Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations waross on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Compressor blowoff valve-reg

Status
Not open for further replies.

vjr0512

Mechanical
Jun 6, 2011
114
IN
A blow-off valve (4" ANSI , Cl 1500 Stainless steel SS316 Globe valve) is installed at the discharge line of a CO2 compressor ( Integrally geared) before its NRV and the compressor discharge pressure is 160 barg. This blow off valve is provided with a fail open actuator. . During HAZOP study a scenario on the Valve Stuck close on demand was brought in and discussed. This led to a recommendation for improving the reliability of the BOV which is basically a spring loaded valve. The fail to open during the trip of the compressor would be detrimental to the compressor due to reverse rotation.

Would like to put forth in this forum whether the above valve struck scenario , which is basically a mechanical struck can be managed with any other arrangement ?

Thanks for your support.
 
Replies continue below

Recommended for you

Presume this blowoff valve is an excess pressure vent valve at compressor discharge. Like any other control valve, agree there may be a remote possibility this PCV may be stuck closed. To address the high pressure scenario here, you may have a PSHH to trip the compressor and a PSV. To prevent backflow at the compressor casing, you should have a high integrity non slam check valve at the immediate discharge of each stage.
With this excess pressure vent PCV in HP CO2 service, there is a higher risk that the valve may not fully close when required due to the presence of solid CO2 within the PCV trim - have you checked simulation results?
 
Thanks georgeverghese for your response

As you know the BOV is provided to ensure the quick depressurization of the high pressure CO2 gases in the discharge side of the compressor. The compressor is provided with 2oo3 high pressure trip at the discharge side. Also we have a non-slam type axial nozzle type check valve to prevent any back flow. PSVs are also provided to protect the discharge side from over pressurization.

The possibility of dry ice formation during the expansion of the gas through the BOV is also checked and ruled out.

But still the question is whether the stuck close position of this fail open valve is a reliable scenario to be evaluated in a HAZOP review.
 
I think evaluating a valve stuck closed scenario in HAZOP is a bit unusual, however IMO if you think it can happen and consequences warrent, it should be evaluated. My experience usually stops at evaluating the possibility for overpressure in pipe or equipment, or the presense of high pressure as related to increasing risk during a fire and stopping with solving those by providing a PSV or Vent, rather than evaluating the extended effect of having a valve failure on top of those situations too. Some would count that as double jepordy (if you believe in that) and stop with simply providing the valve.

You do not say why blow off is required. Is it a safety issue, as in overpressure protection, or a case of venting pressure or product to a safe place in case of fire, etc., or is it a process issue? Safety would be of a higher consequence. It appears that it could be of lesser consequence, simply protecting a compressor from backflow in addition to having already provided a NRV, so maybe the NVR is a sufficient compensation for a nonfunctioning vent and/or the consequences for compressor damage might be relatively low, or its a very small compressor.

1) The reliability of a spring loaded valve is usually relatively high
2) If probability of sticking closed is not low enough, then attack the reasons for sticking closed.
3) If the probability of valve sticking closed remains high, the solution is usually frequent testing, once or twice per year, etc. and/or frequent cycle of observation & maintenance. It is common risk mitigation procedure to require increased freq;uency of testing for safety PSV and vent type valves. If a possibility for extreme consequences exists, two vents might be provided.



 
If you have these non slam check valves on each stage discharge, then there should be no concern with reverse rotation of the rotor from this stuck close failure. Other overpressure concerns would still be valid, but it appears you have adequate protection to minimise the duration and risk of overpressure from this stuck close PCV. By the way , is this an overpressure PCV or a depressurisation BDV?
 
Thank you very much 1503-44 and georgeverghese for your responses

This Blow off valve is provided to ensure quick depressurization of the hold up volume between the compressor discharge and the NRV. As the Anti surge and PSV tapings are upstream of the NRV and these valves being located at a higher platform for accessibility reasons, the inlet piping of these valves are contributing a large hold up volume against the vendor requirement of minimum hold up volume. The high pressure CO2 in this hold up leads to surge in the final stages of the compressor during coast down and to avoid this situation a quick depressurization is required which can be achieved by the quick opening of this BOV.
Currently there is no provision of partial stroke testing for this BOV and being a fail open valve performing the PST during the normal operation of the compressor is not advisable as it may lead to plant upset as the gas will get vented out.

Pl note that we have the NRV only at the final discharge.

 
Indeed, standard practice is to install the ASV with min holdup on the HP side so as to enable a faster dissipation of the high pressure. With increased volume on the HP side in your case, depressurisation of the HP side by the ASV will be slower, and may in some cases, lead to the compressor going into surge in the final stages on the compressor.

But you should have this NRV downstream of the ASV tapping for each stage, and not just for the final stage. Else pls explain.

Presume your ASV for the final stages is fitted with a quick blowoff relay on the ASV valve positioner to improve speed of opening.


 
Thanks george for your response...

Would like to inform that this is an integrally geared compressor with 8 stages. As this compressor loop is not designed for settle out conditions we have not envisaged to have multiple NRVs as given in compressors designed for settle out conditions. So the intermediate NRV should help to isolate the different settle out sections of the compressor loop.

We also would like to inform that here the loop is provided with an intermediate blow off valve downstream of the 4th stage discharge and an ASV and a Blow off valve at the final discharge. The ASV is connected back to the upstream side of the first suction knock out drum. During the trip scenario both the Blow off valves as well as the ASV will get opened and since the BOVs are venting to atmosphere the loop will get depressurized to atmospheric condition.

The ASV is having the feature of quick opening in less than 2 seconds.

As was discussed in the HAZOP, improving the reliability of a spring loaded valve is what has been recorded and this does not allow us to close this point without a strong evidence...

 
I can see the concern with the BOV not opening with the consequence of high settleout pressure on the lower stages. That is not good design. To me, the BOVs' are provided to prevent individual stages from going into surge during a trip ONLY, given that there is only one ASV for the whole unit. But the system designer has also taken undue credit for the action of BOTH BOVs' to enable zero or minimal impact on mechanical design pressure due to high settleout pressure.

You may speak to a field instrumentation engineer who may suggest some ways to improve reliability of these BOVs'. But that still does not remove the concern of high settleout pressure on the lower stages, since you need very high reliability ( SIL 3 or better on both of these blow off valves) to say you have eliminated the possibility of high settleout pressure on the lower stages.
 
Being from Power background I see a similar situation for a turbine steam stop valve vis-a-vis stuck open(not closed as in here) condition. There the valve is power actuated with on line testing facility. The on-line testing is generally to part close the valve momentarily and quickly back to open.

If you can make the valve pneumatically/hydraulically operated you can discuss with vendors to test the valve with crack open and then close quickly, provided no adverse effect on compressor operation can happen as may be confirmed by simulation.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor