Hallo, I'm using scad3 for a few weeks and I'm very happy about the power and speed of simulation. Now, I tried a more complex simualtion with a LT1533 (I modified the sample) and got the following error: Analysis: Timestep too small; time=0.000173642, timestep=1.25e-019: trouble with node "n001" What does that mean and how can I prevent this error. I tried to modify the .tran command to use less steps, but this doesn't help. Thanks, Rainer

I need a model for a gas discharge tube similar to the Siemens A81- C90X, if anyone can help I'd appreciate it. I have just started to use LTSpice and would like to know if there is any information out there that shows how to convert a manufactures model to something that will keep LTSpice happy. I was hoping to use the DIAC that comes with LTSpice but found out I have to supply my own model. Any hints for a day-old user how to do this? So far it's a cool program.

I need a method to bind schematics to cell and block symbols that uses the symbol attributes, so that instantiating and placing a symbol causes that symbol's definition to be included as a subcircuit in the netlist. This needs to work a) when the symbol and schematic have the same name (albeit with differing extensions), b) when the symbol and schematic have different names (e.g., "XXX.asy" bound to "simplified_XXX.asc"), and c) when a generic symbol is bound after placement to a specific differently-named schematic (e.g., when binding different power supply designs into a common test framework). File hierarchy location of the files for the symbols and schematics is not important, as long as they are not placed under the SCad3 Program Files hierarchy. (They aren't LTC's.) The solution could require that all of the symbols and schematics files be in the same folder, probably the folder on which SCad3 is focussed. Or the solution could require replicating a lib hierarchy under the folder on which SCad3 is focussed. Or ??? Requiring the instantiating user to know the correlation between symbol and schematic and add a .lib statement for each such is unacceptable, just as it would be were a .lib to be needed for each LT part and discrete that is added to a schematic. Besides, adding a .lib statement does not solve the problem of binding a symbol with one name to an instance with a different name, as occurs in the LT-provided symbols (e.g., CNY17-3.asy binding to CNY17.sub as in last example below) Placing a pre-bound symbol should be adequate when the symbol has a fixed binding to a subcircuit netlist. Placing an unbound symbol and setting one or two of the symbol instance's attributes to the name of the netlist should be adequate for binding a generic symbol to one of a set of equivalent netlists. (Note that a .lib cannot accomplish this, particularly when two identical generic symbols in the schematic are bound to different instance values, as occurs regularly with generic symbols for discretes.) I've tried binding the schematic to the symbol definition and/or to the symbol instance via the SpiceModel attribute. I've also tried binding it via the SpiceLine2 attribute by putting a "\n.lib " prefix before the schematic name, hoping that this would cause SCad3 to treat this as a separate hidden .lib declaration. Neither of these approaches works. Nor does removing the represented cell's or block's name, which is otherwise appended to the constructed SPICE line after the SpiceModel field. (Incidentally, cell's _do_ need names, so removing the name would not be an acceptable method either. But it was worth trying.:) Has anyone solved this problem? Does anyone have any suggestions? Is there a methodology for adding bogus pins to the symbol to cause the SPICE parse of the constructed SpiceLine to work? Mike? Helmut? Others? Two example symbols and schematics follow: *******attempt to use SpiceModel attribute******* ==============diode bridge symbol================ Version 4 SymbolType CELL LINE Normal -20 16 -20 -16 LINE Normal -44 16 -44 -16 LINE Normal -44 -16 -20 0 LINE Normal -44 16 -20 0 LINE Normal -64 0 -44 0 LINE Normal -20 0 0 0 LINE Normal 60 64 60 32 LINE Normal 36 64 36 32 LINE Normal 36 32 60 48 LINE Normal 36 64 60 48 LINE Normal 0 48 36 48 LINE Normal 60 48 80 48 LINE Normal 60 16 60 -16 LINE Normal 36 16 36 -16 LINE Normal 36 -16 60 0 LINE Normal 36 16 60 0 LINE Normal 16 0 36 0 LINE Normal 60 0 80 0 LINE Normal -20 -32 -20 -64 LINE Normal -44 -32 -44 -64 LINE Normal -44 -64 -20 -48 LINE Normal -44 -32 -20 -48 LINE Normal -64 -48 -44 -48 LINE Normal -20 -48 16 -48 LINE Normal 16 -48 16 -48 LINE Normal 16 0 16 -48 LINE Normal 80 -48 16 -48 LINE Normal 0 48 0 0 LINE Normal -64 48 0 48 RECTANGLE Normal 80 72 -64 -71 TEXT -16 -96 Left 0 BRW40 WINDOW 0 -51 59 Left 0 SYMATTR Prefix X SYMATTR Description Diode bridge SYMATTR SpiceModel BRW40.asc PIN -64 -48 NONE 0 PINATTR PinName A1 PINATTR SpiceOrder 1 PIN -64 0 NONE 0 PINATTR PinName A2 PINATTR SpiceOrder 2 PIN -64 48 NONE 0 PINATTR PinName AC2 PINATTR SpiceOrder 3

