Explain the impossible..How can a temp. of 250 F generate a pressure o 1

[paulnormand]

How can a temp. of 250 F generate a pressure of 8,000psi in a pressure tank?

I'm running a series of hydrostatic tests on a product in a pressure vessel. The test is to be run at 250F and 10,000 psi.

The vessel is made from API certified flanged pipe and flanges. Stands about 6'high and has an i.d. of 3.8 ". Pumping port is on the bottom flange. 3 band heaters provide the heat.

Pressure is monitored using an Omega pressure transmitter (current type 6-20 ma) and using an Agilent data logger.

Tank is filled with tap water. It is pressure tested first to assure it is tight, then returned to atmosphere.

Start the heating cycle and by the time it reaches 250F the logger is showing 8,000 or so psi. This doesn't make any sense to me. I figure at 250F, the most I should be seing is 2 or 3 times atmosphere.

Any explanations?

 

[arto]

Sounds like you're thermally expanding an ~incompressible fluid & the change in volume is being accommodated by dilation of the pressure tank.

If you allowed "ullage" for vapor, then you'd be doing PV=nRT.
Now you're doing alpha'''*V*deltaT &
Delta P = [bulk Modulus]*delta V/V, with the expansion of the vessel making up the difference.

 

[TrevorP]

Do a keyword search on this site on "thermal expansion" and you will find the answer. Assuming the vessel is full of liquid, i.e. no vapour space, then, as a rule of thumb, you can expect a pressure of between 40 and 100 psi pressure rise per °F. If I assume your starting temperature is about 80°F, and your final temperature is 250°F, you have a temperature rise of 47 psi per °F, based on your figure of 8000 psig. This is not atypical. If you do a keyword search, you will find out how to calculate it theoretically, if you're interested...

 

[25362]

Neglecting the expansion of the container, you can estimate the pressure increase by thermal expansion, of any blocked-in liquid, by dividing its thermal expansivity (1/V)(dV/dT) by its isothermal compressivity (1/V)(dV/dP).
These values can be obtained from handbooks in chemical engineering, chemistry or physics.
For water this quotient would be around 55 psi/^oF at ambient conditions.

 

[hacksaw]

you need to check the calibration on your pressure transmitter and have it set up for 4-20 mA.

 

[Montemayor]

paulnormand (Electrical):

I start to worry when you say that it is "impossible" to generate 8,000 psig by heating water to 250 oF! This is elementary Physics - not even engineering. As Arto and TrevorP point out, this is elementary thermal expansion of a fluid. If you don't throroughly understand what is happening when you heat an incompressible fluid (like water) in a vessel that is devoid of any vapor space (filled 100% with liquid), you should not be doing what you describe. Someone could seriously get hurt.

The most important point everyone seems to have forgotten is: What is the MAWP (Maximum Allowable Working Pressure) of the vessel in question??? God help us if it isn't an ASME VIII coded vessel or hasn't any calculations on it!! How does one know the safe working pressure of this vessel? To hell with the "certified" API fittings -- what are the assurances that this vessel has been appropriately designed for the ultimate conditions imposed on it (which are not conventional press & temp.!)???

The next important point (& I can see why this hasn't been described if thermal expansion isn't understood) is the existance (or lack of) of a safety relief device on this "pressure vessel". Is there one or is this thing unprotected? I'm really worried now.

I seriously caution you on proceeding any further until you can address the above points I just brought out. I have done too many hydrostatic tests in the 3,000 - 5,000 psig range and also am very aware of what can go wrong with direct electrical heating pads on steel surfaces. You must have redundant safety devices to protect you from catastrophic, sudden hydrostatic pressure developed within the vessel. You should also heed what hacksaw says and make sure all the instrumentation is 100% in working order before even attempting this operation. I am a senior and experienced professional engineer and know very well the difference between a pneumatic test and a hydrostatic test. I always opt for the hydrostatic because it is less dangerous; however, it is still dangerous if you don't understand nor dominate the principles of thermal expansion.

I sincerely hope this operation continues on a safe course.

Art Montemayor

 

[jay165]

Agree w/ Montemayor. Hydrotesting is EXTREMELY dangerous, even at low pressures, much less the several thousand psi range.

 

[chicopee]

From the description of the high pressure rise w/ temp increase is amazing but not surprising for those who have studied thermodynamics.

All indicators point out to a thick wall (rigid structure) and small volume test container. You can plot the increase temp. and pressure yourself without calculations. Get a steam/water table or a Mollier diagram.

You can safely assume that the original specific volume of the water to remain constant during the increase in temp. and pressure. From that constant line going vertically upward on a Pv diagram you can start plotting the temp's and corresponding pressures.

Starting out at 70 deg. F from the saturated liquid line, there will also be a corresponding saturation pressure value. Extract from the table or Mollier diagram pressure and temp data along that constant specific volume line and you will note that it will not take much temperature to get the increase in pressure that you noted.