16 replies to this topic

[Himanshu Sharma]

Warm Greetings to all !

I have a system which is to be designed for blocked outlet case,both upstream and downstream of this KOD we have control valves.

Now due to constraint in Equipment layout,piping engineer wants to shift the PSV which is on outlet vapour line of KOD to Inlet Feed Line of KOD.

3 percent presure drop criteria for inlet line can still be satisfied, I do not anticipate any problem in such an arrangement but have never seen the same in my company's past Projects.

Please share with me your views if shifting PSV to Inlet line can pose any kinda problem.

[fallah]

Warm Greetings to all !

I have a system which is to be designed for blocked outlet case,both upstream and downstream of this KOD we have control valves.

Now due to constraint in Equipment layout,piping engineer wants to shift the PSV which is on outlet vapour line of KOD to Inlet Feed Line of KOD.

3 percent presure drop criteria for inlet line can still be satisfied, I do not anticipate any problem in such an arrangement but have never seen the same in my company's past Projects.

Please share with me your views if shifting PSV to Inlet line can pose any kinda problem.

Why don't you consider mentioned PSV directly on the vessel shell?

The vessel is horizontal or vertical?

Main concern regarding PSV installation on inlet line may be:
1-Potential of subject to two phase flow leading to complecated PSV sizing precedure and pressure flactuation due to exist distributor at the inlet.

2-In vertical position,inlet line usually connected to KOD at considerable distance below the top,and may being subject to liquid head leading to open earlier.

Edited by fallah, 12 September 2009 - 06:44 AM.

[Himanshu Sharma]

Warm Greetings to all !

I have a system which is to be designed for blocked outlet case,both upstream and downstream of this KOD we have control valves.

Now due to constraint in Equipment layout,piping engineer wants to shift the PSV which is on outlet vapour line of KOD to Inlet Feed Line of KOD.

3 percent presure drop criteria for inlet line can still be satisfied, I do not anticipate any problem in such an arrangement but have never seen the same in my company's past Projects.

Please share with me your views if shifting PSV to Inlet line can pose any kinda problem.

Why don't you consider mentioned PSV directly on the vessel shell?

The vessel is horizontal or vertical?

Main concern regarding PSV installation on inlet line may be:
1-Potential of subject to two phase flow leading to complecated PSV sizing precedure and pressure flactuation due to exist distributor at the inlet.

2-In vertical position,inlet line usually connected to KOD at considerable distance below the top,and may being subject to liquid head leading to open earlier.

Dear Fallah ,Thanks for your precious time first !

I think i should have elaborated it a bit more.

We are talking of a vertical vessel acting as a KOD designed for separating upto 300 microns particle, upstream of a PSA Hydrogen purification unit.

1).There are no chances of a two phase flow ,the max one can expect is HC mist or some very fine droplets of condensate .

2)PSV is proposed to be on Highest Point of Inlet feed Line.

3)Regarding PSV on shell itself;we do that but for skirt supported vessel and this is not supported like that.

Having clarified this ,if still there can be any Problems by this location of PSV???

[fallah]

Dear Fallah ,Thanks for your precious time first !

I think i should have elaborated it a bit more.

We are talking of a vertical vessel acting as a KOD designed for separating upto 300 microns particle, upstream of a PSA Hydrogen purification unit.

1).There are no chances of a two phase flow ,the max one can expect is HC mist or some very fine droplets of condensate .

2)PSV is proposed to be on Highest Point of Inlet feed Line.

3)Regarding PSV on shell itself;we do that but for skirt supported vessel and this is not supported like that.

Having clarified this ,if still there can be any Problems by this location of PSV???

Dear Himanshu Sharma,

Looking at safety point of view,it is better to install the PSV on top of the shell/overhead line.Because of the case is blocked outlet,there is potential of PSV inlet line to be submerged in liquid (ovefilling above liquid HH level),if PSV being installed at drum inlet line.

One more point:please clarify about relation of skirt suporting and lack of possibility of PSV installation on the shell of drum.

