## Masoneilan Control Valve Sizing

This is the orignal Dresser Masoneilan Control Valve Handbook OZ1000.

This control valve handbook on control valve sizing is based on the use of nomenclature and sizing equations from ISA Standard S75.01 and IEC Standard 534-2. Additional explanations and supportive information are provided beyond the content of the standards.

The sizing equations are based on equations for predicting the flow of compressible and incompressible fluids through control valves. The equations are not intended for use when dense slurries, dry solids or non-Newtonian liquids are encountered.

Original equations and methods developed by Dresser Masoneilan are included for two-phase flow, multistage flow, and supercritical fluids. Values of numerical factors are included for commonly encountered systems of units. These are United States customary units and metric units for both kilopascal and bar usage.

The principal use of the equations is to aid in the selection of an appropriate valve size for a specific application. In this procedure, the numbers in the equations consist of values for the fluid and flow conditions and known values for the selected valve at rated opening. With these factors in the equation, the unknown (or product of the unknowns, e.g., Fp Cv) can be computed.

Although these computed numbers are often suitable for selecting a valve from a series of discrete sizes, they do not represent a true operating condition. Some of the factors are for the valve at rated travel, while others relating to the operating conditions are for the partially open valve. Once a valve size has been selected, the remaining unknowns, such as Fp, can be computed and a judgement can be made as to whether the valve size is adequate. It is not usually necessary to carry the calculations further to predict the exact opening.

To do this, all the pertinent sizing factors must be known at fractional valve openings. A computer sizing program having this information in a database can perform this task.

A comprehensive set of Cv factor tables for all Dresser Masoneilan valves is included in one handy reference. This tabulation supersedes all previous listings since many changes were requiered to bring all ratings in line with actual test data taking into account the use of the critical flow factor, Cf.

The formulas in the early chapters are set up in sections to simplify manual calculation for the more common control valve sizing problems.

### Why Dresser Masoneilan Valves Are Calculated?

Control valve sizes are calculated for two reasons:

1. To keep the installation expense down by using as small valve as possible.
2. To select a size which is large enough to handle the maximum required capacity, yet small enough to close down to the minimum required capacity without seating.

While the dollar cost is of great importance to a user, the Instrumentation Engineers, as automatic control specialists, are primarily interested in valve size from the engineering point of view as summarized in (2) above.

There are good reasons why “equal percentage characteristic” control valves (i, e. the 50:1 rangeability) usually recommended by the Instrumentation Engineers are considered the standard against which to compare the flow characteristics of all other styles of inner valves.

The equal percentage curve provides a characteristic distorted less by line pressure losses which rob the valve of pressure drop.

### How Dresser Masoneilan Valves are Sized?

A brief outline of the usual way in which control valve sizes are determined will be of interest to many readers. The numerous details will not be covered here; separate articles treat the subject thoroughly for those who need specific help in actual valve sizing.

All calculations do employ the same basic procedures, however,  and can be set forth in an outline of few words.

An almost universal approach to valve sizing today is through the Cv factor method. The capacity rating,  Cv, (sometimes called flow coefficient) is the expression of a valve’s capability of handling flow as defined in its rate of discharge of water under a fixed pressure drop condition of  1 psi.

All kinds of flowing fluids at various pressures and temperatures can be reduced by computation methods to terms of a required Cv factor for a particular installation, Once this has been done,  it is a simple matter to select the appropriate size of control valve from a list of Cv’s for the style of valve wanted.

There are three basic formulas published in the literature from which a value of Cv factor can be obtained:

• one equation is used for liquids
• one is for use with gases
• and one is for steam and other vapors

All stem from the velocity head component in Bernoulli’s theorem,  that is,  v^2  =  2gh.

The three equations, when used with a thorough understanding  of their limitations, i.e. :

• how the process data  must  be  adjusted  for  varying  temperatures
• densities

, and so forth,  provide  a means  of  accurately  determining the Cv  requirement   for  all  flowing  materials.    No nomograph,  no special slide rule or other short cut means can do anything for the valve calculator that these three equations will not do.

