TfETi UNTIVf2RSi TY OF MI(:TCI-iGANT COllege of'Tg'ineering Depaxrtment of ilecharnical lgineorine Cavitation and i,.ul-tifhase jFlow Laboratacy Report NTo. UICTH 01357-29-I E~r. o p e an T~ r~ R e o r *t F. Go, Ha:i_~' t t Eui-Co~r~en~n rr i n Reiolt - g, 0, T-) l(tft Ye G. TaeT rmn-iLt Professor3 teareharge Cavjitatison an~ld 7I~tulth-Lhsie Flow L-abt I}8vlLortoBr Mecharii c al o ri,;d eeriLs e-FarI+en t Univ e si ty of IIihi Ann Arbor, C ieicrgat - S1 0 3 October 26, 1974

ATL- Trip Report - 1 - 12 74 I INTRODCJ CTI OJ present This report reviews the/state-of-the-art and probable future course of sodium cavitation research with particular application to FBR programs; in Europe and Britain as understood by the authors The to a large extent information has been gleaned,'c/ from conversations ald other unpublished documentation resulting from the authozis visits during the past summer to atomic energy laboratories in France, England, and Holland, This extended visit in Europe was primarily as a Fulbright to the French atomic energy laboratory in Grenoble (CENG), as well as a week spent at the French laboratory at Cadarache. Ijater and shorter visits were Ezii also made to the fast reactor pump groups at Risley and at Neratoom (Holland). The contact with the French has been extended over a period of 5 - 10 years, starting with a Sabbatical 6-month to Electricite de France (EdF), a/leave to the French private hydraulic research company in Grenoble (SOGREAH), and finally the Fulbright to CENTG. In addition for several years my laboratory at Michigan has been engaged in a series of sodium cavitation tests under contract from Cadarache. These tests are now complete,and have examined the So,{ C,.'' effects upon damage rates of varying/temperatures and pressures on a variety of materials of interest including SS-316. The final report is not as yet written, and the data is not yet cleared for release, In connection with this work primarily I have been a consul-tscnt to the French AEC for several years. The present report vill cover the pertinent work in all these countries as I know it, and will list published references where possible. It will start with the work in France)since this is the most extensive. +~~~~~~~~~~- C~.tt t'>rd alo <C t

ANTI Trin Report - 2 - 1G-I 0/26/74 F~ u AS c LCAJIoe-lf-*/&#~ C'c(-L o (P __ -_ __._.__ ~zc. A. French Sodium Cavitation Research 1. General Bcround In both France and England, the FBR programs as compared to that in the U.S., has advanced rapidly toward the production and luse of t'demoureactors", so that, in my view, ~ highly practical problems such as cavitation acheired early prominence and attention. As far as serious research on this problem I know,~kix/seems to have started probably sooner in France than in potential Britain. In any case the/problems vwith cavitation lie primarily under the following headings: coAld a. Cavitation ITcention Can ordinary/\,ater tests on pumps and other components adequately predict inception performance of the sane componentsi in reactor temperature sodium? b, Cavitation Dane - Can limited cavitation He tolerated in a sodium reactor system without prohibitive damageas it normally is in water systems? c. Acoustic Perf-ormanuce - In some rR FBR systems (such as French Rapsodie and Phenixy and British PFR) it is expected that acoustic instrumentation around the core vill be utilized to detect boiling as V an important part of the safety system. Since the acoustic signature of sub-cooled boiling and cavitation bubbles is essentially very much the same, the use of such acoustic boiling detection pre-supposes the complete absense of cavitation during normal operation. This imposes a much more stringent requirement on the design of pumps and other types of components such as flow restrictions than has been faced. previously almost in other flow systems, andmc in fact may be/beyond the present state of the art. 2, Electricite de Franlce and Unive t of Michiian Pro'ramavtion itaf~ - /tests have been conducted with tunnel systems, mostly on venturis, in our lhbbratory at the University of Michigan, in both water and mercury, to determine the of cavitation inception sigma Ad,'~

