Document QJoQpm72bxK4Q2KrB6dn33on8

FILE NAME: Keene (KNE) DATE: 1953 Feb 3 DOC#: KNE021 DOCUMENT DESCRIPTION: Memo RE Rust and Corrosion - 85% Magnesium vs. 'Kaylo' Marshall F. Allen Monoper U S A A THE MAGNESIA INSULATION MANUFACTURERS ASSOCIATION WASHINGTON 4, D. C. 20m 1317 F Street, N. W. Executive 3008 I February 5, IPS3 f K.I.M., Release 5 3 - g TO: II: . i.lrir Ehret, Jr. SUBJECT: Rust and Corrosion - B5 Magnesia vs. "Kajlo* FROL:: Hr. Harry A. Kieselbacb _ ^ , K r * **ue.*leck has requested that the fcllowinr Ci- a c z a bt er sir ur; bniesin. e*o? you in triplica~te. s s * *vc ?lefs* bear Hr Ein= thEt thes& tests were ci tw " laboratory'' type anc can serve or.lv as ind<- e _ Eini.int actual sertvCi,ccehec=oknd*itipornosp.erly cor.trdle: Fumose of lests (1) To check reported clains by Keyio salesaan that: )E \ 2-->- Magnesia Ecceierates rusting of iron pipe. (t) Ilsylc inhibits rusting (2 ) To determine whether B5% Magnesia will cause corrosion (electrolysis) in the absence of oxygen. U ) I>c. Eayio. U) Whftber 65i "agnesia made wit] wax exhibits less ehernie ac- tivity tnan the untreated product. Materials used In Tests Standard BS% Magnesia Block (J-K) Keylc Block. ' Snail cube of *K.R." Magnesia (J-K*. Canadian Asbestos Snorts (J-K) (lor voltage only ) readings e Test Xo. 1 E * RUSTIKS USAA 620128 ` S ^ T e & ' . l & i r e . " cp8r. cf tf 1.- lor 24 h w . Duration of test - 20 days. t c make up for evaporation). (T*F water was used results -- amount of rusting* All samples showed bout the same T est Xo. 2 Thin strips ef iron (from f/C bands) were placed in open test tubes containing: (t) Tap water in which B Magnesia has been allowed tc soak. (d) Bciled'tap water (tc remove air) ir. rubber stoppered test tube. Results -- Verv little difference in the quantity c' * *. * *' `> ks d v cte. Th's test was conducted at an average i f .t. h e . t ^ ^ cor.vectio*. cur. ^ c<* the tine curing the The return line was warm abc^w b.A c. une v-a e dav and about 255? during the r.ignt. F T"fTROlfill (aip rxciriu.) Test Kc. 1 Strips of iron were bolted to strips of brass and mlaced in rubber-stoppered test tubes containing. (a) Tap water in which 855? Magnesia had been allowed tc soak. (t) Dc. Xaylo. (c) Do. H . R . r Magnesia. USAA P?n**>p Test tubes were supported in the manner des cribed for ''Rusting" tests for a period of about two months, -- Kaylo shoved the greatest corrosion deposit at better. of test tube. Magnesia showed about l/S that cf Kaylo. "Y.:.R.''Magnesia shored about 1/10 that of B5 Magnesia. ' These results indicate that with ar. improvement in the "Y.R.process ccrrosior. might be completely elimin ated. They also indicate the reascr. fcr the activity cf the manufacturers cf ILsylc insulation in research on waterrepeller.cy. ELECTROLYSIS AKL OXIDATION The purpose cf this test was tc determine whether these materials rculd form electrolytes when immersed in water and to obtain an approximation cf the voltages de veloped between an iron pipe and a brass valve or fitting exposed to such materials under water. (The composition of the brass cr fcrcr.ze used in the manufacture of valves is probably different from the sheet brass used in this test and might show somehat different results.) were Strips cf iron and brass, connected electrically, suspended in cper. <ars containing tap water and: (a) Lumps of 65% Magnesia (t) Lames cf ILaylc. . (c) Canadian Asbestos Shorts The jars were placed or. tep cf an insulated house heating steam boiler er.d maintained at an average ' temperature cf about 5 deg.F. The liquid level in the jars varied cue tc evaporation and the contents were allowed tc cry completely tpree times during the test. VeItare Fetdlnrs, Brass tc- I (b'.iter gC.O: ohms ter rclt) (a) 0,52 Volts average. (b) 0.4E r " (c) G.5S " " Current Readings. Brass tc Iron (Meter 4 ,lf ohms) (a) 0C microamperes average - about &DC ua/sc, in. cf iron. (t) SOD microamperes average - about 25C ua/sc. in. cf iron. (c) Fee measured. - 4- SAA 620130 Sediment ^ Located at level ct average rater-line. fc (b) Entire surfece heavily covered, v (c) About 50 of surface. Extending dovn frowa wer-iine. ever th. I'i" " J ! * 1*4 ,, General .9^ t5e basis of these crude tests, the sn-'ite'' believes it safe to assume that Keylo rill ot caui* *1. i i i L r 1- c?ri'osi?= (electrolysis) than 65;' Ktgn-eis Ob'1 ba tef !a!rEea '=1Qer Kater- Ir` fEct* * W M l t . may It appears tht accurate laboratory tests, simulating actual service ccn-it.cr.s, coulo be used tc advantage ir. overcoming the claims mace by this competitor* Very truly yours, EAE/th E. A. Kieselbach