In fact, the three equations will do much that short cut means cannot accomplish.   They must  be  resorted  to from time to time to handle calculations that cannot be made correctly otherwise.

### How does one make use of the computed Cv factor?

There are two methods commonly employed,  either of which is acceptable:

1. The first is to use an estimated maximum flow rate with a carefully estimated (see next section) pressure  drop  at this maximum  rate. This method  produces  a maximum required  Cv factor from which the valve size can be selected  from a list of  Cv‘s. The premise  is that  the normal  rate to be handled  is easily provided for by a partly opened valve  along the lines discussed  in the previous  section.
2. A second approach calls for a calculation of the required Cv factor for normal flow rates with  carefully estimated pressure drops at these rates.   This gives the Cv factor requirement for the normally operating valve.  By increasing this value by a reasonable percentage factor the maximum Cv factor is established from which the size is selected.
Characteristic curves of various inner valve styles are often consulted in the second method. By experience, some engineers feel confident in placing their normal operating strokes at a certain fraction of full stem travel. They seem satisfied if the listed Cv factor is something in excess of their maximum Cv as increased from the computed normal. Such a procedure can tend to produce some oversized valve installations, unless much care is exercised.

While either procedure is permissible,  it must be borne in mind that the maximum Cv factor calculation method is more direct and invariably leads to a valve size ample for the process conditions at hand.

## The Question of Pressure Drop

The importance of using carefully estimated, realistic values of pressure drop in valve size computations cannot be overemphasized.  One must constantly remember that pressure drop which is to occur in the control valve plays almost as prominent a part as flow rate. In contrast with the more readily established estimates for rates of flow,  selection of representative pressure differentials is not always easily made.

The function of a Dresser Masoneilan valve is to adjust its area to provide the rate of flow needed for maintaining a set control point. Not only does valve port area depend on the rate to be handled, but upon the excess pressure in the process system –  all or a portion of which will appear across the valve ports. Just how much excess pressure  occurs depends on process equipment  and piping details.

The process designers must necessarily work with many variables and approximations. If the control valve were not installed, the varying pressures would be absorbed in other parts of the equipment and the resulting flow would be excessive, unpredictable and rampant in nature. To remedy such chaotic conditions, automatic controllers are used and the control valve does the chore.

### Is it not logical, then, to study the proposed operating conditions with the object of selecting realistic pressure drops from careful engineering surveys of all factors that can affect the proposed valve?

Dresser Masoneilan valves utilize the surplus pressures even though the popular misconception is that valves create the drop.   In performing the valving function,  the restrictive ports control the flow rate so that the excess pressures become centralized in one place – at the ports of the valve.

Many control valve specifications are written which give questionable pressure drops.  There are probably several reasons for including such arbitrary numbers; but it is obvious that artificial data cannot be used in valve sizing.  The seriousness of computing Cv factor values on the customary “5 psi drop available”is shown in the following table:

For example,   if  the size calculation is made on a 5 psi drop,  but the valve must  actually perform with  a 20 psi drop,  the Cv calculation is high by 100 percent and  an oversized  control valve  results.

In arriving at a suitable value for pressure drop, there is no substitute  for good  engineering  judgment, conscientiousness  and  experience. It is recognized  that, in some valve  applications,   all the pressure  loss in a system will not show up across the valve ports. In such cases, the engineering judgment called  for  in estimating what drops will occur at the valve may require careful analysis of  the system hydraulics.

While such conditions are relatively few, they certainly demand more careful consideration than a casual specification of “5 psi drop available” as a means of side-stepping the issue. There are no fixed rules that always apply, but a critical study of all the process pressure conditions should lead to reasonably representative values for use in computing Cv.

### Definition of Flow Coefficient Cv

The accepted method of valve sizing in the U.S.A. is the Cv factor approach. Cv is a capacity rating coefficient which is defined as the number of U.S. gpm of 60°F water which will flow through a valve at a specified opening with a pressure drop of 1 psi across the valve.

In Europe, the Kv and Av coefficients are more widely used. Econom­ics and control are the two principal reasons for sizing control valves.

