Jump to content



Featured Articles

Check out the latest featured articles.

File Library

Check out the latest downloads available in the File Library.

New Article

Product Viscosity vs. Shear

Featured File

Vertical Tank Selection

New Blog Entry

Low Flow in Pipes- posted in Ankur's blog

Inlet Line Loss For Psv's Not Mounted On Equipment

psv safety valves line sizing relief valves api

This topic has been archived. This means that you cannot reply to this topic.
24 replies to this topic
Share this topic:
| More

#1 MI_Sidd

MI_Sidd

    Junior Member

  • Members
  • 17 posts

Posted 12 March 2013 - 02:13 AM

The query is related to safety valves which are not mounted directly on equipment's but on process lines.

 

For such cases as per API-520 Part II, Section 4.2.2:

 

“When a pressure relief valve is installed on a process line, the 3 percent limit should be applied to the sum of the loss in the normally non-flowing pressure relief inlet pipe and the incremental pressure loss in the process line caused by the flow through the pressure relief valve.”

 

There is a confusion between our colleagues for calculation of inlet line losses based on the interpretation of the above API statement.

 

These are the current steps used to calculate "incremental pressure loss" in the process line.

 

1. Add normal process flow rate and rated relief flow rate.

2. Calculate pressure drop for this total flow rate using relieving properties eg. relieving density etc.

3. Calculate incremental pressure drop by subtracting normal ∆P from ∆P calculated in step 2.

 

The confusion is for step 2, with these two interpretaions:

 

Interpretaion 1:

 

calculate ∆P using relieving density as stated above for total flow rate (Normal + PSV relief)

 

Interpretaion 2:

 

Do not use reliveing density for total flow rate, instead take weighted averages of density of normal flow rate properties and relief load properties and then calculate pressure drop for total flow rate using these new density.

 

This new interpretation has come based on a practical example in which the normal flow rate has density of around 200 kg/m3 and relief flow rate has density of 2.0 kg/m3 (H2 gas blowthrough case). So if pressure drop is calculated using interpretation 1, it calculates using a density of 2.0 kg/m3 for total flow rate and gives a very high ∆P.

 

Interpretaion 3:

 

calculate ∆P using relieving density as stated above only for PSV rated relief rate. do not take normal flow into account.

 

If anything is not clear, kindly ask..

 

Thanks in advance. 


Edited by MI_Sidd, 12 March 2013 - 05:51 AM.


#2 fallah

fallah

    Gold Member

  • ChE Plus Subscriber
  • 4,790 posts

Posted 12 March 2013 - 04:35 AM

Ml_Sidd,

 

IMO, the interpretation 2 will lead to more reasonable result at least for your case...



#3 ankur2061

ankur2061

    Gold Member

  • Forum Moderator
  • 2,484 posts

Posted 12 March 2013 - 04:54 AM

MI_Sidd,

 

Wow, this is something totally new for me. My understanding and some questions related to my understanding:

 

1. During normal conditions there is no flow in the PSV inlet line. If this is a relief system mounted on a liquid flowing line (correct me, if I am wrong) then there may be a stagnant liquid column in the PSV inlet line. This stagnant liquid may be at a lower or higher temperature than the flowing liquid. Lower if the normal flowing temperature is higher than the ambient temperature and higher if the ambient temperature is higher than the normal flowing temperature. In either of these cases the density of the liquid would correspond to the temperature of the stagnant liquid column in the PSV inlet pipe.

 

2. Now you have overpressure due to H2 gas blow through in the liquid flowing line. The PSV inlet line now flows a gas-liquid 2-phase mixture when the PSV pops. Thus during pressure relief, the physical properties correspond to the gas-liquid 2-phase mixture at the relief temperature. So the properties such as density correspond to a 2-phase mixture at the relief temperature. These properties are totally different from the properties of the stagnant liquid in the PSV inlet line during normal flow or from the H2 gas which entered the line due to gas blow-through.

 

To conclude, I would reckon that the inlet line pressure drop would thus be based on the gas-liquid 2-phase mixture flow at the relieving temperature.

