Density and Viscosity


Thread Starter


We developed an ultrasonic measurement device for the Army to measure online the density of fluid to paste-like mediums. It also can measure the viscosity. Does anybody know whether the process industry (or anyone else) needs such a technology or does such a technology already exist?

We are interested in any information that will help us. Thanks for your help!!
Pat, thank you for your reply. But the technology you mention is an in-line measurement device. In-line measurement has the problem that it is not contactless and has to be cleaned regularly to achieve correct results. With our technology we can measure contactless with ultrasonic sound from outside through a mixing tank or pipe. I think that's a big difference. Some more ideas? Thank you!


Dear Kevin

I have requirement of On-line density & viscosity measurment right now, looking for Non-contact type measurement.

If you are interested, would request you to write to me on following id

email id : iconcnl [at]


Andrey Romanenko


Depending on the capabilities of the system, this might be applicable in, for instance, resin and adhesive production. Currently the end of batch condition is determined by manual sampling of the reactor medium and measuring the viscosity in the lab. I can also think of other similar applications.

How can we learn more about your development and is it market ready?

Best Regards,
Andrey Romanenko
Ciengis - Advanced process control and Optimization
I'd be very interested to learn more about this sensor.
Ultrasonic measurement of density and viscosity is not new in itself, as this paper from 2001 suggests, and density measurement by ultrasound has been around longer:<pre>
In-situ ultrasonic viscosity measurement inside of an air-conditioning compressor
Article from: Tribology & Lubrication technology
Article date: October 1, 2001
Author: Oyamada, Tomonaga; Inoue, Youichi; Mizumoto, Muneo</pre>
And what always seems to be a limitation is the accuracy. In this article they claim only 10 and 20% accuracy.

No doubt significant advances have been made but so far as I can discover, the best accuracy so far is around 2% (FSD or reading?).
I assume therefore that this sensor represents a breakthrough in performance, but to what level?

Because of the very variable nature of viscosity, accuracies of between 2 and 5% have good application in behavioural measurement applications such as fuel atomising, ink or paint coating control etc. where the viscosity is only needed at the process temperature, but 2% is not good enough for analytical or quality measurements where the viscosity is needed at a reference temperature.

I also have some reservations about non-intrusive measurements, especially for analytical measurements. But that can be attributed to the influence of sensors which derive the viscosity from the interaction of the fluid with a sensor surface so I am willing to discover if ultrasonic sensors are not so susceptible; such sensors can be thought of in a similar manner to temperature sensors, i.e. the temperature sensor measures its own temperature which you hope is the same temperature as the surrounding medium. So it is with most viscometers, they measure (or are primarily responsive to) the viscosity of the fluid in the boundary layer in contact with the sensor, which you hope represents the viscosity of the surrounding fluid.

With an ultrasonic device we are presumably looking at the viscosity of all the fluid in the cross pipe so we will obtain an accurate measure of the median viscosity. The only problem is, viscosity is very sensitive to temperature and if the fluid is not homogeneous and if the measured temperature is not as well understood across the pipe as the viscosity then some errors interpreting the result may arise.

Vibrating element sensors, inserted into the flow stream, have proven vulnerable to external temperature effects if the installation is not well designed (Temperature, not rheological behaviour, is the more difficult problem to manage in process analytical measurements).
One way to be able to know the viscosity and the corresponding temperature (you need both values accurately known for a quality assessment) may be to ensure the flow is fully homogeneous which, in a non-invasive measurement seems unlikely... to ensure homogeneity would require the intrusion of some sort of mixing system which defeats the point of a non-intrusive sensors, or some means to develop a reliable cross pipe temperature measurement that correlates with the measured viscosity.

It would be natural to expect a temperature gradient and viscosity gradient across the pipe. Non-intrusive measurement suggests we are not able to intervene inside the pipe to develop a fully homogeneous flow at the measurement point. In some applications it is an easier situation with insertion type devices, or such has been the case thus far.

When dealing with lubricants and greases (Sengenuity have been making a play for this market with their SAW - Surface Acoustic Wave sensors, but theirs is an intrusive device) it depends on what you want to know.

If all you are interested in is the behavioural viscosity (viscosity at the process temperature) because you are interested in the actual lubricity, that's fine for a 2% sensor especially because you don't care about the temperature.

But if you want to be able to detect imminent critical failure due to fuel or coolant contamination, or the breakdown of the lubricant as it ages, then you need a measurement that will tell you the viscosity at a reference temperature. That is a far more challenging task.
In these applications, I am not sure what benefit there would be to a non-intrusive sensor over an intrusive sensor. I would think it would be easier to derive a reference temperature measurement from an intrusive sensor. Modern vibrational sensors can handle all sorts of fluid conditions including dirt and particles and, well designed and installed, are virtually maintenance free. They can also deliver the necessary accuracies for analytical measurements.