such parameters as velocity, Reynolds number, diameter, temperature, starting in the early 60(s). gas content, etc/ This work has been extensively reported 41, eg.). and hence provided a basis for comparison for later French tests using precisely the same venturi geometry. Tests were thus made by EdP in choperation with Cadarache, and ourselves in providing the detailed drawrings of the venturis used and the data obtained, The French tests were in both vwater and sodium, the data including the dependence upon vrlocity and Reynolds number i~, agreeing quite closely with our ovmne This agreement was only for the cases of air-saturat led water, which atas,whiere air content was not o-.:c'tt(kcX-t <Ne the situation in the French tests/ Our data, including a wide varia-tior in air content, showed a substantial difference in trend wvi't-h velocity betwceen low and high air content. The pertinent curves are here attached for convenience, and were also attached to my letter to Dr. Hoppenfeld of October 12. The following major conclusions result from this work in my opinion. a. Inception sigma depends substantially on velocity and/or Reynolds number,and there is considerable experimental scatter for both fluids. However, water and sodium fall wnithin the same scatter band. The variation with velocity (or Reynolds numinber) ix m for either fluid is more important than any difference between sodium antd water. b. From an engineering viewpoint all the scatter and variations mentioned above are relatively small, in that the design of-s4 components such as Tpumps would not be substantially affected. The results of the above tests are reported in various references in theopen literath=e starting in 1971x (4-6,eg.), including a 1972 AS3E papery (4). 2and similar ones at Cadarache, A summary of the result of these tests/from the FTench viewpoint is found in refo (7), 1970 ASI3E Cavitation Forum. To quote, "Presently available results seem to indicate no difference (between water and sodium), withiM measurement accuracy, but other tests will be performed to check whether factors such as argon content in sodium, its purity,

AITTL Trip Report 4- GH - 10 2_ and its temperature level may influence inception of cavitation." These later tests at Cadarache will be discussed in the followring. 3., Cadfarache Sodiiuma Cavitation Research Approximately parallel time-wise with the venturi sodium and wvater cavitation inception tests of EdF with the U-LI venturis discussed (late 60s) above, a similar program was undertaken by Cadarache. Bakh Neither EdF nor Cadarache tests involved the building of special loops for cavitation studies, but rather the attachment to existing large loops a of/smaller cavitation bypass loop, into which venturis, orifices, etc. could be installed. In all thes- cases (as in our own mercury tests) cavitation inception wras determined acoustically,, Dio, being the only feasible method. The Cadarache tests included orifices and other components as well as Venlturis, and I believe the comparative water tests were done in well-established water tunnel facilities at SOGREAR in Grenoble. These tests also included investigating the difference between argon and helium as pressurizing gas, and surprisingly it was found that there was a substantial difference betw-een inception sigmas obtained with these difrent gases. This may be due to different solubilities of the two gases in sodium. It is somewhat, reminiscent of results obtained with an MI pump loop at Oak Ridge (8), There different indeption results were obtained depending upon whether argon or vapor were used for pressurizationo( >''%tt Cci) sodium and water To my knowledge the results of the above/cavitation inception tests at Cadaracheat have not been published in the open literature. In a letter to myself in 1970 (in translation),. "it seems that the threshold of cavitation for an orifice depends upon the operating conditions of the loop", i.e., the type of pressurizing gas, and perhaps other factors of which I am unaware. In any case, these results were considered sufficiently unsatisfactory and unsettling, that Cadarache

Am r e 5 -T]_ - 10/26/74 decided to build a new large sodium tunnel especially for the investigation of cavitation (designed by SOGREAH I believe). This facility is now complete, and its operation is expected to commence this autumn., It will be used for both inception and damage tests, It is expected that entrained gas microbubble spectra. will be measured acoustically in a cooperative program mwith the group at Risley. Their work will be described later, The Risley instnrumentation for determination of entrained gas conteznmt in sodium- has already been used inl a cooperative endeavor rith the French on a heat transfer boiling loop at C:ETGG (Grenoble), and the en.traiLned bubbles were successfully measured. These results have not yet been published to my knowledg;e. 4. French and U-1,1 Cavitation Damage Research In the IU, S,)liquid metal cavitation damage tests in vibrator_ y facilities have been made by NASA (Lewvis Research Laboratory) ( 9 ), at Hydronautics Inc. (10), and by our ovm laboratory (11, etc.) over the past decade, Tests have been vwith lithium, ~lead-bismuth alloys and mercury (11), sodiurm i and mercury (9), and sodium (10). Iriuch of this early work was financed by NASA for the SNAP-50 program. Sodium otI.cali liquid metal and /cavitation damnage tests were also made in the U.S. by CALtIL~ Oa Ridge, recently This work is summarized in a report by myself written/for ANL (12) and in the open literature (6). The only recent sodium cavitation damage tests to my knowledge are those contracted to ourselves by Cadarache, conducted in a vibratorfacility over ranges of pertinent temperature and pressure. These results are not yet cleared for publication, but hopefully will be in a few months. However, the general trends found are similar to those also found by NASA (9), and generally applicable to all fluids (sk eg). For all fluids, damage rates decrease very substantially at high temper~ ature (even though materials may also be substantially weakened) due to "thermodynamic"t restraints which become operative ai high vapor

