Cen­trifu­gal Pumps Char­ac­ter­is­tics Curves, Sys­tem Re­sis­tance, Flow Prob­lems & So­lu­tions

Cen­trifu­gal Pumps are the in­dus­try work­horses which are very re­li­able and per­form well with Mean Time be­tween Fail­ures ex­ceed­ing 4 years or more. How­ever, even ex­pe­ri­enced en­gi­neers wrongly di­ag­nose flow re­lated prob­lems and pro­pose ex­pen­sive and in­ef­fect

Chemical Industry Digest - - What’s In? - S Raghava Chari

– S Raghava Chari, In­de­pen­dent Con­sul­tant While cen­trifu­gal pumps and sys­tems are the work­horses of the in­dus­try, widely used, due to im­proper un­der­stand­ing of the prob­lems that come up, many a times ex­pen­sive and in­ef­fec­tive so­lu­tions are proposed even by ex­pe­ri­enced en­gi­neers. The author proves this through many case stud­ies and so­lu­tions.

Char­ac­ter­is­tic Curves

Man­u­fac­tur­ers test each model of their cen­trifu­gal pumps (CCP) and fur­nish char­ac­ter­is­tic curves (CC) sheets. CC is pri­mar­ily a graph show­ing the re­la­tion­ship be­tween the flow rates - usual sym­bol (US) Q - usu­ally in m3/H and the dis­charge pres­sure of­ten men­tioned as the ‘dy­namic head’ – (‘h’) in me­ters of pumped liq­uid (PL) col­umn. In ad­di­tion, the graph de­picts the fol­low­ing re­la­tion­ships also:

1. Q Vs kW in­put to the pump

2. Q Vs Ef­fi­ciency

3. The above 2 for dif­fer­ent im­pel­lor dia

4. The above for a given dia im­pel­lor at dif­fer­ent RPMs(N)

5. Net Pos­i­tive Suc­tion Head (NPSH)

Pump se­lec­tors and man­u­fac­tur­ers se­lect pumps con­sid­er­ing the CC and pur­chasers given Sys­tem Re­sis­tance (SR) curves – ex­plained later - in ad­di­tion to the nu­mer­ous other spec­i­fi­ca­tions that go in pump se­lec­tion. A man­u­fac­turer would han­dle two cus­tomer or­ders ac­cord­ing ta­ble 1:

Fig-1 shows the char­ac­ter­is­tic curves for the ra­dial flow CCP and ta­ble 1 one of its use. It tells fur­ther that the Q Vs h curve for a ra­dial flow pump is rel­a­tively flat and that the head de­creases grad­u­ally as the flow in­creases. Note that the brake horsepower in­creases grad­u­ally over the flow range with the max­i­mum nor­mally near Q=QR the rated flow.

Mixed flow cen­trifu­gal pumps and ax­ial flow or pro­pel­ler pumps have con­sid­er­ably dif­fer­ent char­ac­ter­is­tics. Ra­dial flow CCP shut-off head i.e. the closed dis­charge valve head is usu­ally 120-150% of the de­sign head.

Sys­tem Re­sis­tance: Sys­tem Re­sis­tance (SR) is the re­sis­tance to flow due to pipe fric­tion, pres­sure loss across the equip­ment e.g. heat ex­chang­ers, the suc­tion

lift and the pres­sure of the des­ti­na­tion unit e.g. boiler drum and other causes. Suc­tion lift and des­ti­na­tion pres­sures and sim­i­lar items con­sti­tute con­stant re­sis­tances and flow de­pen­dent re­sis­tances e.g. pipe fric­tion vari­able re­sis­tance. Flow de­pen­dent re­sis­tances are usu­ally pro­por­tional to Q2. It is cus­tom­ary to ex­press SR in terms of me­ters of PL col­umn. Ta­ble 2 shows the SR of the var­i­ous items of the hy­draulic cir­cuit.

Re­view of Few Im­por­tant Pump oper­at­ing Points: Process plants’ and other im­por­tant crit­i­cal pumps e.g. boiler feed water pumps come with a hot standby as in fig 3. Let us re­view few im­por­tant pump oper­at­ing points with the help of fig 3:

Pump Start

up: Op­er­a­tors start­ing the pump A af­ter a turn­around, close pump A dis­charge block valve BV, open suc­tion valve, open cas­ing fill valve and fill the pump cas­ing with water known as prim­ing. Close the prim­ing valve and switch on the motor. The au­to­matic re­cir­cu­lat­ing con­troller (ARC) does the fol­low­ing given func­tions:

1. Check Valve

2. Min flow by­pass valve

3. Pres­sure drop dis­si­pa­ter to prevent high ΔP ero­sion of let­down in­ter­nals

4. Pro­vides warm up flow to keep by­pass line warm

5. Re­fer to the author’s ar­ti­cle ‘Spe­cialty Valves Save’ for de­tailed in­for­ma­tion on ARC and other types of min flow by­pass sys­tems.‘

Pump A ARC by­passes the min flow large CFPs re­quire and Pump B ARC pre­vents B’s re­verse ro­ta­tion. Us­ing dis­charge BV and LCV he es­tab­lishes flow re­quired and puts LCV on au­to­matic. Opens dis­charge block valve fully. Leaves the suc­tion and dis­charge valves of Pump B open. About 200 Kg / H of warm up flow across pump B ARC keeps the pump and pipe­lines warm and ready for pump B in­stant start up.

Trans­fer to Pump B: Op­er­a­tor or auto cut in re­lay starts pump B motor. The LCV will re­store the flows to orig­i­nal val­ues dur­ing op­er­a­tor clos­ing pump A dis­charge block valve. Once pump comes to zero RPM, he closes the suc­tion block valve, drains, and hands over to main­te­nance if re­quired. Pump A ARC pre­vents its re­verse ro­ta­tion even if BV is open.

The above de­scribed pump start up pro­ce­dure and leav­ing the dis­charge block valve open were fol­lowed in two fer­til­izer plants where the author worked. This offers the fol­low­ing ad­van­tages:

1. Avoids the la­bor in­ten­sive task of rush close/open of large block valves. E.g. open­ing a 16” 600# valve in­volves high torque 20 Nos. of hand wheel ro­ta­tions of around 5-10 minutes du­ra­tion

2. Avoids de­layed pump trans­fer and even process

shut downs be­cause of stuck dis­charge BV mak­ing pump trans­fer im­pos­si­ble.

Still the author could not con­vince re­fin­ery en­gi­neers in western In­dia to fol­low this pro­ce­dure. The author would be happy to hear the read­ers’ views.

Min Safe Flow for CFP: CFP CC shows that the pump con­sumes around 20% rated power un­der closed dis­charge pump run. The power churns the pump fill and dis­si­pates enor­mous en­ergy e.g. 20kW for a 100 kW rated pump. This enor­mous en­ergy dis­si­pates as heat and va­por­izes the small vol­ume PL fill. The lu­bri­ca­tion of var­i­ous mat­ing parts and ro­tor buoyancy sup­port dis­ap­pears re­sult­ing in se­vere rub dam­ages. Hence, never run large CFPs with blocked dis­charge for more than ten minutes. This is the rea­son for large CFPs com­ing with ARC or some other min flow by­pass ar­range­ment.

Case Study 1 shows the im­por­tance plant de­sign­ers and CFP mak­ers as­sign to min flow by­pass sys­tem to pro­tect large pumps.

Case Study 1: Fig 3 shows Tur­bine Driven BFW Pump A feed­ing 5 boil­ers with a motor driven hot standby. Af­ter around 3-years run, steam pro­duc­tion dropped by 15% and the hot standby started rolling back­wards!

Op­er­a­tion en­gi­neer started that too and hap­pily re­ported re­sum­ing 100% steam and thus 100% am­mo­nia pro­duc­tion!

The trou­bleshoot­ing process en­gi­neers hastily con­cluded that worn CFP in­ter­nals due to the 3 years of con­tin­u­ous runs as the cause and in­sisted on over­haul­ing the pump.

No shift from the near zero shaft vi­bra­tions made the author to elim­i­nate other causes, be­fore open­ing the crit­i­cal pump. Worn in­ter­nals usu­ally lead to ex­ces­sive ro­tor vi­bra­tions. By pump cir­cuit study, the author found that in ad­di­tion to ARC, the pump has come with man­ual min flow by­pass also (fig 3). No warm up by­pass across the nor­mally closed (NC) by­pass valve puz­zled the author. By draw­ing care­ful re-study, the author found that, the valve ven­dor had drilled a 1.5 mm dia hole in the gate to by­pass ≈200 kg/h hot water. Voila, here is the cul­prit! The author con­cluded that the high ΔP hot water let­down – 120 to 1.3 bars – ero­sion of the 1.5 mm dia hole by­passes sig­nif­i­cant quan­ti­ties of BFW and starves the 5 boil­ers. His in­sis­tence to lift the gate valve bon­net tak­ing a 1 day SD paid well. Let­down had eroded the hole to ≈25 mm and waste­fully by­passed es­ti­mated at 60 T/h BFW.

The author got the hole plugged by strength weld­ing a disc at the hole and grind­ing the weld flush. In ad­di­tion he got a ¾” by­pass across the gate valve with mul­ti­ple ori­fices re­stric­tion to by­pass 200 kg/h warm up flow. Boil­ers re­sumed steam pro­duc­tion within 8 hours. The LCVs as­sumed nor­mal near 60% open. Boil­ers did not get enough BFW even at 100% open LCVs, be­fore the by­pass gate re­pair.

Thus the author’s de­tailed cir­cuit study and log­i­cal think­ing averted an ex­pen­sive pump over­haul and few days lost pro­duc­tion.

A urea plant re­placed the worn out al­ways run­ning CFP with a new one. The new pump did not de­liver the rated quan­tity and hence the hot standby be­came the ‘al­ways run­ning’. The trou­bleshoot­ers came up with no so­lu­tions.

An in­quis­i­tive newly joined grad­u­ate en­gi­neer asked, “Why a dis­charge to suc­tion by­pass, in ad­di­tion to the min flow by­pass re­stric­tion? That opened the eyes of all! The erec­tor had in­cor­rectly in­stalled the drain valves as in fig 4b in­stead of as in 4a to save a valve - an oper­at­ing en­gi­neer sug­ges­tion!

Case Study 3, an­other low through­put prob­lem: A pi­o­neer cen­trifu­gal com­pres­sor tech­nol­ogy based fer­til­izer plant, suf­fered grad­ual ca­pac­ity fall,which went un­no­ticed. Fig 2 shows the hy­draulic cir­cuit. 24 months af­ter com­mis­sion­ing. Ap­pre­cia­ble pro­duc­tion loss hap­pened due to too high exit process gas tem­per­a­tures of var­i­ous water cooled heat ex­chang­ers and the large steam tur­bines’ sur­face to con­densers vac­u­ums drop­ping to < 600 mm Hg from 700 mm Hg. This dis­turbed the top man­ager, who or­dered an im­me­di­ate rem­edy.

The trou­bleshoot­ing process en­gi­neers hastily at­trib­uted the cause to in­creased cool­ing water con­sump­tion due to var­i­ous de-bot­tle­neck­ing tasks dur­ing com­mis­sion­ing. They proposed ad­ding an­other CW cir­cu­la­tion pump - a very dif­fi­cult to im­ple­ment pro­posal cost­ing >Rs 10 mil­lion in the eight­ies.

The author, - main­te­nance man­ager then - re­view­ing the pro­posal ques­tioned the process en­gi­neers, on the SR. Their eva­sive an­swers re­vealed their shaky SR knowl­edge!

He ex­plained SR and took them to the pump. The dis­charge header pr gauge read 342 kPa against 340 the

PFD spec­i­fies. The tri­umphant process en­gi­neers said, “We are right Sir. SR has not in­creased.”

To dou­ble check, the author ze­roed the PG, and found it still show­ing 342 – ob­vi­ously stuck gauge! A new gauge showed 370 kPa close to relief valve setting of 380 kPa. The relief valve must be pass­ing heav­ily due to too low seat­ing pres­sure from the too low mar­gin be­tween oper­at­ing and set pres­sures. Ac­cept­able mar­gin is 10% or more of oper­at­ing pr. The leak, piped back to the cool­ing water basin is not vis­i­ble. In ad­di­tion, he got the shell cov­ers of a small shut down heat ex­changer opened. Foul­ing was so much that one won­dered how water ever flowed in it! Ob­vi­ously in­creased SR is the cause of too low pump delivery and too poor heat ex­changer per­for­mance

The author chided the process en­gi­neers for their in­ad­e­quate SR knowl­edge, con­se­quent hasty wrong con­clu­sion, and re­sult­ing pro­posal for an ex­pen­sive pump ad­di­tion, which would not have solved the prob­lem any­way. Why? The author re­quests the read­ers to solve this puz­zle.

On the author’s sug­ges­tion the plant ad­vanced the turn­around by two months, got all heat ex­chang­ers hy­dro jet cleaned. Pump dis­charge pr was only 280 kPa. In fact, op­er­a­tors had to throt­tle many heat ex­changer water valves to get cor­rect exit gas temps. Fi­nally the pump dis­charge header PG set­tled at 310 kPa and water flow rate as shown in CC. The var­i­ous heat ex­chang­ers’ gas exit temps and sur­face con­densers’ vac­u­ums at­tained nor­mal val­ues, and pro­duc­tion ex­ceeded rated ca­pac­i­ties.

Case Study 4; Why Me­chan­i­cal Seals & Thrust Bear­ings failed of­ten? A tur­bine driven mul­ti­stage CFP (fig 5) with motor driven hot standby draws 2-bar, 110oC potas­sium car­bon­ate (K CO ) so­lu­tion from the

2 3

CO strip­per and feeds to CO ab­sorber at 35 bars pres

2 2 sure to scrub Re­formed Gases.

Af­ter 18 months of ex­ceed­ingly good run since com­mis­sion­ing,the suc­tion and dis­charge me­chan­i­cal seals (MS) de­vel­oped leaks. Thrust an­tifric­tion bear­ing (TB) also showed poor shock pulse me­ter read­ings. The re­newed MSs and TB lasted 3 months only. Crew changed both and lived with both fail­ures with­out propos­ing any so­lu­tion.

The seal sup­plier as­sured that the spare MSs as good as those they supplied to pump man­u­fac­turer and re­quested to look for other causes. To trou­bleshoot, the author noted the fol­low­ing from failed seal, bear­ing in­spec­tion and draw­ing study and posed the Root Cause Anal­y­sis ques­tions of why and how?:

1. The MSs have API 31 seal flush plan: 6 bars 60o C con­den­sate flushes the seals and pre­vents the hot and ag­gres­sive 115o C K CO wet­ting the seal for

2 3 seal pro­tec­tion

2. Eyes of im­pel­lors 1,2&3 face left and those of 4,5&6 right to can­cel ro­tor ax­ial thrust to­wards suc­tion usual in CFPs sub­stan­tially

3. Balance Drum (BD) left side pres­sure is > BD right side pres­sure, as the seal flush­ing line con­nects it to suc­tion seal. Thus the net force on BD to­wards dis­charge can­cels more of the ax­ial thrust.

4. The TB han­dles the very small mag­ni­tude resid­ual thrust 2 and 3 leave

5. Brit­tle seal elas­tomers in­di­cate, hot and ag­gres­sive K CO wet­ting the MSs; pos­si­ble only if con­den­sate

2 3 flush fails? Has it failed?

6. Poor TB lives in­di­cate ex­ces­sive TB load. How is it pos­si­ble? Per­haps, too high BD left side pr →, low ΔP acrossBD → re­duced force on BD to­wards suc­tion → loss of thrust can­cel­la­tion. Why?

7. Per­haps in­creased clearance be­tweenBD OD and BD cham­ber ID. BD right side pres­sure ex­ceed­ing 6-bars would prevent flush­ing con­den­sate en­try, es­sen­tial for long MS lives. This sur­mise sup­ports the ob­served brit­tle elas­tomers – item 5

8. The draw­ing spec­i­fied 3- mm thick stel­lite – a cobalt based – hard fac­ing al­loy – de­posit of 450 BHN hard­ness at WR bores and BD OD to re­al­ize hard­ness dif­fer­ence of min 50 BHN with SS 316 mat­ing parts to prevent galling.

How to con­firm sur­mises 5, 6 and 7? Few minutes of in­tense think­ing sug­gested check­ing the skin temp of the ¾” flush line up­stream and down­stream of the

NRV. With nor­mal seal flush­ing it should be around 60O C.

Near am­bi­ent NRV up­stream and 103OC NRV down­stream con­tact ther­mome­ter read­ings con­firmed the au­thors’ sur­mises of:

1. No seal flush­ing flow be­cause of higher than usual ΔP i.e. higher than usual pr at BD right side

2. Sur­mises 5 to 7 of last para­graph.

What is the rem­edy? The author thought of the rem

edies as un­der:

1. Check BD& Wear Ring clear­ances dur­ing next seal change shut­down

2. Put spare wear rings if ex­ces­sive clear­ances

3. Sleeve the BD to re­al­ize de­sign clear­ances

4. An op­er­a­tor sug­gested ad­ding rota me­ter to mea­sure each seal flush flow.

The crew re­ported 6-mm higher clear­ances at wear rings and BD when they dis­as­sem­bled the CFP two months later for MS and TB change. The con­fir­ma­tion of the author’s Root Cause Anal­y­sis (RCA) based sur­mises de­lighted the author to ec­stasy, be­cause solv­ing the prob­lem now is easy! All over 6-mm clearance in­crease points out that K CO so­lu­tion has cor­roded

2 3 away the stel­lite and for­tu­nately not SS 316. The author de­cided not to stel­lite the mat­ing parts though the mat­ing parts be­come galling prone. Amer­i­can Petroleum In­sti­tute (API) 620 rec­om­mends in­creas­ing the clear­ances by 0.125 mm from those given for non­galling pair to prevent galling of even galling mat­ing parts. Crew mea­sured BD cham­ber bore dia was 500 mm and spec­i­fied clearance with stel­lite hard fac­ing 1.75 mm. He got a 20” Sch40 250 mm long pipe bit from a neigh­bor­ing plant to sleeve the BD OD.

Ma­chin­ists ma­chined the pipe bore for medium in­ter­fer­ence fit­ting on to the pol­ished BD. Soaked the pipe in 200oC oil bath for 5 minutes, wiped oil well and slid the pipe over the BD. Af­ter cool­ing to room temp, they fin­ish ma­chined the pipe bit OD to 500-(1.75+0.125) = 498.125 mm di­ato re­al­ize galling pair clearance and trimmed the ends flush with that of the BD sides. The re­sult­ing sleeve thick­ness of 10 mm ex­ceeds 6 mm, min nec­es­sary for sta­ble shrink fit­ted sleeve. crew as­sem­bled the pump back and handed over in 12 hours af­ter putting new MS and thrust bear­ing be­cause of rea­sons given be­low:Mak­ing new SS WRs and putting new wear rings in­volv­ing ro­tor dis­as­sem­bl­yare time con­sum­ing, and too long­mo­tor driven pump run is risky. The shop made new wear rings from suit­able SS 316 pipe bits tak­ing care to in­crease the clear­anc- es by 0.125 mm – galling pair- in two months leisurely. Though MS and TB fail­ures dis­ap­peared, the author got SS wear rings put to min­i­mize leaks and save power at a suit­able shut­down op­por­tu­nity. In ad­di­tion, he pro­vided a rota me­ter 0-1000 LPH to mea­sure each seal flush flow, and green marked 500 – 700 LPH as nor­mal flow.

The op­er­a­tion en­gi­neer heard the author won­der­ing aloud, “how come, noneno­ticed pump ca­pac­ity loss, de­spite such enor­mous clear­ances in­creases. He re­sponded, “We did no­tice Sir, we just in­creased the tur­bine speed by 25-50 RPM when­ever we could not pump enough to the ab­sorber. Tur­bine RPM was 3400, 18 months ago, it was 3570 till new SS wear rings went in and now it is back to 3400!

Seal lives ex­ceeded 36 M and TBs never failed there­after.

Case Study 5: why car­bon bushes of sub­merged pumps failed of­ten? 21 Nos. of steam tur­bines sur­face con­densers of a Saudi Ara­bian fer­til­izer fac­tory came with a tur­bine driven con­den­sate pit pump mounted on top of a buried tank and a motor driven pit pump (fig 6). Of­ten the pumps’ car­bon bushes sup­port­ing the im­pel­lor eyes and the long shaft reach­ing the driver wrecked. The site en­gi­neer at his wits end re­quested the author to han­dle the prob­lem.

The author watched a pump changeover: Op­er­a­tors opened the 3” water side valve to fill the con­den­sate col­lec­tion tank. They were about to push the motor start PB. Hold, shouted the author and asked, “have you opened the steam side valve?” The per­plexed op­er­a­tor grinned and said, “Steam side valve, Sir, No su­per­vi­sor told me to open it”. The author called the op­er­a­tions en­gi­neer and he too was ig­no­rant of the steam side valve. The author ex­plained, “the con­denser is un­der high vac­uum. Hence, un­less the pit tank top is open to the same vac­uum water can­not flow into the tank. So you must open both the ¾” steam side and the 3” water valves

and wait for 5 minutes to be sure of enough con­den­sate col­lec­tion in the tank and sub­merg­ing the pump. So far you have been wreck­ing pumps be­cause of dry run­ning!”

The sheep­ish op­er­a­tor opened the steam side valve also and waited for 5 minutes and then changed over to the motor driven pump safely. Thus the author pre­vented an­other wreck and all future wrecks!

By a memo the author re­quested the op­er­a­tions man­ager to re­vise the pump start up pro­ce­dure and hold classes for all op­er­a­tors and shift en­gi­neers.

What do the above dis­cus­sions tell us? The above dis­cus­sions tell us that CFPs are more than what meets the eye, es­pe­cially dur­ing com­mis­sion­ing. Most man­u­fac­tur­ers fur­nish sturdy and re­li­able CFPs. Mean Time be­tween Over­hauls ex­ceed­ing 4 years is com­mon for well in­stalled and aligned to the driver CFPs. Most fail­ures are due to ex­ter­nal pump fac­tors as the above case stud­ies show. (see ta­ble 3)

Hence, do not mod­ify pump in­stal­la­tions hastily. Elim­i­nate all ex­ter­nal causes first.

Other im­prove­ments: Squir­rel cage in­duc­tions mo- tors (SCIM) are in­dus­try work­horses be­cause of their fol­low­ing ben­e­fits:

1. IMs are highly re­li­able and run for long pe­ri­ods with prac­ti­cally no main­te­nance due to sim­plic­ity of con­struc­tion; just 3 parts – a sta­tor, a ro­tor and sup­port­ing bear­ings.

2. Runs on the com­mon and eas­ily avail­able 3 O (large mo­tors) or 1O AC power sup­plies.

3. SCIMs self-start on push but­ton press­ing

4. Thanks to avail­abil­ity of mo­tors of volt­age rat­ings up to 13.2 KV their sizes are small even for very large mo­tors say 25 MW.

5. Mo­tors are avail­able for dif­fer­ent fixed speeds from 300 to 3000 RPM for stan­dard 50 Hz sup­ply.

6. Mo­tors of dif­fer­ent en­clo­sures and start­ing torque to suit the driven ma­chin­ery needs are avail­able for use in the open, even in harsh en­vi­ron­ments.

7. Con­struc­tion of haz­ardous area mo­tors is sim­ple and less ex­pen­sive due to the ab­sence of spark pro­duc­ing com­mu­ta­tors. In ad­di­tion, ab­sence of wear- ing and spark­ing com­mu­ta­tors elim­i­nates pe­ri­odic main­te­nance shut­downs. Hence, IMs offer the high­est avail­abil­i­ties amongst all driv­ers.

8. IMs are the most en­vi­ron­ment friendly drives due to no emis­sions

9. IMs are the low­est cost driv­ers.

How­ever, these work­horses suf­fered a se­ri­ous lim­i­ta­tion of non-amenabil­ity to speed vari­a­tion till re­cently. Users var­ied the ca­pac­ity of CFPs, com­pres­sors, and fans, by en­ergy waste­ful dis­charge valve throt­tling.

Ad­vent of Vari­able Speed Drives: The ever-in­creas­ing en­ergy costs and stricter and stricter en­vi­ron­ment reg­u­la­tions con­strained in­dus­try to be en­ergy ef­fi­cient and limit pol­lu­tion, cre­at­ing a cry­ing need to vary SCIM speed vari­a­tion be­sides oth­ers.

The great strides of solid-state power elec­tron­ics and the cry­ing need as above re­sulted in the in­ven­tion in the eight­ies of a de­vice called Vari­able Fre­quency Drive (VFD). We will stick to the nomen­cla­ture VFD, though it goes by var­i­ous names of VSD (Vari­able Speed Drives), ASD (Ad­justable Speed Drives), and FC (Fre­quency Con­verter).

VFDs ac­cept the stan­dard 1 or 3 φ AC power sup­plies and out­puts vari­able fre­quency and volt­age con­di­tioned power to run SCIMs at vari­able RPMs and torques to suit driven ma­chin­ery re­quire­ments. Hav­ing got a rudi­men­tary idea of VFDs, we will study the ben­e­fits of us­ing VFDs for CFPs (cases 1&2) and pis­ton / plunger pumps (PPs, case 3) ca­pac­ity con­trol, and as a ‘de­sign out main­te­nance de­vice’ to re­place the ex­ist­ing en­ergy waste­ful, and low avail­abil­i­ties vari­able speed drives (VSDs) in few ap­pli­ca­tions (Case Stud­ies 7&8).

Case 1: CFP of pre­dom­i­nantly Con­stant SR Hy­draulic Cir­cuits: The re­sis­tances a Boiler Feed Water Pump hy­draulic cir­cuit offers is neg­li­gi­ble com­pared to that the boiler drum pres­sure of 100 bars i.e. ≈1000 m WC. In ad­di­tion it is con­stant. CFP affin­ity law tells us that h = h α N2 where hPD = pump de­vel­oped

PD SR head and Q flow rate. hP ex­po­nen­tially de­creases with de­creas­ing N, whereas, the SR re­mains nearly con­stant. Hence, flow to the boiler drum would cease at lower N. Thus VFD or any other de­vice to

vary motor speed for flow con­trol is use­less. Dis­charge con­trol valve (DCV) throt­tling is the only prac­ti­cal way to vary the flow. How­ever, the en­ergy lost by dis­si­pa­tion across DCV is in­evitable and neg­li­gi­ble for small flow vari­a­tions.

CFP of pre­dom­i­nantly vari­able SR cir­cuit: VFD speed con­trol for CFPs of pre­dom­i­nantly vari­able SR and for Plunger Pumps is very ben­e­fi­cial and al­lows wide flow vari­a­tions at con­sid­er­able en­ergy sav­ings. See ex­am­ples 1 and 2.

Ex­am­ple 1: A plant’s pump house feeds plants A to D (fig 2). Plant A draws 0.3QR when a 3 hour long batch is on and 0 at other times. Thus they draw water for 12 hours a day only. Plants ‘B to D’ are con­tin­u­ous process plants. A flow con­troller FC ma­nip­u­lates CFP dis­charge con­trol valve (FCV) to meet flow re­quire­ments. Is VFD pump speed con­trol in­stead of FCV ben­e­fi­cial?

VFD Retro­fit Anal­y­sis:

Ex­am­ple 2: As­sess the suit­abil­ity of VFD ap­pli­ca­tion for the plunger pumps of data de­tailed be­low:

VFD Retro­fit Anal­y­sis:

Salient Plunger Pumps In­for­ma­tion: Plunger pumps find ap­pli­ca­tions for rel­a­tively smaller through­put at high dis­charge pres­sure ap­pli­ca­tions e.g. pumps to feed liq­uid am­mo­nia and am­mo­nium car­bo­mate at 250 bars pres­sure into Urea Re­ac­tors in the 30 to 40 m3/H flow range. They gen­er­ate the head to over­come the hy­draulic cir­cuit’s SR at all Ns. Hence, a safety valve at dis­charge is nec­es­sary to prevent over­pres­sure bursts in case of in­ad­ver­tently closed dis­charge block valve. The PP de­liv­ers a con­stant vol­ume per rev­o­lu­tion. Hence, by­pass­ing part of the Q back to suc­tion or varying the pump speed is the com­mon flow con­trol method. Urea plants vary the pump speed us­ing a hy­draulic torque con­ver­tor be­tween pump and motor to con­trol re­ac­tor feed car­bo­mate flow rates, though of 30% ef­fi­ciency, due to process re­quire­ments.

