18 replies to this topic

[chemicooool]

Theres a three stage reciprocating compressor with recycle valve from 3rd stage discharge to 1st stage suction. The control of the recycle valve is given on discharge pressure.
The recycle valve is cascaded with discharge pressure on Auto.
Will this control work?

[Bobby Strain]

Maybe.

 

Bobby

[ankur2061]

chemicooool,

 

Recycle control also known as spillback flow control is often employed in reciprocating compressors is often employed for controlling the capacity of reciprocating compressors. There are some advantages as well as disadvantages of recycle or spillback control.

 

1. Spillback from discharge to suction should be used for continuous flow control.

 

2. Spillback control is often used in combination with suction valve unloading and / or clearance pocket. Spillback or recycle causes power wastage.

 

3. Spillback flow rate should basically be rated flow rate of compressor, however, in the case of stepwise control over 2 steps, spillback flow rate can be reduced by considering incremental flow rate. Spillback flow rate shall be greater than the incremental flow rate   determined by valve unloading and / or clearance pocket because a few percent flow is still required to maintain the control valve opening above the operable range even prior to shifting to another incremental step. Refer the sketch below:

 

4. Control system for spillback can be such that both the pressure controllers, the one at the suction and the one at the discharge can control the spillback control valve through a low signal selector. Refer the sketch below for an illustration for spillback control:

 

Regards,

Ankur     

 

 

Attached Files

•  Recip_compressor_spillback_control.jpg   44.04KB   114 downloads

[mohit3040]

Ankur Sir,

 

I have few doubts In the sketch attached. Firstly, why the 2nd stage spill back line is shown to be routed at the downstream of 1st stage discharge cooler. I understand same should be routed at the upstream of cooler.

 

Secondly, what should be the sizing criterion w.r.t flow rate for spill back control valves. I couldn't able to understand the same with the sketch, which is showing spill back flow rate is 55% if the unloading operation is executed with 3 steps and 28% if the unloading operation is executed with 5 steps. Shouldn't we size this spill back control valves for 100% rated capacity.

 

Further, i have one more question in my mind.

Can we have a common spill back (i.e. from downstream of 2nd stage discharge aftercooler to 1st stage suction) for 2 stages of any reciprocating compressor?

 

if yes,

1. How individual stage capacity is controlled with common spillback?
2. What is the control philosophy followed with common spillback system?
3. How the selection is made between individual stage capacity control and common spill back control?

Thanks!!

[chemicooool]

Ankur Sir,

The compressor about which I put up the above question has spillback valve cascaded with suction and discharge pressures on AUTO with low selection.

Also,

1.The compressor does not have loading steps.

2.It operates on 100% load with total recycle during start-up.

3. It has discharge flow transmitter.

Since the compressor is designed to deliver constant pressure of a fixed value, is it possible to change the discharge control on spill back vlave into flow control taking signal from discharge FT? The suction pressure control on spillback will remain the same with low selector on suction pressure and discharge flow.

[Art Montemayor]

I have always used recycle control on multistage reciprocating compressors with caution.  I offer my experience if it can be of help here.  The reason(s) I apply recycle with care is because of the following:

• Recycle control is used under the premise that a reciprocating compressor is running at a steady, continuous speed (& capacity).  This capacity is normally oversized due to variances in compressor feed gas varying at times.  Any deficiency in compressor feed gas is compensated by the recycle valve sending compressed product from the last stage back to the first stage suction.  This is the principle of the energy wastage correctly described by Ankur.  However, the recycle valve cannot control an excess of compressor feed gas; it can only control a feed deficiency.  The same rule applies to any other type of compressor capacity control (such as valve unloaders or clearance pockets): an over abundance of feed gas has to be controlled by other means.  The compressor can only vent any excess feed gas.  You cannot maintain a constant final stage discharge on a reciprocating compressor without doing something with the produced high pressure gas - you either send it downstream or you vent it to flare, waste, or whereever - in order not to stop the compressor.  This is the excess gas feed condition that I describe.
• A reciprocating compressor has a definite maximum rod loading limit on each of its throws or piston rods.  If you exceed the gas feed to a cylinder, you increase its suction pressure and you increase the load on its piston rod.  In effect, you are forcing the cylinder to do more work than it was designed to do.  If you take compressor discharge gas and recycle it to the 2^nd stage suction while the 1^st stage is running at its normal capacity, you will increase the 2^nd stage suction and its rod loading.  You will have compression ratios between stages different from design.
• When you use individual recycle valves on each (or some) of the compressor stages you are increasing the interstage pressures and, therefore increasing the density of the interstage suction gas stream and the mass capacity of the indicated cylinder.  This will raise the rod loading on that throw.  This is the reason I have always used only one recycle valve per multistage compressor – and the control has been to divert final discharge gas back to the first stage suction.  This ensures that the compressor never exceeds its design capacity and rod loads because the first stage suction is kept at its design pressure.  This, of course, applies only to compressors without any intermediate side suction streams

I have never seen any reciprocating compressor with a flow capacity control.  I find it very difficult to believe that one could measure and control the pulsating gas flows produced by the compressor.  It is far more practical and easier to employ pressure control where you need it.  And the usual place where you need to keep a positive, controlled pressure is at the reciprocating compressor’s first stage suction.

