Viscosity of a crude oil blend 1

[alfonsoa]

Would anybody point me to a book/journal/article where I can find a correlation for estimating the viscosity of a blend of two crudes?

Let's say I have a blend of 80% of a 19 API at 160 F and 20% of a 30 API at the same temperature.

Your help would be much appreciated.

Regards,

Alfonso

 

[25362]

If you'd like an exercise to be exemplified please indicate the viscosities of the two components.

 

[TD2K]

You need the viscosity of the two streams rather than their API gravity. There are generalized correlations for viscosity versus API out there for crudes but you are best to work with actual viscosity data.

Maxwell's Data book on Hydrocarbons has information on estimating the viscosity when blending two oils using blending indexes, Cameron's hydraulic data book has charts for estimating the viscosity of a blend and my favorite, Guide to Petroleum Product Blending (you can order it over the Internet). There are lots of other references also for this.

 

[alfonsoa]

Thanks for your prompt answers!

Alfonso

 

[mbelaidy]

These are book aticles whre you can find eithr charts or relationship to estimate the viscosity:
1.Berry's Handbook,seventh edition.
2.Petroleum refining engineering,nelson .
3.Cameron hydraulic data.

 

[25362]

Unlike sulphur content and specific gravity, viscosities cannot be arithmetically related to the % of the components. TD2K already explained that Maxwell's chart uses centistokes (cS) at 100[sup]o[/sup]F to obtain blending indices.

These blending indices are multiplied by the vol % of the components. The sum of these products, divided by 100, results in the blend's own index. One goes back to the chart with this value to obtain the kinematic viscosity of the blend in cS.

 

[jmw]

Try ASTM D341.
Alternatively DNV PS (Det Norske Veritas Petroleum services has a fuel oil blending program which will work for you.
Another alternative is to visit

http://www.steamesteem.com

where you can download a shareware program for this calculation,
These are all calculations primarily used for heavy fuel oil blending but should be applicable for most hydrocarbons. Since fuel oils are blended based on a variety of parameters some of these calculations allow you to determine also the density, sulfur content, etc of the final blend. In crude oil pipelines density limits are applied which once met then allow the fuel to be further belnded to meet viscosity limits.
However, i would add a cautionary note. many of the assumptions that go into these calculations are that the viscosity and density data used are truly representative. All too often the values used are based on lab sample evaluation. Unless these are conducted very carefully there can be significant errors. This assumes that the sample was representative to begin with. Various analysies perormed have indicated that errors in the data entered can result in significant errors. What is significant? In heavy fuel oil terms these errors can be as much as US$5 per ton. In crude oil pipelines where the crude has to be diluted with distilate fuel to meet the pipe line transport specifications the cost of excess distlate due to poor measurement can be phenomenal. Ultimately, if yours is an operation involving significant volumes of crude then you may wish to consider online viscosity measurement or an online blending system to control blending automatically.
For information on blending visit

http://www.jiskoot.com

 

[pazd]

the simplest thing I will do is to go over a HYSYS program where I put a mixer and that's it..you are done

 

[jmw]

PAZD,
can you amplify your comment please?

 

[tickle]

25362

If I remember correctly Kern (heat exchanger book) states you can estimate viscosity of a blend by using the reciprocals of the viscosities

 

[taylorg]

Don't forget the issue of compatibility!!!!!

While every fuel is manufactured to be stable within itself (that is, not having a tendency to produce asphaltenic sludge), it does not necessarily follow that two stable fuels will be compatible when blended. This is especially true of crude oil. As your request is dated 20th March I hope you have not found this out the hard way!?

As a rule of thumb, the risk is greatest where the fuels have a fairly large difference in viscosities, and/or where the two fuels have come from different fields/refineries.

Please be sure of their compatibilty before you blend....

Regards
GT

 

[jmw]

Asphaltenes can be a problem even with compatible fuels. The levels of asphaltenes can be quite high but are usually bonded to resins. When you add a low viscosity cutter to a high viscosity residual oil or even to a heavy crude (as in viscosity control of crudes in pipelines) the resins are soluble in the cutter stock and release the asphaltenes which can then drop out of the oil. (Fuel Field Handbook by Kim B Peyton, Nalco/Exxon... McGraw and Hill). This is somethiong that occurs continuously in some crude blending installations and is an occassional hazard in the fuel supply to large diesles, especially during engine shutdown when the heavy fuel is replaced with diesl in the fuel circuit.

 

[osho1980]

There are some direct correlations being used for viscosity blending one of them is Refutas method. It is easy to use and I have the spreadsheet for this. In case you need it please send me your emain ID.

 

[jmw]

OSHO 1980
The Refutas Method is new to me but a quick seacrh finds only one reference which is to

http://www.cheresources.com

where they have a spreadsheet which uses both the Refutas and Louis Methods, both of which are new to me.
It would be interesting to compare the results with the DNV PS program and the shareware program from

http://www.steamesteem.com.

Not all methods produce the same results. We experience significant problems with some clients who persist in using proprietary programs or programs based on the refeiners own equations for temperature Vs Viscosity which is a key element in these programs. In our own experience the ASTM D341 equation for temperature Vs Viscosity provides extremely good correlation with laboratory results whereas some do not.
All i have to do now is figure out how to give you my e-mail without giving it to everyone!

 

[alfonsoa]

Good discussion here. At the end of the day, I decided to prepare a spreadsheet for calculating the viscosity of a blend of crudes based on the W.A. Wright method, as described on the Guide to Petroleum Product Blending.

This method requires knowing viscosities (cSt) at two different temperatures for each of the crudes to be blended.

Suggestions re where to upload the spreadsheet are welcome.

Alfonso

 

[jmw]

http://www.dnv.com/maritime/fueltesting/aboutdnvps/technicaladvisoryservices.asp

This link is to Det Norske Veritas petroleum Services. I guess you will need to contact them to acquire their belnd optimisation program.

http://www.solartron.com

and follow the bunker fuel link to obtain a spreadsheet for calculation fuel viscosity at different temperatures. (This sheet is pre-loaded with fuel specifications taken from CIMACs fuel standards)

 

[osho1980]

jmw
Interesting to note that you have keen interest in process viscosities, would like to learn more. I am one of the contributers to cheresources for refutas method. I have checked various methods and tried to develop my own equations based on viscosity curves. you can send me your mail at jaisinghk@rediffmail.com