AI\L TL rT-n R eo.rt -6 FGHH 10/26/74 densities existing at high temperature. As a result, cavitation damage pumaps or other does not appear to be a very prevalent phenomenon in/Ai&M;: comoonents cold water handling a-ter mander PTl or B.VR conditions, as compared to/purips. However, the results presently available do not indicate that the same reactor temperatu re vwill be true of/soditn components, since reactor temperature sodium muc h is/more nemarly comparable -to cold water than to P`,R temperature water, 5. French Acoustic 71ork I believe, As previously mentioned the Rapsodie and Phenix reactors have,/ acuas tic been inst~umented, as part of the safety systems, for the/determination of boiling in the core. Such instrumentation, however, probably cannot distinguish with certainty between such sub-cooled boiling and cavitation. I believe that cavitation has recently been detected by sudih near instruments/~ the outer fuel elements of the Phenix core, and apparently the pumps do in fact operate h te: wvith complete absense of cavitation. $ ~ It may be recalled that cavitation also occurrer in the seals at the bottom of the fuel elements in the Fermi reactor, and did produce considerable pitting after relatively short exposure. In this case the stainless steel us. inserts were then replaced by ever Etellite, but no later observation of the stellite was/possible. As a result presumably of these acoustic instrumentation requirements, an acoustic group has been built up at Cadarache, and has worked in the development of instruments for the measurement acoustical ly of entrained gas in sodium (or other Tfluids) by measuring either velocity of sounda or attenuation effects due to the gasA Ahy e of course also beae active in the development of high-temperature transducers for use in sodium. Similar developments heave - also been made at Argonne and at Risley (to be discussed later). The French instrumentation for entrained gas measurement does not seem equal to that developed at Risley, so that the Risley instruments have been used in cooperative programs in France, and I believe this cooperation is now continuing.

MiA Report - /26/74 TqB3 Sodium Cavitation Research at Risle 1. General Background The writer visited the group of Mr. C. Te Booriman at Risley in September, ~A 1974 after the I. Mech. E. Conference on Cavitation at Herriot-Y7att University in Edinburgh. I have been in x3m fairly close contact with this group ford about 5 years since the visit to pertinent U-M of Mr. Boorman, and hence have been exchanging/reports wilth them about since/that date, which is probably about the tile of the commencemen:t of a serious effort on cavitation pertinent um the FBR program at this location, Their effort at first was limited to cavitation in stationary components as venturis, trifices, etc., and involved very careful work in a nwater tunnel vrith highly sophisticated acoustic instrune.tation for the counting of individual cavitation bubble collapse pulses, as well as the develop-ment of a system for measuring the entrained gas size and population densities. The initial work in water has later been compared with sodium cavitation results nm the same components. This has lately been extended to pump tests. In general I believe they have done th most sophisticated and successful Tvokk in the field so far available. As a resultsthey are now able to"commercialize" their instrumentc and use them in joint programs,,vith the French. I have in fact recently determined that they are willing to sell the instrumentation for entrained gas microbubble spectrum measurement to oursel; 2. Cavitation Tests in Water NEXT XX tMr. Boorman's group at Risley (Engineering Technology,consists of at least Messrs: R. F. Taylor, C. Betts, and A. Collinson, and has develosped a vwater-tunne 1 facility for the cavitation-inception testing of various stationary components. They work closely vith the Acoustics group of Mlr. Burton, which includes Messrs: D. Gray, sS McLeod, MdcKnight, and others. The cavitation inception measurements