I recently discovered LTSpice and I love the program. My question is the following: I wish to modify the model parameters of say, the NPN bipolar transistor, so as to see the impact of various parameters on ECL gate delay. I know how to do this on the netlist, but I cannot see how I can do this from the schematic entry - when I right click on the NPN, where can I modify things like BF, Is, etc.? I have not found anything about this in the help. Did I miss it? Suggestion: I appreciate LTSPice is "free", so one can only do so much work on it in terms of filling requests... But here is mine anyway. I would be ecstatic if we could plot/display S-parameters within LTSpice, in a Smith Chart and Bode format. I managed to plot them, but one needs to build this fictitious circuit around the real circuit in order to extract the S-parameters. And then one needs to write out huge expressions to plot H21, MAG/MSG, and U... If Linear Technology is interested in the RF market, it might make sense to add S-parameter tools down the line. Being able to export data w/o the Ltutil.exe would be good too. Thanks for a great tool. Cheers! Bolo

OK where did I go wrong, I'm getting - Could not open include file "MYLIBRARY1.LIB" I copied the switch out of help, and the library is in the sub directory. MYLIBRARY1.lib *SYM=CSW .MODEL CSWITCH1 1 2 W1 out 0 Vsense CSWITCH1 Vsense a b 0. .model CSWITCH1 CSW(Ron=.1 Roff=1Meg Vt=0 Vh=-.5) .ENDS Version 4 SHEET 1 892 692 WIRE 448 384 448 448 WIRE 448 528 448 560 WIRE 448 560 208 560 WIRE 208 560 208 384 WIRE 208 560 208 576 WIRE 208 304 448 304 FLAG 208 576 0 SYMBOL C:\PROGRAM\ FILES\LTC\SWCADIII\lib\sym\voltage 208 288 R0 WINDOW 123 24 132 Left 0 WINDOW 39 24 160 Left 0 WINDOW 0 42 52 Left 0 SYMATTR Value2 AC 20 SYMATTR SpiceLine Rser=1 SYMATTR InstName V1 SYMATTR Value SINE(0 20 5000) SYMBOL C:\PROGRAM\ FILES\LTC\SWCADIII\lib\sym\csw 448 304 R0 WINDOW 0 55 20 Left 0 SYMATTR InstName W1 SYMATTR Value CSWITCH1 SYMBOL C:\PROGRAM\ FILES\LTC\SWCADIII\lib\sym\res 432 432 R0 SYMATTR InstName R1 SYMATTR Value 10 TEXT 488 560 Left 0 !.include MYLIBRARY1.lib TEXT 174 600 Left 0 !.tran 0 .1 .001 .001 * C:\Program Files\LTC\schematics\current.asc V1 N002 0 SINE(0 20 5000) AC 20 Rser=1 W1 N002 N001 CSWITCH1 R1 N001 0 10 .include MYLIBRARY1.lib .tran 0 .1 .001 .001 .backanno .end

What's the best way of creating an opamp model if the manufacturer does not have a spice model for it? Thanx, -Jon

Hello Helmut. Thank you for your advice. I was able to modify the transistor models from the SPICE directive, or even better, by editing the model files for the NPNs. The .STEP is a good suggestion when you wish to play with a single parameter. As for you question regarding S-parameters, I have included a netlist that shows how to measure S11 and S21 for a simple two-port (consisting of a single shunt resistor Rtest2port) which you can substitute by your circuit. Then, you duplicate the circuit and switch the sources from the input to the output to "measure"S22 and S12. Then you plot Vs11, Vs22, etc, which correspond to S11, S22… Once that is done you can go to the plot window, and plot various quantities like the Unilateral power gain by clicking ADD TRACE and then paste the expression for U in the window as (((MAG(V(s21)/V(s12)-1))**2)/2/(((1-(MAG(V(s11)))**2-(MAG(V(s22)))**2+ (MAG(V(s11)*V(s22)-V(s12)*V(s21)))**2)/2/MAG(V(s12)*V(s21)))*MAG(V(s21 )/V(s12)) - Re(V(s21)/V(s12))))**0.5 (note that the square root is needed because LTSPICE takes 20log of everything in the dB scale) NETLIST TO MEASURE S11 and S21 ================================================================== * C:\Program Files\LTC\SwCADIII\CB_Files\FET_model.asc V1 N001 0 0 AC 1 0 Rser=0 Cpar=0 R6 N002 N001 50 E1 N003 0 N002 0 2 R8 S11 0 1G V2 N003 S11 0 AC 1 0 Rser=0 Cpar=0 R7 N002 0 50 E2 S21 0 N002 0 2 R9 S21 0 1G Rtest2port N002 0 100 .ac dec 99 1G 300G .backanno .end

Has anyone migrated any of the vacuum tube models floating around on the 'net to LTspice? Bill

I was looking around for a model of a low voltage MUX like the 74lv4051. I found a a 6 pin fragment in the Philips LV library (http://www.philipslogic.com/support/spice/) called SWI1. I wired it up and it ran fine with DC control input, but when I attempted to toggle the switch with a square wave, once it turned off it never turned on again. Any ideas what's going on here and or pointers to a fully functional part model. Thanks SH

I want to start out by saying that I'm a hobbyist, not an electronics professional. When I first heard of SwitcherCAD III/LT Spice perhaps a year ago, I went to the Linear web site with the intent to download the program and completely failed to understand what I almost had in hand. From the descriptions, it looked like some kind of specialized program for switcher-based power supplies and that was far too limited to be very interesting at the time. It wasn't until later reading of comments by one of the active members of the LT Spice team (Mike) that I began to realize my mistake and attempted to download LT Spice in January. Boy, was that a good decision. In the past, I've tried, and actually purchased, a few other programs claiming to provide electronics simulations. I've had a crashed Windows, substantial memory leaks, and overly complex setup for simulation and overly complex arrangements to observe the signal values I'm interested in (in some cases, having to manually wire up a "meter" in order to get those results, and even then with difficulty.) By comparison, and this isn't simply a false impression because of my building on prior experience, LT Spice was a dream to use. Once I set down to seriously confront a simple circuit, it took me no time at all (and without reading any help files) to create it and get results I was interested in seeing. This stands in stark contrast to past experience. As I've spent a little more time in LT Spice, every moment has been well paid back by discovery of still more useful and interesting features. It's actually such a joy, that I've gone out and purchased an original Spice II PhD thesis from 1975, recommended by Mike, and another similar book and have found all of it very useful. Every minute has been well-paid and I can barely imagine a tool which is easier to use. Of course, I'm still too much a neophyte, both in electronics knowledge as well as in simulation depth, to know whether or not this program provides simulations which are incorrect in some important details and which will lead me into very wrong impressions which will be hard to unlearn. But I'm pretty confortable in my current belief, tested in a few cases I've actually built and used, that it's providing good results. And comforted aslo by the active support of this program. All that is by way of introduction for a question I have: I've taken advantage of .param in order to allow me to specify certain important values for a basic degenerative amplifier design. The components used in this model design are all specified with expressions using the set signs, {}. Simulation results look encouraging to me, now, but I'd like to see what the results of the calculations for these expressions were, in simulation, so that I can now specify real values and build one for testing. I could go use my calculator, of course, and figure these out. But I'd like LT Spice to display the results of its own calculations without my having to use a calculator, by hand. Is this easily doable? My guess is that I could write an expression into the graphical display as another trace and just see the value of the straight line, but that's not a good way to do this. What I'm considering as "nice" would be to right-click over the expression on the schematic and call up a dialog box for extering the expression, but where this dialog box would *also* show the current result of the existing expression. Anyway, that's my question for now. Thanks to all! Jon

