~96T '~ tventrqaa * OKI 'avutiisni QH8VOg JAVHx HaiNImmfodI aamitwod rivdtmhoaj 0,4 q~aodag u IITmlTTata V au-in qslodlaj 9~'-90TT 0r:11 sssniriuvsrid acvalnms cinv CHOWo 'NaJNId 'aaaisht x, uanasav SV xmrii9iofnohdaaa - 1 HVJ SJ 2 NSWallfSVzN Hlmssauc WOYNHcIVIGi omiimaav SaoianVa UTSUODSTM. 'u04etddv fliisiwsho Hacld d A ZlfLTIJSNI MU~ I -'.- ~ ~ ~. P / ½.J a" - -. w~ 0 2-!-
I THE INSTITUTE OF PAPER CHEMISTRY Appleton, Wisconsin FACTORS AFFECTING DIAPHRAGM PRESSURE MEASUREMENTS PART I. REPRODUCIBILITY AS AFFECTED BY TESTER, PLATEN, GAGE, AND SURFACE CLEANLINESS SUMMARY As one phase of the current investigation of factors affecting bursting strength results, an Institute-mill comparison was made with respect to diaphragm pressure measurement. Of the seven companies forwarding data to the Institute as of this date, two reported diaphragm pressures in satisfactory agreement with the Institute results. The remaining five mills reported pressures from 12 to 25% (4.6 to 9.4 p.s.i.g.) lower than the Institute. In general, a procedure for evaluating diaphragm pressure which can be duplicated by everyone appears essential to diaphragm standardization. Therefore, work carried out since the last meeting has been directed toward determining possible causes for the larger differences noted above. Most attention has been centered on the 0.375-inch distention; however, the measurement problems at 0.710- inch distention have also received attention. The initial work was directed toward determining the reproducibility of pressure measurements at the Institute as affected by tester, lower platen, diaphragm and platen cleanliness, etc. In general, the results indicated that at 0.575 inch: 1. Neither tester nor lower platen appeared to markedly affect diaphragm pressures. Some differences occurred between testers but they appeared to be a function of the order of testing and may have been caused merely by repeated removal and insertion of the same diaphragm.
Fourdrinier Kraft Board Institute, Inc. Page 2 2. The differences between trials, testers or conditions were in general, reasonably small (usual range 0-2 p.s.i.) and did not approach the large differences noted in Report Eight between Institute and mill. 3. Thorough cleaning of lower platen and diaphragm in distilled water and acetone to remove possible surface contaminants did not materially improve reproducibility. Somewhat similar trends were obtained at 0.710-inch distention but the differences tended to be larger-both absolutely and on a percentage basis. More work appears necessary in this area. In addition to the above, investigations were made as to (1) the effect of air in the tester manifold and (2) a comparison of results obtained with two size gages. In general, air in the manifold had little or no effect on diaphragm pressure as might be expected. When 60 and 300 p.s.i. gages were used to evaluate diaphragm pressure, equivalent results were obtained. Thus, assuming properly calibrated gages, this factor would not necessarily explain the differences noted in Report Eight.