To sum­ma­rize, run­ning all types of pumps and com­pres­sors at nec­es­sary speeds to meet the flow re­quire­ments saves con­sid­er­able power, wear and tear than run­ning these at fixed motor speeds and re­sort­ing to other flow con­trol meth­ods. Torque con­ver­tors though of high MTBF, are no­to­ri­ously of low ef­fi­cien­cies and most dif­fi­cult to re­pair be­cause over 100 pre­ci­sion small parts as­sem­bled into the cas­ing. The author knows of two in­stances of other plants open­ing a torque con­verter and aban­don­ing it. For­tu­nately 1:1 spare held avoided huge pro­duc­tion loss.

These and the above dis­cus­sions pro­jected potential of very short pay­out pe­riod en­cour­aged the author in 1985 to fit VFD for the 3.3 KV, 450 KW mo­tors pow­er­ing a car­bo­mate pump and its hot standby cou­pled to Motor, GB and TC.

Only one ven­dor of­fered a com­bi­na­tion of 3.3.KV/400 V step down in­put trans­former 1 + VFD + 400.V/ 3.3 KV step up out­put trans­former 2. It looked too com­pli­cated and the ven­dor could not fur­nish names of suc­cess­ful users. Hence, it did not get top man­age­ment ap­proval. The case study be­low shows that now HT VFDs are avail­able and saves power and ma­chine wear and tear:

Case Study 6: VFD pow­ered Screw Com­pres­sor Saves: A mega sized In­dian re­fin­ery came with screw com­pres­sor cou­pled to VFD pow­ered 6.6 KV motor to hold crude dis­til­la­tion tower top pres­sure con­stant. For­merly, re­finer­ies waste­fully by­passed part of the screw com­pres­sor dis­charge back to suc­tion to vary the through­put to main­tain the tower top pres­sures. Ta­ble-4 shows the re­cur­ring sav­ings the re­fin­ery en­joys from the VFD:

VFDs ‘De­sign out Main­te­nance’: The case stud­ies 7 and 8 ex­plain it.

Case Study 7: VFD retro­fit ‘De­signs out Main­te­nance’ & boosts pro­duc­tion: Me­chan­i­cal vary-speed­drives (MVSD) vary the belt speeds of 12 Nos. ofBelt weigh feed­ers’ to vary raw ma­te­ri­als – urea, filler, and potash – feed rates to man­u­fac­ture dif­fer­ent grades of com­plex NPK fer­til­iz­ers at dif­fer­ent pro­duc­tion rates to meet mar­ket re­quire­ments. A 0.7KW Motor flanged to MSVD cas­ing runs MSVD in­ter­nals of nu­mer­ous moving parts to run the in­put shaft of a gear­box flanged to the MSVD cas­ing at 500 to 2500 RPM. The gear­box out­put shaft runs the weigh feeder con­veyor belt drum at speeds of 0.5 to 2.5 m/mt. A pneu­matic ac­tu­a­tor (PA) re­spond­ing to 3-15 psi out­put from con­trol room panel mounted weight record­ing con­troller (WRC) ad­justs a lever of MSVD to vary the out­put shaft speed.

A pneu­matic speed trans­mit­ter ST cou­pled to MSVD out­put shaft out­puts 3 15 psi pneu­matic sig­nals pro­por­tional to belt speeds. A load cell gen­er­ates the sig­nal for belt-load­ing (Kg/Me­ter)-Kg/M; a pneu­matic ana­log com­puter (AC) mul­ti­plies these two sig­nals and out­puts weight rate (Kg/H) sig­nal to WRC.

Prob­lems of the sys­tem: MSVD fail­ure modes are fast wear and tear of var­i­ous pre­ci­sion mat­ing parts, parts’ se­vere cor­ro­sion, con­se­quent seizures, es­pe­cially the ax­i­ally trav­el­ling pul­ley halves on in­put and out­put splined shafts. Failed speed trans­mit­ters due to these rea­sons added to the poor weigh feeder avail­abil­i­ties, very high main­te­nance ef­forts and costs and se­vere pro­duc­tion losses as high as 10%. Prod­uct NPK Qual­ity too was poor draw­ing enor­mous cus­tomer com­plaints. (Fig 10)

So­lu­tion: The author elim­i­nated the MSVD and Speed Trans­mit­ter and thus the poor weigh feed­ers’ poor avail­abil­i­ties.

Crew did the sim­ple trial mod­i­fi­ca­tion tasks: Dis­carded the MSVD; flanged the gear­box and the Motor to a shop made pedestal bolted to the MVSD base plate. In­stalled and set the VFD to out­put 17-85 Hz fre­quency in V/f = con­stant mode to meet belt con­veyor torque re­quire­ments and ac­cord­ing to 4-20 mA cur­rent sig­nals. (Fig 11)

Thus VFD pow­ered mo­tors run the weigh feeder at 0.5 to 2.5 m/min ac­cord­ing to WRC out­puts as be­fore, but at 100 % avail­abil­i­ties.

A pur­chased P-I (pres­sure to cur­rent) con­verter con­verts the 3 15 psi WRC out­puts into 4-20 mA to vary VFD out­put.

The motor and hence the belt speed is pro­por­tional to fre­quency and hence the 4 20 mA i.e. to the 3-15 psi WRC out­put. There­fore WRC out­put it­self serves as belt speed sig­nal also, elim­i­nat­ing the trou­ble­some speed trans­mit­ter too.

As dis­cussed in the above para­graphs, by VFD use, gone are the most trou­ble­some MSVDs and Speed trans­mit­ters boost­ing that weigh feeder avail­abil­ity to 100%. Re­sult­ing pro­jected in­creased pro­duc­tion, im­proved prod­uct qual­ity, less main­te­nance ex­penses and no cus­tomer com­plaints of failed / de­layed de­liv­er­ies etc. at neg­li­gi­ble in­vest­ment pleas­antly sur­prised the man­age­ment im­mensely. The man­ager chased the author to ef­fect the mod­i­fi­ca­tions within 6 weeks – VFDs rush pur­chase 5 wks + 1 wk in­stal­la­tion!

Case Study 8: VFD ‘De­signs out Main­te­nance’ & boosts ce­ment pro­duc­tion: Ro­tary Drum Kilns’ and Dri­ers’ con­ven­tional drives are no­to­ri­ously main­te­nance prone re­quir­ing daily chores of lu­bri­cat­ing, free­ing stuck parts, and clean­ing. Con­se­quent low avail­abil­i­ties re­sult in huge pro­duc­tion loss and enor­mous main­te­nance ex­penses. A typ­i­cal ro­tary kiln drive con­sists of a 1000 kW motor cou­pled to a heavy ra­tio gear­box (fig 12). GB Out­put shaft cou­ples to a be­tween plum­mer blocks

bear­ings pin­ion shaft. The pin­ion mesh­ing with the drum mounted and linked to the drum huge split gear drives the kiln at about 10 RPM. Drive Sys­tem Pro

blems: Be­low listed are drive sys­tem prob­lems and con­se­quent low avail­abil­i­ties

1. The links link­ing the large split gear to the drum break of­ten and shut down the plant

2. Split gear ro­ta­tion throws large lube spills and causes se­ri­ous house­keep­ing prob-

lems and slip haz­ards

3. Of­ten break­ing Links re­pair is te­dious, time con­sum­ing, haz­ard prone, and of­ten of un­sat­is­fac­tory out­comes; these re­duced the avail­abil­i­ties to around 80%

4. Lu­bri­cat­ing the open mesh­ing pin­ion with the split gear and sev­eral rollers sup­port­ing the drum is a daily te­dious chore.

On the other hand a Saudi Ara­bian ce­ment plant ro­tary kiln is a per­fect ex­am­ple of ‘De­sign out main­te­nance’.Gone are the gear­box, open mesh­ing gears, their­lu­bri­ca­tion, haz­ard prone lube spills and the pain in the neck links cou­pling the large sprlit gear to the shell’. The new all elec­tric drive is un­be­liev­ably sim­ple and main­te­nance free! (Fig 13)

The kiln shell has the sil­i­con steel ro­tor lam­i­na­tions keyed to it. Thus the shell dou­bles as ro­tor also. Wind­ings con­cen­tric to the ro­tor lam­i­na­tions con­sti­tute the ‘120 pole sta­tor’. The shell and sta­tor wind­ings thus con­sti­tute an in­duc­tion motor!

Pow­ered by a VFD ac­cept­ing 6.6 KV 60 Hz 3 φ AC and set to out­put 10 Hz 3 φ AC the ‘motor ro­tor’ i.e. the kiln ro­tates at 10 RPM! Look how sim­pli­fied and main­te­nance free the drive has be­come – In­deed a mag­nif­i­cent ‘de­sign out main­te­nance’ ex­am­ple!

The author rec­om­mends all Ro­tary Kiln, Ro­tary Drum Dryer and Ro­tary Gran­u­la­tor users to check and retro­fit mod­ern drive if fea­si­ble.

Grund­fos launches De­wa­ter­ing pumps

Grund­fos In­dia, man­u­fac­turer of in­tel­li­gent & ef­fi­cient pump­ing tech­nolo­gies, has re­cently launched its de­wa­ter­ing pumps DPK and DWK. Grund­fos’ re­li­able de­wa­ter­ing pumps, DPK and DWK, aim at eas­ing the way of life while re­strict­ing the eco­nomic losses. These com­pact sub­mersible pumps can de­wa­ter ef­fec­tively. DPK & DWK can be used in homes, of­fices, gated-com­mu­ni­ties, con­struc­tion sites, ex­ca­va­tions, tun­nels, un­der­ground park­ing, in­dus­trial drainage pits etc. Mak­ing it a per­fect so­lu­tion for all the water log­ging and drainage re­lated is­sues dur­ing the mon­soons.

The DPK and DWK range of de­wa­ter­ing pumps are con­cep­tu­al­ized and en­gi­neered af­ter go­ing through many lev­els of strin­gent tests. They come with in­tel­li­gent fea­tures such as:

• Built-in bi-me­tal sen­sor that pro­tects against over­heat­ing of the motor

• Pump per­for­mance is un­hin­dered by sand or other abra­sives is re­quired, or if the avail­able power sup­ply is lim­ited

• Com­bin­ing in­stal­la­tion flex­i­bil­ity with re­li­a­bil­ity and ease of ser­vice. Speak­ing about Grund­fos’ de­wa­ter­ing pumps, Shankar Ra­jaram, Vice Pres­i­dent – Busi­ness De­vel­op­ment, Grund­fos In­dia said, “In­dian mon­soons are be­com­ing more un­pre­dictable every pass­ing year. It is there­fore es­sen­tial to be proac­tive in ad­dress­ing is­sues such as water log­ging by be­ing equipped with the right pump­ing so­lu­tions. Grund­fos range of de­wa­ter­ing sub­mersible pumps are de­signed to work op­ti­mally in such en­vi­ron­ments. These pumps are easy to in­stall and ser­vice and there­fore are a per­fect so­lu­tion for both com­mer­cial and do­mes­tic cus­tomers”.

For more de­tails con­tact: Grund­fos Pumps In­dia Pvt Ltd, Chen­nai. Tel: 044-45966800 Email: salesin­dia@grund­fos.com Web: www.in.grund­fos.com

Aerotherm’s fluid bed dryer

Con­tainer of Aero Therm Fluid Bed Dryer is made out of MS / SS304 / SS316 /

Alu­minum as per re­quire­ment.

The per­fo­rated sheet & fine wire mesh screen will be pro­vided at the bot­tom for proper air dis­tri­bu­tion.

Heat­ing can be supplied with Elec­tri­cal, Steam, Ther­mic

Fluid or Hot Air Gen­er­a­tor.

