White paper: Originally published in ISA InTech Magazine Page 1

Improving Differential Pressure Diaphragm Seal System Performance and Installed Cost
Tuned-Systems; Deliver the Best Practice Diaphragm Seal Installation To Compensate Errors Caused by Temperature ariations!
Tuned-Systems are the est practice for differential pressure diaphragm seal system configurations. Tuned-Systems contrast significantly to traditional symmetrical configurations! achieving the lowest temperature-induced errors" est time response" and lowest installed cost when measuring level in closed vessels. Differential pressure seal systems have traditionally een specified with identical capillary lengths and seal configurations on oth the high and low pressure process connection. Specifying symmetrical systems was once elieved to achieve est total system performance. #ctually" the asymmetry of TunedSystems compensates for temperature-induced errors. The following discussion will e$plore how a diaphragm seal system wor%s and prove the theory ehind Tuned-Systems. What are Tuned-Systems? Tuned-Systems are an asymmetric configuration of a differential pressure diaphragm seal system. The simplest form of a Tuned-System directly mounts the diaphragm seal to the high pressure process connection. &limination of the e$cess high pressure capillary immediately improves response time" and performance" while reducing installed cost. Total system error is compensated y leveraging diaphragm induced temperature errors against head effect temperature errors. 'urther performance improvements are achieved y ad(usting configuration varia les as detailed elow. Installed cost is reduced y eliminating the e$cess high pressure capillary. )ecause the transmitter is direct mounted to the vessel" neither mounting stand nor mounting rac%et are re*uired for further reduce installed cost. How Do Seal Systems Work? Diaphragm seal systems respond to changes in oth process pressures as the level changes" and in static pressure over the li*uid. These variations in pressure are transmitted through an oil- filled capillary to a differential pressure transmitter-sensor. The capillaries and seals are filled with incompressi le oil compati le with the process temperature" pressure" and media composition. The transmitter is commonly mounted at grade" or in close pro$imity to the high-pressure process connection. 'or applications under vacuum" the transmitter is mounted elow the high-pressure connection to reduce vacuum effects on the transmitter fill fluid. The minimum capillary length is dictated y the distance etween the mounting position of the transmitter and the low-pressure connection. #ll cavities within the assem ly are oil-filled including the diaphragm" capillary" and transmitter ody. #lthough manufacturing techni*ues help ensure a high-*uality fill" temperature-induced errors are inherent to diaphragm seal systems. Diaphragm Seal System Elements: )ecause the transport mechanism of a diaphragm seal system is the fill fluid" it is important to understand the fill fluid physical characteristics. The fill fluid oil is an incompressi le fluid and a change in pressure within the process is directly translated to the transmitter-sensor. Proper fill fluid and assem ly preparation are critical to achieve a high *uality filled system. Proper preparation re*uires removing all gases from oth the fill fluid and the un-filled transmitter-seal assem ly. # successful fill process prevents am ient air from entering the assem ly. #ir or other gases in the system are compressi le fluids and cause erratic transmitter output shifts. &ach fill fluid has it own uni*ue physical characteristics and play the largest role in total system performance. The physical characteristics include+ viscosity" coefficient of thermal e$pansion" and specific gravity.

www.rosemount.com

White paper: Originally published in ISA InTech Magazine Page , 'ill fluid viscosity is a measure of velocity flow rate and dictates the response time of the diaphragm seal system. # temperature increase causes the fill fluid to ecome less viscous and yields a faster response time" while a decrease in temperature slows the response time. Capillary inside diameter and length also impact system response time. # small inside diameter restricts the fill fluid flow causing slower response time. Capillary length relates to the time for a change in pressure to reach the transmitter-sensor.

Low Side Seal

Distance between process taps

High Side Sensor Figure 1. Tuned-System used to measure level in a closed vessel.