An issue with some sensors is there sensitivity to pressure effects. This is not a problem for the vibrating element sensors, but I note from other ultrasonic devices that factors such as the velocity of sound do appear to be pressure sensitive. Is this device insensitive to pressure effects?

As I say, I'd be very interested to learn more about this sensor because I'm sure it must be an advance over previous Ultrasound technologies and there will be some applications where a non-intrusive sensor has a clear advantage.
I don't know the automotive sensor but I'd be suspicious of the Fuji sensor, that is, I think its is probably a vibrating element sensor which is run at relatively high frequencies rather than a sensor that employs ultra-sound techniques.. Emerson's sensors run at around 1kHz, Lemis at 4.5Khz. I would be inclined to question if Fuji aren't simply referring to their operating frequency.

Most vibrating element sensors tend use a mechanical device to shear the fluid and the viscosity is then a function of either the amplitude, the power to sustain a constant amplitude, bandwidth at resonant frequency or decay times, but not a true ultra-sound principle.

The technology you describe is very important for many industries because if you can measure contactless the density and viscosity then you can also measure the concentration, crystallization, temperature, pressure... And if your technology is good enough for the Army then it probably will be good enough for the industry.

It's important for many industries e.g. chemical, biotech, pharma, water and waste water, food industry, sewage plants, rivers, dredging industry,...

I would contact ABB and if they are smart they will collaborate with you.

good luck
Gentlepeople... here's a question concerning a tangent topic:

Has anyone heard the following dimensional-unit used to measure viscosity:

"Aluminum-Ball (or Aluminium-Ball) Seconds?"

How about "Tennis-Ball Bounces?"

Regards, Phil Corso
Modern viscometers are very capable sensors.
For example, tuning fork sensors have success in applications from bitumens and asphalts to PIB and for end point spotting in polymerisation reaction control. An application in point is in the methyl methacrylate production both in batch reactors and inline processing. These are all applications from 8-10years ago.

I am sure that ultrasonic has a great deal to offer and I look forward to seeing this technology find its place because it will undoubtedly have application in pipe line applications.
Viscosity probably has more "oddball" units (pardon the pun) than any other measurement but this is a new one on me. Can you say where you came across it?
JMW... the first, "Aluminum-Ball-Seconds" was a dimensional unit used by Celanese Plastics Company, to measure the viscosity of melted cellephane!

The second, "Tennis-Ball-Bounces" was a dimensional unit used in an LNG-to-Gas plant to monitor the leak-rate in a cryogenic heat exchanger's perlite filled "cold-box" enclosure.

Here's another unusual one, "Firkens per Fortnight!" However, its explanation, "like describing a ratchet without using your hands" is more difficult. So, contact me off-list for details!

Regards, Phil Corso

Ps: If anyone is interested in weird, bizarre, or awesome, but true, stories I have a great one concerning Synchronization of two industrial plants, one the the Celanese facility, the other the Ballantine Brewery, located in the Iron-Bound section of Newark, New Jersey.

Vince Graziano

Mr. Corso,

I would be interested in learning about your sync efforts at the Ballantine Brewery.

Vince... first a little background to my story.

Located in the heart of Newark, NJ, Ironbound community received its name from the fact that it was surrounded on all sides by railroad lines. It contained a few industrialized plants, several of which, like Celanese and Ballantine, had employees residing in the area.

And, now the rest of the story.
One morning I received a call from a representative of a large number of the area's residents. He asked me, as Superintendent of Utilities, if I could arrange to have the power plant "blow" its steam-whistle at 7:00 am, as a reminder to workers employed in local plants. I relayed the request to the Chief Operator of the Power Plant, adding the he could "sync" time using a huge, 10-foot clock, located high up on the Ballantine Brewery tower. Being about a half mile away, it was clearly visible from the Chief’s office.

About 6 months later, during which things worked well, our power plant tripped. At about 10 am, my phone rang and a voice asked, "Is your plant shut down?" I replied, "Yes it is, but why your interest?" He replied, "I’m Ballantine’s Chief Engineer, and we set our clock by your Steam Whistle!"

Best Regards, Phil Corso

I am interested in density and viscosity measurement by ultrasonic means. Please, give me a call at 440 264-2248; that is by Cleveland OH.

we are interested in density and viscosity measurement ultrasonic means. contact us aasthachem at gmail dot com
Dear Kevin,

We are interested in this instrument as it meets our requirements in the textile industry.

Can you please send us a technical catalog for it?

Just ultrasonic or are other technologies considered?

There are now several; vibrating sensors that measure both density and
viscosity in one sensor.


PS you can contact me via if you are open to evaluate different technologies against your application.