B31.1 and safety valve requirements 1

[athomas236]

I am involved in a power and desalination plant that comprises 4 gas turbines and associated HRSGs supplying steam at 85bar 565C to 2 back pressure steam turbines through a HP steam header. Steam from the steam turbine exhausts is supplied at 3bar 160C to 5 desalination units through a LP steam header.

To accommodate for differences in the demand for power and water each HRSG is provided with a bypass valve which connects from HRSG outlet steam pipes (between the superheater outlet header and the boiler stop valve) to the LP steam header. There is an isolation valve in the pipework between the bypass valve and the LP steam header.

All pipework is designed in accordance with ASME B31.1 and this question relates to the interpretation of clause 122.5.1 of this code which states:

“Where pressure-reducing valves are used, one or more relief devices or safety valves shall be provided on the low pressure side of the system. Otherwise, the piping and equipment on the low pressure side of the system shall be designed to withstand the upstream design pressure. The relief or safety devices shall be located adjoining or as close as practical to the reducing valve. The combined relieving capacity provided shall be such that the design pressure of the low pressure system will not be exceeded if the reducing valve failed open”.

To meet the requirements of this the contractor is proposing the following:

(a) To install a safety valve on each bypass pipe between the bypass valve and the isolating valve. This safety valve is rated at 10% of the maximum bypass valve capacity and is intended to relieve any pressure caused by leakage through the closed bypass valve

(b) To install 3 pressure measurements on each bypass pipe between the bypass valve and the isolating valve. These pressure measurements will use a one out of three voting logic to close the bypass valve if pressures higher that the pipework design pressure are detected.

The bypass valves have hydraulic actuators designed to fail closed in the event of the loss of hydraulic actuation.

The contractor recognises that the capacity of safety valve does not meet the requirements of B31.1 but is arguing that the provision of 3 pressure measurements on each bypass pipe that will close the bypass valves meets the requirements for pressure relief devices as stated in clause 122.5.1. My own opinion is that this does not meet the requirements of this clause and that the only way to comply with this clause is to install safety valves or similar integrity mechanical pressure relief devices (bursting discs) rated for the maximum capacity of the bypass valves.

Any comments would be appreciated.

Best Regards

athomas236

 

[Ashereng]

This is just a guess, but it seems that the contractor is using a HIPPS in place of a PSV?

If so, I think the substitution should be allowed.

Of course, please also check with the inspector.

 

[Ashereng]

Did you say 1oo3?

At least the contractor is looking at safety rather than availability.

 

[athomas236]

Ashereng,

I did say 1oo3.

I understand that some European codes allow controls and instrumentation as a substitute for relief valves which I assume is what you mean by HIPPS.

My question, however, relates specifically to interpretation of B31.1 which I think do not allow such an approach.

Thanks

athomas236

 

[fvincent]

athomas236

I am facing the very same question.

My first opinion would be that certain kinds of failure in the reducing valve might mean that you would not succeed closing this valve at all.

However, I am not a specialist: I cannot say whether such occurrence (of an unsuccessful emergential closing of the BP valve) is realistic.

If the answer is yes, obeying the code should lead you to sizing the safety valve with the full capacity of the reducing valve.


Regards





regards

fvincent

 

[davefitz]

athomas236:

The arrangement suggested by the contractor does not meet the B31.1 requeirments.

Only in the special case of a fuel gas pressure relief is there permitted an alternative to 100% casualty flow rated relief valves. In that special instance, one can avoid the 100% capacity relief valve only by incorporating many redundancies, including (a) 2 INDEPENDENT PRV's in series plus(b) a fast acting ( 1 sec stroke period) inlet stop vale that closes upon overpressure due to (c) HARDWIRED to at least one pressure switch plus ( a leakage whistle relief valve. { Independent means actuator power supply AND pressure sensing AND control system are independent of each other}.

The arrangement the contractor suggested is not going to cut it as far as B31.1 goes. A software glitch in the control system will cause a failure , as will someone switching the valve to manual and pushing the wrong button. There are even cases where simply keying in a FM radio Xmitter nearby will set off the actuator.

 

[Ashereng]

athomas236,

The HIPPS alternative is as you indicated. I have seen it in applications REPLACE a PSV - where we could not vent the relief valve.

I am not sure how the client got permission to do this.

I highly suggest you contact your inspector/safety guy to go over this.

HIPPS is still relatively new. My first contact with it is about 5 years ago. Many jurisdictions, as you have mentioned, allow it - and more still, have not come across it. Even if B31.1, as davefitz interpreted it, does not allow it, you may be able to get an exception for your particular application.

The availability/reliability/safety of a HIPPS system can be made to equal (maybe exceed) that of a straight PSV. The exact solution needs to be engineered, in detail of course. The 1oo3 is a wonderful start.

You didn't give the reason for the contractor not going the PSV route. I am assuming that it is because there is some technical difficulties.

I am not advocating one solution over another. I am stating that there may be solution other than the traditional PSV, and depending on the circumstance, may be warranted. BUT, you need to talk to the governing bodies.

 

[davefitz]

Another casualty scenario that we saw 2 yrs ago was: for large bypass valves with thick walled bodies and/or a thickness difference between the plug and the cage, we have seen the plug stick solidly in the cage due to differential expansion. If the valve is undergoing a vast thermal transient and the plug is thinner than the cage or body, then the plug wil heat up much faster than the cage, and it gets stuck. No safety logic or Mother-of-all hydraulic actuator will correct that .

 

[athomas236]

Ashereng,

It is my expectation that cost is a major consideration for the contractor.

In addition, the contractor is European so he is more comfortable with using instrumentation as a way of elimination safety valves.

Regards,

athomas236