AI-'kJ.j'Tj. enor. t TFT. - 810/26 74. are fnow made both visually anc-d acousti.cally. The acous-tic apparatus at present involves th;e couming of individual bubble collapse events, as shown.m i.n Figs w;'attach.ead At constant velocity, for exa.plRe i.t is found thatle as ITPSH is reduced from a very h.Jigh yalue, the numnbe.r of such counts per- second incameases a-t; a fixed rate aas tt'Db conven-ti;on.1al. ca.vitation incepitio:n point is approacchedo,-Tlen ",visible inception" is reached) i.t is fowund that the slope of the cou.ats vs. fS.__'H cuxae o.ncrea ~; p:roxima t ely /as a s-tep t-o a hic,,her antd constant vallue, Thus a very precise definitiono.of cavita-ation inception is afforded. This sezane approach has been used at the University o'f r I:tinnesota:irn water tu-nnel tests (1) To my knowledge this -work at Risley has not yet been 2published:in the open li-terat.ret, In addition, acoustic methods for measuring entrained gas microbubble content in liquids v,:ere developed by the Risley group using bo-th changes 5in velocity of soud d said sound. scat-terli, effec-t,s, Agai n no t;,.ii.. is apparelntly published in the open literature, and they are not will_.n,to disclose the det-ails of -these instruments, since they are cons:i dered. to harve commercial po-t)en tial and hlence are.m*optbietaryo As allready mentioned, theye instLrunents cani apparently be urcha..ed or u.se in our ovia laboratories_ As already mentl;ioned/they are bei.ng,/;uasecd ia cooperative p.rograms with the F,'rench at Cadarache and at Grenoble (GLTG) for boiling and cavitation tests in sodiruam Total gas con-tent is also measured in a conventional manner in th'e water cavitation tuar.nel al Risley., A very effective deaereator is used in conjunction vith this loop, so that total air con'tent is about 10-6 by volumae (saturation is about 10-2)o Even so. there are apparently enough entrained particles that cavitation occurs with relatively conventional sigr-a valuese Thus conventional deaereation.even dow!m to very low valueswill apparently not appreciably suppress cavita'tion in water (or perhaps in sodium), Of course no data to verify this is as

3._ PPR_ Pump Development and Sodium Cavitat;ion After development.of the acoustic techniques for bubble pulse measurement and entrained microbubble spectrum measurement at Risley, the same techniques have been applied to sodiuj~m cavitation tests of some componen-ts such orifices, etc., and are being applied, according to my understandint,to m testsm of the PFR pumps (both in water and sodium)o A description of the pump tests in general has been published (14), but no detailed results concerning t.hbble. collapse some of these counts and entrained gas content spectra as yet. I believe/bhm tests and data reduction are still in progress. Data from these testst and also those at Neratoom for thh the Germsan-aDltch work on SNTR-300 (to be described later),can be used, already along iath other data sources/available, to generate a normalized curve comparing acoustic, visible, and head fall-of-~ ca1vitationr_ inception poJints for centrifugal pumps, toe be used perhaps to speciify and laboratolry vendor/puvmp tests and design for reactor sodium pumps. A crude foim of such a curve,was attached to my let-ter to Dr. Hoppenfeld of October 12, 1974. Hopefully it will be possible to produce a later documxent data incorporating all available/z-a=ir~' into such a curve, and from this specifying probable cavitation-free NPSH for given cases from the standard Head vs. ITPSH curve. Additional >t sophisticated cavitation pump work to support the group at Risley is being continued at National Engineering Laboratory (NEL). The group at ITEL (Dr. I. S. Pearsall) has long experience with, and are world leaders in, cavitation pump ~ research and design (15, 16, eg.), To quote Dr. Pearsall (17) concerning pump cavitation inception scale effects, i.e., variation of inception sigma with model size, speed, etc~; "On cavitation inlception there is so little information that no reliable conclusions can be drawn. There is some evidence that air content alters the trends of scale effects. Mcost of the tests

ANL TriE Renort - 10 - FGH - 1__L2 6/74 on pump perfolnnance breakdoiiwn have shown an improvement wvith higher speeds and larger sizes ( - -- iee., higher suction specific speed). Some tests,however, show the reverse effects,", It thus appears to the present vriter that very careful model tests are necessary- to achieve a given objec-ive regarding cavitation performance, and -that the possible differences between model aind prototype due to speed and size changes, etc., are much greater than acoustically the uncertainties between v.ater and soditm. In any ca.se,/instrumented prototype tests are clearly necessary to verify the cavitation performance of the reactor pumps. Damage testirg, on the other hand, is much more difficult.in that model tests where velocity and/or size are modelled are almost meaningless, even over extremely p. long test periods.: Of course if zero cavitation is, as is apparently the case in the Phenix and PFR designs, damage should not occur. If limited cavitation is allowed (very slight visiblte) as in the SIM-300 design,to be discussed next, damage is certainly a good possibility, Presumably the present lack of damage in EBR pumps is due to the fact that the pump design weas ver_ conservative, so that in fact, zero cavitation vwas achieved, C. Neratoom and SITIR Pumo Develonment and Test 1. G-eneral a cro d After my visit to Risley described above, I visited (1-day) Neratoom at the Hague, and in company with some of the Neratoom pump Erg ineering'Vorks, fDc)-&v &(c= -4-'group engineers, the Stork/amqazn, where the sodium pumps for SITR-300 have been manufactured and designed. Tbe sodium tests were conducted in West Germany/ This project is quite well documented in the published literature (18-24), and the Dutch group is quite willing to discuss the pump desegns in detail. The leader of the group s/\BR H. Palkkel, but it appearsthant he has been incapacitated for some time, and it