[mohit3040]

@Art Montemayor

"When you use individual recycle valves on each (or some) of the compressor stages you are increasing the interstage pressures and, therefore increasing the density of the interstage suction gas stream and the mass capacity of the indicated cylinder.  This will raise the rod loading on that throw.  This is the reason I have always used only one recycle valve per multistage compressor – and the control has been to divert final discharge gas back to the first stage suction.  This ensures that the compressor never exceeds its design capacity and rod loads because the first stage suction is kept at its design pressure.  This, of course, applies only to compressors without any intermediate side suction streams."

 

As per your statement, which is reporduced above, i couldn't able to understand that using individual recycle valves on each of the compressor stages, how the the interstage pressures will increase and raise the rod loading on that throw.

Because as per my understanding the multi stage compressor with individual stage recycle valves, we always have each stage suction pressure control through spillback valve. Refer the file attached for your reference.

 

Can you please clarify further that when to use common spillback and when to use individual spillback.

Attached Files

•  3 stage compressor control.jpg   35.48KB   35 downloads

[Art Montemayor]

mohit3040:

 

I don't see what the purpose of recycling to the intermediate stages is.  If you have a reason for applying this type of reciprocating control, please tell us the circumstances or reasons for doing it that way.   I have applied a lot of reciprocating recycle control valves in my past and everyone of them was from the discharge of the last stage to the suction of the first stage.

 

The reason I applied that type of capacity control was that it suited the application, it was the simplest, the cheapest, and it protected the compressor suction from going into vacuum.  I never found a need to add more capacity control valves around the other compressor stages.  (I've installed and operated reciprocating compressors with up to 5 stages of compression).  Therefore, if I can't justify using more than one recycle control valve, I won't do it - unless there is an engineering need or reason for doing it. 

[chemicooool]

I agree with Art. If there's no withdrawal of gas from intermediate stage, there need not be a recycle valve in those stages. Only one recycle valve from discharge of compressor to its suction will serve the purpose.

[mohit3040]

@Art Montemayor
Sir, In the makeup gas compressors i have seen that almost all the licensors specify individual stage spill back control. There may be one thing in different from normal reciprocating compressors that these compressors are now provided with hydrocom control for energy saving instead of loader/unloader valves. But I don't think that hydrocom makes any difference in selection of individual stage spillback and common spillback, as hydrocom controls capacity in between 30%-100% during H2 run without any action of spillback valves.
 
Few of the reasons i have in mind for having individual stage spillback control as mentioned below, please correct me if i am wrong,

·         As the reciprocating compressor is a constant volume machine. Each stage of the compressor will be designed based on some constant volumetric flow rate and size of the stages will decrease with increase in pressure for the same mass flow rate (i.e. 1st stage capacity> 2nd stage capacity> 3rd stage capacity). Volumetric flow rate at each stage suction will vary depending upon the suction pressure. And without any individual stage spillback control, suction pressure at intermediate stages will vary and we will not be able to control the same. If the intermediate stage pressure drops down there will not be any problem as such and it can be controlled even with common spillback with increase in volumetic flow rate at first stage suction but if the intermediate stage pressure will increase there is a chance that it can reach to design pressure and chance of PSV popping.

·         Secondly, It is always better to recycle the remaining gas after pressure control from intermediate stage rather than last stage, as in this case one need not to compress the additional gas in the 2nd and 3rd stages which is to be recycled back and hence saves energy.

[Art Montemayor]

mohit3040:

 

I don’t know why you introduce the subject of “hydrocom”.  I believe you are trying to refer to Hydrocom – a hydraulic valve unloader system sold by Hoerbiger compressor valves, an Austrian company.  I am very familiar with Hoerbiger and their products.  Their Hydrocom system is nothing more than the “lifting” of suction valve plates.  This basic system is as old as the reciprocating compressor.  I do not favor this method because of the wear and tear it lends to suction valves.  Before using this system I would employ clearance pockets for capacity control.  But this has nothing to do with recycle capacity control, so I don’t know why you bring it up.

 

You ask for me to correct you if you are wrong in what you present.  I believe you are wrong in what you write in this last post, and I base myself on my past experience in operating many reciprocating compressors in many gas services.