Fourdrinier Kraft Board Institute, Inc. Page 3 INTRODUCTION As one phase of the current investigation of factors affecting bursting strength results, the Institute and B. F. Perkins and Son, Inc. initiated a study designed to determine the variability in commercially manufactured diaphragms. For this purpose, the Institute evaluated 65 diaphragms from a recent manufacturing order and the results were reported to the Technical Committee in Report Seven dated September 1, 1962 (1). In brief summary, it was noted that the diaphragms exhibited pressures ranging from 36 to 40 p.s.i. gage at 3/8-inch distention. After reviewing the results, the Committee requested the Institute to forward a number of the diaphragms to each member for mill evaluation. Eight to the Technical Committee summarized the replies received (2). Report Of the six companies forwarding data to the Institute, two reported diaphragm pressures in satisfactory agreement with the Institute results. The remaining mills reported pressures from 12 to 25% lower than those obtained by the Institute. Since Report Eight was issued, one additional reply has been received and the results for this mill were about 12% lower than the Institute results. These results indicate that a procedure for evaluating diaphragm pressure which can be duplicated by everyone must be established if any progress in diaphragm standardization is to be made. In the interim period, the Institute investigated the reproducibility of its own procedure for evaluating diaphragm pressure under a number of conditions. These results are reported herein. I
Fourdrinier Kraft Board Institute, Inc. Page 4 PROCEDURES Part 1. Reproducibility of diaphragm pressure measurements as affected by by tester and lower platen. A. Diaphragms: Select six diaphragms from stock and identify by number. B. Lower clamping platens: 1. As used at present on DE 35 and DE 105. (Clean platen before using to remove loose graphite, dust, etc.) 2. New platen purchased from B. F. Perkins and Son, Inc. C. Diaphragm pressure measurements: Obtain the diaphragm pressure for each condition as follows: 1. Install the diaphragm as scheduled in Part D and adjust the diaphragm height in its retracted position so that its upper surface is flush with the lover platen. 2. Distend the diaphragm to approximately 1.8 cm. ten times. 3. Check to determine if the diaphragm in its retracted position is still flush with the lower platen and adjust if necessary. 4. Determine the pressure required to distend the diaphragm to 3/8 inch and to 1.8 cm. D. Test conditions: 1. DE 35 with a. Present lower platen b. New lower platen 2. DE 105 with a. Present lower platen b. New lower platen Evaluate each diaphragm at each condition using procedure described in Part C. Repeat the evaluation at a later time: E. Order of test: 1. For the first four diaphragms, begin the evaluation on DE 105 with the new platen. 2. For the last two diaphragms, begin the evaluation on DE 35 with the new platen. I
Fourdrinier Kraft Board Institute, Inc. Page 5 Part 2. Effect of cleaning diaphragm surface and platen on diaphragm pressure. A. Tester: DE 105 with 60 and 300 p.s.i. gages. B. Lower platen: As used at present. C. Diaphragms: Select four diaphragms from stock. D. Procedure 1. Wash each diaphragm in (1) distilled water and (2) acetone and dry using a clean towel. After washing, handle with plastic or rubber gloves which have been similarly cleaned. 2. Wash the lower platen as in 1 above. 35. Insert the diaphragm in the tester and determine pressure at 3/8 and 0.71 inch as described in Part C of Part 1. 4. Remove the diaphragm and: a. Blot excess glycerin on towel b. Wash in distilled water c. Wash in acetone d. Handle only with clean gloves 5. Clean the lower platen as in "D" above. 6. Insert the diaphragm and redetermine pressure at 3/8 and 0.71 inch. 7. Remove diaphragm and wash as in "D" above. 8. Repeat for each diaphragm. 9. Repeat for each diaphragm. Part 3. Effect of surface contamination with glycerin on diaphragm pressure. A. Tester: DE 105 with 60 and 300 p.s.i. gages. B. Procedure 1. Immerse a diaphragm in glycerin and then wipe the glycerin off the diaphragm using a fresh paper towel. 2. Insert the diaphragm in the tester and determine pressure at 0.375 and 0.71 inch as specified in Part 1C. 3. Repeat steps 1 and 2 four times.
Fourdrinier Kraft Board Institute, Inc. Page 6 Part 4. Effect of air in manifold on diaphragm pressure. A. Tester: DE 105 B. Diaphragm: Select three from stock C. Procedure: 1. Clean diaphragm and lower platen using distilled water and acetone. 2. Remove any air present in the manifold. 3. Insert a diaphragm and determine pressures at 3/8 and 0.71 inch. 4. Add approximately 0.5 cc. of air at manifold and determine pressures as described in Part 1C. 5. Add approximately 0.5 cc. of air at manifold and determine pressures as described in Part 1C. 6. Repeat with two additional diaphragms. 7. Note: Air was added at the manifold vent plug. Part 5. Effect of gage capacity on diaphragm pressure reproducibility. A. Tester: DE 35 and 105 with 60 and 300 p.s.i. gages. B. Diaphragm: Use diaphragms currently in the two testers. C. Procedure: 1. Determine the pressure at 0.71 inch using first the 60 p.s.i. gage and second the 300 p.s.i. gage on Tester DE 35. 2. Repeat at 0.5375-inch distention. 3. Repeat (1) and (2) using Tester DE 105.
