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

Pcv Fail Open Upstream Of A Control Valve


This topic has been archived.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

#1 J_Leo

J_Leo

    Gold Member

  • Members
  • 262 posts

Posted 14 June 2016 - 04:05 PM

Hello, As shown in the attached sketch, HP steam is used as the heating medium for an exchanger with Tube design pressure of 100psig. A self-contained PCV is used to let the pressure down to 50 psig before a temperature control valve. A PSV at the inlet to the tube side is set at the tube design pressure. In the case of PCV failure, what status of the downstream TV should be considered for the relief load calculation? Should we consider fail open for the TV as well? If so, is this a double jeopardy? TV is driven by instrument air. Thank you in advance for your inputs. Regards, Leo

Attached Files



#2 Bobby Strain

Bobby Strain

    Gold Member

  • Members
  • 3,529 posts

Posted 14 June 2016 - 07:12 PM

You don't know what the position of the TV valve will be at the time of the failure. And the bypass around the TV may be open, too.

 

Bobby



#3 S J

S J

    Veteran Member

  • Members
  • 46 posts

Posted 14 June 2016 - 09:11 PM

J_Leo

Seeing your sketch, control fail open shall not be considered.

Considering two independent contol failures is double jeopardy. (as per API 521)

In sketch, one is auto control valve controlled by IA and DCS signal and the other is just self-regulator.

If both are control valves using IA and DCS signal, you should consider both wide open case because both valves have same failure source.

If TV bypass is installed and administrative control such as Locked-open/close is done, simultaneous failure of control valve and bypass is not considered normally in my experience. But, I think you need to discuss with HAZOP chairman for simultaneous failure of bypass and control valve and administrative control because simultaneous failure with bypass opening is not definite in API 521. Or you need to find this in the specification.


Edited by S J, 15 June 2016 - 12:15 AM.


#4 fallah

fallah

    Gold Member

  • ChE Plus Subscriber
  • 4,930 posts

Posted 15 June 2016 - 01:22 AM

Hello, As shown in the attached sketch, HP steam is used as the heating medium for an exchanger with Tube design pressure of 100psig. A self-contained PCV is used to let the pressure down to 50 psig before a temperature control valve. A PSV at the inlet to the tube side is set at the tube design pressure. In the case of PCV failure, what status of the downstream TV should be considered for the relief load calculation? Should we consider fail open for the TV as well? If so, is this a double jeopardy? TV is driven by instrument air. Thank you in advance for your inputs. Regards, Leo

 

J_Leo,

 

Assuming there is no by pass line around the TV, because no credit should be taken for any favorable instrument response, you should consider the TV position as required for maximum processing flow in the case of PCV failure leading it to be in wide open position...
 



#5 J_Leo

J_Leo

    Gold Member

  • Members
  • 262 posts

Posted 15 June 2016 - 07:58 AM

Thank you all for your inputs. The approach I am thing to take is to calculate the flow through failed open PCV first. This flow will be the max for the relief load. The downstream TV should have a much bigger Cv because the normal pressure drop is much less than that across PCV. Even when TV is wide open, the flow through the TV will be restricted by the max flow through PCV. The relief rate will be much less than the relief rate through stand-alone TV without upstream PCV. I will do the calculation first and see what the relief rate will be. Regards, Leo

#6 S J

S J

    Veteran Member

  • Members
  • 46 posts

Posted 15 June 2016 - 07:13 PM

J_Leo

It looks good. As I said above, simultaneous wide open of control valve and PCV will not be considered as per API 521.

Actually, during HAZOP, to mitigate the case of control valve fail open , we do install PCV at upstream or downstream of control valve sometimes.

Anyway, you mean maximum TV outlet pressure is higher than 100 psig in case of PCV fail open.

I think you need to receive fail open flowrate of PCV or maximum TV flowrate with higher inlet pressure from vendor.

Only vendor could supply correct flowrate and it will take a few minute to calcuate the flowrate by vendor because they have their own calculation program.


Edited by S J, 15 June 2016 - 07:14 PM.


#7 cea

cea

    Gold Member

  • Members
  • 108 posts

Posted 16 June 2016 - 05:40 AM

In my opinion, simultaneous failure of both control valve need not be considered, since one the valve is self controlling & independent of common causes of control valve failure, such as instrument air failure, power failure or control failure etc.

 

The only instance of concern if TV bypass (if available) is open at the time of PSV failure. In that case, lower CV value of valve (either PSV or TV bypass) will govern for flow at relieving condition.

 

Well, I am surprised how the steam that is depressurized from  >100 psig stepped down to 50 psig, can be used for heating, which ideally would be superheated steam. (Just academic curiosity).



#8 fallah

fallah

    Gold Member

  • ChE Plus Subscriber
  • 4,930 posts

Posted 16 June 2016 - 07:00 AM

 

Well, I am surprised how the steam that is depressurized from  >100 psig stepped down to 50 psig, can be used for heating, which ideally would be superheated steam. (Just academic curiosity).

