7 replies to this topic

[deltaChe]

Dear Sir/Madam
sorry to bother your all this trivial.
Thermal expansion : a process fluid result from energy absorbed from an external source.
The increase in fluid temperature brings about a decrease in density and an
increase in pressure that may require the provision for thermal relief.
The energy source may be a high temperature process stream or a relatively cool
source such as ambient heat gain. In certain instances, such as with cryogenic fluids
(LNG, ethylene, or ethane for example) or chlorine liquid, provision for thermal relief due
to ambient heat gain may be required.
The analysis of a thermal expansion scenario will fall into one of the following three
categories: (1) simple hydraulic expansion with a liquid only relief,
(2)thermal expansion of a liquid with vaporization at relief conditions,
and (3) thermal expansion of a compressible vapor.
Thermal expansion relief valves are required in liquid-full systems if the system can be blocked in and/or subjected to heat input from the atmosphere or process that results in overpressure.

The question is below (please see attached file):
Q1.Is PSV set pressure should follow the design pressure that the equipment be protected (in this example Heat exchanger DP= 22 kg/cm2g)
or follow the downstream header pressure design pressure(in this example hot oil return header = 11kg/cm2g)
Q2.This Heat exchanger located around 7 m. Is the fire case scenario or thermal expansion scenario should be considered? But, in this example, the fluid(hot oil) will not vaporized
and hence thermal expansion of a liquid with vaporization may not be in the consideration.

Q3. About thermal expansion case, is the method in my post attachment correct or could you please provide more clear shortcut method for solving this case?

Thank you so much.

Attached Files

•  thermo expansion psv .xls   43KB   380 downloads

[paulhorth]

Steven,

I find this is an interesting post, I will try to answer your questions.

Q1.Is PSV set pressure should follow the design pressure that the equipment be protected (in this example Heat exchanger DP= 22 kg/cm2g)

or follow the downstream header pressure design pressure(in this example hot oil return header = 11kg/cm2g)

If you have isolation valves on the hot oil supply and return, then the PSV set pressure should be 22 kg.cm2g for the design pressure of the exchanger shell, and this design pressure should be taken into the shellside piping up to these isolation valves.

If there are no isolation valves, then there is no overpressure case, because the oil will expand into the header (IF your FCV has a minimum stop fitted). But there are good reasons to fit isolation valves, including tube rupture.

Q2.This Heat exchanger located around 7 m. Is the fire case scenario or thermal expansion scenario should be considered?

The height of a fire which is usually considered, I believe, is 8 m, so in this case you should consider it.

But, in this example, the fluid(hot oil) will not vaporized

and hence thermal expansion of a liquid with vaporization may not be in the consideration.

Yes, I agree.

Now for Q3, looking at your calculation, I agree with your basis, with the following comments.

Step 2: watch your units! This expression gives heat input in BTU/hr, while your other data is in metric units

The expression for thermal expansion coefficient comes out like this, I believe:

specific volume V = 1/(rho) where rho is DENSITY (mass per unit volume), not SG.

Then the expansion coefficient , dV/dT = 1/(rho)^2 . d(rho)/dT

So the expression is in Density not SG and has reciprocal of density squared in it. This is different from your expression in Step 3 on the spreadsheet. I have checked this with a worked example.

Then, the expression in Step 4 yields the VOLUMETRIC relieving rate, in what ever consistent volume and time units you have used.

What is surprising in this result is that the mass of fluid in the exchanger cancels out, so that the relieving rate is independent of the amount of oil in the exchanger! So you don't need to know the number and size of tubes, just the external area of the shell. i did a similar calculation (thermal expansion relief in a heat exchanger) a few weeks ago but I did not notice this.

Paul

Edited by paulhorth, 13 September 2011 - 12:41 PM.

• Back to top

---

#3 paulhorth

paulhorth

Gold Member

• 
• ChE Plus Subscriber
• 
• 396 posts

Posted 13 September 2011 - 04:13 PM

Steven,
Thinking about this a little more (always a good idea in my case) I realise that for your first question, there is absolutely no objection to setting the PSV for the system design pressure of 11 kg/cm2g. Having the exchanger shell design pressure at 22 kg/cm2g just means that the PSV does not need to be sized for the tube rupture case because the DP is more than 75% of the tube side DP.

Paul

• Back to top

---

#4 fallah

fallah

Gold Member

• 
• ChE Plus Subscriber
• 
• 5,019 posts

Posted 14 September 2011 - 04:47 AM

Having the exchanger shell design pressure at 22 kg/cm2g just means that the PSV does not need to be sized for the tube rupture case because the DP is more than 75% of the tube side

Paul,

Just to support your right conclusion regarding no need the PSV to be sized for the tube rupture case:

As per API 521 Fifth Edition, it doesn't need to be sized for the tube rupture case because the corrected hydro test pressure of the shell side (22*1.3=28.6 kg/cm2g) is higher than DP of tube side (28 kg/cm2g). Even it could be said, it doesn't need to PSV for tube rupture case.

