19 replies to this topic

[jessie]

Hi all,

I am currently evaluating the adequacy of the existing relief valves at high pressure gas processes. I found it is a little confusing at where the flow will choke and the corresponding conditions. After digging through some materials, I would like to check if my understanding is right or not. Let's say:

Fluid: high pressure ethylene gas
Controlling Scenario: Blocked Outlets
Relief Valve: 1.5"x2.5" D orifice
Rated Capacity: 38700 lb/hr
P set= 5000 psig
P operating= 4500 psig
T operating= 200 F
P relieving= 5500 psig
T relieving= T1*(P2/P1)^(k-1)/k = 224 F (polytropic process, with k=1.221)

then the pipe segments should be,

(1) - Valve -> (2) - outlet flange, 2.5" -> (3) - Enlarger -> (4)- 3" Pipe,2ft -> (5) ->Flare header, 10 psig

Calculating backwards (assume isenthalpic):
(5) *15 psig, -55 F *choked
(4) 19 psig, -54 F
(3) *42 psig, -49 F
(2) *Nozzle exit
(1) 5500 psig, 224 F, valve inlet condition

Questions:
1. The flow is choked at nozzle exit(2), outlet flange exit(3), and 3" pipe exit(5), is it right?
2. The back pressure of the relief valve is the outlet flange pressure at 42 psig, right? (not the 3" pipe inlet)
3. If answer to question 2 is "yes", is this an example of body flow choke (choked even with no outlet piping)?
4. If answer to question 2 is "yes", we shall check every gas service valve to see if the outlet flange will be choked or not for back pressure calculation, right?

Sorry, I had only checked if the outlet piping is choked or not before, not realizing the outlet flange will experience choke flow in some cases (a dangerous mistake), so I just want to make sure if my current reasoning is correct.

Any Comment will be appreciated.

[JoeWong]

Jessie,
Welcome.

Choked at nozzle inside relief valve.

[rxnarang]

Hi all,

I am currently evaluating the adequacy of the existing relief valves at high pressure gas processes. I found it is a little confusing at where the flow will choke and the corresponding conditions. After digging through some materials, I would like to check if my understanding is right or not. Let's say:

Fluid: high pressure ethylene gas
Controlling Scenario: Blocked Outlets
Relief Valve: 1.5"x2.5" D orifice
Rated Capacity: 38700 lb/hr
P set= 5000 psig
P operating= 4500 psig
T operating= 200 F
P relieving= 5500 psig
T relieving= T1*(P2/P1)^(k-1)/k = 224 F (polytropic process, with k=1.221)

then the pipe segments should be,

(1) - Valve -> (2) - outlet flange, 2.5" -> (3) - Enlarger -> (4)- 3" Pipe,2ft -> (5) ->Flare header, 10 psig

Calculating backwards (assume isenthalpic):
(5) *15 psig, -55 F *choked Is this the downstream pressure of choked flow? I think Yes, else the downstream pressure should be 10 psig.
(4) 19 psig, -54 F
(3) *42 psig, -49 F Yes, if the outlet flange is choked, the reducer can be choked too
(2) *Nozzle exit
(1) 5500 psig, 224 F, valve inlet condition

Questions:
1. The flow is choked at nozzle exit(2), outlet flange exit(3), and 3" pipe exit(5), is it right? Looks right
2. The back pressure of the relief valve is the outlet flange pressure at 42 psig, right? (not the 3" pipe inlet) Correct
3. If answer to question 2 is "yes", is this an example of body flow choke (choked even with no outlet piping)? Correct
4. If answer to question 2 is "yes", we shall check every gas service valve to see if the outlet flange will be choked or not for back pressure calculation, right? Please see post link below Rajiv

Sorry, I had only checked if the outlet piping is choked or not before, not realizing the outlet flange will experience choke flow in some cases (a dangerous mistake), so I just want to make sure if my current reasoning is correct.

Any Comment will be appreciated.

http://www.cheresour...803

[JoeWong]

Jessie,
This could be an interesting discussion...

I wonder should choked flow occurs outlet of PRV. As choked flow occurs, back pressure on the choked piece will begin to increase. This may potentially affects flow passing PRV as well as choked point (secondary). Spring acting on the PRV seeing lower forward force and starts to re-seat. Possibly lead to closure of PRV. Then pressure in the vessel continue to rise and results PRV to open again...Cyclic and chattering scenario occurs which seriously damage PRV seats. In additional to this, metastable condition and cyclic effect from PRV inlet pipe to choked piece may results cyclic response in the pipe and increase the likelihood of fatigue failure (of course someone may say that PRV operation frequency and time is rather low. Therefore, fatigue failure may not be an issue. Nevertheless, this is still uncertain if you really add up the online testing time, PRV passing, mal-operation, etc.)

If you have balanced-PRV or pilot operated PRV, probably it reduces the chattering risk. But somehow this needs to be checked particularly those with low set pressure.

Besides that secondary choke lead to intensified dynamic strain in pipe and mechanical integrity of the outlet pipe shall be ensure sufficient to resist dynamic stress.

I would suggest try to avoid secondary choked downstream PRV as much as you can...

This may have been discussed other forum. If any of you aware of, please provide the link in subsequent post.
Appreciate other experienced engineer and expert like Mr. Montemayor, Phil, Doug, Rajiv, Ankur and those i have not named here contribute your opinion.

[jessie]

Thank you, Joe and Rajiv,

With the reference provided by Rajiv in the other post, CCPS Guidelines for Pressure Relief and Effluent Handling Systems §2.4.2.2.3 Body Flow Limitation, it is possible to attain choked-flow conditions in the pipe size of the outlet flange if the calculated flow from the valve nozzle is predicted high enough. It says choking itself is not the problem; the problem only arises if the choked-flow pressure exceeds the allowable back pressure limit for the particular valve.

In my case as the first post, the back pressure is acceptable ~1%. However, there are some conventional valves that I had seen have excessive back pressure due to the choking at the outlet flange. I think this question (asked by Bill) had been discussed at the end of the other post (link provided by Rajiv), but it really had no conclusion.

If the built-up back pressure exceeds 10% in the case of choking at the valve outlet flange, one would;
1. Select a larger valve outlet size with the orifice size unchanged (3K6 instead of 3K4)
2. Reduce the orifice size only if the valve is oversized originally
3. Select a bellow or pilot valve

So why the manufacture provide the conventional valve with a high capacity and the outlet flange size that will exceeds its allowable back pressure of 10%? Sorry, I just address the same question that Bill had.


In CCPS Guidelines for Pressure Relief and Effluent Handling Systems,

Quote

For the case of gas flow, the problem may well only reflect the conservatism of the calculation models for zero-length pipe and of the criterion for allowable built-up back pressure. There is no test data nor evidence in industrial experience for valves with no tail pipe to show that a real operational problem is indicated.......

The dilemma then is whether or not to change the selection of the device to obtain calculated results in accord with the nonmandatory Code requirement. Present experience does not show the need for such a change in gas flow service, though the designer may choose to do so in extreme cases.

Unquote

So I guess it is only a matter if we want to select such a valve using our own judgment. (No code specified)

Thanks for the inputs

[JoeWong]

So why the manufacture provide the conventional valve with a high capacity and the outlet flange size that will exceeds its allowable back pressure of 10%? Sorry, I just address the same question that Bill had.

No one medicine cure all sickness... No single solution solve all problems...

Chinese proverbs say
Out of all solution proposed by most intelligent people, there will be at least one USELESS...
Out of all solution proposed by least intelligent people, there will be at least one USEFUL...

There will be some matching cases allow these PRV to perform his duty with cost reduction.

[rxnarang]

I have wriiten to Tyco/Crosby with this question as well as link to Jessie's post. If the reply comes in, I will post it.

Rajiv

[JoeWong]

I have wriiten to Tyco/Crosby with this question as well as link to Jessie's post. If the reply comes in, I will post it.

Rajiv

Rajiv,
Thanks. Good approach.

[fallah]

As choked flow occurs, back pressure on the choked piece will begin to increase.

Hi Joe,

Which back pressure will begin to increase?Why? May you mean build-up back pressure.

Please clarify/confirm,
Regards

[fallah]

If the built-up back pressure exceeds 10% in the case of choking at the valve outlet flange, one would;
1. Select a larger valve outlet size with the orifice size unchanged (3K6 instead of 3K4)
2. Reduce the orifice size only if the valve is oversized originally
3. Select a bellow or pilot valve

So why the manufacture provide the conventional valve with a high capacity and the outlet flange size that will exceeds its allowable back pressure of 10%? Sorry, I just address the same question that Bill had.

Hi Jessie,

As far as i know,under critical flow conditions (chocked flow) the actual pressure at the nozzle exit of a PRV would be equal to critical flow pressure (or higher in non critical).It is obvious,if a much lower pressure (than critical flow pressure) exists downstream of a PRV nozzle exit,the expantion would be occured (from critical flow pressure to downstream pressure) irreversibly in the form of shock and waves often named as pressure discontinuity.
Thus,as you know,chocked flow conditions would occur according to pressure already exists in downstream of PRV nozzle exit and sizing of the valve would be done based on pressure value in downstream of PRV nozzle.If the downstream pressure be equal to or lower than critical flow pressure sizing would be based on critical conditions.If the downstream pressure be higher than critical flow pressure sizing would be based on noncritical.

As per above seems build-up back pressure should be limited by piping configuration/routing and its sizing because there is no parameter of piping configuration in PRV sizing.In case on no possibility of meeting 10% build-up back pressure seems 3rd item of your above solutions should be considered.

Hope you submit your comments in my above statements and clarify if your mean contradicted with my understanding.

Regards

[JoeWong]

As choked flow occurs, back pressure on the choked piece will begin to increase.

Hi Joe,

Which back pressure will begin to increase?Why? May you mean build-up back pressure.

Please clarify/confirm,
Regards

Fallah,
Once choked flow occur, the pressure at choked will increase quickly so that more mass flow passing the choked point. This is physic...

Edited by JoeWong, 05 March 2010 - 05:13 PM.

[fallah]

Fallah,
Once choked flow occur, the pressure at choked will increase quickly so that more mass flow passing the choked point. This is physic...

Hi Joe,

Would you please submit your explanations using upstream pressure/pressure at VC/critical flow pressure/existing back pressure?

As far as i know once choked flow occurs (when back pressure is equal to or lower than critical flow pressure),at first pressure decrease up to corresponding pressure of local sonic velocity (in VC or a place like that),then will be recovered to critical flow pressure and finally match with existing back pressure along with energy dissipation in the case of back pressure lower than critical flow pressure.If downstream piping configuration would be such that pressure loss would be considerable,pressure begins to build-up in upstream of tail pipe may pass 10% limitation of build-up back pressure and even changes current choked flow to nonchoked flow if build-up back pressure pass a value of around 50% of PRV upstream pressure.

If you agree with my above statement,with increasing build-up back pressure not only there isn't more mass flow but also there may be mass flow decreasing.

Please comment on above,

Regards

[JoeWong]

Compressible fluid passing a restriction e.g. hole, nozzle, tube, pipe, etc, if downstream pressure is lower than critical pressure, then further decrease in downstream pressure will not affect the flow. Theorectically, there will be marginally change in flow due to the fact that downstream pressure affecting vena contracta (VC) cross sectional area and location of VC. However, it is not noticeable from engineering aspect. Generally we "say' it is no effect.

If upstream pressure increase, it tends to increase density of fluid, change the pressure profile from inlet, opening, VC and outlet section. The will result higher mass flow passing the restriction. Nevertheless, choked flow still maintain at vena contracta but at different pressure profile.

Similarly, a PRV with pipe & reducer downstream, choked flow occurred at PRV downstream piping. The choked pipe section will restrict the flow. However the pressure at the inlet of PRV will continue to open the RPV in order to pass more mass flow. You will get the situation where PRV wants to pass more flow but Choked pipe section tends to limit the flow. The only way to balance the situation is that pressure upstream of choked pipe seciton (downstream of PRV) to increase to allow more mass flow passing the choked pipe. New dynaimic equilibrium (infact to me it is metastabil) is achieved between PRV outlet and choked pipe.

Above dynamic equilibrium tends to increase the probability of PRV opennig and reseating and further increase the chances of PRV chattering. Besides there is danger of increase in PRV outlet pressure resulting reduction in PRV flow...

There are several postings in http://webwormcpt.blogspot.com/ related to Chattering, backpressure, choked flow, etc which could be intersting for further reading...Try to search for them...

[fallah]

Similarly, a PRV with pipe & reducer downstream, choked flow occurred at PRV downstream piping. The choked pipe section will restrict the flow. However the pressure at the inlet of PRV will continue to open the RPV in order to pass more mass flow. You will get the situation where PRV wants to pass more flow but Choked pipe section tends to limit the flow. The only way to balance the situation is that pressure upstream of choked pipe seciton (downstream of PRV) to increase to allow more mass flow passing the choked pipe. New dynaimic equilibrium (infact to me it is metastabil) is achieved between PRV outlet and choked pipe.

Above dynamic equilibrium tends to increase the probability of PRV opennig and reseating and further increase the chances of PRV chattering. Besides there is danger of increase in PRV outlet pressure resulting reduction in PRV flow...

Thanks for your reply,

As i understood and hope being correct,what you mentioned is a detail description of Build-up Back Pressure occuring at the same time of the choked flow in PSV relief that if being over than 10% of set pressure may causes chattering of the PSV.
It could be prevented during discharge line sizing/routihg.

Your comment on my understanding would be welcomed.

Edited by fallah, 17 March 2010 - 03:30 AM.

[JoeWong]

"10% of set pressure..." being a general guideline / criteria in avoiding/minimizing PRV (spring loaded without balanced bellow nor piston) chattering.

No doubt proper design of PRV discharge piping and routing help to avoid PRV chattering. However, choked flow occurs at the PRV mating pipe section will significantly increase the pressure drop (e.g. more than 10% of set pressure) and hence built-up backpressure. Whatever you do on the downstream piping can not helps...

Edited by JoeWong, 17 March 2010 - 09:47 AM.

[fallah]

choked flow occurs at the PRV mating pipe section will significantly increase the pressure drop (e.g. more than 10% of set pressure) and hence built-up backpressure. Whatever you do on the downstream piping can not helps...

Thanks a lot for your explanations,

As i think the key point in discussions performed in current post is choked flow occuring at the PRV mating pipe section (it isn't occured normally) that customer disables to cope with its consequences such as higher build up back pressure.

Don't you think it is the responsibility of the vendor to indicate such a problem (if any) in its proposal for the cases may subject to mentioned issue?

Edited by fallah, 18 March 2010 - 10:06 AM.

[JoeWong]

Don't you think it is the responsibility of the vendor to indicate such a problem (if any) in its proposal for the cases may subject to mentioned issue?

Generally the backpressure at the PRV mating flange will be low. The choked point occurs just downstream of the mating flange before it is expecanded to tail pipe. So the pressure drop is large in this short piece between PRV mating flange and the expander. Therefore it becomes the responsibility of EPC engineer to ensure proper operation of PRV. Under this kind of situation, a balanced PRV or pilot operated PRV should be considered.

[fallah]

Therefore it becomes the responsibility of EPC engineer to ensure proper operation of PRV.

The procedure/way by which EPC engineer ensures proper operation of PRV?

Edited by fallah, 18 March 2010 - 02:50 PM.

[JoeWong]

EPC engineer to select correct type of PRV, good tail pipe design, etc so that PRV is operate correctly...

[sheiko]

I have wriiten to Tyco/Crosby with this question as well as link to Jessie's post. If the reply comes in, I will post it.

Rajiv

Dear Rajiv. Had you received any answer from TYCO/CROSBY?