 

cea,

 

The system is probably equipped with a desuperheater before entering to the heat exchanger which hasn't been shown by OP beacuse isn't among the main mater's components...
 



#9 J_Leo

J_Leo

    Gold Member

  • Members
  • 262 posts

Posted 16 June 2016 - 08:26 AM

Thank you all for your inputs.

 

Actually, the normal heating is LP steam with desuperheater. I am considering the scenarios related to this HP steam stream only, so I didn't show the LP steam. This HP steam loop is only used as a backup. It is better to equip a desuperheater but currently it is not there maybe because it is a backup only.

 

I have another PSV in a closed methanol loop on the other side of the exchanger. Please see the attached sketch. I am wondering if the PCV fail open case will cause the relief of the PSV on the methanol loop. If so, is it acceptable if I assume the methanol generated is proportional to the total HP steam available for the PCV fail open case. The relief load is the total methanol generated minus the normal methanol vapor rate in the loop.

 

Regards,

Leo

Attached Files



#10 S J

S J

    Veteran Member

  • Members
  • 46 posts

Posted 16 June 2016 - 09:15 PM

J_Leo

 

It looks same design pressure is applied to shell and tube. So, tube rupture shall not be considered.

So, PSV at methanol loop shall not be designed for tube ruptue.

Anyway you need to confirm if single control fail open of TV or PCV will have higher pressure than 100 psig.

If not, control fail open case will not be applied to PSV at steam loop.

Additionally, if you fail to keep API rule during HAZOP, you could need to design PSV for simultaneous both fail open.

Good luck.



#11 J_Leo

J_Leo

    Gold Member

  • Members
  • 262 posts

Posted 20 June 2016 - 09:13 PM

S J,

Thank you very much for your inputs. I think only the PCV fail open case will cause over pressure.

Regards,
Leo

#12 Bobby Strain

Bobby Strain

    Gold Member

  • Members
  • 3,529 posts

Posted 20 June 2016 - 10:22 PM

No one ever doubted this conclusion. The question remains: What is the position of the TC valve and its bypass? Maybe it makes no difference if the flow through the PCV is critical in any case.

 

Bobby



#13 J_Leo

J_Leo

    Gold Member

  • Members
  • 262 posts

Posted 22 June 2016 - 11:08 AM

Hello Bobby,

 

I am not sure about the position of TV and bypass at the time of PCV failure. Anyway, the max can pass through is limited by the flow through PCV because it should be much smaller, because the normal pressure drop through PCV is much higher than that through TV.

 

Regards,

Leo



#14 Bobby Strain

Bobby Strain

    Gold Member

  • Members
  • 3,529 posts

Posted 22 June 2016 - 01:51 PM

Leo,

       You are postulating, not engineering. In a system such as yours, any values must be confirmed with calculated results. Otherwise, there is little confidence in what you do. This happens much to often.

 

Bobby



#15 J_Leo

J_Leo

    Gold Member

  • Members
  • 262 posts

Posted 22 June 2016 - 04:30 PM

Bobby,

 

I have calculations to support the relief loads I got. The Cv numbers for TV and PCV are estimated based on normal flow rate and pressure drop, assuming 70% opening at normal conditions. The relief loads will have to be confirmed during next engineering phase when the valves are selected by the vendor.

 

Many times engineering is based on assumptions, especially for PSV calculations.

 

Regards,



#16 Bobby Strain

Bobby Strain

    Gold Member

  • Members
  • 3,529 posts

Posted 22 June 2016 - 05:28 PM

Leo,

      My experience (50 years) suggests that engineering should never be based on assumptions. Especially with regards to safety systems. Codes, standards, and recommended practices are essential.

 

Bobby



#17 J_Leo

J_Leo

    Gold Member

  • Members
  • 262 posts

Posted 22 June 2016 - 05:42 PM

Bobby,

 

We always make assumptions for PSV calculations during PreFEED or FEED phase because there are many things have to be confirmed when the vendors are selected. For example, Cv for control valves, pump shut off pressure, some equipment sizes such as filters, coalescers, special type heat exchangers. These have to be confirmed when vendors are selected during detailed engineering phase. If you don't make assumptions you cannot proceed.  I don't have that many years experience but I have worked for multiple EPC firms and this is how the PSV calculations are done.

 

We do follows standards, such as API 520, 521, 526.

 

Regards,

Leo



#18 Bobby Strain

Bobby Strain

    Gold Member

  • Members
  • 3,529 posts

Posted 22 June 2016 - 08:06 PM

Unfortunately, I believe you are right about how the EPC firms proceed.

 

Bobby



#19 S J

S J

    Veteran Member

  • Members
  • 46 posts

Posted 22 June 2016 - 08:43 PM

Gentlemen

 

In the latest API 521,

4.2.3 Double Jeopardy....The simultaneous occurence of two or more unrelated causes of overpressure

(also known as double or multiple jeopardy) is not a basis for design....instrument air failure during fire exposure may

be considered single jeopardy if the fire exposure causes local air line failure....

