Document ganvORODgX5Xd8x66a3aJXKEa

FILE NAME: Power Generation (POW) DATE: 1950 Jan DOC#: POWOIO DOCUMENT DESCRIPTION: Published Article from Trade Journal P W S W A IN Editor L N ROWLEY F A ANNETT E W F FELLER j C McCABE B G A SKROTZKI J J O'CONNOR C F SANDERS C F MARSCHALEK Executive Editor Associate Editor Associate Editor Associate Editor Associate Editor Assistanf Editor Assisfarti Editor Art Editor SHELTON FISHER Publisher W J HARING Sales Manager W a shin g to n : GEORGE DOYING A ffo co : EARLE MAULDIN WORLD NEWS OFFICES: Lo n d o n , P a ris , B e r lin , M o sco w, Tokyo, M elbourne, Bom bay, Rio de Ja n e iro , Buenos A ires CROSS-INDEX O F SUBJECTS FOR THE MONTH JANUARY 1950 VOL 94 NO. 1 STEAM GENERATION, FUELS, FIRING AND COMBUSTION 'PC'- j l \ j t h i DIESELS, GAS ENGINES AND ACCESSORIES 7 f,I AIR CONDITIONING AND REFRIGERATION ELECTRICAL EQUIPMENT, ELEVATORS PIPIN G ,.V A LV ES AND SPECIALTIES . t j TRA N SM ISSIO N , LUBRICATION j \ !# f' s f < i; j 1 j STEAM SER V IC ES, KEATING INSTRUMENTS. CONTROLS COMPRESSORS Hickling Station completes its first year of service Combine know-how and tools for good erection.................................................................................... ........................ ........................ 74 79 i ,4 : 4 Here are the working tools for flue-dust collection: Part II ........................ 82 Simplifying plant lubrication............................................................................................................................................................................................ . . . 86 * 4 <: 4 i 4 4' 4 j 4 j 4 j How much torque is needed to start centrifugal pumps? ........................ 88 ;< Can slag deposits be prevented?.............................................................. 91 Accumulator plus controls evens out steam peaks....................................................................................... ........................ 94 Low-cost producer gas from bituminous c o a l............................................................................................................... ........................ 9 7 Midget gas-turbine unit will do many chores M T-. . 98 4 What the future holds in turbines, generators and motors..................................................... 102 ........................ t New water-hardness test is faster, gives more accurate results................................... 105 ........................ Letters from a turbine specialist.............................................................................................................................................................................. 1 09 ........................ Electric control adds new look to old hydraulic elevators 1 10 1 ........................ Hoover Dam tops 50 billion kilowatt-hours................................................................................................................... 113 ........................ -f " ^ "j -4 A., : >5 > r ' <1 ; ': 4 .i'l 'j i (: .4 > >j :: . - b p v -4 "4- j < ; 4' 4 I 4 2,;V 4:j i -1 4'j 4 Electronics: 21 -- How electronic oscillators operate.................................................................... . . . 1 1 6 Power engineer's notebook of plant-tested ideas................................. .........120 Problems fresh from the plant................................................................... .........124 Technical briefs for busy power men....................................................... .........128 News of the power field............................................................................ .........132 Data sheet: Handy viscosity-conversion chart....................................... .136 Plant equipment news: information center for new products............... .........138 George Edwards: A look into the future............................................... 146 < 00 V-- Reader's service page for new products and literature.......................... y. i F : 1 i 4 4 4; 4 fi . 4 4 4 \ 4' 4- 4 4 < 4 4 ; <<4 <4 < 4 4 < 4 < 4 /- 4 V 4 yi. 4' 4 4 , $: i . '4. 4 4 < tr-4:< 4' 4' < * <V 4 ' -4 j 4 j 411 4 4 4y C o v e r p h o to , c o u r t e s y o f N o r d s t r o m l 'a h e D i v , R o c k z v e ll M f g C o : .r ________ -- Ni ,, fcviliiii -- VOLUME 94, NUMBER 1, ABC ABP COPYRIGHT 1950 by McGRAW-HILL PUBLISHING C O ., IN C. ALL RIGHTS RESERVED PretuLlit Mcr?w (1840-1948), Founder; James H. McGraw Jr, President; Curtis W McGraw, Vice-President and Treasurer; Eugene Dufficid, Senior Viceand n i, . Fub,lcS.hon? Division; Nelson Bond, Vice-President and Director of Advertising; Joseph A Gerardi, Secretary; J E Blackburn Jr, Vice-President J L y; ' ,, D ? f Cl,r? !S * lon DISTRICT MANAGERS; New York, W W Quarles; Philadelphia, C R Long; Cleveland, N 0 Wynkoop Jr; Chicago, C Boughton. 330 w ?,0l t0nJ .J E Slater; St. Louis, C Boughton PUBLICATION OFFICE: 418 W est 25th St. New York 1. N. Y. EDITORIAL AND EXECUTIVE OFFICE, Fron)!c,,.ndL,?i_h,e.wJ fork !?'. N .Y. Coble Address: "McGraw-Hill New York DISTRICT OFFICES: 520 N Michigan Ave, CHICAGO I I ; 8 Post St, San inriov i-uuuuii ytl m r u i r c A a m c i e s a r e m a e x e a in d ojii m e in a u s r r i a i atts. x annually; copies are available on request to the editorial offices, 330 .W 42nd St, New York 18! P OWE R ",,,,January 1950 69 MECHANICAL SEPARATION ELECTROSTATIC DESIGN. 5 Flue gas can pass through lam inations, above; dust is pulled out narrow end Cone-shaped laminated-steel dust sep arato r takes flue gas in at wide end Horizontal flow unit has perforated plate electrodes; 8 3/4 in. centers 0' Here Are the Working Tools for By L N ROWLEY, Executive Editor And J C McCABE, Associate Editor Keeping your plant stack with in the nuisance limits of to day's smoke ordinance means intelligent use of the dust col lectors on the market. This confclusion of a 2-part survey helps you make your selection 82 Vi e began our discussion of dustcollection methods and designs available back in November P ow er, pages 78-81. We had progressed from the simple mechanical baffle and cinder traps to Ihe latest in cyclone-style collectors. There is still another mechanical removal method, Fig 1, 2. Separating element is a series of laminated-steel plates made up in a cone. This cone goes into a steel casing normally placed between boiler flue-gas outlet and in duced-draft fan, Fig. 2. Dust-carrying gas can then enter the cone's wide opening. Once inside, gas escapes through the laminations and proceeds to inlet of induced-draft fan. But to pass through laminations, gas has to change direction sharply. En trained solids stay behind in the cone under influence of the motion from the entering gas stream. As the dust particles approach the top of the cone they are caught up in the suction of a secondary fan and pulled out at the top. This fan delivers flue dust either to a secondary separator or to a recirculating system for return to the furnace zone. Electrostatic Precipitation. A widely used method of flue-dust removal, par ticularly for pulverized-coal furnaces, is the electrostatic precipitator, Fig. 3-7. This device performs several basic func tions: It (1) electrically charges dust particles traveling in the gas stream P O W E R January POWER Jf ,0AS LONG reco rd of high-e f f ic ie n c y c o llec tio n on la r g e -s iz e d b o ile r s p g f f ijg " '^ H igh-voltage electrodes G as flow I -G -r-o--u--n--d--e--d----e--l-e--c-t--r-o--d---e s 6 First stage, Fig. 5, builds up particle charge the next has uniform field to speed precipitation rapidly; of dust High-voltage Inlet bushing C o lle ctin g electrode rapper Gas out Series combination of mechanical and electrostatic col lectors has become quite important in many applications Flue-J)ust Collection-- Part II 12) drives them to a ground or collect ing electrode, where it discharges them 13) removes collected material from electrodes. Performing all these operations on a fast-moving gas stream is not easy. A high unidirectional voltage has to be et up. Opposite the high-voltage dis charge electrode, usually a small-diam eter wire or a twisted square steel rod, a grounded or collecting electrode. These are bare essentials, Fig. 3. For any given wire diameter and electrode spacing there is a critical eoltage. Above this voltage, an electrical discharge forms in the air or gas sur rounding the high-voltage element. A timhardment between gas molecules begins. Electrons are torn loose from gas atoi$)|)r molecules. Once loose, the electrons combine with other atoms to produce negative ions. These follow electrostatic fields of force to the grounded electrode. Any suspended dust particles in the gas stream become a target for these loose ions because the particles have a higher dielectric constant than gas or air. So the ions attach themselves to these dust particles and in doing so give them a negative charge. Once you establish a sufficient charge on a particle the electrostatic field force regulates its movements. Force exerted is proportional to product of charge on the particle and electrical-field strength between high-voltage and grounded electrodes. Let's say now that all dust particles in the flue gas carry a charge and are being pulled out of the gas stream. Next step is to collect the dust. Usually when the charged particles reach the grounded or collecting electrodes the charge leaks off to ground. Dust hangs on by mechanical adhesion. Some designers, Fig. 5 and 6, feel the jobs of charging dust particles and then collecting them should take place separately. So they divide the precipi tator into zones. High-voltage zone is designed to give maximum charge to all particles as fast as possible. Then the collection zone can be engineered for January 81 a uniform high intensity to give effective collection at reduced power. The higher the voltage applied, the more thorough the charge saturation on the particles and the stronger the elec trical field--all of which means the shorter the time needed. But there is a definite practical limit to how high applied voltage can go. Some of the limiting factors, in addition to electrode spacing, are temperature, density, chem ical composition of gas passing through collector. How do you go about producing re quired unidirectional high voltage? It has to be anywhere from 30,000 to 100,- 000 v. Besides a liigh-voltage trans former, input ac energy has to be rec tified. Here there is a choice of rectifier methods--mechanical or electronic. Mechanical rectifier usually consists of a disk of insulating material. On the disk are 90-deg conducting arcs set 180 deg apart. Four stationary contacts, 90 deg apart, are: one high-voltage pre cipitator cable terminal, one grounded terminal, a pair of transformer sec ondary terminals. Disk is motor driven at synchronous speed so voltage at the stationary high-voltage terminal is rectified. Electronic rectifier employs the high- voltage Kenotron tube. For large pre cipitators a 4-tube high-vacuum hot- cathode diode circuit is often used. Designs on the market, based on long tube life with certain circuit safeguards, assure good tube performance. Practical Aspects. Now that we have the elements of an electrostatic pre cipitator. how about practical angles? One of the most critical needs for good performance is uniform gas-velocity distribution through the collector. A small section, of a precipitator operat ing with a gas flow considerably above rating suffers because already-collected particles are swept back into the gas stream. Other areas handling a gas velocity below rating and turning in excellent performance cannot make up for this loss. .Am Suitable design overcomes two other inherent difficulties of electrostatic pre cipitators: their failure to collect or hold particles that (1) are large (2) have high carbon content. The answer to the large-particle prob lem is to: (1) allow enough time in the precipitating field for the electrostatic force to overcome the inertia of the particle in the direction of gas flow (2) design a collecting electrode to prevent the large particle from "balling up" with its neighbors and rolling along the electrode and out through the unit. The high carbon-content particle presents quite another problem. Here the job is largely one of holding the carbon particle when first precipitated. Since carbon is somewhat of an elec trical conductor, the carbon particle takes on its charge quickly enough. In fact, its saturation charge approaches three times that of a normal ash particle. But because the carbon is conduct ing, it behaves far differently from normal collected dust. Carbon loses its charge rapidly to the grounded elec trode. But the now uncharged particle is still under the influence of the highvoltage elements opposite. Result is the carbon particle gives up some of its own negative ions to the grounded elec trode. This action, charging by induc tion, leaves the collected carbon parti cle with a slight positive charge. This weakly charged positive particle feels the pull of the negative highvoltage electrode. So it leaves the grounded electrode and reenters the gas stream, where it comes into a field filled with negative ions. Once again the particle takes on a negative charge, returns to the collecting electrode. But the same charging-by-induction phe nomenon occurs and the cycle begins again. This zigzag path of the carbon particle may continue until it -passes completely out of collector. Solution for this difficulty---charging by induction--must be through some method of reducing influence of highvoltage elements on the collected parti cle. Reduction mav be either through a reduced field intensity in the collector element's neighborhood or a collector element that makes reentrainment phys ically impossible. The reentrainment problem is most aggravated when time comes to remove any collected material from the elec trodes. Today's power plants can't shut down just to clean out dust collectors. Yet the material must be constantly removed to permit top collection effi ciency. Both high-voltage and collector elec trodes take on coatings of flue dust. The high-voltage element, if a wire, has its effective diameter increased by the coating. This reduces the effective corona discharge and lowers operating efficiency. In addition, dust build-up in creases normal rate of flashovers within the precipitator. Just how to effect this removal is the subject of much research. In general, electrodes, both high-voltage and col lecting, are rapped. But methods, areas rapped and controls over reentrainment vary considerably. Proper vibration of electrodes can remove practically all dusts. So they must not be too rigid or too heavy in cross section. Any rapping must be timed so successive impacts do not in terfere with electrode vibrations al ready set up. This means automatic rapping, by air- or motor-driven rappers. Contactors or motor starters can be so set that no two electrodes are rapped together. Or dampers can be closed to isolate sections undergoing rapping. Fig. 3 shows the bare essentials of a horizontal gas-flow collector. The highvoltage ionizing electrodes are sup ported by insulators so there is no danger of arcing over to the precipitator casing. Further, these electrodes are accurately centered between collecting electrodes so peak voltage and peak 84 P O W E R Ja n u a ry 1950 IIBBERS. Steam supply j <ts 'team `halo \ 'Water curtain ' Sludge ou tlets Jbber directs ore discharge (left) has gas ubbing water the collector a collector : nment phys- lem is most :s to remove m the elec ts can't shut ;t collectors. ! constantly llection effi- Hector electe dust. The vire, has its ed by the te effective s operating build-up in fe rs within uoval is the In general, ,e and colliods, areas mtrainment trodes can :s. So they J o heavy in ; must be do not in rations al- automatic en rappers, can be so 1 ire rapped 3 closed to apping. :ntials of a The high- are suplere is no >recipitator trodes are collecting and peak lu a ry 1950 WELL IN WET COLLECTION OF GAS-BORNE SOLIDS |iri lunnance can be maintained. Should ilit- high-discharge electrode be closer to one collecting electrode than to its neighbor, there is danger of too fre quent arcing over. Thus, it is important that the support ing structure be stable. So single- and 2-point suspension is sometimes used by hanging weights on the bottom ends of tin- di-charge electrodes. These weights picvcnl swinging. Collecting electrode has several de signs. It must not buckle or warp be cause this defect reduces electrical clearance, causes the same trouble as a winging high - discharge electrode. Physical shapes of collecting electrodes vary widely: (1) rod-curtain designs made up of curtains of small-diameter pipes hung vertically in a horizontal- flow precipitator (2) corrugated-plate curtains of corrugated sheet iron (3) dual plates made by welding steel sheets to the two sides of a frame with retain ing holes punched out in the sheets (4) pocket electrodes with V-shaped strips hung verticaHy so the V's apex faces the gas stream (5) special high-duty col lectors using bent steel strips hanging from each side of a frame to make a dust spout between electrode's sides. Still another design, Fig. 5 and 6, has its pocket electrode employ a flat plate with a 1%-in. mesh screen spaced l l/2 in. from plate. This arrangement permits concentrating the electrostatic lines of force on the screen. As a result, eld intensity between collecting elec trode plate and screen runs much less than outside the screen. So the tendency ? car_hon particles to be charged by - "Jduction is prevented and their reen- ttamtnent stopped. ~ *r,es Arrangements. A growing dency combines mechanical with trostatic collectors, puts them in Ties, Fig. 7. There is considerable dis cussion over which should lead--me chanical or electrostatic. Here are per tinent facts. At the moment the strongest single point in favor of the mechanical col lector as a leader is that it works on the "whole" dust and at typical dust grain loadings. So its performance is as good as the mechanical collector alone. Then the electrostatic need handle only that dust portion on which it is most capable ---the fines. Further, during soot blow ing when grain loadings reach 200 to 300 gr per cu ft the mechanical col lector operates at about the same effi ciency. It again relieves the electrical unit, which then does not have to supply the required negative ions for adequate ly charging the heavier dust loading. One difficulty this arrangement, Fig. 7, creates is. emptying out the electro static precipitator hoppers. Dust is so fine it may bridge, but proper hopperdesign can correct this. With the electrostatic unit trailing the mechanical, material lost during rapping may be emitted from the stack. This is the strongest claim made by those who advocate putting the me chanical collector in the "back-up" spot. How important this consideration may be depends on the way the rapping is carried out and how much of a prob lem, if any, it creates. Gas Scrubbers. Fig. 8-10 show equip ment employing a very different system for cleaning up a gas stream. In these the dirt-laden gas is passed through an apparatus designed to trap suspended solids in a scrubbing liquid, usually water. In each scrubber, there are one or more basic principles: (1) impingement of gas stream against water droplets or wetted surfaces (2) centrifugal action to separate particles from the gas stream (3) a breaking up of gas into small streams. Separation then occurs. Fig. 8 receives gas at lower section of scrubber. The gas rises through a curtain of water created by a number of low-pressure sprays. These humidify the gas, strip it of coarse dust. Then the impingement plate (600 to 3000 holes per sq ft) and baffles above each plate receive the gas. As the gas goes through the perfora tions it speeds up and at the highest velocity it strikes the water-shielded baffle. Here the greatest cleaning action takes place. All suspended matter tends to travel in a straight line and impinges against the under surface of the wetted baffle. It is then trapped in the scrub bing water and carried out. Fig. 10 illustrates a newcomer to the gas-scrubbing field -- the venturi scrubber. It offers a means for inject ing a low-velocity scrubbing liquid into a high-velocity stream of dirty gas. The gas impinges upon and disrupts the rather large jets of liquid so the drops of liquid accelerate and disrupt still further. Some liquid atomizes, at least briefly. Then the gas and vapor mixture slows down and the two are separated in an eliminator, Fig. 10. Four collection methods are possible in the venturi setup: (1) collision (2) diffusion or molecular bombardment (3) electrostatic effects (4) condensa tion. Collision and diffusion, the manu facturer feels, are the most important. Relative velocities of suspended parti cles to the water droplets go from 12540Q sips at fir-t impact. This drops rapidly as the water accelerates and opportunities for collision are excellent. Downstream in the expansion section the gas slows down. Water droplets may for a time have higher velocities than the solids, which increases opportunity for collision. 1lie droplets coalesce and much of the liquid leaves the venturi as solid water, the rest as fog or vapor. As soon as gas impinges on the liquid, and right afterward, resulting turbulence produces 5 to 25 sq .ft of droplet sur face per cu ft gas. This makes collection by diffusion an important part of over all performance. Further, these liquid droplets are known to be charged; hence electrostatic action plays some part. Although no installations of this equipment have been made on boiler flue gas, the manufacturer feels such an installation can be made economically. Fig. 9 represents a relatively lowcost method for scrubbing a boiler flue gas. Here water runs over a plate through which the gas passes. The water then continues over the sides, making a fairly solid water curtain, within which the entering flue gas is channeled. A steam spray is introduced near the outlet to overcome draft loss through the collector. , 0 VVER Ja-nuary 1950 85 tK' t* ; Iiti. I te: Ik. P W SWAIN Editor L N ROWLEY ExecutiveEditor F A ANNETT AssociateEditor E W F FELLER Associate Editor J C McCABE AssociateEditor B S A SKROTZKI AssociateEditor J J O'CONNOR Assistant Editor C F SANDERS AssistantEditor C F MARSCHALEK ArtEditor SHELTON FISHER Publisher W J HARING SalesManager W a sh in g to n : GEORGE B BRYANT JR A tla n ta , EARLE MAULDIN WORLD NEWS OFFICES: L o n d o n , P a ris , B e rlin ,. M o sco w , Tokyo, Melbourne, Bombay, Rio de Janeiro, Buenos Aires FEBRUARY 1950 VOL 94 NO. 2 plant o p k a S gT S S m s , Mi'.miNANa > m T S r STEAM GENERATION, FUELS, FIRING AND COMBUSTION TURBINES, ENGINES, AUXILIARIES; HYDRO POWER ^ 'V. \ S-Xif si. v,T.'^..*v'\s4vs^'vfssV'i, ss S-^ v.' , A'A/vl> ' Cs vf- xi vt"*. J-.wi-.iv.' to:' mm DIESELS, GAS EN GIN ES AND A CCESSORIES AIR CONDITIONING AND R.E..F..R..I.G.ER..A..T..I.O...N... ii ELECTRICAL EQUIPMENT, ELEVATORS sv i s s ' ' s P IP IN G , S.SS , * V . VALVES ^ S. . ' AND , S P E * C I - A * L TI E S ' ` 1\ 5 ?! fvtV TRAN SM ISSION , LUBRICATION STEAM S ER V IC ES, HEAT|NG INSTRUMENTS. CONTROLS PUMPS, COMPRESSORS Hk MEASURING ELECTRICITy....................................................................... 73 Instruments for measuring current and voltage ............................................... 74 Measuring resistance. .................................................................................... 78 Power, energy and demand metering............................................................ 80 Power factor, synchroscope and frequency meters...................................... 86 Recording instruments........................................................................................ 90 Instrument selection............................................................................................ 92 Silica removal by ion exchange........................................................................... 93 Huey gas turbine ticks off 3000 hours.............................................................. 96 4 Diesels in hydro country save stored water, purchased peak power.............102 L i Barbadoes Station-- a picture report.....................................................................105 tej| Letters from a turbine specialist.............................................................................109 Ej aper-plant modernization proves that survey pays...........................................110 N Insulation handbook: 1 -- How heat^flows....................................................... 114 f-Y" N i 4 i r i ? H 'i ! s ji. .1? r4 (, I 4 ? -1 !4 k . \ U j Wood firing in pulverized^coal units...................................................................117 u Ninemile Point -- new outdoor plant................................................................. 118 \ 4 Electronics: 22 -- Oscillating-circuit applications........................................... 122 Power engineer's notebook of plant-tested idgas-.^. . .....................................126 Problems from the Plant. 130 Plant-equipment news ........ 146 | ;r Technical briefs..................... 134 Engineer's bookshelf . . . . . . .154 H | News of the power field . . .140 George Edwards............... .........156 111 Data sheet.............................. . 1 4 4 Reader's service page.. . . . . . .192a p H 4" N ;' 4; I I 4 4 4 Ni 4; 44 :4 .- 4 -- u .' I N k 4 ;4 4 i4 .4 ' 4 i ' ' 1r. ,-t.... VOLUME 94, NUMBER 2 ABC ABP COPYRIGHT 1950 by McGRAW-HIlL PUBLISHING C O ., INC. ALL RIGHTS RESERVED Aldwych House, Aldwych I-------- . . -- _ . ------------ . index an nually; copies are available on request to the editorial offices, 330 W 42nd S t, New York 18. P O W E R February 1950 69 RADIATION Radiant heat waves, exactly like radio waves but much shorter, travel through air or vacuum with the speed of light CONDUCTION Heat flows through a solid by conduc tion only. Fast-moving molecules at hot end speed cold ones by bumping process CONVECTION Hot air is lighter than cold, so air heated by radiator at left rises, then cools and falls again to floor Insulation Handbook: Part I This handbook, to be published in six parts, starts with the fundamentals of heat flow. Later articles answer practical ques tions--What kind of heat insulation meets specified conditions? How thick? What saving? How applied? How protected? By PHILIP SWAIN, Editor T h i s s e r ie s serves as a condensed handbook of heat insulation for power engineers. It will help them save heat units and dollars by the intelligent selec tion and application of helit insulation to the hot and cold surfaces of piping, tanks and many other types of powerservice equipment. Before turning to this practical side, here are a few scien tific fundamentals. Heat Flow. Throughout these articles we shall measure heat in Btu, but re member that heat can be measured in other units as well. For example, one kilowatt-hour is 3413 Btu. Therefore, one watt hour is 3.4 Btu. Thus any insulation completely surrounding a 40-watt electric lamp bulb will necessarily transmit 3.4 x 40 -- 136 Btu per hour after the lamp and the insulation have warmed up to a steady state. And this will hold true regardless of the thickness or efficiency of the insula tion, since the lamp is a constant-rate source of energy. Heat Travels Through Vacuum. Heat can traverse a vacuum, as from the sun to the earth, in one way only -- by electromagnetic radiation, exactly like radio waves and light waves in kind, except that the waves are much shorter than the shortest radio waves. All speed through space at the rate of 186,000 miles per sec. This radiant heat >will pass through air, as it will through a vacuum, but it has no need for the air. You might say radiant heat flows in spite of the air and loses something through absorption by the air. Heat Flow Through Solids. Just as heat can flow in one way only (by radiation) through a vacuum, so also there is only one way for it to flow through a solid--that is, by conduction. In any solid, at any temperature above absolute zero (--460 F), the molecules are in constant vibration. The higher the temperature, the faster they vibrate. Each molecule bounces back and forth within a limited range of its normal position, as though it were fastened to a stake by a rubber cord. Each moving molecule is constantly bumping into its neighbors. Naturally, when fast-moving molecules bump into slow-moving ones, the fast are slowed down and the slow are speeded up, on the average. That is how heat flows by conduction from the hotter parts of the solid to the cooler parts. Thus tempera tures tend to equalize throughout ffr solid object unless heat is constantly supplied to the hot points, and taka away from the cold points. Convection. Convection, the third, way to transmit heat, commonly occurs in liquids, gases and vapors, but nevin solids. In convection the .liquid or gas ( vapor) actually moves from the surface, where the heat enters, to cold surface where it is removed. Th" the heat is carried bodily from one po! to another. Of course, convection always involr some conduction also, at the poin where the h%i|t 4& "loaded" into a "unloaded" from the' moving fluid. This convection movement may produced "naturally" by temperatu differences, or by a pump or fan. air circulation set up by a radiator a room (see sketch above) is a C` mon example of natural circulation, is the circulation of water heated by-; submerged steam coil. In both cas. the heated fluid expands and then rr because it is lighter than the surro ` ing air or water. Later it cools, beco heavier and falls, thereby comple the convection circuit. Much higher fluid-flow rates are tainable with forced circulation by or pumps, and much is gained by increased flow. In fact, the flow of increases much faster than the flo gas or water. Not only is more circulated, but its high velocity away the surface film, and thereb duces the resistance. That is why fans are widely us industry to blow air over heating and pumps to push water or 114 PO W ER February FORCED CONVECTION SPEEDS DP HEAT TRANSFER Forced convection, by fans or by pumps, scrubs away much of the insulating fluid film that retards the flow of .heat, thereby greatly increasing the heat-transfer capacity of steam-coil air heaters, liquid heat exchangers and similar equipment How Heat Flows liquids at high speeds through heat exchangers. In this way a given invest ment in heating surface can transmit far more per hour than would otherwise be possible. Most of the practical applications of heat transfer and heat insulation in volve a combination of two or three of these methods of heat moving--conduc tion, convection and radiation. The role of true radiation in heat transmission is often misunderstood. It may be important, or it may be un important, depending largely on the temperature levels involved, because the heat radiated from the surface varies as the fourth power of the absolute temperature. The rate at which these fourth powers build up is truly aston ishing (see sketches at right). There is no such simple mathematical rule for heat transmitted by convection. True, figuring heat delivery would be easy if we knew not only the rate of air circulation but also its temperature leaving and returning for a new load of heat. In that case heat moved would be the weight of air moved, multiplied by the temperature difference and by the specific heat of air. The practical difficulty is that of pre dicting just how much air will flow (in the case of natural circulation), and (in all cases) what its initial and final temperatures will be. The field of convection heat transfer is so full of theory and guesswork that the average practical man is often forced to solve his heat-transfer prob lems by cut and try. He adds or sub tracts heating surfaces until he ob tains the desired heat-transfer rate. But we must not jump from this fact to the unwarranted conclusion that it is difficult to make a reasonably close estimate of the heat loss from insulated surfaces. The opposite is true. The reason is that film resistance on the steam and air sides of a metal plate, as well as the resistance of the plate itself, becomes less important when the sur face is insulated. For example, the heat loss from a hare healed surface may vary consider ably. depending on the position, and on the nature of the surrounding air currents because both have a big effect on the outside air film. Yet as soon as the insulation is applied, air currents and position become relatively unim portant (see final illustration on p 116). That is why the heat loss through in sulated surfaces can be accurately fig ured from the known properties of the insulation. . . .-. y... Conduction. In practically all cases where insulation is applied in substan tial thicknesses, conduction through the insulation itself is the controlling factor in heat transmission. Here, fortunately, the theory is sim ple. Consider (sketch, top p 116) a square slab of a given insulating mate rial, 1 ft square and 1 in. thick. As sume one face is one degree hotter than the other. Then heat will flow from the "hot" face to the "cold" face, and its amount, in Btu per hour, will be the so-called K factor, or conductivity. Once the K factor is known for a given material, it is easy to figure how much heat will be transmitted through any thickness of it over any area and with any temperature difference. Total heat flow will be in direct proportion to P O W E R February 1950 2500 abs 3 0 0 0 abs 625 1296 RADIATION SKYROCKETS AS FOURTH POWER OF ADSOLUTE TEMPERATURE For example, take 500 abs (40 F) as standard of comparison. If you dou ble temperature to 1000 abs, radia tion jumps to 2 X 2 X 2 X 2 = 1 6 times. At triple temperature (1500 abs) radiation i s 3 X 3 X 3 X l 81 times that at 500 abs, and so on, as indicated above. That is why radi ant heat " is far more important at the higher temperatures 115 A LL INSULATION TRAPS DEAD AIR IN TINY POCKETS CELL ACTION 00000 0oo0o 00000 0o0oo 0 0 0 0 . ..O. 000o0 H eat flow----------------------> - CRYSTALS FILAMENTS Fur SHAPED CELLS A /V V /v A A g g g & Sg g i* /V W \ A / < S F e lte d PAPER LAYERS DEAD AIR SPACE DOES THE JOB in most types of insulation. P rim ary path fo r h eat tra n sfe r across an a ir space is by con- vection. If air spaces are tiny, and there are many, cot vection curren ts c a n 't really get going and resistance is hig Insulation Handbook: Part 2 Types and Shapes of Heat Insulation Here Editor Phil Swain explains how insulation does its job, lists data on forms, temperature limits. Future articles will cover heat loss, economic thickness, selection, application T h i s a r t ic l e e x p l a in s why certain materials are good insulators. Then it briefly describes materials and shapes in common commercial use for insula ting heated surfaces. W h y In su la tio n In su la te s. It is gen erally understood that insulation de pends on "dead air," but just how dead air works is not always clear. Heat flow through air is mainly by convection be cause air is a poor conductor. That is why every good heat insulator must: (1) lock air in small pockets so con vection can hardly operate (2) leave only very little solid material running in lieat-flow direction. The sketch at top left shows the ideal condition if the cells pictured are as sumed to be microscopic in size, and the solid walls running in the direction of heat flow very thin. View each cell as a hurdle the heat must overcome. At each cell the heat, after passing easily through ..the left thin wall must pass to the air in the cell, then circulate to the right wall by natural convection, and there give up its heat. Repeating this operation many times per inch of thickness can build up a great total resistance to heat flow even if the material itself (as with aluminumfoil insulation) is a very good heat conductor. For aluminum, note that the bright reflecting surfaces are additional important barriers, this time to radiant beat. Stiff or Efficient? Most commercial insulations are a compromise between insulating qualities and other qualities needed for specific applications. For example, where insulation is com pressed, or binder added, to increase strength and stiffness, some insulating efficiency is almost always lost. In general almost anything to in crease stiffness and strength (other than some kind of protective shield) in creases weight, reduces insulating value. Thus insulating brick and "hardfinish" asbestos cements have lower values than the usual 85% magnesia and mineral-wool insulations. Every form of insulation has some upper temperature limit beyond which it cannot be used without shortened life o.r early destruction. Thus insulation of hair and wool is rarely used above 200 F. In ordinary "air-cell" asbestos insulation the binder often sets a safe upper limit of 300 F; 85% magnei calcines if applied direct to pij hotter than 600 F. Even asbestos, oft viewed as a near refractory, lo: strength far below red heat. Table on facing page roughly clai fies insulations by substance and sha Complete consistency isn't possible this sort of table. To avoid confusi and reduce generalities to a stric practical level, future articles v consider properties of magnesia, m eral wool and oLlier insulations--a methods of applying them to varii surfaces and equipment types. Also, special problems in insulati cool and cold-Surfaces will he covei in a separate article. Here are a fewr high-spot commen Organic Fibers. Fibers of hair, a of wool, can be felted (loosely int laced). This felt can be formed i sectional pipe covering, blocks, or it sheets pliable enough to fit over cun or irregular surfaces. This organic sulation should not be used at tempe tores above 200 F. Asbestos Air Cell. This low-cost sulation (layers of corrugated asbes paper) is used chiefly on domes steam and hot-water lines--rarely permanent industrial lines. Seldom u above 300 F. Other Asbestos. Asbestos fibers n be "felted" to form sheets and bio and pipe covering. Any of these sha can be made stiffer and stronger adding binder, but always at the 118 POWER March I COMBINATION INSULATION MINERAL WOOL uny, coiv -e is high for steam pipes above 6 0 0 F, diatom aceous-silica inner '!\er protects outer of more efficier.: 35o magnesia insulation MINERAL W OOL is supplied in m any form s, including boards, left; loose and noduled wool, right, b lan kets, b atts, cem ents magnesia f to pipes istos, often ory, loses rhly classi-,, and shape. ~ possible in confusion a strictly tides will lesia, mintions--also to various ;s. insulating be covered comments: E hair, and osely interormed into ;ks, or into over curved organic in- at tempera- cost in*-, asbestos j lomestic rely fc_ om used ers may [ blocks# iic-n.-t- of some loss in the etlicicncy. The same holds true of asbestos cement. The greater its cementing qual ity (hardness, strength, weight) the poorer its insulating quality. Thus soft cements of high insulating quality are commonly given a thin finishing coat of hanl-finish cement, applied mainly as a protection. Mineral Wool is made by exposing a thin stream of molten slag (or glass) a high-velocity siearn jet. Primary product is a woolly mass of tangled mineral fibers. In this form--poured loose, laid out flat as a halt or blanket la layer of mineral wool sandwiched between wire mesh or other supporting layer)--it lias high efficiency. This effi ciency is somewhat reduced when ma terial is strengthened and stiffened by compacting, felting and by adding hinder to form boards and blocks. 85% Magnesia, a nonproprietary induct, consists, of 15% by weight s-be.-tos fiber reinforcement and 85% carbonate of m ag n e sia, c h e m ic a lly 'hi'hed from dolomite rock. Insulating 'able comes entirely from microscopic air spaces between magnesia crystals. because of its character 85% mag nesia cannot be felted. It is formed by "et molding. Commercial forms are 'cctinnal pipe covering, blocks, seg ments and cement. For pipe temperatures above mag nesia temperature limit of 600 F in sulation is applied in two sectional ayers inner layer of temperaturenesisting diatomaceous silica, and outer 0 the more efficient 85% magnesia. High-temperature diatomaceous silica usulations are made up to 1200 F-- even up to 1900 F. Insulating "rfracries are obtainable (for backing urnace firebrick) for temperatures up 'o 2500 F. March 1950 119 I J ( S \J f Is k W V, V ' A blending tank containing a liquid a t 300 F. is located only 4 /r from a w all. T h e ta n k is to be covered w ith sta n d a rd thick 85% M agnesia block insulation. How would you cover the back of the tank in this lim ited working space? IkL / %. r , \ I \u v r i `t \ 4 U U P . v i i: I HERE'S WHAT THE ARMSTRONG ENGINEER RECOMMENDED: stro n g engineer. T h e nex t tim e y o u In this case, the tank was com pletely set up, with all pipe con nections m ade, when the insulation men were called on the job. It was im practical to disconnect all pipes size as the back of the tank. Then an angle was welded to the bottom of the tank back, and the insula tion was slipped in place. The angle supported the weight, and need a heat insulation job done, try A rm stro n g 's com plete C o n tract Service, for dependable engineer ing advice, quality m aterials, and expert workm anship. and move the heavy tank far wires were used to hold the insula enough from the wall to get behind tion firm ly in place. s FR EE B O O K L E T it and apply the insulation. The T his problem is typical of the Send today for the new booklet, "A rm f|; 300 te m p e ra tu re h eld in th e ta n k m a n y q u estio n s t h a t arise on a n y strong's I n d u s t r i a l was critical, and om itting the in heat insulation job. A rm strong en Insulations." It de sulation on the back would have m ade tem perature control difficult. To solve the problem of applying gineers and installation mechanics experience them every day. T hat experience has taught them the scribes c o m p le te ly the m aterials and services offered to users of insulations the insulation, the A rm strong en answers to alm ost any insulation for all tem peratures. ARMSTRONGS gineer recom m ended th a t blocks of 85% M agnesia be wired to ex panded m etal lath, cut to the same problem, large or small. W hen you h av e an in su latio n q u estion y o u 're in doubt about, consult an A rm W rite today to Arm strong C ork Co., 7003 M aple Avenue, Lan caster, Pennsylvania. INDUSTRIAL INSULATIONS PO W ER March 1950 165