Fallah

Edited by fallah, 14 September 2011 - 05:05 AM.

• Back to top

---

#5 fallah

fallah

Gold Member

• 
• ChE Plus Subscriber
• 
• 5,019 posts

Posted 14 September 2011 - 06:01 AM

Q2.This Heat exchanger located around 7 m. Is the fire case scenario or thermal expansion scenario should be considered? But, in this example, the fluid(hot oil) will not vaporized
and hence thermal expansion of a liquid with vaporization may not be in the consideration.

steven,

I don't think you can say the hot oil will not vaporized in the vessel surrounded by a pool fire!

In general, when a vessel is located in a fire zone (or there is the possibility it may subject to fire) you should consider fire case for it regardless of its elevation (of course, equipment located in high elevation is another story and isn't included in this discussion) with respect to flame source.

But specifically, for determination of vapor generation rate due to pool fire as per API 521, you should specify the portion of the wetted (by internal liquid) wall of the vessel (here the exchanger) that is less than 7.6 m above the source of the flame.

Fallah

• Back to top

---

#6 deltaChe

deltaChe

Gold Member

• 
• ChE Plus Subscriber
• 
• 153 posts

Posted 14 September 2011 - 10:07 AM

Steven,
Thinking about this a little more (always a good idea in my case) I realise that for your first question, there is absolutely no objection to setting the PSV for the system design pressure of 11 kg/cm2g. Having the exchanger shell design pressure at 22 kg/cm2g just means that the PSV does not need to be sized for the tube rupture case because the DP is more than 75% of the tube side DP.

Paul

Thank you for Pauhorth and fallah's incisive reply. The cliet perfer that PSV is set
for the system design pressure of 11 kg/cm2g instead of Heat exchanger DP 22kg/cm2g.
Their resaon for this is :
IF PSV set pressure is 22kg/cm2g, the pipe will be ruptured at 22 kg/cm2g and PSV
can't function properly.

Thank you very much.

Edited by steven7788, 14 September 2011 - 10:11 AM.

• Back to top

---

#7 paulhorth

paulhorth

Gold Member

• 
• ChE Plus Subscriber
• 
• 396 posts

Posted 14 September 2011 - 06:14 PM

Steven,

CORRECTION ALERT
I have thought about this yet again. On the question of the set pressure and sizing basis of the PSV, I now believe that my second answer was wrong.

I said

there is absolutely no objection to setting the PSV for the system design pressure of 11 kg/cm2g. Having the exchanger shell design pressure at 22 kg/cm2g just means that the PSV does not need to be sized for the tube rupture case

This is not correct, because, although the exchanger is OK, there is some piping with design pressure of 11 kg/cm2g which could be exposed to overpressure from the tube rupture if it is isolated. The PSV should be sized for the tube rupture case, to protect the isolated section of piping. Alternatively, the section of piping which can be isolated should be raised to 22 kg/cm2g design pressure (as in my first answer), then the tube rupture case can be eliminated.

I think the rest of the hot oil piping can stay at 11 kg/cm2g, so long as there is a free path out for a volume flowrate equal to the tube rupture flow, so that the pressure never exceeds 11 kg/cm2g.

Paul.

• Back to top

---

#8 fallah

fallah

Gold Member

• 
• ChE Plus Subscriber
• 
• 5,019 posts

Posted 14 September 2011 - 11:43 PM

Thank you for Pauhorth and fallah's incisive reply. The cliet perfer that PSV is set
for the system design pressure of 11 kg/cm2g instead of Heat exchanger DP 22kg/cm2g.
Their resaon for this is :
IF PSV set pressure is 22kg/cm2g, the pipe will be ruptured at 22 kg/cm2g and PSV
can't function properly.

steven,

In your first post you mentioned:

....downstream header pressure design pressure(in this example hot oil return header = 11kg/cm2g).

All design practices regarding piping design around equipment would rightly suggest and insist to keep the design pressure of equipment and its around piping (all inlet/outlet lines) till first isolation valve equal.

When DP of the return header is 11 kg/cm2g and that of the exchanger's shell is 22 kg/cm2g, it certainly means there is an isolation valve between exchanger outlet and return header (hasn't been observed in your sketch) based on it spec break between piping upstream and downstream of that valve would be considered. Indeed, you mentioned DP of header is 11 kg/cm2g not exchanger outlet line itself.

Thus, i think your client decision should be checked based on above statement. Of course, because of no need to consider tube rupture case, probably the PSV sizing scenario would be blocked outlet.

Fallah

Edited by fallah, 14 September 2011 - 11:45 PM.

