I have a heat exchanger that is heating 40% propelene glycol with steam. I used the equation from API 521 section 5.14 for pressure rise due to simultaneous heating of the pipe and blocked-in liquid. For the pressure rise I am getting a pressure in the thousands for PSI. This does not make much sense so I am wondering if this is correct.
I have attached the file with the equations that I used. Could someone review my equation and find my problem. For the isothermal compressability I could not find the value so I substituted the eqaution in and solved for final gauge pressure.

What exactly do you mean by "blocked in" liquid? Does this mean that your heating the liquid and it has no place to expand. If that is the case, then what do you expect to happen to the pressure?

Hopefully this is just a computer model at this point. If it's real life, install a relief valve before it explodes and someone gets hurt. Please.

Patricia Lougheed

******

Please see FAQ731-376: Eng-Tips.com Forum Policies: Eng-Tips.com Forum Policies for tips on how to make the best use of the Eng-Tips Forums.

I meant that the liquid is isolated by two closed valves. It is not in operation yet. I already had the meeting about this calculation and we have confirmed that it is accuarate and we will instal a pressure safety valve.
Thanks for the reply.

The rule of thumb that I use for a reality check is 100 psi/°F. I've seen that relationship work in real life for over 40 years.

So for your 120°F temperature change, pressure should increase on the order of 12,000 psi. Your calculation is 3 times that value. If I got your result I would assume that I had fat fingered something. I didn't dig through your spreadsheet because I would never try to do a calculation that dense in excel (I use MathCad for that kind of stuff so I don't have to deal with unit conversions). The answer should be somewhere within about 500 psi of 12,000 psi.

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.

Pardon me if I don't open the spreadsheet.

Seems you should check for expansion of the pipe, diameter and length, due to internal pressure, expansion due to temperature change, and adjust the liquid density due to pressure and temperature. If it's a mix of liquids, that could be a problem to get the properties for the mixture.

If the pressure calculates too high, look for yielding in the pipe, which could relieve much of the pressure. Not necessarily a good thing, though.

If you're calculating 36,000 psi pressure, what happens to those valves? Or are there any flanges or other fittings that would give way first?

I have incorperated the liquid density with linear aproximation. Also the eqaution that I used involved the poisson ratio, modulus of elasticity, wall thickness, and pipe diameter. So the pipe yeilding is incoperated in this equation. Also I added leakage but I know that the leakage will increase as pressure increases which is something that I have not included yet so I will probably try that next.
Thanks for all of your help. I am just an intern so this is helping me better understand this career.

zdas, isn't that for pure water? He's got a glycol mixture.

I admit I didn't open the spreadsheet, so I didn't check the calculation. I've just seen the results in real life. Not pretty. Luckily no one was hurt, the utility just had to replace the heat exchanger (they knew more than I did, until it blew up. I refrained from saying "told you so.")

Patricia Lougheed

******

Please see FAQ731-376: Eng-Tips.com Forum Policies: Eng-Tips.com Forum Policies for tips on how to make the best use of the Eng-Tips Forums.

"In the real world" zdas04's 100 psi rule-of-thumb is plenty accurate.

Valves leak at the packing, usually 'weep' at the seat, etc. You will loose some containment and pressure. Also, the temperature can only be approximated in a dynamically heated system - you have 'hot' spots and 'cool' spots until diffusion gets everything into equilibrium. But nothing 'holds still' long enough to equilibrate.

The pressure rise for a blocked in fluid being heated as Zdas04 has pointed out can be very high, it's typically impractical if even possible to 'engineer out' the case by selecting a suitably high design pressure.

I've seen relying on leakage through valves, operating procedures (you don't block in both ends) or conventional thermal relief valves etc to prevent overpressure. There are usually caveats where you can use these. For example, if you specify bubble tight shut-off valves, you can't take credit for normal leakage to prevent overpressure if a section of piping was to be shut-in with the bubble tight valves.

There's no single answer that fits everywhere. In a refinery, to simply say that you have to have a thermal relief valve between every pair of block valves that 'could' be closed would be a nightmare of small bore collection piping for hydrocarbon services or for that fact, any service that you didn't want to allow discharge to grade.