Fourdrinier Kraft Board Institute, Inc. Page 7 DISCUSSION OF RESULTS As mentioned previously, the results in Report Eight indicate that differences in techniques of measuring diaphragm pressure may give quite different results. For example, in Report Eight it was noted that the average difference between Institute and mill evaluation of diaphragm pressure at 0.575 inch ranged from +1.2 to -9.4 p.s.i.g. Under these conditions, it was felt that some standardized method of measuring diaphragm pressure must be established if progress in diaphragm standardization is to be made. The reasons for the above differences were not known. Therefore, it was felt that work should be initiated in an effort to determine possible causes for the differences. With this in mind, variables which might affect diaphragm pressure measurements are listed below: A. Instrumental variables 1. Lower platen a. Dimensions and design b. Smoothness and frictional characteristics around orifice 2. Gage a. Capacity b. Calibration 5. Distention height stop a. Manual b. Automatic (as used in Report Seven and described in Report Five) B. Diaphragm insertion 1. Centering 2. Pressure applied to lower platen when tightening clamping ring (note: Institute used 150 p.s.i.).
Fourdrinier Kraft Board Institute, Inc. Page 8 B. Diaphragm insertion (continued) 3. Glycerin or other lubricants applied to top surface of diaphragm or to lower platen. 4. Air in tester. With the above in mind, it was thought desirable to intercompare diaphragm pressure measurements using the Institute's two bursting strength testers. As part of the comparison, results obtained using a new platen just purchased from B. F. Perkins and Son, Inc. were compared with corresponding values obtained using the platens presently in use-termed "old" platens in the following discussion. Two trials (separated by at least a day) were made at each condition with a number of diaphragms. The results obtained are tabulated in Table I. Referring to the table it may be noted that at 0.375-inch distention: 1. Approximately equal pressures were obtained with old and new platens although there seemed to be some tendency for the new platen to give slightly lower results than the "old" platen on DE 35. 2. The differences between trials ranged from +1.3 to -2.0 p.s.i.g. on Tester DE 105 and from +1.3 to -1.4 p.s.i.g. on Tester DE 35 for a given platen. There appears to be some tendency for the results in the second trial to be somewhat lower than in the first trial, which may reflect changes in the diaphragms merely due to repeated removal and insertion. Thus, while the differences were somewhat larger than might be desired, it was felt that they were not excessively large. 3. There appears to be some tendency for Tester DE 35 to yield somewhat lower diaphragm pressures than DE 105 for the first four diaphragms where testing started on DE 105. The trend tends to reverse for the last two diaphragms
s asalj;u~aapd Rt{4 UT PRWC3TPuT ST ea PuB 6 sm2-ejwdvtp 10J TST-fl L13B UT paut~qqo anati s~~uipeajaq;w qtaqtl UT Iapao GtLJq 'T WT'WI~ '0a utbj~qdebtp SU-EOTTOT SaS~Tw4U~avd 9 M% UT PSVRDTPUT ST 9 PURe 'ta 'LT 'Oa SwI&BJM&TjPp Soj TeTJ41- tu UT PRUTeqqo OSRA s~utpeaj RTl4 tldtwm UT sep10 RLUf 0'T+ 6 9-1 cyi Z3&084?/ 96 0 3-0,99 9'9+ ciytt- 0, 91 o, 69 091 0'T- F~e- or 6L 9,3+ T 0-i &' T- o e 6 o- O 17'T+ 09.i ',JJTCI 9'4T+ F& 0-999q e&6- (C)929g (T)o~e6 o0 99 o% - 0, 1 y9 O'eJ. 91)- (1)a99 (4)w41 9'T4 IfT 60E P49 ~ 910- ira-?'t+ *JJTQ evt-' + ) (a)cotj9 T0O+ (9)EB% (9)T4 q 992T+ 1-ira- F&9T+ Fa09 0I179 91)- F& O+ 9,99 t 1 99 <0 + 9*4a- t 1 1T- ly. g o 69 a&tl 9* 99 a'a± 4.0O+ g,3 P F 90 e T0O+ T'T 9'0-' 9-0- IJiTQl 4a Tae H *0 H co 0 0mL- -4- ct 4 0. T- m 1+ oc 6-i 0,0 La0-009- OYOT- mae- 9 a- O1 L- ma - JJiTCl 0.09 0'T9 I<0+ < 9e O'99 T0O+ 9, T T T ~ QO+ LUi~ 0m O cr99 CMT- o% 9 omo6 LfTl IflT T 0- cr9-0- 9 3-½ 1 +.01 1, if - Oe+ f7ga a- 9, 19 cr0o6 9'4T- < 9 &glt+ 919L T T+4 O019 9%T4 i769 91L9 T* 0- oe F (L)0r99 (4)our9q 9r0- (9)O4~9 (T)9499 4.T 60T T 0-0 I e i7fo oa o1 &~T ma0, 9* 0- Q,-( 3 a -ite- 9¾LT- )2<0-0' - ¾ISTQC Q'T PT± ea 'J-jcTa T0O+ (9)9'C (9)4'Be 9'T+ (tr)9 e (B)9'T C '5JJ~ Ua4vfd UaRWTJ 'J5Th UaR4WcJ UR4STd 'JJTcE ugwbeta Uecved IJJTQ UR4BTJ UGR4eTJ ON f PTO MGM PTO MGM PTO ARM PTO - arm TeiaJ aidb1mdtgdi 4 act.i?4sr 4oT auaa0 a 4 auaasg 40T 1c Ha TS-9I Ts'd 'ROuI 12<0 we alunssoad ma nrqdbtqj Tsd -'Tqz'u1 Le 1 o4b aansszaa~ uffib:gcibtq x mnliai cnualc1oh a HIanSSaUi w ovmvini iw n MO trzrlvrra Hmncn cmv ~ jsai Jo oagdas I TIEVtl, O+ T, IT LT F
Fourdrinier Kraft Board Institute, Inc. Page 10 where the order of testing was reversed-i.e., testing started on DE 35. Therefore, the effect is probably caused by factors associated with the diaphragm. The trends obtained at 0.710 inch are somewhat similar to the above although considerably larger differences were encountered-particularly between trials (on DE 105) and between testers. In the latter case, the differences reversed as the order of testing reversed and it may be concluded that factors other than test machine actually caused the between tester differences. In carrying out the above tests, normal precautions were taken to wipe glycerin off the diaphragm before insertion and to wipe the surfaces of the lower platens. It is known, however, that diaphragm pressure is sensitive to the frictional characteristics of the diaphragm and lower platen surfaces (_). It was thought possible, therefore, that some contamination of the surfaces might have occurred during the previous testing which contributed to some of the observed differences. To test this possibility, four diaphragms were evaluated in one of the testers after thoroughly cleaning both diaphragm and lower platen with (1) distilled water and (2) acetone. After washing, any handling was performed using clean gloves. Four trials were made on each diaphragm and the results obtained are tabulated in Table II. Referring to the table, it may be noted that: 1. At 0.375-inch deflection, diaphragm pressure appeared to decrease slightly on each trial relative to the first trial. This was apparently due to the stressing of the diaphragm during previous trials. Some of the differences -noted in Table I apparently arose from this cause.
- aouajajaj Sa 2OU2JG~J TWFJal S~T Un l~2sb2 9C-0 fft- 09-6 9'0-0*0 T ~ 21 ~ T X9qT- X IT- X'9T- 0*0 0'i9 o' 69 Cl OCT LBE- L'T- C1T- 0*1*~ L,~ L ' E e B T 9 TT - 91-T 2 lye ly' 0+ *fy*0+ 9'96 W,96 <'96 I~ T- ff 0- T La T' IT 629-9.9-9* 99 9* L9 11 96 LPT- ff-t- EB'T- LleE i2 4L E~ A~ 'S'-s'd qo"t 4LL'o '~ou~fljjtq[ S'-r'srI TIOUI TL'C I'aDUaflJJTQ 'Ts- uot4tu4stq pa,4,et.puj ;e G.flss2JJ Luej~tjd&eTQC.ON O ThSTJI ms-.rqetci - HaLIVM (INV anoj 2 0DV H11IM WDVHUcdYTQ CENV NaJmVqa ONINVaID EffLAV M~flaIKflQO&da aanssaaal wovma~cvi IT armvj, OUTN ;sodag TT a~u 'OUI '2;n4T4SUI PaLog ;jafaax au~papnog
Fourdrinier Kraft Board Institute, Inc. Page 12 2. At 0.710 inch, larger differences tended to be obtained-either on an absolute basis or percentagewise. At this stage it appears that improvements in technique are needed to improve reproducibility at this distention level. 5. The use of the acetone-water wash did not particularly improve the reproducibility between trials. In the previous evaluation, considerable effort was made to improve reproducibility by cleaning possible friction-reducing agents from the diaphragm or lower platen. As an opposite approach, one diaphragm was washed in glycerin and merely wiped dry using clean paper toweling. The diaphragm pressure measurements in Table III indicate that: a. At 0.375 inch, the differences between trials were not appreciably larger than those obtained in Table II after careful washing in water and acetone. b. At 0.710 inch, larger and more erratic differences were encountered between trials as in the previous evaluations. During the course of the previous testing, there were some indications that air in the manifold of the tester led to erratic pressure reading. In an effort to check this possibility, small quantities of air were purposely introduced into the gage manifold at the manifold vent plug. Diaphragm pressure measurements taken before and after admitting the air are shown in Table IV. As may be noted, the air had only a minor effect, if any, on the pressure measurements at 0.575 inch. This would be expected.
Fourdrinier Kraft Board Institute, Inc. Page 13 TABLE III DIAPHRAGM PRESSURE REPRODUCIBILITY AFTER WASHING DIAPHRAGM IN GLYCERIN Trial No. Diaphragm Pressure at Indicated Deflection, p.s.i. Difference Difference 0.375 Inch p.s.i.g. a 0.71 Inch p.s.i.g. % 1 2 3 4 36.2 35.3 54.7 35.6-0.9-1.5-0.6-2.5-4.1-1.7-96.8 101.4 90.8 72.4 +4.6-6.0-24.4-6.2-25.2 abased on Trial 1 as reference. TABLE IV EFFECT OF AIR IN MANIFOLD ON DIAPHRAGM PRESSURE Diaphragm No. No Air Diaphragm Pressure, p.s.i.g. 0.5 cc. Diff., %a 1.0 cc. Diff., a At 0.375-Inch Deflection EP20 35.0 34.3-2.0 34.6-1.1 EP22 34.4 33.8-1.7 33.9-1.5 EP24 32.8 51.6-5.7 31.9-2.7 At 0.71-Inch Deflection EP20 86.4 89. +3.0 90.8 +5.1 EP22 84.6 91.0 +7.6 72.0-14.9 EP24 88.0 85.4-3.0 82.6-6.1 abased on "No Air" condition as reference.
... I Fourdrinier Kraft Board Institute, Inc. Page 14 In the Institute-mill comparisons in Report Eight, most of the mills used 200 p.s.i. gages whereas the Institute used a 60 p.s.i. gage. It is believed that the lower capacity gage is helpful in making the relatively low pressure measurements required. At pressures in the 23 to 30 p.s.i. range, the 60 p.s.i. gage should be more accurate. However, if properly calibrated, the differences due to gage may be small. As a check on this, the results in Table V taken on a 60 and 300 p.s.i. gage are in good agreement at both 0.375 and 0.710 distentions. TABLE V EFFECT OF GAGE CAPACITY ON DIAPHRAGM PRESSURE REPRODUCIBILITY Pressure Gage Capacity, p.s.i.g. Diaphragm Pressure at Indicated Deflection, p.s.i.g. Tester DE 55 Tester DE 105 At 0.575-Inch Deflection 6o 300 Diff., 60 300 Diff., %a 25.2 25.0 25.3 25.0 +0.4 0.0 At 0.71-Inch Deflection 40.9 43.5 40.3 43.0-1.5-1.1 Based on 60 p.s.i. gage as reference. Based on Go p.s.i. gage as reference. When either Institute bursting strength tester is checked with standard samples in connection with the instrumentation program, diaphragm pressure measurements are made using the apparatus described in reference (3). A summary of results obtained during the past two months is shown in Table VI. As may be noted in the table, pressure readings at either distention level were within tolerances-23 to
GRVJaAV 0.9F E9-TE-TT O'E Fg-9F-TT 0.9a 2;9-98-TT 0.9a 0,t 0.9F F9- -FT 0.9F E;9- -FT O'e 0.9a E9-L -at O't 0.9"?9-L -FT O'" 0.93 89-OT-BT Otm O.qc F9-TT-3T O't 0.9E Fg-TT-K 0-Tt 0-17a 39-FT-FT 0.8 O'F Fg-LT-K O.Ft 0,a F9-LT-FT O, F E9-9T-ET O't 0'F 39-6T-E;T O, fj 0,a ag-oa-k 0,t 0,F 39-OB-FT O' V7 0-9c: E;q-TF-E;T O' 0.E 9-T?-ET O't O' E;9-9E;-?T O' Ti7 0,!;F 89-9B-?T 0-Tt 0.9F 9-L -T 0.0 0,F 9-6 -T 0.0 0,F 9-OT-T O' TIT O' O- Ft, 0, 9-tT-T O.F'7 0,F 9-9T-T 0.9F e9-ta-l O- 92 9-cz-T 0.0, itf, 9-F-T IIOUI OT) 'O T40UI L'O llz)ui OTL, 0 TqouI L, O,419cr OT a(i E act -9--c-s-d 'uo-1409tjag P94130-1puI q-e aanssaaj uovaqdst(j NOTJNSISI(I HOMI OTL'O (INV LPO IV SOMIGVEH HMSSaUd WOVIIHdVIG JO MOSIMIKOD IA EFIE[VI T Slecl -OUI laqnqtqsui Pauog qj:,aj:x T;DTU-llpalloq
Fourdrinier Kraft Board Institute, Inc. Page 16 30 p.s.i. at 0.375 inch and 40 to 45 p.s.i. at 0.71 inch. While pressures at 0.71 inch varied through the entire tolerance range, it may be interesting to note that the corresponding readings at 0.375 inch varied only from 23 to 26 p.s.i.g. In summary, the results appear to indicate that at 0.375-inch distention, differences in diaphragm pressure as great as 2 p.s.i.g. can occur. These small differences appear to be associated, in part, with the stressing of the diaphragm. In any event, no cause for the large Institute-mill differences was found. As would be expected, larger differences were encountered at 0.71-inch distention. However, the results seemed more erratic than would be expected. In an effort to (a) determine possible causes for the large Institutemill differences at 0.375 inch and to (b) improve measurements at 0.710 inch, the following work is planned: pressure. 1. Comparison of automatic and manual methods of evaluating diaphragm 2. Effect of errors in distention height on pressure readings. apparatus. 3. Recheck the distention height reproducibility of the automatic LITERATURE CITED 1. Variability of commercial diaphragms. Part I. Between cavity variance..preliminary Report 7 to the F.K.B.I. Sept. 1, 1962. 2. Variability of commercial diaphragms. Part II. Comparison of diaphragm pressure evaluations by Institute and mill.. Preliminary Report 8 to the F.K.B.I. Dec. 4, 1962. 3. Effect of diaphragm lubrication on diaphragm pressure and bursting strength.. Preliminary Report 6 to the F.K.B.I. Jan. 30, 1962.