4.4.8.3 Inlet Control Devices and Bypass Valve

....The scenario to consider is that one inlet valve is in a fully opened position regardless of the control valve

failure position....If the system has multiple inlets, the position of any control device in those remaining lines shall be

assumed to remain in its normal operating position. Therefore, the required relieving relieving rate is the difference

between the maximum expected inlet flow and the normal outlet flow,....

 

So, as per API 521, we consider simultaneous several control valves fail open as a local or general failure (IA failure),

and we do not consider simultaneous fail open for unreated control valves. "difference between the maximum expected inlet flow

and the norma outlet flow..." also means that API does not consider simultaneous fail open of unrelated controls.

Actually, regardless of API, whenever I discuss overpressure scenario for control fail open with client in HAZOP,

I feel protection could change because client process engineer or chairman could have some different experience and they insist severe protection strongly.

 

Regards.



#20 J_Leo

J_Leo

    Gold Member

  • Members
  • 262 posts

Posted 22 June 2016 - 09:19 PM

S J,

I think I followed this quoted by you
"The scenario to consider is that one inlet valve is in a fully opened position regardless of the control valve
failure position....If the system has multiple inlets, the position of any control device in those remaining lines shall be assumed to remain in its normal operating position. Therefore, the required relieving relieving rate is the difference between the maximum expected inlet flow and the normal outlet flow,...."

For the case we are discussing, HP steam is used as backup. When the HP steam flows to the exchanger, the LP steam is closed. So there is only one inlet. The approach I used is the max through PCV minus 30% of normal considering the unit might run at turndown capacity. If without PCV, the Max flow could be TV wide open plus bypass 50% open.

Regards,
Leo

Edited by J_Leo, 22 June 2016 - 09:22 PM.


#21 J_Leo

J_Leo

    Gold Member

  • Members
  • 262 posts

Posted 22 June 2016 - 09:32 PM

Bobby,

Could you please explain what approach you would use to come up with the relief rate on the HP steam side?

Also, suppose the calculated HP steam flow rate is 180% of normal flow for the design of the methanol vaporizer. The rated flow is 110% of normal. What approach would you use to come up with the relief rate on the methanol side?

Thank you,
Leo

Edited by J_Leo, 22 June 2016 - 09:33 PM.


#22 S J

S J

    Veteran Member

  • Members
  • 46 posts

Posted 23 June 2016 - 12:08 AM

J_Leo

 

In API 521, 4.4.8.3, "The potential for the bypass valve to be inadvertently opened (e.g. during normal operations, control valve maintenance, start-up, shutdown, or special operations) while the control valve is operating (both bypass and control valve wide open) should also be considered unless administrative controls are in place." When I performed ExxonMobil Project, specification indicated, if a control valve is fitted with a bypass , the larger of the flow through either the wide-open control valve or the full-open bypass shall be taken as the basis....It is assumed that the bypass will not be operated simultaneously with the control valve...

I think, without PCV, either of TV wide open or bypass full open with Lock device could be considered. Finally, it will be up to your decision.

For the case of installation of PCV and TV, I'm wondering if maximum pressure through TV would be higher than PSV set pressure.

Because either of PCV or TV will function well when malfunction of either happens, final maximum pressure through TV needs to be checked by vendor or some calculation package. So, I think the scenario of control fail open would be mitigated. But, PSV for blocked outlet case is still required.

 

Regards.



#23 Bobby Strain

Bobby Strain

    Gold Member

  • Members
  • 3,529 posts

Posted 23 June 2016 - 10:03 AM

Leo,

       Show all the details for the piping and controls. Including the normal steam supply. A P&ID if you have it.

 

Bobby



#24 ChemEng01

ChemEng01

    Gold Member

  • Members
  • 84 posts

Posted 25 June 2016 - 03:20 AM

TCV should be considered in the maximum opening position during normal operation. So I would assume it was fully open. 

 

Your PCV shown in the sketch is shown controlling the upstream pressure to 50 psig? 



#25 J_Leo

J_Leo

    Gold Member

  • Members
  • 262 posts

Posted 25 June 2016 - 10:05 AM

S J,

 

I agree with you there will be on relief on the steam side if you consider it is double jeopardy for the simultaneous failure of TV and PCV. However, I will still consider TV fail open when PCV fails. There are two reasons: (1) It is HAZOP recommendation to add PCV fail open case. (2) Sudden increase of upstream pressure caused by PCV failure might increase the chance of the TV or bypass valve failure. The HP steam design pressure is 700psig, which is might higher than the normal upstream pressure of the TV. The max flow rate through the fail TV and bypass will be the flow through failed open PCV. 

 

There is no relief for blocked outlet. (1) There is no block valve downstream (2) Even there is a block valve downstream, there will be no relief because the normal operating pressure downstream of TV is about 35psig, far away from the accumulation pressure 110psig for the PSV.

 

Thank you,

Leo






Similar Topics