Viscosity Measurement

[jmalter]

I read in a recent Control magazine (April 2003) that it is possible to calculate viscosity of a fluid by using a coriolis flowmeter and differential pressure. Does anyone have experience doing this and is it possible with crude oil?

Thanks in advance.

Joe Malter

 

[CARF]

Hi Joe,

If you have a Newtonian fluid (water, oil, beer) and laminar flow, YES you can calculate viscosity from pressure drop and flow.

The whole trick is that pressure drop is linear with pressure differential divided by flow.

viscosity = C * dP / flow [Engineering Unit]

I know Endress + Hauser has Coriolis mass-flow meters in protoype which give viscosity as an output.

Post your e-mail and I can sent you a writing on it.

Cheers,
MVD

http://www.pressure-drop-calculator.com/

 

[jmw]

Yes and no.
The new Endress and Hauser viscosity measurement system is based, i believe, on the change in signal amplitude according to the viscosity of the fluid.
Micromotion used to, and possibly still do in some cases, supply their mass meter with a dP cell to measure the pressure drop. Pressure drop isthen a function of the viscosity at any given flow rate. (see.

http://www.viscometer.com)

The problem with process viscosity measurement, when you are talking about crude oil, is that temperature is avariable and refinery measurements are analytical. the process capilary viscometer has held its position as the industry standard for viscosity measurement for over 40 years because of the difficulty of making process analytical viscosity measurements. Most process viscometers are suited to behavioural measurements. The difference is that a behavioural measurement is where you need the viscosity at the process temperature e.g. fuel oil heating for atyomisation control and analytical measurement requires the viscosity at a reference temperature. In pipe line applications the reference temperature posted by the pipe line operators is a moving target!. The problem is that the point at which you measure viscosity is at a different (variable) temperature to the pipeline reference temperature. To make suitable online contiuous process viscosity measurment requires a sophisticated measurement system which is not just a viscometer. Solartron Mobrey (

http://www.solartronmobrey.com)

usually use a system based on two viscometers operating at different temperatures to map the temperature Vs viscosity relationship (continuously) and then calculate the viscosity at the reference temperature using the ASTM D341 calculation. This gives favourable results compared to a lab viscometer (ASTM D445) which is a capillary immersed ina temperature bath operating at constant temperature. These lab measurements are usually analysed ststistically (API 555)... in some refineries this means unning as many as 8 parralele lab measurements to obtain a mean value. The lab capillary relies on the same flow rate is a function of viscosity theory as the mass flow meters with dP transmitters except they operate as apurose designed instrument under precisely controlled conditions. Once you move away from this you are making compromises.
However, i have assumed you are wanyiong to look at the quality of the crude oil. If you are wanting a simple identity of the crude for, say, adding the correct amount of drag reducer then you can use quite simple viscosity measurement techniques.

 

[jmalter]

The subject just came up again for my project. We are looking for a means of determining the viscosity of the crude oil in a pipeline. We are interested in the value to compare against the pump curves to possibly make adjustments to the pipeline operation to maximize throughput.

I'm an instrument and controls guy and when I saw the possibility of calculating the viscosity by installing a DP transmitter across our flow meter I thought I had something. Now I'm trying to see if it will work and give us meaningful results.

Joe Malter
jmalter@pipesys.com

 

[RdJ]

While on the subject.

Can anybody help me with some information on how to build a fast and effective instrument, that can help me to calculate viscosity (home made viscometer).

 

[jmw]

In pipeline applications such as you describe you probably only need a behavioural viscometer (measures the viscosity at the operating tempertature) so a direct insertion viscometer will be fine. The 7829 Visconic from solartron is used for this purpose. It is inserted into the side of the pipe. It is now available with a long stem so you can make the measurement away from the pipe wall and there are plans for a hot tap version. I don't know at what stage they are with this yet but if your pigging, you may need this.
Readings are instantaneous. You may even be able to get a rough temperature correction calculation from it ie. to estimate the viscosity at the mean pipe line temperature. It is also used for product differentiation where drag reducers are doesd according to which crude is in the pipeline.

 

[hacksaw]

Rdj- you need to post you own question to keep from confusing everyone...read the literature, there are some really simple viscosity measurements that might satisfy your needs.

 

[jmw]

Rdj- since you've asked the question, and used the term "simple" try any cup method. I've seen this done with a shampoo bottle and a stop watch. The simpler it is the more error you can expect but basically all you do is fill an open container with a small hole in the bottom and measure the efflux time. These viscometers are real cheap to buy. e.g. ford cup, DIN cup etc. Visit

http://www.viscometer.com

technical section for comparative data i.e. cup efflux time to mPs.