Orifice Flow Meter For Viscous Liquid 1

[Shahin71]

Hello All :

We Used to specify PD flow meters for such applications (Heavy Fuel Oil of 380 cSt@50 Deg C)

This time the flow is large (900 m3/hr), PD meter will be expensive. Did any one come across such a case; orifice flow meter with such a fluid? Accuracy is important.

If orifice is not the right choice, what can be the other option?

Thanks

Shahin

 

[TD2K]

What are your flowing conditions of the fuel oil and its density and viscosity? What accuracy do you need/want/can live with.

That's going to help people give you suggestions.

 

[jmw]

A lot depends on where you measure it. If you are a refinery and measuring as it is blended at, say 90-130degC, then the viscosity is going to be in the range where a helical bladed turbin meter can handle it. Examples include Faure Herman or FMC. You may need to use viscosity compensation even though these are better than straight blade turbines for viscous fluids.
If you down around terminal temperatures e.g. 50degC (or if blending, 30-60degC) then I expect the viscosity will be too high for a turbine. You might then want to consider mass flow. These should be cheaper than a large PD meter.
The problem you may face with orifiace plate is that your viscosity will vary significantly as yyour process temperature varies and that's going to mess up your sizing calcs. The advantage of PD and mass flow is relative insensitivity to viscosity change. Helical bladed turbines are pretty good with viscosity but they do allow the use of viscosity to compensate the calibration.
If your are off shore (which at 900m3/hr i doubt,) then i would check to see if the mass flow meter can handle the vibrations. Mass flow meters are used for heavy fuel oil in some power stations burning burning bunker C but i have yet to see one on board ship.

 

[Shahin71]

Thank you

TD2K

Viscosity will be 360 to 380 cSt @ the pumping temp. ; 50 Deg C.

Client requires not to exceed 0.3% error of the full scale.

jmw

This is a terminal , temp. will be controlled at 50 Deg C

Thank you for the options given.

How about Aultra Sonic flow meters?

Regards

Shahin

 

[jmw]

Your client may be pushing his luck a bit; or at the least, worrying about the wrong parameter, in a fue;l transfer application.
There are lots of problems with fuel quality here. Most (or many) transactions are via a PD meter and corrected using a density value. A lot of disputes centre on consistency, i.e the viscosity first then all the other parameters. Then there are problems with density. Many transactions are based on volumetric measurement but billed according to mass.
The density (and the other parameters) are usually taken from the fuel sample and samples are later sent for analysis. Many disputes centre on being quoted too high a density for the sample by the supplier. However, if you are in a large pipe line scenario e.g. a tank farm or terminal, then on line density measurement is an affordable option if you use volumetric meters and flow computers. So too are online viscosity analysers giving you a number of other quality factors.
However, 0.3% for mass or volume is achievable for mass or volume measurement but i can't comment on whether it is achieved in fuel transfer applications. I would recomend you visit the technical information sections of the web sites for

http://www.IBIA.net,

http://www.bunkerworld.com,

http://www.Worldbunkering.com

and, a good site,

http://www.viswalab.com

etc.
Ultrasonic meters are used for crudes, certainly. I have worked on a pipeline application in Canada where they used a multi-chord ultrasonic meter in a 16" pipe (between the tank farm and the main pipeline) and had an analyser cabin installed by the pipe with a fiscal density meter (tube type) and a dual viscometer (for viscosity at reference temperatures). But i don't know what accuracy they were obtaining from the meter. However, some Ultrasonic meters and some mass meters have custody transfer accreditation.

 

[PMarinshaw]

A wedge meter could be ideal for such an application. Suggest you contact the following individual to discuss the application:
Dave Bell - Accuwedge
832-746-7179

 

[jmw]

This site is pretty useful for wedge meter information.

http://www.pfsflowproducts.com/wedgetyp.htm

Note the expected accuracy is 0.5%.of the coefficient accuracy for the calibrated range, whatever that means.
While suitable for the process conditions, i would hesitate to believe the accuracy will approach that of turbines and certainly not of a PD meter. What is needed is an understanding of the linearity, repeatability etc. or, in short, a % reading accuracy and claification of what that means in an operation where denisty and viscosity will will change and may have an effect on the volumetric accuracy of the meter, unless I am mistaken, the wedge will be susceptible to these errors as are other dP devices. I stand to be educated here and would be pleased to discover its real capability.
PD meters are relatively viscosity insentive. The viscosity of a 390cst fuel will change from 365 to 415cst for a +/-1degC temp change. A properly sized PD meter will be pretty well fully positive at these viscosities, or at the least, the slip flow will not vary significantly. Hence, so long as it dies not regularly suffer air being blown through it, its perfoemance is pretty accurate and stable, providing a very good volumetric reading.
My comment on accuracy in my previous reply questions the client achieving his 0.3% accuracy which will be the mass value. This is because heavy fuels are usually measured with volume meters and sold by weight. This means the density has to be obtained from somewhere and factored in. Note that some PD meters, because they may be sized for the worst case, do require a viscosity correction (the universal constant approach) because if sized for high viscosity (rotor clearances,)at low viscosity slip flow can be significant in custody transfer applications.

 

[jmw]

PS one reason the PD meters are expensive is that there is often one in the cutter stock and one in the resid line. This may help when re-calulating the blend properties (assuming the properties of the ingredient fuels are accurately know to begin with) but how you blend is important. Blend calculators predict the properties of the blended fuel from the viscosity, density, ash content, sulphur content etc of the cutter stock and of the resid. The problem is that there are many afctors that affect the quality during blending. One operator took top middle and bottom samples from his 380cst tank, had them analysed and found that they varied from 250 to 420cst.
Getting the flow ratio right, doesn't mean the quality will be right. A good site to visit for information on blenders is

http://www.Jiskoot.com,

and you can visit CBI Engineering of Denmarks web site (sorry, i don't have a web address) for information on other blending systems.ting the volume right and then getting a good representative density value to determine the mass delivered is only part of the problem. Getting the blend quality right is another matter.