Document M4JwgM7bZLMnw3kz800wg4evV

.: 'Vj , *'OtotroHnq ',,y. heater IS min capacity S4QfiOO/b/hr 170F to boiler I ^ feed wmps\ Jan Feb Mar Apr May June O fiardntss = - Chtoridt --------- Wei/ Mo.747 ----------- Well Ho.746--------------Well Mo. Worer composition, once (be plant got under woy In 1935, underwent rapid changes in chlorides and hardness (obove) Clarifying equipment Includes 40,000-gal tank to coIh and mix well and city water. It then goes to the clariffl Filtered water from plant process pumped through metered, healer on first-stage hot-process unit. Steam from deaerat Inlet'Woter controller end vent condensor goes to the sproy ing heater raises the first-ttogo makeup woter to 200 F Internal Conditions of Old BoilerSGuide Water Treatment For New Units Here's o unique approach in planning treatment for new units: A 10-year-experience serves as spadework in pro ducing a sound installation By i N WELSH, H M RIVERS and C J KILDUFF* > A specially designed feedwatertreatment system represents o major part of the power modernisation pro* gram at U. S. Industrial Chemicals, The, Curtis Bay, Md. plant (Power, Tab 1948, pp 64-67). Design for this water-treatment plant drew heavily on a 10-year operating record of internal phosphate treatment - of raw-water makeup for two 125,000-lb-per-hr 650pslg 645-F steam generators. When plans called for adding to steam gen eration with two 240,000-lb-per-hr 6SOpsig 700-F generators (one already Is In operation) this 10-year experience furnished many influencing -factors. Wll-Wofe Composition. Back In 1935 when the 125,000-lh-per-hr boilers went into service, condensate returns were estimated to run 10 to 30% of feedwater needs. Makeup came from nine wells, as shown in Table I, p. 90. Water from Well 1326 had 15-ppre calcium and 12-ppm magnesium, mak ing it fairly hard. Water from Wells 747, 750, 132S, 1468 and 1469 ran quite soft. Considering the capacities of all wells, a mixture of woter from roost, excluding 1326, gave a water as low in hardness as hot-process, lime-sodatreated-water. Concentrations of iron, aluminum, dissolved solids and suspended ma terial in the well waters figured in early considerations. The over-ell pic ture in a mixture was not alarming except for Well 1326 water, which was relatively high in iron and dissolved solids, end hod a concentration of sus pended material of 3 ppm, the maximum found. Aluminum concentration was not more than a trace. As for silica, it was recognised that, its concentration was high in relation to other consti tuents, but back in' 1935 silica removal was not understood as it is today. Also, barring localised excessive heat input to the boilers, silica was not expected to be troublesome. Once the plant got under way water composition changed rapidly for the worse, Fig. 1. As timo went on further deleterious changes developed in the water. Food-line Deposits. Table I shows the low pH of the well waters. Also, a substantial part of water hardneas was due to magnesium, as the ralio'of mag nesium to celcium was higher than in most natural waters. Despite thorough deaeration of f water, corrosion evidence appeared! feed pumps and regulating v*1tot the (all of 1935. Some coTtwbn.H been anticipated so it was detected^ fore much damage was done. An menta were made to change feeding! caustic soda for maintenance of bejS water alkalinity to the olkali cben would be introduced continuously.^ the boiler feedwater. Following this change depoE formed in the feed lines and regnliy valves. Although magnesium bydron ferric-iron oxide and ignition lot*;* substantial in amount, magnesion/B cote was the main deposit eonstitg This compound separates solid;! from High-tempersture feedwalirjR telnlng a fair amount of magnesiums silicate if pH is in excess of '* 8.0.. Subsequent adjustments cf *8 chemical feed indicated deposition cf be prevented by holding pH to and from mid-1941 on the Mg lem was no longer serious. But chsF in row-water composition roidsfa control of pH difficult. A reduction magnesium concentration in.the^ water was desirable. -.VTM Anotelio Scale. Ansleito (M aluminum silicate) first app* the boilers In Sept 1935. From'J until pretreating equipment vtfl stalled a constant battle was wtpqj stop formation of this complex-* scale. Attempts were made to more condensate, but conuadojul 88 (547) POWER SapUmbW] -from slop evaporators ruled this out. JHsgnesium sulphate was Ted to one ^boiler to reduce silica concentration, but.magnesium precipitated as magfC-nesium phosphate, and phosphate con1 sumption increased too greatly, r' Even tannin-feeding was instituted. It Kdid not aflect rate of scale formation | .but it did seem to render deposits more . brittle and hence more easily turbined' s .from tubes. After several years its conr linue^ ma did not seem justified. f..\ Scpro relief from onelcite deposits was gained by (1) discontinuing use of 'xster from some wells (2) Increasing ^capacity of others by rehabilitation (3) . installing new pumps (4) making up F net loss in water with city water, Viable I. .^i.Ansleite scale cannot form if alumi-. bourn docs'not get into the boiler water. I*tWe a'um'num did not appear In filtered makeup-water samples its k-souree was recognized to be suspended fvPtsterial. I,, Sept 1944. Well 3700 l lT? ^ccerne heavily laden with susI pended matter and demonstrated the r-SL*116*4 of lhU thlnkiD* Boiler hrtled from analcite-seale formaTii*? >^er a boiler had generated only L?S.iUicn lb of steam. Irb*/* cverol years boiler tubes had y 'isn *,> ,urbined after every 125 to fc broil > ^ ,,eam t0 T^d over- ptlea.0ftee<U*e anftlch-scnf forma* ttrmll v ' *mProvementa In roekeup PaiM** i eitCfldd runs to 400 to 500 ft[9^0n_Jh f steam between 1940 and Ji*.t . cama the 1944 experience. l&Pvy t a ** Ssptamber 1948 Analyses of raw-water samples in need for clarification. (2) The new Oct 1944 (Table I) showed less bus-, 240,000-lb-per-hr steam generator was pended mater In Well 3700 water than to be installed, and troublefree opera was * present on occasions in Sept. tion demanded the best possible water. Nevertheless the 74 ppm of suspended Selection el equipment. The wealth material was sufficient to add 2 ppm of experience gained in ten years' op of aluminum and 33 ppm of silica to eration of the 125,000-lb-per hr boilers the woter. The other well Waters con simplified selection of equipment Burns tained less but far from insignificant and Roe, Ine, consulting engineers. amounts when silica and aluminum Hall Laboratories, Inc., consultants concentrations of filtered samples are on'water, and engineers and chemists compared with total concentrations of -of U. 5. Industrial.ChemlesJs, Inc co these constituents. Even city 'water operated in reaching conclusions. - had a substantial omount of suspended Except for less suspended solids in solids containing silica and aluminum. water from Well 3700 as the result ' X-ray examination identified hluml- of strainer repairs, Table I, 1944 list num silicate at eilllmanite in the sus ing. shows representative analyses on pended matter of Well 3700 water. which conclusions were based. Total The boiler feedwater must have carried capacity of Wells 1323 and 1326 was considerable suspended matter at the 125 gpm at the lime as compared with time showed because the ecale formed 300 gpm each for 3700 and 4241. Thus aluminum eilicate in relatively small the mixture to be treated would be amounts. This would not normally re mainly water from the last two wells main as such very long in high-tem plus city water. perature, alkaline boiler water kinless Well 3929 water was not to be used it was there in considerable amount. but its inclusion was necessary in all Pretroatmeiit Found Necestory. Rapid considerations because of (1) possi changes in well-water..compositlon in ble failure, of other wells (2) radical dicated the need for pretreating equip change in water composition. ment. Removal of suspended material Table I indicated much of the ailica .was considered In 1937. Always the - and all the aluminum could be removed hope that further contemplated water- . from well waters and city water by re supply changes would obviate the need moval of suspended material. II base- ' for pretreatment delayed action.- exchange softening was to be considered Hope was abandoned in 1944, and a tjiis suspended matter had to be re decision made to install suitable equip moved if only to prevent fouling ex ment for two reasons:- (1) Abnormal, change materials. With hot-process contamination of Well 3700 water with pretreatment silica removal might be suspended material demonstrates the satisfactory. But the high pH, hot (543| 89