. But if they are too small, they won’t scatter enough to be seen by the transducer. There is no magical formula to calculate the minimal size given the wavelength. Arbitrarily and given our experience, we could say that 2% of the wavelength should be a good limit.

Ubertone has selected and characterised polyamide particles which fit perfectly to our devices and transducers. For more information, you can have a look at the particles product page.

What is pulse coding? How does it improve the accuracy of velocity measurements?

Pulse coding is a unique technique implemented in Ubertone’s profilers that removes the effect of parasitic noises on the velocity profiles.

These parasitic noises may be of different natures:

  • ultrasonic noise generated by external sources (pumps, waterfall, pipe vibrations, etc.),
  • electromagnetic noise generated by external sources (electrical engines, variable speed systems using a frequency inverter, etc.),
  • ghost echoes from the previous acoustic pulses generated by the profiler itself.
    • Ghost echoes (also known as phantom echoes in ultrasonic testing) come from intense scattering area or reflection on wall or objects. They particularly occur when working:

      • at low emitting frequencies,
      • in closed tanks,
      • in small pipes and channels with rough surface,
      • at a high PRF (pulse repetition frequency).
        • When any of these noises has a frequency signature (and it is usually the case), a standard profiler will interpret this as a Doppler shift and will cause an important bias in the velocity profile. This is shown by the red profile on the graph below where a ghost echo introduces an error in the velocity. When the pulse coding is activated, these noises are not taken into account by the signal processing of Ubertone’s profilers so that the estimated velocity corresponds accurately to the velocity of the flow at the given position (green profile in the graph).

          This technique is part of the intellectual property of Ubertone. Each emitted pulse is coded. This allows the receiving electronic to distinguish the signal coming from the last emitted pulse from any other parasitic noise.

What does the static echo filter do?

When a cell (i.e. single measurement volume) contains both a static wall (or any motionless object) and moving particles, the estimated velocity will be the average between the wall velocity (which is zero) and the particle velocities (considered equal to the liquid flow velocity). This average is weighted by the energy of the respective echoes. As the echo of a wall may be much stronger as the echoes from scattering particles, the wall will introduce an important bias in the velocity estimation.

The static echo filter implemented in Ubertone’s devices is a numerical processing that removes the effect of motionless objects in order to get only the velocity of the moving particles.

This filter is very useful when measuring through a wall.

How does the automatic gain control work?

Ubertone’s devices are equipped with an automatic gain control algorithm that optimizes the gain over the full observed window with a logarithmic law. It is thus recommended to limit the number of cells to the area of interest so that the gain is well adapted to this area (and not influenced by the echoes after an interface for example).