Direct-mount High Side Seal

Low side Sensor

Seal Temperature Effect Error: The fill fluid coefficient of thermal e$pansion is the rate a fill fluid volume e$pands or contracts in response to temperature changes. # larger coefficient of thermal e$pansion factor e*uates to higher response rate to the change. The fill fluid volume e$pands to an increase in temperature and contracts to a decrease in temperature. The larger fill fluid volume within the seal system" the greater the total volume e$pansion or contraction. System volume is highly dependent on capillary inside diameter" capillary length" and seal cavity volume. )ecause a diaphragm seal assem ly is a closed system" the e$panding fill fluid volume presses against the seal diaphragm. The seal diaphragm restricts the e$pansion causing a ac%-pressure on the fill fluid. The diaphragm ac%-pressure is highly dependent on diaphragm stiffness" or spring rate. Diaphragm spring rate is a function of the diaphragm pattern" thic%ness" material modulus of elasticity" and diameter. # more fle$i le diaphragm with a high spring rate minimi-es the ac%-pressure e$erted on the transmittersensor. The variations in ac%-pressure e$erted on the transmitter-sensor are commonly referred to as Seal Temperature &ffect &rror. Head Temperature Effect Error: The fill fluid specific gravity is the ratio of the fill fluid density compared to the density of water. #s temperature changes" the specific gravity of the fill fluid changes! an increase in temperature lowers the specific gravity while a decrease in temperature increases the specific gravity. The seal elevation e$erts pressure on the differential pressure transmitter-sensor and is referred to as .ead Pressure. The fill fluid specific gravity" com ined with seal elevation" are the primary varia les re*uired to determine the head pressure /.ead Pressure 0 Specific 1ravity $ .eight2. The initial head pressure can e calculated and is -eroed out of the differential pressure system during cali ration and commissioning. .owever" variations in temperature cause changes in fill fluid specific gravity and su se*uent variations in head pressure from original commissioning. These variations in head pressure are commonly referred to as .ead Temperature &ffect &rror. Harnessing System Elements:

www.rosemount.com

White paper: Originally published in ISA InTech Magazine Page 3 Traditional systems apply e*uivalent seals and capillary lengths to either side of the differential pressure transmitter-sensor" this creates identical /or nearly identical2 pressure changes due to seal temperature effect error. )ecause the pressure changes are e*ual" and on opposite sides of the transmitter-sensor" the net seal temperature effect error is cancelled. Therefore" the net error is completely a function of the head temperature effect error. The symmetric design was considered to deliver the est total system performance. .owever" traditional symmetric configurations ignore the largest source of error! head temperature effect error does not reali-e the opportunity to reduce total system error. Tuned-Systems harness the physical characteristics of the fill fluid" and the mechanical design features of the diaphragm seal system to deliver est total system performance. )oth head and seal temperature effect errors occur simultaneously within the system in response to temperature changes. The differential transmitter-sensor cannot differentiate the error type. Therefore" the Total System &rror is the sum of the seal temperature effect plus the head temperature error" and represents the error transmitted to the differential pressure transmitter-sensor. It can e concluded that differential pressure diaphragm seals must e viewed as a system to effectively compensate for total system errors induced y temperature changes. Consider a Tuned-System that eliminates e$cess high-pressure capillary" and e$periences an increase in temperature from original -ero. The head temperature error causes a net positive error and is identical in magnitude to a symmetrical system installation" under the same temperature variation conditions. .owever" the Tuned-System seal temperature effect errors yield a net negative error. The fill fluid volume on the high-pressure side of the system has less volume compared to the low- pressure side of the system. Therefore" the volumetric displacement" and resulting diaphragm ac%-pressure" is dominated y the low pressure side" thus proving total system error for Tuned-Systems is less than traditional systems. The detailed mathematical proof %ead Pressure Error on Transmitter Sensor : is outlined elow.
Total System Error &TSE': Minimi ing Total System Errors: Total systemerrors seal error is eliminated when headand pressure To ta%e the theory of Tuned-Systems a step further" total system can e compensated" in error e*uals some cases eliminated. 4inimi-ing total system errorsseal re*uires creating temperature effectperformance errors temperature error.seal Therefore" Tuned-Systems 9ow-side Sensor Pressure /.,2+ ., in 0 /D T2 , $ Sg $ & $ to that are .ead e*ual" and opposite magnitude head temperature effect errors. addition to reducing highimprovement is proven y In solving the Total System &rror e*uation .ead Pressure 0 .capillary 1 - ., pressure fill fluid volume" the following can e varied elow. for additional performance improvements! decrease high-pressure diaphragm stiffness" increase low-pressure fill fluid volume" increase fill fluid e$pansion and5or increase low-pressure diaphragm stiffness. Due to the num er of varia les" .ead Pressure 0 /Dcoefficient" T2 - /D, $ Sg $ & $ T2 1 $ Sg $ & $ TS& 0 -D $ Sg $ & $ T - { 71 - 7, }&T achieving a fully compensated differential pressure seal system re*uires an automated software tool" 0 /D1 - D,2 /Sg $ & $ T2 6*uic%ly 6, and easily 1 such as 6osemount Instrument Tool%it. Instrument Tool%it has the capa ility to 'actor out li%e varia les and *ualitatively compare results of the calculate numerous potential compensated seal systems for any given application condition. .igh-side Sensor .ead Pressure /.12+ .10 /D1 $ Sg $ & $ T2 D, : D1 therefore" D1 - D, 0 -D traditional system scenario versus Tuned-System scenarios+