 

You have failed to mention or describe what compression scenario you are dealing with: is the first stage suction flow rate uncontrolled from another process – as is often the process case – or is the discharge of the compressor to be kept at a constant flow rate (dependent on the first stage suction flow rate?  Without you fixing the process requirements or basis, you really don’t have the basis for setting a capacity control on the compressor – at any of its compression stages.  You first have to state why you are controlling the capacity of the compressor in question.

 

I have already stated why I control the way I recommend: in order to maintain the suction pressure (and temperature) at a constant value.  This ensures that the mass flow rate going through the first stage – as well as all other subsequent stages – is kept constant.  It does not mean that the NET produced gas mass flow rate produced by the compressor is kept constant.  When less gas is available to be fed to the first compression stage, less net high pressure gas will be produced by the compressor.  Since I have no side streams entering or leaving any of the intermediate stages, I have no need to introduce more than one recycle valve that will recycle a portion of the final discharge gas back to the first stage suction.

[ruihui]

@Art, you may help me on the below conditions.

 

There is a VRU compressor package in one gas plant, totally two stages recip compressor, the 1st stage suction pres. 265kPag, 2nd stage discharge of 1810 kPag, suction flow rate as showed below, unit is Sn3/h,

 

design: 551, normal: 284, min: 96.4, and there will be 710 Sm3/h @ 800 kPag side stream will be added into interstage 9.5 hours every 24 hours.

 

We got performance run reports from Ariel, for the min case,

 

1. 1st stage suction of 96.4 Sm3/h, 2nd suction with 710 Sm3/h added is 806.4 Sm3/h, to keep inter stage pres not exceed 800kPag( exceed this, the 710 Sm3/h can not injected into inter stage), Ariel requires 1st stage suction has to maintain 279 Sm3/h flow, which means there is 279-96.4=182.6 Sm3/h from 710 come to 1st stage suction. 1st cylinder of 6-3/8", 2nd cylinder 5.5", compression ratio 2.645/2.175, speed of 617.5 rpm.

 

2. both 1st stage and 2nd stage flow of 96.4 Sm3/h, same cylinder configuration, 1st / 2nd compression ratio changed to 1.639/3.56, speed 600 rpm.

 

My questions:

 

1. why the inter pressure for case with 710 sm3/h added is higher than the case without 710?

2. why increase 1st stage suction flow can help to reduce the inter pressure?

3. as the current both stages cylinder are double acting, can we use single acting for 2nd stage without 710?

4. do you have good solution for this conditions?

 

Thanks a lot

 

Ray

[Bobby Strain]

You can download software from Ariel and run cases to your hearts content.

 

Bobby

[Anshu18]

Hi,

I have gone through your discussion about the spillback for reciprocating compressor.

I believe that this spillback valve should be strictly treated like a capacity control valve & not like anti-surge valve. Came across few P & Ids for Recip compressors where the spillback valve is 'FO' type & recip first stage suction drum & piping is designed for settle-out conditions between first & last stage!! The reason cited is the spillback valve is opening after every trip of the compressor!! i personally find it very strange.

May I request Art Sir to give us some insight. 

[Art Montemayor]

Anshu18:

 

The only insight I can offer based on your post is the following:

• The topic of this thread is clearly reciprocating compressor capacity control through use of a recycle valve that allows gas flow from the final discharge to the 1^st stage suction when the 1^st stage suction pressure starts to decrease due to a feed gas volume deficiency.  Therefore, I don’t understand your comment about this recycle valve being treated as an “anti-surge valve”.  There is no such thing as “surge” related to a reciprocating compressor.  Are you confusing a reciprocating compressor with a centrifugal model?
• You are correct in finding fault with a recycle valve that is installed to fail in the open position.  What engineering contractor developed and approved those P&IDs?  Without access to the P&IDs and the ability to study the actual process I have to depend on your claim and I would advise you to contact the engineering contractor and argue your case about an unsafe and potentially dangerous situation caused by them.  I would certainly put in formal writing advising them of your discovery and a request for a safe resolution.
• Since this is a reciprocating compressor, there is also no such thing as a “settling out” condition related to this type of compressor.  A reciprocating compressor has both suction and discharge valves that actuate like check valves - there is no back flow allowed through the cylinders.  A centrifugal machine (which is devoid of check valves) does suffer a “settle-out” pressure event which is a merging of the discharge and suction pressures when the machine is stopped.  If you believe there is a design for settling out the pressures in your application you should challenge the design with the engineering contractor who designed it that way.

[Anshu18]

Thanks for your elaborate response!

• My comment regarding treating recycle valve like antisurge valve was based on my experience with some consultants. They even build a logic making this valve open after every trip of the compressor. As you explained, this is not a correct method of designing. I did argue with them & challenge their design. I did it too many times. Hence, thought of putting it in this forum.