In my PC running Linux (SuSe 8.1) I installed wine, then I installed SwitcherCADIII. The program seems to work well, I have only a little problem, I am not able to update the program using internet.If I try to sync release I get the the message "could not access web ..." I have the connection to internet active, but I get this message. I dont know if using wine it is possible to update the program. Someone can help me ? Stefano Delfiore

Are there any special considerations with using P-spice op-amp macro models with LTspice? I'm trying to simulate a circuit that uses a Burr Brown/TI electrometer op-amp (OPA128 or OPA129) and I'm getting incorrect results. I downloaded the models from TI's website. The circuit's been built and thoroughly debugged, so I'm confident in the design. When I couldn't get the circuit to operate correctly, I entered a simple test circuit to verify some of the basic op-amp characteristics. I found that the input bias current was way to high (about 12nA instead of 100fA), and the gain-bandwidth of the OPA129 was high by about a factor of 10. I suspect that the models are just screwed up, but since I'm new to LTspice I'm not completely confident with my conclusion. Any help would be greatly appreciated. Thanks, John

I know that I can create my own library files and use .include to specify them. But is it possible to extend the dialog box list for the NPN transistors, for example, so that I can simply select from the list rather than overriding the "Value" on the schematic, directly? It's nice to look over the parameter list, sometimes, to look for the best of some particular parameter, without having to print out my library file. And if there is a way, can it be secure against the automatic update process, so that I keep my additions in the face of regular updates? Or if I should decide to forget about the standard components included with LT Spice and start building my own large library of parts with trusted parameters, can I get them to be popped into these selection lists? Jon

I'm interested in calculating iterative values for resistors. For example, an emitter resistor's value may be based on the emitter voltage divided by some current I need to have. However, the emitter voltage is, of course, dependent on the emitter/collector current which will depend on the emitter resistor value. An iterative process could track down this value closely, but I don't know how to write the expression. Here's an example of what I mean: Vcc | ,------x | | \ | / R2 | \ | | | | |/e Q2 x----| PNP | |\c | | | '-----> to the load |/c Q1 control |\e signal | at Vdd | or 0V, \ not at / R1 Vcc \ | | gnd In this case, R1 needs to be: R1 = (Vdd - Vbe(Q1))/(I(load)/10+Vbe(Q2)/R2+Ib(Q1)) I want to set up I(load) to some value and have R1 computed from that. When I type V(N001,N003) as part of the expression, in order to get the Vbe(Q1) for example, it's rejected with "Unable to find definition of model v - default assumed." This isn't what I want, naturally. Is there any way to set up the expression for R1, appropriately? Jon

Hello, I'm trying to use the National LMC6484A opamp model in LTSpice. I've downloaded their model and placed it into ..\LTC\SWCADIII\lib\sub directory under the name LMC6484A.sub. Then I created a simple test schematic where I used opamp2 symbol with Prefix = X, Spice Model = LMC6484A.sub, Value = LMC6484A.sub properties. I've also added the .inc LMC6484A.sub directive to the netlist. Running the simulation produces the following error message: Error: Unknown subckt called in: xu1 ...... lmc6484a.sub lmc 6484a.sub What am I doing wrong? Thanks, Eugene.

I am trying to run a Burr Brown Op Amp, the OPA336. The transient analysis "sort of" runs but LTSpice notes a couple of problems, most notably unrecognized parameters, jssw and vfb. The AC analysis fails to converge. The comments in the deck offer some advice with regard to modifyng ITL and gmin via .OPTION which doesn't seem to help. Are these more PSPICE problems? Thanks

I'd appreciate any help on modeling valves. For example, I copy what I have for the triode 12AX7A: * 12AX7A Triode PSpice Model 8/96, Rev. 1.0 (fp) * * ------------------------------------------------------------------- * This model is provided "as is", with no warranty of any kind, * either expressed or implied, about the suitability or fitness * of this model for any particular purpose. Use of this model * shall be entirely at the user's own risk. * * For a discussion about vacuum tube modeling please refer to: * W. Marshall Leach, jr: "SPICE Models for Vacuum-Tube Amplifiers"; * J. Audio Eng. Soc., Vol 43, No 3, March 1995. * ------------------------------------------------------------------- * * This model is valid for the following tubes: * 12AX7A/ECC83, 7025, 6EU7, 6681, 6AV6, 12DW7/7247 (Unit #1); * at the following conditions: * Plate voltage : 25..400V * Grid voltage : 0..-3.5V * Cathode current: 0..8mA * * * Connections: Plate * | Grid * | | Cathode * | | | .SUBCKT 12AX7A P G K E1 2 0 VALUE={45+V(P,K)+95.43*V(G,K)} R1 2 0 1.0K Gp P K VALUE={1.147E-6*(PWR(V(2),1.5)+PWRS(V(2),1.5))/2} Cgk G K 1.6P Cgp G P 1.7P Cpk P K 0.46P .ENDS 12AX7A.SUBCKT 12AX7A P G K E1 2 0 VALUE={45+V(P,K)+95.43*V(G,K)} R1 2 0 1.0K Gp P K VALUE={1.147E-6*(PWR(V(2),1.5)+PWRS(V(2),1.5))/2} Cgk G K 1.6P Cgp G P 1.7P Cpk P K 0.46P ... and the obvious question: How do I relate the previous info with the symbol. Thanks in advance, Guillermo

Thanks for the help on the Burr Brown amp. Noise analysis is a nice feature of LTSpice. It's helpful to poke around a circuit to see where noise is being generated. However, I didn't put alot of credence into the actual predictions because of potential limitations in the SPICE models. Out of curosity, I compared the RMS noise in an actual circuit using a couple of different op amps, including the LT2078A. I found that the predicted noise was fairly close to the actual measured values. The circuit is basically DC-coupled so 1/f noise is expected to be significant. Is this a fluke? Is there any way to tell if a model will predict noise performance in general and 1/f noise in particular. Steve H.

Hi! I too am new to spice. I heard that you were supposed to export a output-waveform to a Wav-file somehow. Can any-one tell me how? Would be cool. Cheers, Sean

Guten Tag, was sind das fr Fehler in LT-Spice? Gerechnet mit LT-Spice und Winspice3. Hello, what sort of errors is this in LT-Spice? The Schematic is computed with LT-Spice and Winspice3 from Mike Smith. I didn't find any control to max Time Steps in the .trans analysis. Arnold BEGRE00 Begrenzer mit Transistoren *************************************************************************** * * * B E G R E N Z E R T R A N S I S T O R E N 0 0 . C I R * * * * Begrenzer mit Transistoren und Dioden in der Gegenkopplung * * Benutzter OPA: TL 051 * * * * 20.03.2003 Arnold Esper * * * *************************************************************************** *************************************************************************** * * * Definition der Eingangsspannung VIN zwischen Knoten 1 und 0 mit AC * * und Puls, AC mit 1VOLT, der Puls wird festgelegt durch : * * * * PULS(U1 U2 T_VERZOEGER T_ANSTIEG T_ABFALL T_WEITE T_PERIODE) * * * * U2_|_ _ ______________ ____ * * | / \ / * * | / \ / * * | / \ / * * U1-|-------- - - - - - ------------------------ * * | * * * * T_VERZ |T_AN| T_WEITE |T_AB| * * | T_PERIODE | * * * *************************************************************************** * * * Definition einer Polygonquelle (piece-wise-linear) * * * * PULS(U1 U2 T_VERZOEGER T_ANSTIEG T_ABFALL T_WEITE T_PERIODE) * * * * _|_ ______________ * * | / \ * * | / \ * * | / \ * * u0-|------- - - - - - \- - - - - - - - - - * * | \ * * | \____________________________ * * * * | | | | | * * t0 u0 t1 u1 t2 u2 t3 u3 t4 u4 * * * *************************************************************************** *V1 1 0 DC 0 AC 1 PULSE(0 .6 100u 1m 1m 1n 1s) ************************ Polygon-Quelle ******************************** V1 1 0 DC 0 AC 1 PWL(0 0 1m 0V 5m -.6V 13m +.8V 17m 0V) *************************************************************************** R1 1 2 22K R2 2 4 22K R3 4 6 100K R4 6 7 22K D1 2 3 DI D2 7 5 DI Q1 3 6 7 BC550C Q2 5 4 2 BC550C *E0 7 0 0 2 100K X1 0 2 60 70 7 TL051/TI ********************* Betriebsspannungen VP VN *********************** VP 60 0 DC 15 VN 70 0 DC -15 ***************************** Analysen ********************************** *.OPTIONS LIMPTS=10000 *.AC DEC 100 10 20000 *.PRINT AC VDB(7) .TRAN 10u 20m .PRINT TRAN V(7) *.DC V1 -1 1 0.001 *.PRINT DC V(7) .model DI D *************************************************************************** .model BC550C NPN(Is=7.049f Xti=3 Eg=1.11 Vaf=23.89 Bf=493.2 Ise=99.2f + Ne=1.829 Ikf=.1542 Xtb=1.5 Br=2.886 Isc=7.371p + Nc=1.508 Ikr=5.426 Rc=1.175 Cjc=5.5p Mjc=.3132 Vjc=.4924 Fc=.5 + Cje=11.5p Mje=.6558 Vje=.5 Tr=10n Tf=420.3p Itf=1.374 Xtf=39.42 + Vtf=10) * PHILIPS pid=bc549c case=TO92 * 91-07-31 dsq **************************************************************************** * TL051 operational amplifier "macromodel" subcircuit * created using Parts release 4.01 on 04/12/89 at 09:57 * (REV N/A) * connections: non-inverting input * | inverting input * | | positive power supply * | | | negative power supply * | | | | output * | | | | | .subckt TL051/TI 1 2 3 4 5 * c1 11 12 3.988E-12 c2 6 7 15.00E-12 dc 5 53 dx de 54 5 dx dlp 90 91 dx dln 92 90 dx dp 4 3 dx egnd 99 0 poly(2) (3,0) (4,0) 0 .5 .5 fb 7 99 poly(5) vb vc ve vlp vln 0 2.875E6 -3E6 3E6 3E6 -3E6 ga 6 0 11 12 292.2E-6 gcm 0 6 10 99 6.542E-9 iss 3 10 dc 300.0E-6 hlim 90 0 vlim 1K j1 11 2 10 jx j2 12 1 10 jx r2 6 9 100.0E3 rd1 4 11 3.422E3 rd2 4 12 3.422E3 ro1 8 5 125 ro2 7 99 125 rp 3 4 11.11E3 rss 10 99 666.7E3 vb 9 0 dc 0 vc 3 53 dc 3 ve 54 4 dc 3.700 vlim 7 8 dc 0 vlp 91 0 dc 28 vln 0 92 dc 28 .model dx D(Is=800.0E-18) .model jx PJF(Is=15.00E-12 Beta=185.2E-6 Vto=-1) .ends **************************************************************************** .END