Regards

Edited by fallah, 13 September 2009 - 01:00 AM.

[Himanshu Sharma]

Dear Fallah ,Thanks for your precious time first !

I think i should have elaborated it a bit more.

We are talking of a vertical vessel acting as a KOD designed for separating upto 300 microns particle, upstream of a PSA Hydrogen purification unit.

1).There are no chances of a two phase flow ,the max one can expect is HC mist or some very fine droplets of condensate .

2)PSV is proposed to be on Highest Point of Inlet feed Line.

3)Regarding PSV on shell itself;we do that but for skirt supported vessel and this is not supported like that.

Having clarified this ,if still there can be any Problems by this location of PSV???

Dear Himanshu Sharma,

Looking at safety point of view,it is better to install the PSV on top of the shell/overhead line.Because of the case is blocked outlet,there is potential of PSV inlet line to be submerged in liquid (ovefilling above liquid HH level),if PSV being installed at drum inlet line.

One more point:please clarify about relation of skirt suporting and lack of possibility of PSV installation on the shell of drum.

Regards

Sir, can you please elaborate a bit ,what i could understand is that whole of the inlet line may get filled till the source of isolation and then the vertical (at an elevation w.r.t PSV) portion of the line from vessel to PSV gives a static head.

Its a valid point of concern but if this PSV is installed on Highest point then there may be no static head problem.

Regarding that skirt supported vessel thing ,as you know that in the event of popping of PSV there will be a down thrust in the opposite direction of PSV upthrust.Vessel support has to be designed taking care of this load.In case of small vessels that are skirt supported PSV can be installed on shell top itself, for bigger vessels PSV are either on Pipe-Racks or a separate platform.

Problem with this KOD is that PSV can not be on shell because of above explanation and if installed on the Overhead line a platform has to be made specially for this in the already crowded layout.

Thats why my engineering group is suggesting it to be on inlet side.

[fallah]

Sir, can you please elaborate a bit ,what i could understand is that whole of the inlet line may get filled till the source of isolation and then the vertical (at an elevation w.r.t PSV) portion of the line from vessel to PSV gives a static head.

Its a valid point of concern but if this PSV is installed on Highest point then there may be no static head problem.

Regarding that skirt supported vessel thing ,as you know that in the event of popping of PSV there will be a down thrust in the opposite direction of PSV upthrust.Vessel support has to be designed taking care of this load.In case of small vessels that are skirt supported PSV can be installed on shell top itself, for bigger vessels PSV are either on Pipe-Racks or a separate platform.

Problem with this KOD is that PSV can not be on shell because of above explanation and if installed on the Overhead line a platform has to be made specially for this in the already crowded layout.

Thats why my engineering group is suggesting it to be on inlet side.

Highest point in a horizontal inlet line doesn't make the sense.Indeed, i have never seen (don't say there is not at all) a PSV to be installed at inlet line of a vessel.

If i were you i preferred to install PSV on the overhead line with a dedicated platform for access to it.

[Himanshu Sharma]

ya infact i too have never specified a PSV on inlet line but the problem is that i have to state a good reason for not installing it there !

My piping guy is an experienced one and has good knowledge of process system ,i have to convince him for this,can you suggest me one good reason ???

one more thing , i have seen PSV on inlet line in 'Linde' Designs.

Edited by Himanshu Sharma, 13 September 2009 - 11:07 AM.

[JoeWong]

ya infact i too have never specified a PSV on inlet line but the problem is that i have to state a good reason for not installing it there !

My piping guy is an experienced one and has good knowledge of process system ,i have to convince him for this,can you suggest me one good reason ???

one more thing , i have seen PSV on inlet line in 'Linde' Designs.

Himanshu / Fallah,

Just to cut in to add some point of views.

I have seen and installed quite a number of PSVs at the inlet line. There is not definite harm of doing so if you completely understand the inlet fluid characteristic, contaminants and risk of two phase flow, solid precipitation, potential blockage, etc. There are advantages and disadvantages for each options i.e. Vapor outlet line, vessel itself (upper or lower compartment) and inlet line. It is upto your own analysis. I have searched for old post and you may be interested. "Where to Locate the PSV Inlet Nozzle ?". One of the advantages of not having PSV on vessel is less nozzle on pressure vessel.