!onclusion: .ead Pressure 0 - D $ Sg $ & $ T Temperature-induced errors are inherent to differential TS& pressure diaphragm seal systems. The errors are 0 - DSg - { 71 - 7 , }&T caused primarily y the fill fluid physical characteristics responding to a 6 change in temperature. Total 6 1 , system error is also a function of the distance etween vessel process connections and the mechanical Seal Pressure Error on Transmitter Sensor: design of the diaphragm seal system. Total system error is uncompensated in traditional symmetrical Symmetrical System (ase: system configurations. .igh-Pressure Seal+

The asymmetry of Tuned-Systems compensates total system error y reducing the high pressure side 71 0 e 7,achieved and 61 0 6 then" TS& 0 - DSg&spring T capillary volume. 'urther performance improvements can y, varying diaphragm rate" Pressure 0 71and $ & $fill T system fillChange volumes" fluid type. The reduced volume and configuration variations compensate for 61 Assumed #ariables: changes in fill fluid specific gravity" there y providing improved performance in pressure #ll of the total system error is differential caused y head pressure error acting on 9ow-Pressure Seal+ seal applications. diaphragm D1 0 Distance etween high-pressure connection and transmitter the transmitter-sensor. 4ichelle #. .ursh" 4ar%eting 4anager" &merson 7 Process 4anagement 1 0 'ill fluid volume within high-pressure capillary-seal assem ly Tuned)Systems (ase:
Pressure Change 0 7, $ & $ T Mathematic Proof: 6
,

7olumetric Displacement 0 71 $ & $ T

If"

7olumetric Displacement 0 7, $ & $ T

D, 0 Distance etween low-pressure connection and transmitter

Total Seal Pressure &rror+

E$! Sg$! "$" #$

Seal Pressure &rror 0 71 $ & $ T - 7, $ & $ T 61 6,

www.rosemount.com

7, 0 'illIf" fluid volume within low-pressure capillary-seal assem ly T 0 6elative -ero 71 ; 7change or in temperature 71 ; 7, from or original 71 0 7, , Sg 0 'ill 'luid Specific 1ravity! #ssume Sg 1 0 Sg, 61 0 6 , 61 : 6 , 61 : 6 , & 0 'ill 'luid Coefficient of Thermal &$pansion! #ssume &1 0 &, 61 0 Spring rate of seal diaphragm on high-pressure connection. Then" Tuned-Systems are proven. #ll scenarios have improved total 6, 0 Spring rate of seal diaphragm on low-pressure connection. system error. Instrument Tool%it provides further proof and *uantitatively solves e*uation. 8ote+ Spring rate is the fill fluid volume change divided y corresponding change in diaphragm ac%-pressure. Therefore" a diaphragm with a larger 6value is more fle$i le.

0 { 71 - 7, }&T 61 6, E ! Sg ! " ! #

White paper: Originally published in ISA InTech Magazine Page < In summary" apply the following asic rules to prove additional compensation of diaphragm seal systems+ #pply an asymmetrical system configuration 6educe the high pressure side fill fluid volume 4inimi-e the high pressure diaphragm stiffness /increase spring rate2 Increase the low pressure side fill fluid volume Increase the low pressure diaphragm stiffness /decrease spring rate2 =se 6osemount Instrument Tool%it to optimi-e total system performance

www.rosemount.com