• I will defiinitely discuss regarding 'FO' recycle valve with the Compressor vendor who supplied it. I may not be able to disclose the names of the consultants who approved these P & Ids due to confidentiality clause. May I request you to elaborate the potential hazard it may pose to the compressor if the recycle valve is opening after every trip?

• I completely agree with you about settle-out condition applied to recip machines & have challenged the design.

[Art Montemayor]

Anshu18:

 

Your reluctance to tell us all the actual background for this query is understandable, taking into consideration a member’s reluctance to reveal certain personal information on the internet.  But the forum needs to know some of the basic reasons and some background in order to furnish a serious, accurate, and successful response to a need.

 

Based on what you now tell us, this thread takes on a more realistic and important role in helping not only you, also a lot of Forum readers and members.  Unfortunately, yours is not a rare case.  Today - unlike my time when engineering mentorship was still a custom - a lot of young engineers are being asked and requested to participate in projects and/or designs that may contain some inherent unsafe or hazardous procedures.  I have personally seen this developing throughout the latter part of my career and it is a definite threat to the practice of chemical engineering.  Our Forum, with its membership of some very experienced and field-tested engineers can help in avoiding what are called “bear traps” for young engineers - endorsing or participating in a faulty or unsafe design that can pose a potential threat to human lives or capital investment.  In either case, such a bear trap can trigger the damage to a chemical engineering career  - plus cause personal traumatic results.  This goes contrary to what should be the experience of a young engineer and has been the principle goal and reason for our Forums.  I write this paragraph as an alert for all young engineers who may be reading this.  Your reported actions are commendable and I will proceed to answer your specific questions, based on my professional experience and practice.  I request all our senior Forum members to jump into this thread with their valued comments.

 

The potential hazard posed to both personnel and machinery should the compressor’s recycle valve fail in open position after a compressor driver’s trip is a serious one should:

• The suction separator vessel and all associated piping, instruments, etc. are not protected by an appropriate safety relief device - rupture disc or relief valve - for the total, instantaneous flow rate of the recycle valve plus that of any continuing suction stream, depending on the process design and piping.
• If the discharge of the compressor is devoid of a successful operation of a discharge check valve downstream of the reciprocating compressor (which should not be the case), the total, high pressure gas volume down stream will have to be vented through the suction side safety device.  Even if a check valve is located in the compressor’s discharge, a correctly-led Hazop will identify this as a credible possibility for check valve failure and subsequent safety problems.
• If the recycle flow rate is not being cooled (as is sometimes the case), the safety discharge of the suction side safety device will be a hot one and pose another unsafe situation.
• If the gas being compressed is a combustible or toxic one, this is a practice that definitely goes against Best Practices and probably against local environmental and safety regulations.
• If the compressor is being started subsequently in automatic fashion, this will pose an overload condition for the compressor because of the higher density suction gas.
• Before allowing a potential over-pressure to enter the suction side of a multi-stage reciprocating compressor, the MAWP of the 1st stage cylinder should be carefully taken into consideration.  Many compressor manufacturers do not stamp the MAWP of their cylinders on each one - unless told to do so.  These MAWP values should be documented in compressor engineering files.

I hope this helps you out.  Never stop challenging another engineer when you detect a possible unsafe or hazardous situation.  A truly professional and serious engineer will never take a challenge as an insult to his/her capability or knowledge.  If they do, they don’t deserve to have a young, striving, and dedicated engineer working under them or with them.  A true professional engineering supervisor will value constructive and intelligent challenges as a means to improve projects and to increase the capability and value of his/her team.

[Bobby Strain]

"professional engineer"

 

In Texas, this is one who has a valid license from the state board to practice. I never bothered with the registration process in 41 years of practice. But this is today's view. There is not much emphasis on learning.

 

Bobby

[Art Montemayor]

I used the term “professional engineer” in this thread in the same sense as the title of this Forum - Industrial Professionals.  In other words, I meant a person who practices engineering as his profession - not a registered professional engineer.

 

Although registered as a Professional Engineer in the State of Texas at one time, I gave the practice up when I developed the perception that all the registration amounted to was nothing but a political ploy to force engineers to pay for a bureaucratic organization run by appointed individuals and a group of state employees.  This organization delivered nothing of value to engineers but still demanded cash yearly fees that were always increased on a timely basis.  I considered that practice a political exploitation of engineers - something that is not done to other professions such as bankers, financiers, and politicians.  If employers of engineers were legally forced to pay for the registration fees required of engineers it would be a different matter; but engineers are expected to shoulder the fees on their own as a “license” to practice engineering.  This is just another selective tax imposed on engineers without their having an opinion on the matter.