8 replies to this topic

[Anup87]

Dear All,

I have a question related to positive displacement compressors. Here case is of kick back valve (which is on recirculation line from discharge dampener to suction dampener). I am very confused to perform analysis on suction vessel. MAWP of suction vessel is 720 psig while discharge pressure of this PD compressor is 900 psig. I am not sure do I need to provide PSV on suction vessel of compressor to protect it in the event of failure opening of this kick back valve. Or I can write this case off by doing settle out calculations. Please suggest.

[Dacs]

I'd approach this by doing the settling-out calculation and set the suction manifold design pressure based on settling-out pressure.

[Anup87]

Thanks Dacs.

But how we can do settle out while compressor is still running.....

[fallah]

But how we can do settle out while compressor is still running.....

Anup87,

You are right, and settle out calculation isn't applicable for your case. IMO, with having the kick back valve characteristics (size, CV, maximum flowrate,..) and relevant line isometric you should do the hydraulic analysis to find if, in worst case and while the compressor is running with discharge pressure of 900 psig, the back flow through the mentioned valve can pressurize the suction side and if so how much?

Fallah

Edited by fallah, 08 September 2012 - 01:54 AM.

[Dacs]

My experience with kickback valves is such that they're not sized for full flow (maybe 50% of rated flow).

And there should be some temperature alarm/interlock at the discharge which will trip the compressor (and normally they're 2oo3). It can be tripped if there exists prolonged recirculation of fluid.

Anyway that's my take on the issue. Other people might have different insights on this.

[Bobby Strain]

What you described sounds like it is a manual valve that is used for starting the compressor. For overpressure protection of the suction side, you must consider the failure of the checkvalve downstream of the compressor. In your case, you should not have serious problems because the suction vessel is designed for a pressure not far from the discharge pressure.

Bobby

[Art Montemayor]

Anup87:

I have designed and applied this type of application many times in the past on reciprocating and screw compressors. This process operation application requires the following information before a recommendation can be made:

• A simplified flow diagram – or better still, a P&ID;
• Identification of the gas source for the suction gas – complete with explanation of how this is being controlled and safeguarded against over-pressure;
• Is this a multi-stage compressor and the stage being considered an intermediate stage?
• How is the discharge, upstream line of the compressor (presumably, the only or last stage of compression) being protected from back flow from the downstream target(s)?
• Is the gas recycled back to suction being done AFTER the compressor’s discharge cooler?

ASME (and all other pressure vessel codes I have known) mandates a pressure vessel to be protected by a PSV. I don’t know what country or vessel code you honor, but I would apply the ASME mandate. Your suction separator vessel (as well as your suction and discharge vessels) is going to have to be protected by a PSV – regardless. It is only a matter of selecting the right relief case.

A recycle control valve on a reciprocating (or PD) compressor is a control valve placed there to allow for capacity variances caused by necessity or by scarcity of suction gas. It controls flow between the discharge and the suction of a compressor. I normally design the recycle valve for 100% of rated compressor capacity. This is done in order to allow a startup and shutdown with little (or nil) load on the machine while the suction and downstream line is blocked off. This may or not be in your scope of work. A similar situation can occur when you have an emergency plant shutdown and the compressor suction and discharge lines are blocked shut. Under this type of scope, you require that the compressor operate on 100% unloading (recycle valve wide open) and a gradual, stepped shut down.

Therefore, you need to furnish complete basic data and scope of work in order to analyze and decide what the capacity should be for a recycle control valve. You may already have capacity controls such as valve lifters, clearance pockets, variable spreed, etc. We don’t know any of that. You have to furnish sufficient information to allow the Forum members to logically and accurately attack your problem.

[Anup87]

Thanks All.

Art, this is a single stage compressor. And I am working on US based refinery, so ASME will be applicable. I am doing revalidation of pressure safety devices.

This case is for single stage compressor. In this compressor (as normal flow path) system also, there is one suction dampener, then compressor then comes discharge dampener and then cooler. After cooler there is kick back control valve, which is discharging to suction dampener. There is a PSV on suction bottle (dampener), which is sized for fire case (previously). But, I am not sure why previously they did not taken failure of kick back control valve in to account. Is there any reason or they just missed it. Please guide.

Edited by Anup87, 14 September 2012 - 04:59 AM.

[Art Montemayor]

Anup87:

The reason I request Basic Data and at least a decent, detailed sketch is because we need it in order to avoid guesses and speculation while trying to furnish decent, detailed, SAFE engineering help or recommendations. Additionally, the sketch says a thousand words and is the best avenue for accurate, specific communications. Since you haven’t responded with the requested sketch, I’ll have to do it in order to ensure that the communication is accurate, documented, and well-understood. Refer to the attached workbook.

This is a very simple PSV application that is done frequently in the Up Stream Hydrocarbon industry – especially in Recovery Gas compressors. That is why I was speculating on it being a multi-stage unit. Fallah has correctly commented on your concern. There is no “settle-out” pressure to be reckoned with when dealing with positive displacement machines. That is a trait specific to dynamic machines (centrifugal compressors) – thank God. I have described the detailed operation of a reciprocating compressor many times in past years on these Forums. A SEARCH will resuscitate most of them. A reciprocating compressor’s main tool in achieving gas compression is CHECK VALVES – both on the suction and the discharge ports of each cylinder. Gas cannot flow back through the discharge port, the cylinder and the suction port without defeating the two check valves. Therefore, there normally cannot be any sort of “settle-out” with respect to system pressure. At least not through the cylinder.

The recycle control valve, as I have explained, is normally there for a specific reason – for capacity control. You are dealing with a POSITIVE DISPLACEMENT machine. It will merrily displace the same volume on each stroke, whether there is suction gas available or not. Therefore, during those times of suction gas flow scarcity (or variations), the compressor will attempt to pull a vacuum in the absence of suction gas. It is during these times that the recycle control valve cracks open and maintains a set, positive gas pressure in the suction drum. This valve should normally be specified on its Data Sheet to FAIL CLOSED. This action is specified to protect the suction drum that is normally designed to suction conditions, not discharge conditions.

The fail action and the characteristics of the operation are predicated on what your operating and shutdown philosophy and procedures are. For example, I would not design the suction drum PSV for the Pool Fire Case. The design of this PSV depends, in large part, to the controls upstream of the compressor suction. You may undergo a blow-through case; we really don’t know. We don’t even have a P&ID to resort to.

As Bobby Strain mentions, there MUST be some check valves downstream of the compressor aftercooler and the take-off for the recycle control valve. This has to be done to ensure that a stepped-down, safe, engineered shutdown of the compressor can be done – either manually or automatically. Usually you have a series of at least 2 check valves to mitigate the blow-back case. When the recycle control valve opens to recycle gas back to suction, it is the downstream target that is the potential source of any pressure in excess of 720 psig to the suction drum. If the line suction valves hold, only suction gas is recycled around the compressor, making it impossible for the suction to rise beyond its source. Should the check valves fail at the same time the recycle valve fails open, then you have a very serious problem for the suction drum – and its PSV. However, this is considered double jeopardy and this blow-back scenario is usually deemed not credible.

If you make the recycle control valve Fail Closed, then the compressor tries to pull a vacuum in the case of scarce suction gas. A series of low pressure switches should protect the compressor and allow for a controlled shutdown.

Attached Files

•  Reciprocating Compressor Capacity Control2.xlsx   23.96KB   315 downloads