 

All of what I have mentioned is based on the assumption that the normal flow through the line is liquid flow and the relief case is related to gas blow-through this line causing overpressure. You need to verify whether what I have assumed is right or wrong.

 

Please let me know immediately that I have interpreted your normal flow and relief scenario correctly.

 

Regards,

Ankur.


Edited by ankur2061, 12 March 2013 - 04:56 AM.


#4 fahood

fahood

    Junior Member

  • Members
  • 18 posts

Posted 12 March 2013 - 05:32 AM

Based on the assumption made by Mr.Ankur that the l

 

2. Now you have overpressure due to H2 gas blow through in the liquid flowing line. The PSV inlet line now flows a gas-liquid 2-phase mixture when the PSV pops. Thus during pressure relief, the physical properties correspond to the gas-liquid 2-phase mixture at the relief temperature. So the properties such as density correspond to a 2-phase mixture at the relief temperature. These properties are totally different from the properties of the stagnant liquid in the PSV inlet line during normal flow or from the H2 gas which entered the line due to gas blow-through.

 

To conclude, I would reckon that the inlet line pressure drop would thus be based on the gas-liquid 2-phase mixture flow at the relieving temperature.

 

All of what I have mentioned is based on the assumption that the normal flow through the line is liquid flow and the relief case is related to gas blow-through this line causing overpressure. You need to verify whether what I have assumed is right or wrong.

 

Please let me know immediately that I have interpreted your normal flow and relief scenario correctly.

 

Regards,

Ankur.

 

Mr. Ankur, I believe at the first instance of gas blow, you would expect the two phase relief. Later the gas blow throught would be dominante and virtually all the relief is of a single phase gas relief. I understand that we size the PSV based on steady state flow, therefore, I would go with sizing the relief valve for 2 kg/m3.

 

Pls correct me if I am wrong



#5 MI_Sidd

MI_Sidd

    Junior Member

  • Members
  • 17 posts

Posted 12 March 2013 - 05:54 AM

Thanks for your replies  fallah, Ankur, fahood

 

Dear Ankur,

 

Letme give some background of the problem, hope it will be clear then.

 

I have a column with safety valves mounted on column top outlet line, the density of fluid through this line is 200 kg/m3 during normal operation.

 

Now, during relief scenario (H2 gas blowthrough case). For verification of 3% line losses, i will check pressure drop in PSV inlet line & process line from column to psv inlet line tap.

 

what is the correct method to calculate the pressure drop in the process line from column to psv inlet line tap based on API-520 Part II, Section 4.2.2:?

 

“When a pressure relief valve is installed on a process line, the 3 percent limit should be applied to the sum of the loss in the normally non-flowing pressure relief inlet pipe and the incremental pressure loss in the process line caused by the flow through the pressure relief valve.”

 

Is it interpretation-1 or 2 or 3 from my previous post?

 

Hope it is clear now,

 

Regards,

MI Siddiqui


Edited by MI_Sidd, 12 March 2013 - 05:55 AM.


#6 fallah

fallah

    Gold Member

  • ChE Plus Subscriber
  • 4,790 posts

Posted 12 March 2013 - 06:34 AM

MI_Sidd,

 

I still believe the interpretation 2 should be preferred.

 

We shouldn't forget we are discussing:  Calculation of "incremental pressure loss" in the process line.

 

One question:

 

Don't you think the density of 200 kg/m3 for flow of column overhead line in normal operation to be very high?


Edited by fallah, 12 March 2013 - 06:41 AM.


#7 ankur2061

ankur2061

    Gold Member

  • Forum Moderator
  • 2,484 posts

Posted 12 March 2013 - 06:42 AM

For verification of 3% line losses, i will check pressure drop in PSV inlet line & process line from column to psv inlet line tap.

 

what is the correct method to calculate the pressure drop in the process line from column to psv inlet line tap based on API-520 Part II, Section 4.2.2:?

Siddiqui,

 

As per my understanding the 3% pressure drop criteria is only applicable for the PSV inlet line and not from the column to the tap-off for the PSV. If going by what you say, most PSVs mounted on long process lines will fail the 3% criteria and specifically in cases where the set-pressure of the PSV is not very high.

 

Based on the assumption made by Mr.Ankur that the l

 

Mr. Ankur, I believe at the first instance of gas blow, you would expect the two phase relief. Later the gas blow throught would be dominante and virtually all the relief is of a single phase gas relief. I understand that we size the PSV based on steady state flow, therefore, I would go with sizing the relief valve for 2 kg/m3.

 

Pls correct me if I am wrong

Fahood,

 

You explanation is logical. The two-phase flow would be for a short period of time and subsequently the gas blow-through will dominate. However, when sizing inlet / outlet lines for PSVs the sizing or pressure drop needs to be based on the worst case scenario i.e. the maximum instantaneous flow rate that can occur. Also understand that the inlet line size can be bigger than the PSV inlet connection and a reducer provided at the inlet of the PSV. I have done a lot of PSV inlet line pressure drop calculations and when we could not satisfy the 3% criteria, we provided one line size bigger than the PSV inlet connection and then provided a reducer at the PSV inlet. The extra cost of providing one line size bigger than the PSV inlet connection and a reducer was insignificant but it did ensure that we were able to comply with the 3% critera as required by API 520.

 

Regards,

Ankur.   



#8 fallah

fallah

    Gold Member

  • ChE Plus Subscriber
  • 4,790 posts

Posted 12 March 2013 - 07:10 AM

,

As per my understanding the 3% pressure drop criteria is only applicable for the PSV inlet line and not from the column to the tap-off for the PSV. If going by what you say, most PSVs mounted on long process lines will fail the 3% criteria and specifically in cases where the set-pressure of the PSV is not very high.

Ankur,

 

Above understanding might lead to a wrong result, because the PSV is located on outlet line to protect the column against overpressure and if we ignore the pressure drop from column to tap-off branch for PSV, especially if it would be considerable, the PSV might be faced with chattering and..during relieving condition..



#9 ankur2061

ankur2061

    Gold Member

  • Forum Moderator
  • 2,484 posts

Posted 12 March 2013 - 07:42 AM

Fallah,

 

If that is the case (overpressure protection of column) than why not mount the PSV directly on the column top or route it to the PSV through a line taken from a nozzle mounted on top of the column instead of providing it on the vapor line and having to account for the additional pressure drop up to the PSV tap-off.

 

A density of 200 kg/m3 would indicate 2-phase flow and not just vapor phase. The OP needs to clarify that.

 

Regards,

Ankur.  



#10 latexman

latexman

    Gold Member

  • Admin
  • 1,174 posts

Posted 12 March 2013 - 07:54 AM

Or use a pilot operated PSV with remote sensor on the column.



#11 fallah

fallah

    Gold Member

  • ChE Plus Subscriber
  • 4,790 posts

Posted 12 March 2013 - 08:02 AM

Fallah,

 

If that is the case (overpressure protection of column) than why not mount the PSV directly on the column top or route it to the PSV through a line taken from a nozzle mounted on top of the column instead of providing it on the vapor line and having to account for the additional pressure drop up to the PSV tap-off.

 

Ankur,

 

Why not mount the PSV directly on the column top:

 

1- Because it is intended to reduce the nozzles number on the column

2- Because the overhead lines of the columns are normally the lines with higher size, then lower pressure loss

 

Of course, in order to minimize pressure drop the PSV tap-off is normally to be taken at the point as close as possible to the column...



#12 MI_Sidd

MI_Sidd

    Junior Member

  • Members
  • 17 posts

Posted 12 March 2013 - 08:38 AM

Dear Fallah,

 

MI_Sidd,

 

I still believe the interpretation 2 should be preferred.

 

We shouldn't forget we are discussing:  Calculation of "incremental pressure loss" in the process line.

 

One question:

 

Don't you think the density of 200 kg/m3 for flow of column overhead line in normal operation to be very high?

 

 

I also agree with Interpretation-2, but do u by any chance have any guideline which would support this?

 

Also, yes, Density of 200 kg/m3 is very high for column overhead line.  A high value was mentioned just to stress the density difference between two cases. It should have been between 10-20 kg/m3. Sorry for all the confusion caused.

 

Dear Ankur,

 

As per my understanding the 3% pressure drop criteria is only applicable for the PSV inlet line and not from the column to the tap-off for the PSV. If going by what you say, most PSVs mounted on long process lines will fail the 3% criteria and specifically in cases where the set-pressure of the PSV is not very high.

 

Initially, i was of the opinion that only rated flow rate had to be considered through process line & PSV inlet line and no normal flow as there can be zero process flow in many scenarios.

 

But API clearly mentions in the clause, so i think it needs to be considered,

 

 API-520 Part II, Section 4.2.2:?

 

“When a pressure relief valve is installed on a process line, the 3 percent limit should be applied to the sum of the loss in the normally non-flowing pressure relief inlet pipe and the incremental pressure loss in the process line caused by the flow through the pressure relief valve.”

 

Also, agree with fallah, if this ∆P is ignored, it can cause chattering..

 

Thanks for your responses..appreciate everyone's prompt response as always..

 

Regards,

MI Siddiqui


Edited by MI_Sidd, 12 March 2013 - 08:49 AM.


#13 th.amitkumar

th.amitkumar

    Junior Member

  • Members
  • 13 posts

Posted 25 August 2014 - 07:19 AM

For Pilot operated PSVs mounted on a column if inlet line hydraulic loss is greater than 3% than can we take line size criteria 3~5% pressure drop of set pressure. Please clarify.



#14 Marc-Andre Leblanc

Marc-Andre Leblanc

    Junior Member

  • Members
  • 22 posts

Posted 25 August 2014 - 10:42 AM

Hello,

 

I agree that the wording chosen by API is quite misleading when it come to PSV installed on process line.

 

My understanding and what followed more closely the wording of API is the following :

 

Consider A : The pressure drop, at normal operating conditions (flow, temperature, composition ...), in the process line from the protected equipement to the junction with the PSV inlet line. 

 

Consider B : The pressure drop, at relieving conditions (rated capacity, temperature, composition ...), in the PSV inlet line (junction from process line to PSV)

 

A + B shall be less than 3% of the PSV set pressure.

 

This method is more conservative than only considering  the pressure drop in the PSV inlet line and follow more closely the wording and what I think is the idea behind the wording.

 

Regards,

 

Marc-Andre


Edited by Marc-Andre Leblanc, 25 August 2014 - 10:43 AM.


#15 fallah

fallah

    Gold Member

  • ChE Plus Subscriber
  • 4,790 posts

Posted 25 August 2014 - 01:10 PM

th amitkumar,

 

It might higher pressure loss than 3% of the set pressure would be accepted, provided that the sensing line of POSV has been taken directly from top of the column. Of course, subsequent capacity reduction due to lower upstream pressure of the POSV should be considered...


Edited by fallah, 25 August 2014 - 01:11 PM.


#16 fallah

fallah

    Gold Member

  • ChE Plus Subscriber
  • 4,790 posts

Posted 25 August 2014 - 01:17 PM

Marc-Andre,

 

The definition you submitted for A cannot be acceptable. Actually it should be incremental pressure loss in the process line i.e. pressure loss at relieving flow conditions minus pressure loss at normal flow conditions...



#17 Marc-Andre Leblanc

Marc-Andre Leblanc

    Junior Member

  • Members
  • 22 posts

Posted 25 August 2014 - 02:38 PM

Hello,

 

Like i said, I think the wording chosen by API is misleading.

 

I think we both agree that what need to be done is add the pressure drop inside the PSV inlet line to a pressure drop attributed to the process line.

 

You can use an incremental like you propose but I dislike making a comparison between what can be very different sets of conditions in the same pipe especially when these sets can be completely unrelated.

 

This can quickly lead to the need for further hypothesis, Complex calculation based on relief scenarios and increase complexity to a problem that does not warrant this level of consideration. 

 

This whole post is a demonstration on how quicly many hypothesis and interpretation will arise depending on a multitude of factors, and how easy we can get lost in trying to figure out the best approach to the problem.

 

Like Annur said, somethime we do not even have to take the process line pressure drop into consideration and it can be Okay to do so!

 

Why?

 

API clearly state and the wording SHOULD denote a recommandation and not a requirement to conform to the standard. (see Foreword)

 

The principal reason for the 3% value is to try to avoid chattering in the valve. 

 

In conclusion

 

I think the method I propose is simple, not too time consumming, representative of the process, and a fair compromise between a conservative and simple approach

 

(Edit: I am not saying the method you propose is incorrect, it can be used and will indeed follow the wording of the standard, but the wording of 4.2.2 is clear on the matter that it is a recommandation, and a good basis to avoid chattering. It is left to the engineer to decide what inlet pressure drop he can really allow (depending on his process and what method of calculation and interpretation he prefer to avoid the chaterring. 

 

Some client even prefer that calculation be done at required capacities instead of rated capacities in some cases!)

 

Regards

 

Marc-Andre


Edited by Marc-Andre Leblanc, 25 August 2014 - 03:08 PM.


#18 th.amitkumar

th.amitkumar

    Junior Member

  • Members
  • 13 posts

Posted 26 August 2014 - 03:41 AM

Dear Fallah Sir
My case assemby (attached herewith) is a multiple POSV mounted on a column. Here hydraulic loss is > 3% for inlet line assembly. Here the pressure drop due in line A+ line B + line C is >3%. Please advice in this case. can i take line size criteria (line loss) >3%,if then how much?

 

Set Pressure 5 kg/cm2g 

Relieving State : Gas

End Point Pressure; 1.72 kg/cm2g

Relieving Inlet Temp. ℃ :126

 Cp/Cv : 1.06

Mol wt. : 81.4

 Viscosity, cP : 0.009

 Comp. Factor (Z) : 0.9

Total Required Flow : 830000 kg/hr

Attached Files



#19 fallah

fallah

    Gold Member

  • ChE Plus Subscriber
  • 4,790 posts

Posted 26 August 2014 - 03:45 AM

Marc-Andre,

 

We are discussing on the basic of calculating pressure loss of PSV inlet line where it has been installed on a process line; then we should adhere to the relevant concept. It's another story if in any case the difference between incremental pressure drop (as I mentioned) and pressure drop at normal operating conditions (as you mentioned) isn't so considerable; then in which to go based on normal operating conditions might not lead to effective error...



#20 fallah

fallah

    Gold Member

  • ChE Plus Subscriber
  • 4,790 posts

Posted 26 August 2014 - 03:53 AM

th amitkumar,

 

The B appears not to be a piping line but it's a connecting line of arrows to show the line numbers...

 

Please clarify what is "end point pressure" also specify the "Blowdown" value of the PSVs from relevant data sheet...



#21 th.amitkumar

th.amitkumar

    Junior Member

  • Members
  • 13 posts

Posted 26 August 2014 - 05:10 AM

Sir,

Yes u are right. Now here is the correct representation. Please advice as per the case.

Attached Files



#22 fallah

fallah

    Gold Member

  • ChE Plus Subscriber
  • 4,790 posts

Posted 26 August 2014 - 07:02 AM

Hi,

 

Line B is related to PSV by pass line, also clarify "end point pressure" and specify "Blowdown" value...



#23 th.amitkumar

th.amitkumar

    Junior Member

  • Members
  • 13 posts

Posted 26 August 2014 - 11:29 PM

Fallah Sir,

The Blowdown value is not given in datasheet but we can go with 6% and end point pressure is 1.72 kg/cm2g. Also the line B is the line related to bypass valve and from this line the connection for inlet lines to each POSV is given. Please advice about Inlet Line sizing criteria (with refrence to above post #21 & #18)


Edited by th.amitkumar, 27 August 2014 - 02:34 AM.


#24 th.amitkumar

th.amitkumar

    Junior Member

  • Members
  • 13 posts

Posted 28 August 2014 - 02:22 AM

Anybody please advice (Regarding #18, #21 & #23)



#25 fallah

fallah

    Gold Member

  • ChE Plus Subscriber
  • 4,790 posts

Posted 28 August 2014 - 11:53 AM

th.amitkumar,

 

Again please specify what do you mean by"end point pressure"? Are the multiple PSVs with staggered set points?






Similar Topics