Specific Heat and Vapor Pressure Info on Hydraulic Fluid 1

[tc7]

Hello all,
I'm trying to perform a calculation of temp rise through a throttling valve. My fluid is Mil-PRF-83282, a synthetic hydrocarbon base hydraulic oil. The basic operating temp upstream of the valve is 110 deg F. Viscosity at this temp is ~13.3cSt.

I'm using an equation derived by Henke which requires specific gravity and specific heat. I cannot find these values anywhere, not from the manufacturer (Anderol) and not the Govern office in charge of this specification. Web search not fruitful either. The MSDS sheet does not include this info either.
Can anyone advise on this information please? Also in need of Vapor pressure data. Other advice appreciated.
Thanks

 

[UmeshMathur]

It is possible to estimate the properties you need, once you have the normal boiling point and specific gravity.

If the oil is a wide boiling mixture, you also need the breakdown into narrow boiling pseudocomponents. This is done normally using a TBP or ASTM distillation curve. Then, you can apply the methods in the API Technical Data Book to get the pseudocomponent fraction properties. Finally, use various mixing rules to get back to the mixture properties.

I would contact the manufacturer directly and ask him for the required information. It would be apalling if he cannot provide even such basic information as outlined above.

 

[25362]

EXXON-MOBIL offers oils answering this spec as per their

http://www.exxonmobil.com/USA-English/Aviation/PDS/GLXXENAVIEMMobil_Aero_HF.asp

They probably could give you the requested info.

 

[tc7]

Umesh-
I agree with the appalling part, our most recent supplier could not provide this data to me. After considerable shuffling and phone tag, I was able to find good info from Shell Oil and Exxon (thanks 25362)

Exxon provide specific values, and Shell provided useful equations, applicable to all hydrocarbon fluids (or so I have been told!). I have posted this info below in case anyone in the future is interested in a similar problem:

Sp Heat = 1/D x (.388 + .00045 x T)

where D = sq rt(Spec Grav at the temp of interest)
and T = temp of interest in degrees F.

Also useful:

Specific gravity varies as +/- .0004 per each -/+ deg F.


Thankyou for your concern on my question.
Best Regards,
tc7

 

[UmeshMathur]

tc7:

In general, the liquid specific heat is dependent not only on the flowing density but also the molecular type. The latter is expressed for hydrocarbons typically through either the normal boiling point (at 1 atmosphere) or the "Watson K" which is defined as:

Kw = 1/SG60*Tb^(1/3)

where

Kw = Watson K
SG60 = specific gravity at 60 F
Tb = Normal boiling point [R]

The normal value of Kw varies from about 10 for aromatics to 13 for paraffins.

For hydrocarbon liquids, the following equation for liquid heat capacity is a more general one than what you quoted. The original reference is:

Standards of TEMA, 108 (1952). See figure in W. F. Bland and R. L. Davidson, "Petroleum Processing Handbook", (page 12-43, McGraw-Hill, 1967).

An equation describing this curve is:

Cpl = [(0.6811 - 0.308*SG60) + T*(0.000815 - 0.000306*SG60)]*(0.055*Kw + 0.351)

Where:

Cpl = liquid specific heat at T [F]
T = temperature [F]
SG60 = specific gravity at 60 F
Kw = Watson K factor, defined above.

Note that, unlike the equation in Bland & Davidson's figure, the Kw correction for Cpl should be applied to the ENTIRE term within the [] brackets, otherwise there will be no predicted effect of oil type on Cp at 0 degrees F.

The equation you have provided incorporates the effect of changing the molecular type of the oil only indirectly (via the flowing specific gravity). I evaluated this equation against the correlation quoted above and found that the discrepancy in the liquid specific heat increases dramatically as temperature goes up. Errors of 50% or more in the liquid specific heat were noted for many tested values. (I varied oil API gravity, Kw, and temperature systematically for both correlations).

An example of an unreasonable Cp value from the Shell equation:

At 500 F, for SG60=0.8762 (=30 degrees API), for Kw=10, Cp from TEMA=0.6169, v/s 0.8065 from the Shell equation. At higher temperatures, the Shell equation can even predict Cp values of 1.5 or higher.

Thanks for raising an interesting question.

 

[tc7]

Umesh-
Boiling point seems to be another elusive value!

Your equation for Kw uses "normal" boiling point temp; however, since in my throttling application, the fluid upstream is ~1800 psi and downstream is ~200 psi, will these conditions affect the Tb that I should use?

Your equation has me use SG at 60 deg F whereas the Shell equation provides for an adjustment for SG at the temperature of interest, in my case ~110 deg F. Is there any inconsistency that should be addressed?

By the way, I am not a ChemE so thankyou for your simplifying explanations.
Regards,
tc

 

[UmeshMathur]

tc7:

The "normal" boiling point refers to the boiling point at standard atmospheric pressure (760 mm Hg).

The nice thing about the equation I provided is that it doesn't require any guesswork concerning the effect of temperature on flowing density.

Another point: throttling a liquid releases very little energy (equal to v.dP). For present purposes, we can assume liquids are incompressible, so v is nearly constant.

Let's do a quick check.

Assume your oil has a specific gravity of 0.88, i.e., density = 54.9[lb/cuft], or specific volume = 1/54.9[cuft/lb].

Then, v.dP = 1/54.9[cuft/lb] * 1600[lbf/sqin] * 144[sqin/sqft] / 778[ft lbf/BTU] = 5.4[BTU/lb]

Assume oil Cp = 0.8[BTU/lb/deg F]

Therefore, temperature change = v.dP/Cp = 5.4/0.8 = 6.75[deg F]

This is a very small temperature change, even with a 1600 psi pressure drop. It will change somewhat if the Cp is different, but is a very accurate Cp really necessary?

 

[UmeshMathur]

tc7:

I forgot to address the question you raised about the use of a flowing SG@T v/s SG@60F in the two correlations.

In the equation I provided, the SG@60F and Kw factor are both part of the "process characterization" of the oil.

The Shell equation shows no effect of molecular type (accounted for by Kw). However, this equation also has a separate temperature effect, in addition to the effect on flowing SG. This may be responsible for the "far out" predictions for many of the cases I described earlier.

In my opinion, of the two, the equation I provided is more consistent and also easier to use. Of course, I have no pride of authorship here.

 

[tc7]

Umesh-
You asked in your 6 Apr posting if an accurate Cp is really necessary, and my answer is apparently not, since the Shell equation yielded Cp of 0.483 and your equation posted on 4 Apr with the Watson correction yielded a Cp of 0.477 which ultimately leads to only ~0.75 deg F difference in my throttling calculation. However it is important from the point of view that one needs to at least know what a "reasonable" Cp is, which at the start of this I did not. As can be seen from your example and assumed value for Cp of 0.8, the better value of 0.477 (or even 0.483) will make a dramatic difference (almost 68%) in the final dT.

This has been a very informative exchange and I thankyou very much for your help.
Best regards,
tc7