Elec­tri­cally heated Dryer will be pro­vided with heat­ing el­e­ment of suit­able rat­ing hav­ing

MS gal­va­nized / Stain­less steel con­struc­tion. Steam / Ther­mic

Fluid Ra­di­a­tors are made out of MS / SS Tubes with MS / SS / AL Fins as per re­quire­ment.

Con­trol Panel will con­sist of Dig­i­tal Temp. Con­troller, Starter for Blower, Syn­chro­nous Timer for batch time setting, In­di­cat­ing Lamps, Fuses, Main Iso­la­tor switch etc.

Fea­tures :

• Range 15 to 500 Kg

• Heat­ing: Elec­tri­cal, Steam or Ther­mic Fluid Ra­di­a­tor

• Uni­form and Quick Dry­ing

• Dry­ing at Low Tem­per­a­ture.

For more de­tails con­tact :

Aerotherm Sys­tems Pvt Ltd

Phone :+ 91 79 25890158

Email: aerothermsys­tems.cm

Web­site :aerothermsys­tems.com

Dexmet’s Mi­cro­Grid ® pre­ci­sion ex­panded me­tal and me­tal foil

Mi­cro­Grid pre­ci­sion-ex­panded foils from Dexmet

® are used in bat­ter­ies, elec­tron­ics, aero­space, med­i­cal, pack­ag­ing, and wher­ever mesh and per­fo­rated foils with high pre­ci­sion, me­chan­i­cal and elec­tri­cal prop­er­ties or EMI/ RFI shield­ing ca­pa­bil­i­ties are re­quired. They are avail­able in most me­tals and poly­mers, or we

can work with your pro­pri­etary ma­te­ri­als.

Me­tals we reg­u­larly pro­duce in­clude: alu­minum, brass, cop­per, Monel™, nickel, steel, stain­less steel, and zinc.


Ex­panded Me­tals Con­fig­u­ra­tions Ma­te­ri­als and Spec­i­fi­ca­tions

One-piece, sin­gle-unit struc­ture Elim­i­nates un­rav­el­ing and con­tact re­sis­tance of woven mesh

Su­pe­rior shield­ing, elec­tri­cal and heat trans­fer prop­er­ties

Wide range of sizes, pat­terns, an­gles and ma­te­ri­als Spe­cial­ized process vari­a­tions, such as flat­ten­ing, pulling, and an­neal­ing avail­able. For de­tails con­tact: Dexmet, USA Tel: (800) 714-8736 Fax: (203) 294-7899 Web­site :sales@dexmet.com

NSL Cen­trifu­gal Pump by DESMI

The NSL range of pumps by DESMSI rep­re­sents high ef­fi­ciency, low NPSH val­ues, easy to in­stal­la­tion and low main­te­nance. The NSL se­ries is widely used within dif­fer­ent ap­pli­ca­tions and mar­kets.

De­sign Fea­tures

The pump is an in-line, ra­di­ally split, sin­gle-stage cen­trifu­gal pump with con­nect­ing flanges ac­cord­ing to in­ter­na­tional stan­dards. The pump is de­signed for mount­ing with elec­tric mo­tors hav­ing dif­fer­ent in­ter­na­tional flange di­men­sions.

The cas­ing is equipped with a re­place­able seal­ing ring. The im­peller is made with dou­ble-curved blades to en­sure low NPSH-val­ues and high ef­fi­ciency. The bear­ing unit is equipped with sturdy ball bear­ings and the small types are fit­ted with life­time-lu­bri­cated bear­ings. In the larger types the lower bear­ing is a dou­ble bear­ing for which a lu­bri­ca­tion point is pro­vided. These pumps are suit­able for :

• In­dus­try water cir­cu­la­tion

• Cool­ing tower dis­tri­bu­tion

• Diesel trans­fer

• Dis­trict heat­ing

• Dis­trict cool­ing

For more de­tails con­tact:

DESMI In­dia LLP, Te­lan­gana, In­dia Mo­bile No : +91-9949339054 Email : desmi@desmi.com Web­site: www.desmi.com

Koch Mem­brane So­lu­tions offers spi­ral nanofil­tra­tion and re­verse os­mo­sis prod­ucts

Koch Mem­brane Sys­tems (KMS), a global leader in mem­brane fil­tra­tion tech­nolo­gies, has an­nounced the re­turn of the FLUID SYS­TEMS nano

® fil­tra­tion (NF) and re­verse os­mo­sis (RO) prod­uct lines. The FLUID SYS­TEMS NF

® and RO prod­ucts are avail­able in stan­dard 8” FRP hard over­wrap con­fig­u­ra­tion (8040) in stan­dard and high area con­struc­tion, and are listed un­der ANSI/NSF Stan­dard 61. The FLUID SYS­TEMS line in­cludes mul­ti­ple prod­ucts en­gi­neered to serve in potable water and in­dus­trial water ap­pli­ca­tions in­clud­ing:

• TFC SW: High re­jec­tion sea­wa­ter RO mem­branes ® for treat­ment of high TDS in­dus­trial streams and sea­wa­ter de­sali­na­tion

• TFC HR: Ro­bust high-re­jec­tion and low-foul­ing ® brack­ish water RO prod­ucts for con­sis­tent op­er­a­tion when treat­ing in­dus­trial streams and waste­water ef­flu­ents

• TFC SR: Low en­ergy NF prod­ucts for water soften® ing, sea­wa­ter sul­fate re­moval and or­gan­ics re­moval In ad­di­tion to NF and RO el­e­ments, KMS offers a line of stan­dard sys­tems, which in­cludes a high re­cov­ery op­tion.

For more de­tails con­tact:

Koch Mem­brane Sys­tems, Wilm­ing­ton, Delaware Phone: +1-978-694-7000 Web­site: www.kochmem­brane.com

Grund­fos DPK

Grund­fos DWK

Newspapers in English

Newspapers from India

© PressReader. All rights reserved.