[fallah]

Himanshu / Fallah,

Just to cut in to add some point of views.

I have seen and installed quite a number of PSVs at the inlet line. There is not definite harm of doing so if you completely understand the inlet fluid characteristic, contaminants and risk of two phase flow, solid precipitation, potential blockage, etc. There are advantages and disadvantages for each options i.e. Vapor outlet line, vessel itself (upper or lower compartment) and inlet line. It is upto your own analysis. I have searched for old post and you may be interested. "Where to Locate the PSV Inlet Nozzle ?". One of the advantages of not having PSV on vessel is less nozzle on pressure vessel.

Joe Wong

Thanks for your explanation,but as you can see in "Where to Locate the PSV Inlet Nozzle ?" we can conclude that normally the PSV may tapped from inlet line of a vessel in the cases that inlet nozzle fitted with inlet device.

[daryon]

[quote name='daryon' date='15 September 2009 - 04:41 AM' timestamp='1253008287' post='32622']
[quote name='fallah' date='14 September 2009 - 04:37 AM' timestamp='1252921672' post='32607']
[quote name='JoeWong' date='13 September 2009 - 11:23 AM' timestamp='1252859585' post='32596']
Himanshu / Fallah,

Just to cut in to add some point of views.

I have seen and installed quite a number of PSVs at the inlet line. There is not definite harm of doing so if you completely understand the inlet fluid characteristic, contaminants and risk of two phase flow, solid precipitation, potential blockage, etc. There are advantages and disadvantages for each options i.e. Vapor outlet line, vessel itself (upper or lower compartment) and inlet line. It is upto your own analysis. I have searched for old post and you may be interested. "Where to Locate the PSV Inlet Nozzle ?". One of the advantages of not having PSV on vessel is less nozzle on pressure vessel.
[/quote]

Joe Wong

Thanks for your explanation,but as you can see in "Where to Locate the PSV Inlet Nozzle ?" we can conclude that normally the PSV may tapped from inlet line of a vessel in the cases that inlet nozzle fitted with inlet device.
[/quote]

I too have seen quite a few PSV installations on gas compression and conditioning systems that are on the inlet lines to Scrubbers/KODs. Also, i remember being in a HAZOP where it was requested that the PSV inlet location be moved from the vapour outlet line of a scrubber to the side of the vessel or inlet line, to avoid any possiblity of the wire mesh type demister getting blocked are potentially isolating the vessel from the PSV. (not sure actually how likely that is? but its more likely for a wire mesh demister to get blocked than a vane pack or cyclonic demister otulet)

Edited by daryon, 15 September 2009 - 04:54 AM.

[djack77494]

I too have seen quite a few PSV installations on gas compression and conditioning systems that are on the inlet lines to Scrubbers/KODs. Also, i remember being in a HAZOP where it was requested that the PSV inlet location be moved from the vapour outlet line of a scrubber to the side of the vessel or inlet line, to avoid any possiblity of the wire mesh type demister getting blocked are potentially isolating the vessel from the PSV. (not sure actually how likely that is? but its more likely for a wire mesh demister to get blocked than a vane pack or cyclonic demister otulet)

I think a problem could arise concerning a demister, if installed. Demisters and their supports are designed for flow in one direction. For a vertical vessels, this is normally upflow (or horizontal). If a PSV on an inlet line were to pop, it could cause a large, sudden reverse flow through the demister, possibly damaging it. Also, you must consider any vessel inlet pressure losses, which may be substantial depending on the installation. As long as these factors are properly respected, I see no problem with an inlet line PSV.

[daryon]

I think a problem could arise concerning a demister, if installed. Demisters and their supports are designed for flow in one direction. For a vertical vessels, this is normally upflow (or horizontal). If a PSV on an inlet line were to pop, it could cause a large, sudden reverse flow through the demister, possibly damaging it. Also, you must consider any vessel inlet pressure losses, which may be substantial depending on the installation. As long as these factors are properly respected, I see no problem with an inlet line PSV.