#### Economics

If a valve is too small, it will not pass the required flow arid will have to be discarded and re­placed by a larger, properly sized valve. Similarly, if the valve is too large, it will obviously pass the required flow, but it will be more expensive than a properly sized, smaller valve.

#### Control

An undersized valve will never deliver the full flow rate, thus  it will sharply  narrow  the controllable flow range. An oversized valve will be throttling near the closed position, and the full control range of the valve will not be utilized.

When the plug throttles very close to the seat, high fluid velocities occur which can cause erosive damage. The ideal valve is one that will function between 40 and 70% of its operating range, going neither wide open un­der maximum flow rates, nor closing down too near its seated position under minimum conditions.

The Dresser Masoneilan Cv factor tables in this post will allow the user to select the ideal valve for the re­quired  application.

### General

When the Cv, or capacity of a valve, versus the percent valve stroke from 0 to 100% is plotted, a curve is generated. The shape of  this curve can be varied by varying the valve capacity or gain at some predetermined rate.

The curve generated is called the valve de­ sign characteristic.

#### Equal Percentage Characteristic

For an equal in­crement of valve stem position, an equal percent change in valve capacity will occur.

This means that the same percent increase will occur between a 30 and 40% valve stem position as between an 80 and 90% valve stem position.

For example,  for a nominal valve with a rated Cv factor of 15.0 and having an equal percentage characteristic, we find the following:

Only in theory is the gain exactly the same for equal increments of valve stem position. In practice, the gain can and does vary from the theoretical.

#### Linear Characteristic

For equal increments of valve stem position, there is a corresponding equal increment of flow. This means that flow is proportional to valve stem position and that the same change in valve capac­ity will occur between a 30 and 40% valve stem posi­tion as belween a 70 and 80% valve stem position.

For example, referring a valve wilt a rated Cv of 34.0, and having a linear characleristic, we find lhe following:

Although in this example the flow increments are identical, in practice, these values can and do vary from the theoretical.

## Control Valve Selection Guide Characteristic

The control valve selection with the proper operating characteristics is one of the most important phases of designing a control loop,   It is not enough to assume that a wide proportional band plus integral con­troller will overcome a serious mismatch between the process and the valve characteristic.  An exact match would usually require a valve custom designed for the process.  However, one or the other of two standard characteristics (linear or equal percentage) is practically always suitable for any given process.  This is important because a serious mismatch might cause the system to be unstable and difficult to control effectively.

In actual practice,  a valve has two characteristics:

1. One is the design or inherent characteristic , determined by laboratory test and defined as the relationship between flow and stroke with constant pressure drop.
2. The other, the resultant characteristic, is more significant.This is the relationship between flow and stroke when the valve is subjected to the pressure conditions of the process.

Cv factor is an expression for liquid flow at a constant pressure drop,  The graphic display of flow versus lift shows how the design or inherent characteristic is changed by variations in pressure drop.

This occurs as the process changes from a condition where most of the pressure drop takes place at the control valve (usually at low flow rates) to a condition where most of the pres­sure drop is generally distributed through the rest of the system (usually at high flow rates).

This variation in where most of the total drop takes place is one of the most important aspects in choosing the proper valve characteristic for a given process.

There are two other very important considerations in choosing  the proper  valve characteristic. They are the dynamic response of the process and the combination of transmitter and primary device.

To exactly determine the dynamic response of a process requires a complete dynamic analysis for each control loop. This is difficult and usually impractical. It is possible to establish general guidelines based upon the extensive experience in field startups and troubleshooting.

The following paragraphs are discussions of normal control applications encountered in industrial instrurmentation. Proper valve characteristic recommendations are included :

### FLOW RATE CONTROL

One of the two types of flow measurement signals

• linear with differential pressure or
• linear with flow rate

are incorporated in flow rate control.   These two types of measurement require separate treatment since they seriously affect dynamic response.