ANI Tri Renort - 11 G - 10/26/74 may be that he vw-ll be transferred to other work wihen he returns, C. J. Hoornweg seems to -be prominent as a replacement at the moment, This group is$ also using acoustic techniques for the detection of cavitation inception as well asifan indication of gas contentrm,both in water and sodium (21). I did not discuss this part of the work:b with anyone familiar with it, but have the general impression that it is much less advanced than that at Risley. 2. STR-300 PUI-rmi Desin The SIT7R-300 pumps, desig~r primarily by the Stork Engineering Viorks, who have long experience in the design of large drainage -pumps where suction heads are low and cavitation a problem, are designed to accept minimum visible cavitation (at a plume of about 1 mm length on blade leading edges according to mater model tests) during normal operation.* They are convinced by their experience with the drainage pumps that this will not create important cavitation erosion. A 6.000 hour test of the sodium prototype rjumrp for SITR-300 convinces them'that this is the case. Apparently- no cavitation pitting xas observed in these tests, though there vas- some relatively smooth erosion ibn apparently non-cavitating regionsv which they attribute to some sort of chemical effects. They state that this erosion (19) is a result of "washing out" of silicon particles originally embedded in the metallic skin of the impeller. The STR-300 pump development (18-21) was started with model tests in water in a 1/2 scale model, and using transparent windows such that cavitation could be observed and correlated with the acoustic detection detailed The necessary/changes in impeller design were made in this model to attain the desired cavitation perfanance. A rather similarl method was *This changed design philosophy from the French and British designs appears to indicate that SiTR-300 does not expect to use acoustic boilini detection as a portion of the safety system.

ANT Trip Retort 12 - H /26/7 employed in the desi'g of the Risley pumps, and seems to me to be necessary in generalif the final product is to be reasonably suited to the conditions, i.eo, neither too conservative, nor vice versa. Once the impeller design had been.settled, a full-scale model was built and tested in water using aaoustic detection of cavitation, as well as the measurement of the normal flow parameters. The fullthe already discussed was scale model was then tested in sodium, and a/6000-hour endurance rum/ made. Acoustic instrumentation vas used to verify the performance with regard to cavitationx inception,and the full head vs. STPSH curates lun. Thus a direct comparison in a prototype machine of cavitation perforxwma cold reactor temperature (19) in/vJater and/sodium is afforded/ I believe that this is the only presently published comparison for reactor prototype-size pumps. A slightly greater IxPSH (about 10%) was required for the sddiuxma than for the water tests The authors believe that thZs was due test primarily to detailed differences in the/loops and experimental error, To quote their second conclusion (19): "There is no deviation in IF~SH behavior between w.Eater at room temperature and liquid sodium at 580~Ct. However, this of course does not means that there is no difference in cavitation damage potential between these fluid conditions, Present evidence in fact does indicate that the damage. potential of cold water and reactor temperature sodium may be of the same order,'r,\that this is a great deal greater tlhan the cavitation damage potential of PSR temperature waterP- A significant The above result) that there is no/engineering difference in NPSH behavior between cold water and reactor ctemperature sodium,3 isthe same as that reached from previous tests in the u.S. for pumps at CANEL and at Oak Ridge. Thess results were reviewed in a previous report by the present author (22). *To provide sodium closer in properties of Go PJR-tenlperature water with regard to cavitation damage, and thus reduce subsivantia]lly cavitation dajnage potential, it is likely that thee sodium pumps should be located in t;lheiho- leg of FBR reactors * 1MW citz't I t~vv\W