Hi djack,

I didn't realise demisters could be damaged by reverse flow. I agree they are designed for flow in one direction but I thought this would only affect liquid removal performance. The mesh demisters i've seen in scrubbers are fixed to a mesh support ring and it didn't matter which way up you put the mesh pad in. The vane packs i've seen are securly bolted to clips welded to the scrubber shell wall. Both arrangents could handle flow of gas in either direction without damage. I appreciate packed beds and other internals could get damaged by reseverse flow but can't really see why a demister would?

I think the relief case in question was blocked outlet, so the volume downstream of the demister pad (gas outlet side) is probably not going to be that large.

[fallah]

Hi djack,

I didn't realise demisters could be damaged by reverse flow. I agree they are designed for flow in one direction but I thought this would only affect liquid removal performance. The mesh demisters i've seen in scrubbers are fixed to a mesh support ring and it didn't matter which way up you put the mesh pad in. The vane packs i've seen are securly bolted to clips welded to the scrubber shell wall. Both arrangents could handle flow of gas in either direction without damage. I appreciate packed beds and other internals could get damaged by reseverse flow but can't really see why a demister would?

I think the relief case in question was blocked outlet, so the volume downstream of the demister pad (gas outlet side) is probably not going to be that large.

Agreed...

Even in fire case usually the subjected vessel to be isolated from upstream and downstream by SDVs,the most of trapped gas would be existed in upstream of demister.Then "a large, sudden reverse flow through the demister" would not be happened at all.

Edited by fallah, 17 September 2009 - 02:25 AM.

[dhns]

Hi djack,

I didn't realise demisters could be damaged by reverse flow. I agree they are designed for flow in one direction but I thought this would only affect liquid removal performance. The mesh demisters i've seen in scrubbers are fixed to a mesh support ring and it didn't matter which way up you put the mesh pad in. The vane packs i've seen are securly bolted to clips welded to the scrubber shell wall. Both arrangents could handle flow of gas in either direction without damage. I appreciate packed beds and other internals could get damaged by reseverse flow but can't really see why a demister would?

I think the relief case in question was blocked outlet, so the volume downstream of the demister pad (gas outlet side) is probably not going to be that large.

Agreed...

Even in fire case usually the subjected vessel to be isolated from upstream and downstream by SDVs,the most of trapped gas would be existed in upstream of demister.Then "a large, sudden reverse flow through the demister" would not be happened at all.

Mr Fallah,

please confirm the availability of inlet distribution devices(like cyclones)
if its is there think about the reverse flow charecteristics.

More possibly tha expectation of inlet distributer or diverter for 300 mic. particles seperation is not mandatory.

if any changes in my understood please inform.

thanking you.

[fallah]

please confirm the availability of inlet distribution devices(like cyclones)
if its is there think about the reverse flow charecteristics.

It doesn't make any sense regarding reverse flow.

Edited by fallah, 17 September 2009 - 06:08 AM.

[Himanshu Sharma]

This separator does has a demister but i would like to add that i am not convinced with the 'large reverse flow theory damaging Demister'.

I read the mentioned article but still need to understand why we normally have PSV on outlet lines rather than inlet line.

Can anybody bail me out of this situation !

[joerd]

Our practice is to connect the PSV to a nozzle on the vessel, upstream the demister, to avoid a situation where the demister or the outlet line could get blocked and restrict the flow to the PSV.
Re. the inlet line, some considerations could be: valves at the vessel inlet, two-phase flow (it is much better if you can ensure vapor flow to the PSV at all times), inlet devices that could restrict the flow to the PSV (if the high pressure originates in the vessel, or downstream of the vessel), spades or spectacle blinds (for fire case if the vessel is isolated for maintenance).
Bottom line, the HAZOP team should decide if the placement of the PSV is adequate, based on a detailed review of the system.