Case 1

1. Measurement  signal:   linear with flow  rate
2. Maximum  to minimum  valve pressure drop with varying flow rates: Greater than 2 1/2 to 1
3. Valve pressure drop at maximum  flow:Less than 40 percent of system drop including valve

Recommendation:   Equal percentage valve

Case 2

1. Measurement  signal:   linear with flow  rate
2. Maximum  to minimum  valve pressure drop with varying flow rates: Less than 2 1/2 to 1
3. Valve pressure drop at maximum  flow: More than 40 percent of system drop including valve

Recommendation:   Linear valve

Case 3

1. Measurement  signal:  Linear  with  differential pressure
2. Maximum  to minimum  valve pressure drop with varying flow rates: Greater than 5 to 1
3. Valve pressure drop at maximum  flow: Less than 20 percent of system drop including valve

Recommendation:   Equal percentage valve

Case 4

1. Measurement  signal:  Linear  with  differential pressure
2. Maximum  to minimum  valve pressure drop with varying flow rates: Less than 5 to 1
3. Valve pressure drop at maximum  flow: More than 20 percent of system drop including valve

Recommendation:   Linear valve

### PRESSURE  CONTROL

Pressure control is subdivided into two classes de­ pending upon the pressure conditions of the system.  When the control valve is to maintain a constant pressure drop to satisfy the control requirements,  a linear valve should be used.   When the valve is to operate with a varying pressure drop of over 2 to 1 to satisfy the control require­ments of the hydraulic system, an equal percentage valve should be used.

### TEMPERATURE CONTROL

In just about every case, temperature control re­quires an equal percentage valve for optimum control. Foxboro designed the first equal percentage valve specifically for temperature control when adequate con­trol could not be maintained by existing valves.   A tem­perature control system should be such that the ratio of maximum to minimum valve pressure drop is less than 2 to 1.

### LIQUID LEVEL CONTROL

Liquid level control is best satisfied by a valve with an equal percentage characteristic when the ratio of maximum to minimum valve pressure drop is greater than 2 1/2 to 1. When this ratio is less than 2 1/2 to 1, a linear valve characteristic gives better performance.

### pH CONTROL

Concentration, as measured by analysis such as pH, is best controlled by a valve with an equal percentage characteristic When the ratio of maximum to minimum valve pressure drop at full flow is greater than 2 to 1. When this ratio is less than 2 to 1, a linear valve charac­teristic gives better performance.

### GENERAL RULES OF APPLICATION

For stability over the entire control range,  most control loops require flow to be manipulated in uniform proportion to controller output. Frictional losses through pipes and fittings increase and pump output pressures decrease with increasing  flow. Therefore,  the available pressure drop at the control valve usually diminishes as the flow rate increases. To maintain  the desired uniform proponionality,   an  equal percentage characteristic is often required  in these varying pressure  drop applications. A linear  characteristic  is preferable on all but tempera­ture control applications when a nearly constant pressure drop exists at the control valve.

A misapplied equal percentage valve characteristic results in increasing valve gain at high flow rates.  It can cause instability unless the controller proportional gain is adjusted at the high flow rate. The control loop then tends to be overdamped at the low flow rates with correspondingly sluggish response.

A misapplied linear valve exhibits the opposite effect.  A controller properly adjusted at the high flow rate has too narrow a proportional band setting for stability at low flows.  The proportional gain has to be adjusted at the low flow rates,  and the control loop is then overdamped and sluggish at higher  flow rates.

The above recommendations are only as good as the hydraulic analysis of the pressure conditions in the system.  Maximum pressure drop can usually be deter­mined fairly exactly since it is close to the shutoff pressure.  Minimum valve pressure drop at full flow is occasionally estimated without thorough analysis, and such estimates are often much too low.   Such a low minimum drop figure erroneously favors the selection of an equal percentage valve.

The end result is the selection of an oversized equal percentage valve that operates in the low portions of lift even at maximum flow.  Consequently, rangeability and response at low flow conditions are lost.   It is important to use realistic pressure drop data for both valve sizing and characteristic selection.

This article has covered the basic aspects of Dresser Masoneilan Valves Sizing  concepts.

## References

Feel free to download the Dresser Masoneilan control valve handbook in pdf format.

MASONEILAN (1998) Control Valve Sizing Handbook
Tagged on: