Density measurement for slurries

[itdepends]

I'm looking for experience/suggestions on density measurements for abrasive slurries.

The unit can be tank or pipe mounted. Slurry densities are approximately 20-25wt% (lime slurry).
Nucleonic gauges are the preferred method but will not be used in this facility.
A straight through coriolis meter has been suggested
A differential pressure transmitter on the tank is also an option but the tank is only 3.5m high and we don't want to lose too much of the live volume.

Suggestions appreciated.

Cheers

 

[Ashereng]

Load cells under the tank, assuming you can get accurate volume data? At 3.5 meters, the tank should be sufficiently small enough to put on load cells with good accruacies. The largest tank/vessle that I have put on load cells was about 50ft tall x 15 ft diameter. We put a lot of structural steel around it to steady it and keep it from moving due to wind (outside tank). Anyway, it works, but was very expensive.

I have not used a straight through coriolis (seen them), only U's, but they should work. How abrasive is the slurries? You may want to ask the vendor about that. I don't know the options for materials of the vibrating tube.

 

[itdepends]

Depending on the velocity of course the material can be quite abrasive as it contains up to 15% silica sand.

Unfortunately I don't think I'll be able to use load cells because of the cost. I'd also be concerned about the mounting method of the tank. The tank is to be mounted on a skid with other equipment and is self bunded. I'd expect a load cell system to need stable (isolated?) foundations.

Because the system is a test rig as a backup we will have a mini lab at the unit and will be able to take fixed volume slurry samples to weigh on a set of laboratory scales (i.e. mini load cells!).

Cheers

 

[Ashereng]

Yes, the foundation needs to be flat, and stable, for the load cells to read accurately.

Yes, the cost will be high.

I thought it was worth suggesting.

 

[m777182]

Search the web. There is really a lot of information there about density measurements.
m777182

 

[jmw]

The principal options would appear to be:
[ul][li]ultrasonic[/li]
[li]vibrating element[/li]
[li]nucleonic[/li][/ul]

I can't advise on ultrasonic and I do understand that many users prefer not to use nucleonic due to the gravation associated with the Health and Safety aspects though others are very happy with them.

I can assist with some info on the vibrating element types:

Vibrating element sensors include all sorts of rod, fork, tube and crystal types. Coriolis meters are simply a mass meter which, because it operates at the resonant frequency, can also measure density. Some are very good at density measurement but any advantage may be lost once in a slip-stream or by-pass sample loop and beome very expensive in the main line.

Tube types are often slip-stream or sample loop installed and care has to be taken to ensure that there is sufficient flow to prevent blockages but not so much that erossion occurs. They may also require frost protection in some climates. Periodically they may require cleaning either by roding them with a "bottle brush" or using a water flush.

However, they have a very successful applications history with slurries going back over 40 years and provide accurate measurement (0.2kg/m3 typical).
Typical applications include china clays, calcium carbonate, calsicum silicates, and so on. Some Tube types have an advantage that they can be used with aerated slurries though because they operate in a different harmonic, the accuracy is then less.

The obvious requirement of tube types for slurries is a single straight tube device but in many applications where a lesser accuracy can be tolerated, tuning forks are used in both pipelines and tanks (+/-1.0kg.m3)

These include the above applications but also chalk slurries (cement industry: the slurry is transferred from the quarry to the cement works and solids content is maximied by dosing with sodium lignosulphonate to reduce the viscosity) which contain silicates, flints and the odd bit of machinery. They are also now used downhole for density and viscosity and in towed arrays for measuring river mud for navigation purposes.

Tuning forks can be installed in the main pipeline and can be installed in tanks. The most flexible approach is to top mount using long stem versions (up to 4 metres) which allows the operator to adjust the depth to find the optimum measurement location.

Most density meters should include temperature correction to determine the density at a reference temperature from which the percent mass or volume is then determined.

Much depends on what you want to do with the measurement, what accuracy you can tolerate and how maintenance free you need it. A search will find a variety of manufacturers, types and performances.

E.g. Controlotron, Mobrey, Dynatrol, ThermoElectron etc.

PS to see what a fork looks like when badly installed in a chalk slurry application, see here:

http://www.viscoanalyser.com/solids2.html

You can, periodically checkinmg the air point (frequency in air) determine if erossion or deposition is occuring and compensate the calibration very simply with an offset value.


JMW

http://www.ViscoAnalyser.com

 

[khan101]

Bubbler is another option. It is cheap, easy to maintain. Accuracy is around + - 1-2%

 

[JimCasey]

Khan is right. The simplest answer is often the best.

Set up a bubbler with two dip tibes. Make one dip tube 2.307 feet longer than the other. Install a DP instrument between the two dip tubes. Bubble a small flow of air ( around 2 cfh)through the dip tubes and measure the differential pressure. With a 2.307 foot difference, the differential pressure in PSI will be the specific gravity of the slurry.

Meriam, Midwest, and Barton make DP indicators. Or you could use electronic DP transmitter to a DCS, and have the computer calculate the density of the slurry.

 

[jmw]

Simplest is indeed often the best, all other things being equal.
It depends on if 1-2% accuracy is sufficient.
1-2% is 10-20kg/m3 for water.
Vibrating fork will give you 1kg/m3 and vibrating tube will get you 0.1-0.2kg/m3.
Accuracy, maintenance, etc. all have cost/benefits to be determined relative to the process.

JMW

http://www.ViscoAnalyser.com

 

[jmw]

There is a PS with vibrating element and with other modern sensors.
No moving parts, virtually no ,maintenance or calibration and no computers... on-node calculation of % solids is standard.
Just give it 24VDC and pull out 4-20mA proporional to %solids.

JMW

http://www.ViscoAnalyser.com

 

[JimCasey]

I do not disagree with JMW. The vibrating element densitometer gives a very accurate reading of what is touching the element. The dip tube at least averages the reading through the zone between the dip tubes. If there is a water suspension of silica sand, it is very likely not to be homogenous and would need to be energetically and continuously mixed. There is a limit to how accurately the glop can be measured because it does not have a unique density value. Even load cells which would average the density by measuring the entire vessel wouldn't necessarily be describing what is pouring out of the overflow port.

 

[itdepends]

Thanks for all the valuable suggestions.

We are going with a straight through coriolis mass flow meter on the (lime) ring main loop as our primary density measurement. In the meant time we are also investigating the availability of vibrating densiometers here in Oz. If all else fails the tank drawing has a couple of new nozzles so that we can put a bubble tube on it and we're setting up a field lab that will be able to perform spot checks.

Hopefully that should cover our bases.

Cheers.