• Back to top

---

[paulhorth]

Steven,
Thinking about this a little more (always a good idea in my case) I realise that for your first question, there is absolutely no objection to setting the PSV for the system design pressure of 11 kg/cm2g. Having the exchanger shell design pressure at 22 kg/cm2g just means that the PSV does not need to be sized for the tube rupture case because the DP is more than 75% of the tube side DP.

Paul

[fallah]

Having the exchanger shell design pressure at 22 kg/cm2g just means that the PSV does not need to be sized for the tube rupture case because the DP is more than 75% of the tube side

Paul,

Just to support your right conclusion regarding no need the PSV to be sized for the tube rupture case:

As per API 521 Fifth Edition, it doesn't need to be sized for the tube rupture case because the corrected hydro test pressure of the shell side (22*1.3=28.6 kg/cm2g) is higher than DP of tube side (28 kg/cm2g). Even it could be said, it doesn't need to PSV for tube rupture case.

Fallah

Edited by fallah, 14 September 2011 - 05:05 AM.

[fallah]

Q2.This Heat exchanger located around 7 m. Is the fire case scenario or thermal expansion scenario should be considered? But, in this example, the fluid(hot oil) will not vaporized
and hence thermal expansion of a liquid with vaporization may not be in the consideration.

steven,

I don't think you can say the hot oil will not vaporized in the vessel surrounded by a pool fire!

In general, when a vessel is located in a fire zone (or there is the possibility it may subject to fire) you should consider fire case for it regardless of its elevation (of course, equipment located in high elevation is another story and isn't included in this discussion) with respect to flame source.

But specifically, for determination of vapor generation rate due to pool fire as per API 521, you should specify the portion of the wetted (by internal liquid) wall of the vessel (here the exchanger) that is less than 7.6 m above the source of the flame.

Fallah

[deltaChe]

Steven,
Thinking about this a little more (always a good idea in my case) I realise that for your first question, there is absolutely no objection to setting the PSV for the system design pressure of 11 kg/cm2g. Having the exchanger shell design pressure at 22 kg/cm2g just means that the PSV does not need to be sized for the tube rupture case because the DP is more than 75% of the tube side DP.

Paul

Thank you for Pauhorth and fallah's incisive reply. The cliet perfer that PSV is set
for the system design pressure of 11 kg/cm2g instead of Heat exchanger DP 22kg/cm2g.
Their resaon for this is :
IF PSV set pressure is 22kg/cm2g, the pipe will be ruptured at 22 kg/cm2g and PSV
can't function properly.

Thank you very much.

Edited by steven7788, 14 September 2011 - 10:11 AM.

[paulhorth]

Steven,

CORRECTION ALERT
I have thought about this yet again. On the question of the set pressure and sizing basis of the PSV, I now believe that my second answer was wrong.

I said

there is absolutely no objection to setting the PSV for the system design pressure of 11 kg/cm2g. Having the exchanger shell design pressure at 22 kg/cm2g just means that the PSV does not need to be sized for the tube rupture case

This is not correct, because, although the exchanger is OK, there is some piping with design pressure of 11 kg/cm2g which could be exposed to overpressure from the tube rupture if it is isolated. The PSV should be sized for the tube rupture case, to protect the isolated section of piping. Alternatively, the section of piping which can be isolated should be raised to 22 kg/cm2g design pressure (as in my first answer), then the tube rupture case can be eliminated.

I think the rest of the hot oil piping can stay at 11 kg/cm2g, so long as there is a free path out for a volume flowrate equal to the tube rupture flow, so that the pressure never exceeds 11 kg/cm2g.

Paul.

[fallah]

Thank you for Pauhorth and fallah's incisive reply. The cliet perfer that PSV is set
for the system design pressure of 11 kg/cm2g instead of Heat exchanger DP 22kg/cm2g.
Their resaon for this is :
IF PSV set pressure is 22kg/cm2g, the pipe will be ruptured at 22 kg/cm2g and PSV
can't function properly.

steven,

In your first post you mentioned:

....downstream header pressure design pressure(in this example hot oil return header = 11kg/cm2g).

All design practices regarding piping design around equipment would rightly suggest and insist to keep the design pressure of equipment and its around piping (all inlet/outlet lines) till first isolation valve equal.

When DP of the return header is 11 kg/cm2g and that of the exchanger's shell is 22 kg/cm2g, it certainly means there is an isolation valve between exchanger outlet and return header (hasn't been observed in your sketch) based on it spec break between piping upstream and downstream of that valve would be considered. Indeed, you mentioned DP of header is 11 kg/cm2g not exchanger outlet line itself.

Thus, i think your client decision should be checked based on above statement. Of course, because of no need to consider tube rupture case, probably the PSV sizing scenario would be blocked outlet.

Fallah

Edited by fallah, 14 September 2011 - 11:45 PM.