III. Bibliography 1. F. G. Hammitt, "Observations of Cavitation Scale and Thermodynamic Effects in Stationary and Rotating Components, " Trans. ASME, J. Basic Engr., D, March, 1963, v. 85, p. 1-16. 2.F.G. Hammitt, D.M. Ericson, Jr., M.J. Robinson, J.F. Lafferty, "Gas Content, Size, Temperature, and Velocity Effects on Cavitation Inception in a Venturi, " ASME Paper No. 67-WA/FE-22. 3. J. Bonnin, R. Bonnafoux, J. Gicquel, "Comparaison des Seuils d'Apparition de la Cavitation dans un Tube de Venturi dans l'Eau et le Sodium Liquide, " E. D. F., Bulletin de la Direction des Etudes et Recherches - Serie A Nucleaire, Hydraulique, Thermique, n. 1, 1971, p. 5-12. J. Bonnin, "Theoretical and Experimental Investigations of Incipient Cavitation in Different Liquids, " ASME Paper No. 72-WA,.FE-31, 1972. J. Bonnin, "Thermodynamic Effects in Cavitation," Proc. Conf. on Cavitation, Institutuion of Mechanical Engineers, Fluid Machinery Group, Edinburgh, Scotland, September 3-5, 1974, p. 355-362. F. G. -Haammitt, A. Keller, O. S.M. Ahmed, J. J. Pyun, E. Yilmaz, "Cavitation Threshold and Superheat in Various Fluids, " Proc. Conf. on Cavitation, Institution of Mechanical Engineers, Fluid Machinery Group, Edinburgh, Scotland, September 3-5, 1974, p. 341-354. J. C. Duquesne, X. Elie, J. P. Constantin, "Cavitation in Flow Distribution Devices of Fast Reactor Cores - Problems Related to Phenix, " ASME 1970 Cavitation Forum, p. 33-34. W.R. Huntley, A.G. Grindell, "The Cavitating Characteristics of Two Types of Electromagnetic Pumps in Potassium," 1966 ASME Cavitation Forum, p. 15-16. S. G. Young, J. R. Johnston, "Accelerated Cavitation Damage of Steel and Superalloys in Sodium and Mercury, " ASTM STP No. 408, p. 186-219, 1967. A. Thiruvengadam, H.S. Preiser, S.L. Rudy, "Cavitation Damage in Liquid Metals, Tech. Progress Reports 467-2 and 467-3 (NASA CR-54391 and CR-54459), Hydronautics, Inc., Laurel, Maryland, 1965. R. Garcia, F. G. Hammitt, "Cavitation Damage and Correlations with Material and Fluid Properties, " Trans. ASME, J. Basic Engr., D, 89, 4, Dec. 1967, p. 753-763. F. G. Hammitt, N. R. Bhatt, "Cavitation Damage at Elevated Temperature and Pressure," ASME 1972 Polyphase Forum, p. 11-13. Edward Silberman, F.R. Schiebe, "A Method for Determining the Relative Cavitation * Susceptibility of Water," Proc. Conf. on Cavitation, Institution of Mechanical Engineers, Fluid Machinery Group, Edinburgh, Scotland, September 3-5, 1974, p. 10] 108.

Bibliography (cont. ) 14. G. Seed, L.F. Bowles, I.D. Macleod, "Design, Testing and Commissioning of Sodium Pumps for the 600 MVW(T) Prototype Fast Reactor, " Proc. Conf. on Reactor Pumps, Institution of Mechanical Engineers, Bath, England, 1973, p. 173-185. 5. I. S. Pearsall, "The Supercavitating Pump, " Proc. 1973, Institution.of Mechanical Engineers, v. 187, 54/73, p. 649-665. 6. I. S. Pearsall, "Design of Pump Impellers for Optimum Cavitation Performance," Proc. 1973, Institution of Mechanical Engineers, v. 187, 55/73, p. 667-678. 7. I. S. Pearsall, "A Review of Cavitation Scale Effects in Hydraulic Machines, " IAHR, Working Group No. 1: Cavitation Scale Effects, January 1974. 8. R.H. Fakkel, C.J. Hoornweg, B. Kamerling, W.K. Mendte, J.H. Bunjies, TJ. Ten Wolde, M. W. Heslenfeld, "Comparison of Cavitation Tests on the SNR 300 Prototype Sodium Pump, Carried Out Using Water at Room Temperature and Liquid Sodium at 580 Deg. C.'' Proc. Conf. on Cavitation, Institution of Mechanical Engineers, Edinburgh, Scotland, September 3-5, 1974, p. 193-202. 9.R.H. Fakkel, C.J. Hoornweg, W.K. Mendte, M. Jansing, H.J. Lameris, J.P. Vrooo M. H. Heslenfeld, "Development, Design, Construct ion and Full Scale Sodium Testinm of a Prototype Sodium Pump for a LMFBR-Power Plant, " Proc. Conf. on Reactor Pumps, Institution of Mechanical Engineers, Bath, England, 1973, p. 197-205. 0. R.H. Fakkel, J. P. Vroom, K. Mendte, H.J. Lameris, "Testing a Sodiumn Pump," Neratoom, Netherlands, Nuclear Engineering International, Dec. 1971. 1. T. Ten Wolde, W. H. Moelker, K. W. Mendte, "Experiences with Acoustical Methods for the Detection of Cavitation in Pumps," Proc. Conf. on Cavitation, Institution of Mechanical Engineers, Edinburgh, Scotland, September 3-5, 1974, p. 363-372. 2. F. G. Hammitt, "Pump and Other Component Cavitation Comparisons Between Alkali Liquid Metals and Water, " UMICH - 01357-34-T, University of Michigan, Dec. 1973.

CONVERGE iT CQHVrRPNT COL M 4916 Fig. I Schematic of Venturi Tube (Dimensions in min.) Electricite de France Test- same internal flow path as U-M venturi ti__, —R __ I.. __ ____ _1 _'0,0 - 0, _-__ ~ - ]sooium 0,04 E',. —-- 0,02 -.' i.....1 1-. 1 _ I I 1 d Fig. zE W Iaion In c p t l I Ies I - 0,02 _____ U04 5t: Throat Reynoids Number 4917 Fig. 2 Water Cavitation Inception Tests Flectricite de France

.n~~~~-.;;.-.._ —.t. o,/.. __, s"___ _ i 0,08.....- -'..........- _. _ _ 0,0 Nt==-'m _ I04 _I2. o - 2it Ii Tests- _ _' —_ -... — --- 0 c -_._......-0,04 _ - ae 160 C - - -- - - -0,06 N /A 200 C t-I v-. -. W M 300C -_- -- -3- 0 0 08 -0,1 o 450 "SI C;:'.,,I I....e..... I I 1 I I I T 4918 Fig. 3 Sodium Cavitation Inception Tests Electricite de France

O.14.__ _. 0 12.....-' 33% (Air contentE by Volum() 0.10. \ _ { [ i I. | i I~t - (1/z " t nturi) Z 0 04 X _ __ o.02 TP~ii~p~ I I 1 P i i (1/4' throat) FIG3 4 C A VIATO0 SMY L IN VI F CO. 02_ _ E0. 0 -O* 06 _, 8 9 10 2 3 4 5 6 7 8 9 10 Throat Rt r nolds Nunmber, P e 3499 FIG. 4 CAVITATION SIGMA VS. REYNOLDS NUMBER IN VENTURI FOR WATER AND SODIM CONTENTS Electricite de France" and University of Michigan.o~~~o~ ~ ~!:O/~ —'C~~ —~~ ~ ~ ~ ~ ~

LIQUID TEMP gpm P. bpinn a9i d A X WATER?0a C 35 I VARIAEL 1 Na gas content300 147 VARIAknownE o i 10' \'Water has very low air content Na gas content unknown 10 to 10 /vol. 10W he % ~o~~ x \ FI1o x C _U W10AR (0\ Courtesy of Risley Engineering and Material Laboratory UKAEA Reactor Group 333 0 1 2 3 FIG. 5 CAVITATION CHARACTERISTIC OF I.S.A. NOZZLE IN SODI UM'-AND IN WATE'R (m r 0-o.5J

SO-Ow-) z Do o iv 1 vY nlGOS NI 31ZZON VS] 1O3nlD33 A.ll013A'A 9'91I P~' E z I o 3ON 01 I~>//%~~ J O8'NS' 0%, 9- -01 o 0 SO I 2.6ul bq4 1 6- = ad

2 O DII Q R.E... TEbS LAi (3R.E.M.L. TESTS 0r r5X KOBAYASHI | Courtesy of Risley Engineering and Material Laboratory UKAEA Reactor Group 0 O 01 02 0.3 m 335 FIG. 7 INCIPIENT CAVITATION (~3 NUMBER VS. m FOR I.S.A. NO ZZLES