BEGIN NEW DATA CASE C BENCHMARK DC-33 C Type-96 hysteretic inductor with TACS logic to give flux by integrating v C Batch-mode EMTP CalComp plotting, including "X-Y PLOT", is illustrated. C For split version with TACS postponed until postprocessing, see DC-45 and C DC-46. As for data ordering, this is intentionally shuffled from what C would be natural in order to illustrate the use of "/" cards for sorting. C Initial condition & load flow classes were added during November of 1991 C after Prof. Juan Martinez complained from Barcelona, Spain, that such C features were not working. He was right. One year later, he complained C again (it still did not work right. Somehow, work was lost; DC-33 had C not been changed; the optimistic report of the January, 1992, newsletter C was incorrect). So, sorting by class was repaired again on 28 Novemb 92. C 1st of 9 stacked subcases total. .000050 .020 1 1 1 1 1 -1 5 5 20 20 TACS HYBRID /SOURCE 14GEN 377. 60. -1. /TACS 1FLUX +GEN 1.0 0.0 1.0 90GEN 91GROUND 99CURR -1.0* GROUND 33FLUX GEN GROUNDCURR /OUTPUT GEN /PLOT C Last step: 400 .02 116.4994069 2.765761424 -2.76576141 .9416257618 C Last step continued .....: 116.4994069 2.765761424 -2.76576142 C Final 5 max. : 3.499909256 .9905492104 376.9925583 3.499613141 3.499909256 C Associated times : .0125 .00415 .01665 .00415 .0125 2Arbitrary 78-character case title text of which this is an example, I hope. First of two lines of 78-byte graph subheading text. Second and final such line of graph subheading text. 194 2. 0.0 20. TACS CURR ABCDEFGHIJKLMNOP1234567890123456 X-Y PLOT Horizontal Axis label123 10. -1.1 1.1 8. -4.0 4.0 194 4. 0.0 20. TACS FLUX TACS CURR Graph heading---Vertical axis la X-Y PLOT Horizontal Axis label123 9999. -1.1 1.1 "9999." returns to conventional (vs. time) plotting 8. -4.0 4.0 PRINTER PLOT 194 4. 0.0 20. TACS CURR { Axis limits: (-3.500, 3.500) C Now that all plotting is done with, let's illustrate the three declarations C that will choose the destination of any subsequent "CALCOMP PLOT" use: SCREEN PLOT { If a vector plot were to follow, it would go only to the screen PEN PLOT { If a vector plot were to follow, ... go only to CalComp plotter SCREEN PEN { If a vector plot were to follow, .. go to both screen & plotter C The preceding 3 declarations really could only be tested by a user who C had both a vector-graphic screen and a CalComp plotter. Hence no use. /LOAD FLOW C Comment of /LOAD FLOW. Since DC-33 does not really involve the load flow, C Comment of /LOAD FLOW. we can not use real load flow data cards. In their C Comment of /LOAD FLOW. place, we use these three comment cards. End test /BRANCH 96GROUNDGEN 8888. 1.E-9 1 1.0 -0.7 2.0 0.9 3.5 1.0 9999. /REQUEST PRINTED NUMBER WIDTH, 13, 2, { Request maximum precision (for 8 output columns) /INITIAL 2GROUND 0. { Node voltage initial condition in fact changes nothing C Initial conditions really are not a part of this problem, although we do want C to illustrate that they, too, can be sorted, and inserted after /LOADFLOW C The preceding redefinition of node voltage at GROUND changes zero to zero. /SWITCH GROUND MEASURING 1 BLANK card terminates all TACS data BLANK card ending all BRANCH cards BLANK card ending all SWITCH cards BLANK terminates the last SOURCE card C Total network loss P-loss by summing injections = 7.106450000000E-06 C Inject: GEN 377. 377. .377E-7 2.6786345332877 C Inject: 0.0 0.0 -2.678634533288 -89.9999992 C ---- Initial flux of coil "GROUND" to "GEN " = 1.00000000E-09 C C Step Time GEN GROUND GROUND TACS TACS C TERRA GEN FLUX GEN C *** Phasor I(0) = 0.3770000E-07 Switch "GROUND" to " " closed C 0 0.0 377. .377E-7 0.0 0.0 0.0 C 1 .5E-4 376.9330268 .5235646391 -.523564601 .0094233257 376.9330268 C 2 .1E-3 376.7321312 1.04694322 -1.04694318 .0282649546 376.7321312 BLANK card ends OUTPUT variable requests BLANK card ending all batch-mode PLOT cards BEGIN NEW DATA CASE C BENCHMARK DC-33 C 2nd of 9 subcases is the same as the first except that control system C modeling is to be done using MODELS rather than the older TACS. Jerry C Almos of BPA converted the data on 18 Nov 1991 after WSM changed the code C to allow such sorting using the same /TACS declaration. .000050 .020 1 1 1 1 1 -1 5 5 20 20 MODELS { Note the change; the 1st subcase had "TACS HYBRID" here /SOURCE 14GEN 377. 60. -1. /MODELS { 9 July 1995, the former /TACS was changed to this new class INPUT GEN {V(GEN)}, GROUND {I(GROUND)} MODEL DC33 INPUT gen, ground VAR flux, curr HISTORY gen {DFLT: 0}, flux {DFLT: 0}, ground {DFLT: 0} INIT curr:=0 ENDINIT EXEC COMBINE AS INTEGRATOR LAPLACE(flux/gen):=(1.0|S0)/(1.0|S1) ENDCOMBINE curr:=-ground ENDEXEC ENDMODEL USE DC33 AS DC33 INPUT GEN:=GEN, GROUND:=GROUND ENDUSE RECORD DC33.FLUX AS FLUX DC33.GEN AS GEN DC33.GROUND AS GROUND DC33.CURR AS CURR ENDRECORD ENDMODELS /OUTPUT GEN /PLOT C Last step: 400 .02 116.4994069 2.765761424 -2.76576141 .9604757618 C Variable maxima : 377. 3.499613141 3.499909256 1.00939921 C Times of maxima : 0.0 .00415 .0125 .00415 C Variable minima : -376.992558 -3.49990926 -3.49961314 -.990568949 C Times of minima : .00835 .0125 .00415 .0125 2Arbitrary 78-character case title text of which this is an example, I hope. First of two lines of 78-byte graph subheading text. Second and final such line of graph subheading text. CALCOMP PLOT SCREEN PLOT 194 2. 0.0 20. MODELSCURR ABCDEFGHIJKLMNOP1234567890123456 X-Y PLOT Horizontal Axis label123 10. -1.1 1.1 8. -4.0 4.0 194 4. 0.0 20. MODELSFLUX MODELSCURR Graph heading---Vertical axis la X-Y PLOT Horizontal Axis label123 9999. -1.1 1.1 "9999." returns to conventional (vs. time) plotting 8. -4.0 4.0 /BRANCH 96GROUNDGEN 8888. 1.E-9 1 1.0 -0.7 2.0 0.9 3.5 1.0 9999. /REQUEST PRINTED NUMBER WIDTH, 13, 2, { Request maximum precision (for 8 output columns) /SWITCH GROUND MEASURING 1 BLANK card terminates all MODELS data { 9 July 1995, MODELS replaces old BLANK card ending all BRANCH cards BLANK card ending all SWITCH cards BLANK terminates the last SOURCE card C Total network loss P-loss by summing injections = 7.106450000000E-06 C Inject: GEN 377. 377. .377E-7 2.6786345332877 C Inject: 0.0 0.0 -2.678634533288 -89.9999992 C ---- Initial flux of coil "GROUND" to "GEN " = 1.00000000E-09 C C Step Time GEN GROUND GROUND TACS TACS C TERRA GEN FLUX GEN C *** Phasor I(0) = 0.3770000E-07 Switch "GROUND" to " " closed C 0 0.0 377. .377E-7 -.25696E-16 .009425 377. .377E-7 -.377E-7 C 1 .5E-4 376.9330268 -1.42571976 1.425719797 .0282733257 376.9330268 -1.42571976 1.425719759 C 2 .1E-3 376.7321312 -1.41394374 1.413943779 .0471149546 376.7321312 -1.41394374 1.413943741 C 3 .15E-3 376.3973844 -1.40217609 1.40217613 .0659431925 376.3973844 -1.40217609 1.402176093 BLANK card ends OUTPUT variable requests BLANK card ending all batch-mode PLOT cards BEGIN NEW DATA CASE C 3rd of 9 subcases ends the original series. 4-9 added 5 March 2007 C Illustration of user-supplied residual flux of Type-96 hysteretic C inductor was supplied by Prof. Mustafa Kizilcay of FH Osnabrueck in C Germany (see E-mail dated 23 March 2001). This data is added to show C that non-zero (but very small) flux of the phasor solution now is C tolerated. Prior to the change several days ago, it was not. Prior C to the change, residual flux needed to be exactly zero for the user- C supplied flux to be used. No longer. Phasor flux < EPSILN is all that C now is required. POWER FREQUENCY 50. PRINTED NUMBER WIDTH, 13, 2, { dT loop columns have width 13 including 2 blanks ZERO FLUX TOLERANCE 1.E-8 { Columns 33-40 carry Type-96 tolerance C Note about preceding declaration, which was added 29 March 2001. After C reading later discussion about trouble with the IDEAL TRANSFORMER data, C Prof. Kizilcay agreed that definition of the flux tolerance independent of C EPSILN would be a good idea. The user can define whatever value he wants C although here nothing will change (1.E-8 = value of EPSILN in STARTUP). .000200 .200 50. 40 1 1 1 1 C Following TACS data is not required. It is used only to integrate the C inductor voltage, thereby indicating the change in inductor flux --- C change with respect to the user-supplied initial value of 0.2 volt-sec. TACS HYBRID 1FLUSSA +MA 1. 1. { Numerator is : 1 1. { Denominator is : 0 + 1 * s 90MA { Type-96 voltage of electric network defines this TACS source 1. 33FLUSSA { Output of integrator, the change of flux, is only TACS output BLANK card ending TACS data QA P1A .0172 .1722 0 P3A .0172 .1722 0 MA 1.8 0 96MA 8888 0.20 1 -0.1199 -0.3754 { Preceding 0.2 is residual flux in [volt-sec] -0.06264 -0.3643 { 2nd of many current-flux pairs -0.0353 -0.3536 { 3rd of many. These define hysteresis loop. -0.02025 -0.3428 -0.00732 -0.3214 -0.002869 -0.3 -0.000538 -0.2786 0.01345 0.0 0.02744 0.2786 0.02977 0.3 0.03422 0.3214 0.04715 0.3428 0.0622 0.3536 0.08954 0.3643 0.1468 0.3754 0.2254 0.3826 0.5494 0.3968 9999 { Terminate last point of Type-96 hysteresis element P2A 1.E10 BLANK card ending all branch cards P2A P3A .020 1.0 { Switch closes at 20 msec } 1 BLANK card ending switches 14QA 8165. 50. -90. -1. 11P1A 1E-10 { 1st half of ungrounded voltage source and ideal Xformer 18P2A .0025MA { 2nd half of ungrounded voltage source and ideal Xformer C Note about preceding two source cards, which represent an ideal transformer. C Yes, there is the ungrounded voltage source, but for engineering purposes, C this has value zero (the 1.E-10 in columns 11-20 of the 1st source card). C The second source card, a Type-18 source, defines turns ratio and 4 names. C In theory, these two source cards could be replaced by the following three C branch cards, which are more readable: C IDEAL TRANSFORMER C 1P1A P2A 400.0 { Rated voltage of high side C 2MA 1.0 { Rated voltage of low side C The resulting phasor solution would be nearly the same, too. Nearly, C but not quite. Prof. Kizilcay's Type-11 source value is small, note C --- small enough to make the phasor flux at time zero < EPSILN. If C instead the IDEAL TRANSFORMER is used, it would appear that the EPSILN C threshold is exceeded. In place of the user-desired residual flux of 0.2 C volt-seconds, the .LIS file will show the noise from the phasor solution: C ---- Initial flux of coil "MA " to " " = -1.87122597E-06 C This then is carried into the dT loop. Without meaningful residual flux, C there is no significant in-rush current. Instead of Prof. Kizilcay's C 30 amps, the PRINTER PLOT shows only about 1.243E-2 amps. For engineering C purposes, the simulation will be completely different (and wrong). BLANK card ending sources C MA .1797146122E-11 .36742498645E-8 -.998414547E-12 .9984146668E-12 .8600609455E-40 .8600609455E-40 C -.3674249425E-8 -89.9719755 -.488343774E-15 -179.9719755 .1834212477E-20 0.0000000 C Total network loss P-loss by summing injections = 3.333364056743E-03 C ---- Initial flux of coil "MA " to " " = 2.00000000E-01 C The preceding result of phasor solution confirms correct operation. The C voltage across the Type-96 branch is not zero, but it is small enough C compared with floating-point miscellaneous data parameter EPSILN so no C special precautions must be taken. Understand what is happening. The C initial flux is equal to the imaginary part of the phasor voltage divided C by the frequency, and this is small: .367E-8 / OMEGA < EPSILN QA MA { Names of nodes for node-voltage output C First 2 output variables are electric-network voltage differences (upper voltage minus lower voltage); C Next 2 output variables are branch currents (flowing from the upper node to the lower node); C Next 1 output variables belong to TACS (with "TACS" an internally-added upper name of pair). C Step Time QA MA P2A MA TACS C P3A TERRA FLUSSA C 0 0.0 .499946E-12 .179715E-11 0.0 -.99841E-12 0.0 C 40 .008 4799.266585 .2154661E-8 0.0 .278303E-11 .211415E-10 C 80 .016 -7765.37646 -.349521E-8 0.0 .438944E-12 .803576E-11 C *** Close switch "P2A " to "P3A " after 2.02000000E-02 sec. C 120 .024 7765.376456 70.13133651 15591.82659 .0177128714 .0990291974 C 160 .032 -4799.26658 110.5331176 33643.29393 22.70094726 1.167639417 BLANK card ending node voltage outputs C 1000 0.2 .9430083E-8 -105.223163 -23383.3158 -.97653E-3 .0484350031 C Variable maxima : 8165. 159.787186 40889.62376 30.82522183 1.523703303 C Times of maxima : .005 .0288 .0294 .054 .054 C Variable minima : -8165. -118.702745 -23490.9996 -.16538292 -.198576963 C Times of minima : .015 .0792 .1996 .1794 .1846 CALCOMP PLOT 193.02 0.0 0.2 -10. 30.MA { Plot Type-96 current over entire time span PRINTER PLOT 194 4. 20. 60. MA { Axis limits: (0.000 3.083) BLANK card ending plot cards BEGIN NEW DATA CASE C 4th of 9 subcases begins a series using /UM on DC-35 U.M. data C 1st of 6 subcases that have same solution but slightly different data. C Begin with /SOURCE (not /UM for U.M.) and full Class-1 U.M. data. C Output has been reduced from 2 lines per dT to one which consists of C just the U.M. variables. For the 10 dT that are taken, signals agree C exactly with DC37.LIS and are documented on comment cards of /OUTPUT .000400 .004 { Take only 10 time steps -- enough to begin motor rotation 5 -1 { No plotting (which has nothing special to do with /UM use) /OUTPUT C The following shows dT-loop output for each of the subcases of the /UM set: C Step Time UM-1 UM-1 UM-1 UM-1 UM-1 UM-1 UM-1 UM-1 UM-1 C TQGEN OMEGM THETAM IPA IPB IPC IE1 IE2 IE3 C 0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 C 5 .002 -.6898285 0.0 0.0 -150.98163 16.0278591 134.953771 15.5970274 131.9385 -147.53553 C 10 .004 -11.779008 -.21882419 -.42361E-3 -203.70113 -91.734138 295.43527 -90.126277 287.718333 -197.59206 /SOURCE 14BUS-A0 180.0 60.0 0.0 14BUS-B0 180.0 60.0 -120.0 14BUS-C0 180.0 60.0 +120.0 11BUS-OM-1 -6.00 /BRANCH 0 BUS-A0 1.0 0 BUS-B0 1.0 0 BUS-C0 1.0 0BUS-A1BUS-A0 0.005 1.0 0BUS-B1BUS-B0BUS-A1BUS-A0 0BUS-C1BUS-C0BUS-A1BUS-A0 0BUSRA1 1.E-5 0BUSRB1 1.E-5 0BUSRC1 1.E-5 0BUS-OM 6.00E4 0 BUS-OM 33.333 /REQUEST PRINTED NUMBER WIDTH, 12, 2, { Request maximum precision (for 8 output columns) /SOURCE 19 { Data for U.M. begins with source type 19 keyed in columns 1-2 0 { The zero does nothing. In fact, a blank card would suffice for DC-35 BLANK card ending U.M. header (3rd of 3 non-comment cards of Class-1 U.M. data) 4 1 1111BUS-OM 2 0.06 0.18850 { 1st machine table 0.0 0.020291 0 0.0 0.020291 0 0.063 0.0001911 BUS-A1 1 0.0 { 1st winding card 0.063 0.0003925 BUS-B1 1 0.0 0.063 0.0003925 BUS-C1 1 0.0 0.083 0.0003925 BUSRB1 1 0.0 0.083 0.0003925 BUSRC1 1 0.0 0.083 0.0003925 BUSRA1 1 0.0 { 6th and final winding card /PLOT 2Arbitrary 78-character case title text of which this is an example, I hope. First of two lines of 78-byte graph subheading text. Second and final such line of graph subheading text. BLANK card ending all BRANCH cards BLANK card ending all SWITCH cards BLANK terminates the last SOURCE card BLANK card ends OUTPUT variable requests BLANK card ending all batch-mode PLOT cards BEGIN NEW DATA CASE C 5th of 9 subcases continues series using /UM on DC-35 U.M. data C 2nd of 6 examples that have the same solution but slightly different data C Replace /SOURCE by /UM and omit all 3 non-comment header cards. But C then U.M. HEADER CARD (UMHC) must be added to define Class-1 data. .000400 .004 { Take only 10 time steps -- enough to begin motor rotation 5 -1 { No plotting (which has nothing special to do with /UM use) /OUTPUT C BUS-A1BUS-B1BUS-C1BUS-A0BUS-OM { Shorten output by eliminating node voltages /SOURCE 14BUS-A0 180.0 60.0 0.0 14BUS-B0 180.0 60.0 -120.0 14BUS-C0 180.0 60.0 +120.0 11BUS-OM-1 -6.00 /BRANCH 0 BUS-A0 1.0 0 BUS-B0 1.0 0 BUS-C0 1.0 0BUS-A1BUS-A0 0.005 1.0 0BUS-B1BUS-B0BUS-A1BUS-A0 0BUS-C1BUS-C0BUS-A1BUS-A0 0BUSRA1 1.E-5 0BUSRB1 1.E-5 0BUSRC1 1.E-5 0BUS-OM 6.00E4 0 BUS-OM 33.333 /REQUEST PRINTED NUMBER WIDTH, 12, 2, { Request maximum precision (for 8 output columns) /UM C Note that there is no Class-1 U.M. data; ATP will supply the Type-19 card. 4 1 1111BUS-OM 2 0.06 0.18850 { 1st machine table 0.0 0.020291 0 0.0 0.020291 0 0.063 0.0001911 BUS-A1 1 0.0 { 1st winding card 0.063 0.0003925 BUS-B1 1 0.0 0.063 0.0003925 BUS-C1 1 0.0 0.083 0.0003925 BUSRB1 1 0.0 0.083 0.0003925 BUSRC1 1 0.0 0.083 0.0003925 BUSRA1 1 0.0 { 6th and final winding card C 14BUSRA1 180.0 60.0 0.0 /PLOT PRINTER PLOT { Batch-mode plotting is for the line printer not vector-graphic /REQUEST U.M. HEADER CARD { To be followed by significant middle card of U.M. Class 1 0 { In fact, a blank card would suffice for DC-35 BLANK card ending all BRANCH cards BLANK card ending all SWITCH cards BLANK terminates the last SOURCE card BLANK card ends OUTPUT variable requests BLANK card ending all batch-mode PLOT cards BEGIN NEW DATA CASE --- NEWSORT --- C 6th of 9 subcases continues series using /UM on DC-35 U.M. data C 3rd of 6 examples that have the same solution but slightly different data C Return to use of /SOURCE and full Class-1 U.M. data as in 1st example. C But add U.M. HEADER CARD (UMHC) which overrides the normal Class-1 data. C See explanation of the preceding NEWSORT within DCNEW-25. There C is old (normal) data sorting logic and there is new data sorting C logic as summarized in the January, 1999, Can/Am newsletter. Putting C NEWSORT on the case-separator card as shown has the effect of placing C a minus sign on SZBED of STARTUP, and this selects the newer sorting C alternative. Either should handle /UM sorting. The NEWSORT could C be added to each subcase without significant change to the .LIS file. C Text of BLANK cards will vary a little, but that should be all. .000400 .004 { Take only 10 time steps -- enough to begin motor rotation 5 -1 { No plotting (which has nothing special to do with /UM use) /OUTPUT C BUS-A1BUS-B1BUS-C1BUS-A0BUS-OM { Shorten output by eliminating node voltages /SOURCE 14BUS-A0 180.0 60.0 0.0 14BUS-B0 180.0 60.0 -120.0 14BUS-C0 180.0 60.0 +120.0 11BUS-OM-1 -6.00 /BRANCH 0 BUS-A0 1.0 0 BUS-B0 1.0 0 BUS-C0 1.0 0BUS-A1BUS-A0 0.005 1.0 0BUS-B1BUS-B0BUS-A1BUS-A0 0BUS-C1BUS-C0BUS-A1BUS-A0 0BUSRA1 1.E-5 0BUSRB1 1.E-5 0BUSRC1 1.E-5 0BUS-OM 6.00E4 0 BUS-OM 33.333 /REQUEST PRINTED NUMBER WIDTH, 12, 2, { Request maximum precision (for 8 output columns) U.M. HEADER CARD { To be followed by significant middle card of U.M. Class 1 0 { In fact, a blank card would suffice for DC-35 /SOURCE C Any number of comments should be legal anywhere within Class-1 U.M. data C as now begins. Because of UMHC (U.M. HEADER CARD) use, only the first of C the 3 non-comment lines of Class-1 U.M. data ever will be seen. The crucial C 2nd will be replaced by the card following the UMHC declaration, and the C blank 3rd card is read and discarded, so will not been seen in the .LIS C file. Yet, it is required. Even though not-used, the 3rd is checked C for being blank, and execution will be halted if it is not so. Add "XXX" C at the start of the 2nd to emphasize that content is arbitrary: 19 { Data for U.M. begins with source type 19 keyed in columns 1-2 XXX 0 { 2nd of 3 non-comment header lines for U.M. will be replaced by UMHC crd BLANK card ending U.M. header (3rd of 3 never will been seen in .LIS file) 4 1 1111BUS-OM 2 0.06 0.18850 { 1st machine table 0.0 0.020291 0 0.0 0.020291 0 0.063 0.0001911 BUS-A1 1 0.0 { 1st winding card 0.063 0.0003925 BUS-B1 1 0.0 0.063 0.0003925 BUS-C1 1 0.0 0.083 0.0003925 BUSRB1 1 0.0 0.083 0.0003925 BUSRC1 1 0.0 0.083 0.0003925 BUSRA1 1 0.0 { 6th and final winding card /PLOT 2Arbitrary 78-character case title text of which this is an example, I hope. First of two lines of 78-byte graph subheading text. Second and final such line of graph subheading text. BLANK card ending all BRANCH cards BLANK card ending all SWITCH cards BLANK terminates the last SOURCE card BLANK card ends OUTPUT variable requests BLANK card ending all batch-mode PLOT cards BEGIN NEW DATA CASE C 7th of 9 subcases continues series using /UM on DC-35 U.M. data C 4th of 6 examples that have the same solution but slightly different data C Data of the preceding 3rd example has been massaged. First, "/" cards C for sorting have been removed to demonstrate that UMHC use does not C require sorting. The full 3-card Class-1 U.M. data has been replaced C by the special Type-19 summary line which will refer to UMHC for the C missing 2nd nonblank card of Class-1 data. Finally, since no sorting C to insert the blank card ending U.M. data, this must be added manually. PRINTED NUMBER WIDTH, 12, 2, { Request maximum precision (for 8 output columns) U.M. HEADER CARD { To be followed by significant middle card of U.M. Class 1 0 { In fact, a blank card would suffice for DC-35 .000400 .004 { Take only 10 time steps -- enough to begin motor rotation 5 -1 { No plotting (which has nothing special to do with /UM use) 0 BUS-A0 1.0 0 BUS-B0 1.0 0 BUS-C0 1.0 0BUS-A1BUS-A0 0.005 1.0 0BUS-B1BUS-B0BUS-A1BUS-A0 0BUS-C1BUS-C0BUS-A1BUS-A0 0BUSRA1 1.E-5 0BUSRB1 1.E-5 0BUSRC1 1.E-5 0BUS-OM 6.00E4 0 BUS-OM 33.333 BLANK card ending all BRANCH cards BLANK card ending all SWITCH cards 14BUS-A0 180.0 60.0 0.0 14BUS-B0 180.0 60.0 -120.0 14BUS-C0 180.0 60.0 +120.0 11BUS-OM-1 -6.00 19 { Class-1 U.M. data has been omitted WSM 4 1 1111BUS-OM 2 0.06 0.18850 { 1st machine table 0.0 0.020291 0 0.0 0.020291 0 0.063 0.0001911 BUS-A1 1 0.0 { 1st winding card 0.063 0.0003925 BUS-B1 1 0.0 0.063 0.0003925 BUS-C1 1 0.0 0.083 0.0003925 BUSRB1 1 0.0 0.083 0.0003925 BUSRC1 1 0.0 0.083 0.0003925 BUSRA1 1 0.0 { 6th and final winding card BLANK card ends U.M. data BLANK terminates the last SOURCE card C BUS-A1BUS-B1BUS-C1BUS-A0BUS-OM { Shorten output by eliminating node voltages BLANK card ends OUTPUT variable requests 2Arbitrary 78-character case title text of which this is an example, I hope. First of two lines of 78-byte graph subheading text. Second and final such line of graph subheading text. BLANK card ending all batch-mode PLOT cards BEGIN NEW DATA CASE C 8th of 9 subcases continues series using /UM on DC-35 U.M. data C 5th of 6 subcases that have the same solution but slightly different data C Data is like the 2nd example except that a second and third disconnected C machine has been added to illustrate multi-machine use for 3 machines. .000400 .004 { Take only 10 time steps -- enough to begin motor rotation 5 -1 { No plotting (which has nothing special to do with /UM use) /OUTPUT C BUS-A1BUS-B1BUS-C1BUS-A0BUS-OM { Shorten output by eliminating node voltages /SOURCE 14BUS-A0 180.0 60.0 0.0 14BUS-B0 180.0 60.0 -120.0 14BUS-C0 180.0 60.0 +120.0 11BUS-OM-1 -6.00 /BRANCH 0 BUS-A0 1.0 0 BUS-B0 1.0 0 BUS-C0 1.0 0BUS-A1BUS-A0 0.005 1.0 0BUS-B1BUS-B0BUS-A1BUS-A0 0BUS-C1BUS-C0BUS-A1BUS-A0 0BUSRA1 1.E-5 0BUSRB1 1.E-5 0BUSRC1 1.E-5 0BUS-OM 6.00E4 0 BUS-OM 33.333 /REQUEST U.M. HEADER CARD { To be followed by significant middle card of U.M. Class 1 0 { In fact, a blank card would suffice for DC-35 PRINTED NUMBER WIDTH, 12, 2, { Request maximum precision (for 8 output columns) /SOURCE 19 { Class-1 U.M. data has been omitted (this card is needed because of /SOURCE) 4 1 1111BUS-OM 2 0.06 0.18850 { 1st machine table 0.0 0.020291 0 0.0 0.020291 0 0.063 0.0001911 BUS-A1 1 0.0 { 1st winding card 0.063 0.0003925 BUS-B1 1 0.0 0.063 0.0003925 BUS-C1 1 0.0 0.083 0.0003925 BUSRB1 1 0.0 0.083 0.0003925 BUSRC1 1 0.0 0.083 0.0003925 BUSRA1 1 0.0 { 6th and final winding card C 14BUSRA1 180.0 60.0 0.0 /PLOT PRINTER PLOT { Batch-mode plotting is for the line printer not vector-graphic C ---------------- Done with data of 1st (real) U.M. Begin data of 2nd: C 2nd, unused U.M. has the same data as the 1st, but it is disconnected, C so should not affect the solution. Outputs have been suppressed. For C variety, this U.M. still uses /UM whereas the first has been converted C to /SOURCE followed by the special Type-19 header card. Note that the C Type-19 card is mandatory because any source might follow /SOURCE whereas C only U.M. data can follow /UM (i.e., the "19" is needed for recognition C as U.M. data). For the 2nd, disconnected machine, note that the original C A3 text BUS has become GUS everywhere. To restore the 9 outputs of C the 2nd machine, add "111" to cols. 7-9 of the card immediately following C the /UM declaaration, and change "0" to "1" in column 47 of each of C the 6 winding cards that end machine data. If this is done, output consists C of 2 lines each time step, and the two will be identical. /UM 4 1 1 GUS-OM 2 0.06 0.18850 { 1st machine table 0.0 0.020291 0 0.0 0.020291 0 0.063 0.0001911 GUS-A1 0 0.0 { 1st winding card 0.063 0.0003925 GUS-B1 0 0.0 0.063 0.0003925 GUS-C1 0 0.0 0.083 0.0003925 GUSRB1 0 0.0 0.083 0.0003925 GUSRC1 0 0.0 0.083 0.0003925 GUSRA1 0 0.0 { 6th and final winding card /BRANCH 0 GUS-A0 1.0 0 GUS-B0 1.0 0 GUS-C0 1.0 0GUS-A1GUS-A0 0.005 1.0 0GUS-B1GUS-B0GUS-A1GUS-A0 0GUS-C1GUS-C0GUS-A1GUS-A0 0GUSRA1 1.E-5 0GUSRB1 1.E-5 0GUSRC1 1.E-5 0GUS-OM 6.00E4 0 GUS-OM 33.333 /SOURCE 14GUS-A0 180.0 60.0 0.0 14GUS-B0 180.0 60.0 -120.0 14GUS-C0 180.0 60.0 +120.0 11GUS-OM-1 -6.00 C ---------------- Done with data of 2nd U.M. Begin data of 3rd: C 3rd, unused U.M. has the same data as the 1st, but it is disconnected, C so should not affect the solution. Outputs have been suppressed. Etc. C (like the 2nd except that the original BUS ---> MUS here). /UM 4 1 1 MUS-OM 2 0.06 0.18850 { 1st machine table 0.0 0.020291 0 0.0 0.020291 0 0.063 0.0001911 MUS-A1 0 0.0 { 1st winding card 0.063 0.0003925 MUS-B1 0 0.0 0.063 0.0003925 MUS-C1 0 0.0 0.083 0.0003925 MUSRB1 0 0.0 0.083 0.0003925 MUSRC1 0 0.0 0.083 0.0003925 MUSRA1 0 0.0 { 6th and final winding card /BRANCH 0 MUS-A0 1.0 0 MUS-B0 1.0 0 MUS-C0 1.0 0MUS-A1MUS-A0 0.005 1.0 0MUS-B1MUS-B0MUS-A1MUS-A0 0MUS-C1MUS-C0MUS-A1MUS-A0 0MUSRA1 1.E-5 0MUSRB1 1.E-5 0MUSRC1 1.E-5 0MUS-OM 6.00E4 0 MUS-OM 33.333 /SOURCE 14MUS-A0 180.0 60.0 0.0 14MUS-B0 180.0 60.0 -120.0 14MUS-C0 180.0 60.0 +120.0 11MUS-OM-1 -6.00 BLANK card ending all BRANCH cards BLANK card ending all SWITCH cards BLANK terminates the last SOURCE card BLANK card ends OUTPUT variable requests BLANK card ending all batch-mode PLOT cards BEGIN NEW DATA CASE C 9th of 9 subcases ends the series using /UM on DC-35 U.M. data C 6th of 6 examples that have the same solution but slightly different data C Data is the same as the 2nd example except that /UM data has been moved C to a $INCLUDE file. Use or non-use of $INCLUDE should have no effect, C and this demonstrates none for this simplest, first level of $INCLUDE. C Enhancement by Gabor Furst. WSM's single $INCLUDE was replaced by C two for more generality. A second induction motor was added to show C that this is possible. Mr. Furst observed the following on 9 April C 2007: "I thought we could have this sub case with two rather than C one $INCLUDE file. I attach my version of this, which demonstrates C how adding another machine is simply adding another $INCLUDE file." C Mr. Furst simulated to Tmax = 1.044 seconds but program developers C truncate this drastically (to 10 steps) to speed execution. This is C just enough to start the rotors and show approximate balance. To see C the entire settling into the steady state, restore the longer T-max, C remove the minus sign from IPLOT immediately below T-max, and C activate the batch-mode plot card (now omitted by $DISABLE). Finally, C control printout (every 5th step for IOUT is a lot, for realistic use). $PREFIX, [] { $INCLUDE files are located in same place as this main data file .000400 .004 { Take only 10 time steps -- enough to begin motor rotation 5 -1 { No plotting (which has nothing special to do with /UM use) /OUTPUT C BUS-A1BUS-B1BUS-C1BUS-A0BUS-OM { Shorten output by eliminating node voltages /SOURCE 14BUS-A0 180.0 60.0 0.0 14BUS-B0 180.0 60.0 -120.0 14BUS-C0 180.0 60.0 +120.0 11BUS-OM-1 -6.00 14BUS-A3 180.0 60.0 0.0 14BUS-B3 180.0 60.0 -120.0 14BUS-C3 180.0 60.0 +120.0 11BUS-3M-1 -6.00 /BRANCH C the following branch records are transferred to $INCLUDE dc33inc1.dat C 0 BUS-A0 1.0 C 0 BUS-B0 1.0 C 0 BUS-C0 1.0 C 0BUS-A1BUS-A0 0.005 1.0C C 0BUS-B1BUS-B0BUS-A1BUS-A0 C 0BUS-C1BUS-C0BUS-A1BUS-A0 C 0BUSRA1 1.E-5 C 0BUSRB1 1.E-5 C 0BUSRC1 1.E-5 C 0BUS-OM 6.00E4 C 0 BUS-OM 33.33 C 0 BUS-A3 1.0 0 BUS-B3 1.0 0 BUS-C3 1.0 0BUS-A4BUS-A3 0.005 1.0 0BUS-B4BUS-B3BUS-A4BUS-A3 0BUS-C4BUS-C0BUS-A4BUS-A3 0BUSRA4 1.E-5 0BUSRB4 1.E-5 0BUSRC4 1.E-5 0BUS-3M 6.00E4 0 BUS-3M 33.333 $INCLUDE, dc33inc1.dat, { Separate file for /UM followed by all data for U.M. 1 $INCLUDE, dc33inc2.dat, { Separate file for /UM followed by all data for U.M. 2 C About the preceding data modularization, there is lack of balance. This C is intentional. Whereas DC33INC1 has /BRANCH data, DC33INC2 has none. C Each also involves /REQUEST which precedes a UMHC request. Since only C one has meaning, expect rejection accompanied by a warning message for the C 2nd of the two. BLANK card ending all BRANCH cards BLANK card ending all SWITCH cards BLANK terminates the last SOURCE card BLANK card ends OUTPUT variable requests CALCOMP PLOT $DISABLE { The following plot card assumes use of Gabor Furst's longer T-max 193 .1 0.0 1.0 UM-1 OMEGM UM-2 OMEGM Induction motor w in [rad/sec] $ENABLE BLANK card ending all batch-mode PLOT cards BEGIN NEW DATA CASE BLANK