Up to this point you have done everything correctly. I assume you have actually saved Draft1.asc in the directory e:\...spice\ so that LTspice knows where it is.
As the file you want to include is in the same directory as the .asc file, you do not need to specify a full directory for the .inc statement, the file name alone is sufficient. (There are a lot of reasons why saving the model file in the same directory as the circuit .asc file is a good idea.)
Now, suppose you have a model in the file which you want to use in your circuit. Suppose it starts:
MODEL 2N1234C npn .......
Place an NPN transistor on the circuit. Next to the symbol are the "Reference Designator" Q1 (or Q2 Q3 etc) and the "Value" NPN. Right click on the Value NPN (not the transistor body) and a dialogue box opens "Enter new Value for Q1". Change the "NPN" to "2N1234C" exactly as shown in the model statement and LTspice will find the model.
If you right click on the transistor symbol after changing the Value, the Transistor Properties box shows which transistor model the symbol refers to; obviously it should show 2N1234C.
LTspice isn't really that fussy about the extensions of included files. I have a file called Models.txt which I include with the simple spice directive ".inc Models.txt". Anything can go in an included file, not just models. For example, if you have loads of .param, .option, .four etc etc statements cluttering up your circuit, they can all be put into a text file, say "params.txt", which can then be included in the normal way.
I'm not sure if I'm missing something here, but this is unusual.
I click on the simulation cmd, set the parameters for transient, ac analysis and noise.
All like I have done countless times in IV. XVII places them on the schematic like always.
However when I go to simulation cmd again to select the sim I want, they are all blank, and parameters need to be entered all over again.
I'm using one of the supplied examples, but it does this for all schematics I've tried so far.
I click on the simulation cmd, set the parameters for transient, ac analysis and noise.
All like I have done countless times in IV. XVII places them on the schematic like always.
However when I go to simulation cmd again to select the sim I want, they are all blank, and parameters need to be entered all over again.
I'm using one of the supplied examples, but it does this for all schematics I've tried so far.
Don
For better or worse the way of working with simulation commands was changed, by design, from v. IV to v. XVII. It caused heartache for many users at first, and was the subject of threads both here and at the LTspice Users' Group.
Mike Englehardt's idea seems to have been to make the simulation command editor something one uses to help create a command, not something to use to switch between commands. In his own words:
Schematic comment and SPICE directive may look similar to LTspice IV, but
it is very different. On XVII, the idea is to have an editor for multi-line
comments and invoke GUI's for each line that you chose to have help composing.
That is, you point at a line of text in the GUI that sets justification and
font size, right click and choose one of the "Help me Edit" items. As such,
there are no longer multiple stored commands in one line. If, from the schematic,
you point a particular line of a multi-line comment and right click, you jump
to that helper GUI but if you press escape, to fall back to the plain text
editor. You can CTRL-right click on text and directly enter the plain text
editor. []ltspice xv11 oddity
If you have multiple simulation commands, then you have to comment out, manually, those that you do not want to run, leaving only the one command active. To comment out a command you need to go to the plain text editor. So, either Ctrl-right click on a command to open the editor, or right-click on it to open the "Help me edit" dialogue box, then press Esc to go to the text editor. Once in the plain text editor you can click on "Comment" to comment out a command, or "Spice directive" to reactivate it.
One just has to get used to this way of working. You get used to it after a while...
For better or worse the way of working with simulation commands was changed, by design, from v. IV to v. XVII. It caused heartache for many users at first, and was the subject of threads both here and at the LTspice Users' Group.
Mike Englehardt's idea seems to have been to make the simulation command editor something one uses to help create a command, not something to use to switch between commands. In his own words:
Schematic comment and SPICE directive may look similar to LTspice IV, but
it is very different. On XVII, the idea is to have an editor for multi-line
comments and invoke GUI's for each line that you chose to have help composing.
That is, you point at a line of text in the GUI that sets justification and
font size, right click and choose one of the "Help me Edit" items. As such,
there are no longer multiple stored commands in one line. If, from the schematic,
you point a particular line of a multi-line comment and right click, you jump
to that helper GUI but if you press escape, to fall back to the plain text
editor. You can CTRL-right click on text and directly enter the plain text
editor. []ltspice xv11 oddity
If you have multiple simulation commands, then you have to comment out, manually, those that you do not want to run, leaving only the one command active. To comment out a command you need to go to the plain text editor. So, either Ctrl-right click on a command to open the editor, or right-click on it to open the "Help me edit" dialogue box, then press Esc to go to the text editor. Once in the plain text editor you can click on "Comment" to comment out a command, or "Spice directive" to reactivate it.
One just has to get used to this way of working. You get used to it after a while...
I ran into this when the new version came out and had a brief email exchange with Mike. I tried to make the case that the old LTspice IV dialogs were very useful in that they behaved as session variables, allowing one to make changes and quickly assess their effects within a session without having to re-enter data with each new simulation. I also argued (probably not the right word) that old LTspice already allowed the method which he proposed as the "correct" way and that what he really did was remove a useful feature without replacing it with something better. But Mike was quite inflexible on the matter.
I'm grateful that LTspice is available to hobbyists for free and realize that I should just be thankful for having access to it. But given the number of people that have questioned this decision it is a little disappointing that user opinion wasn't given more weight. The new LTspice is so clumsy in this respect that I don't use it; I'm still using LTspice IV, even though it is no longer officially supported.
I'm grateful that LTspice is available to hobbyists for free and realize that I should just be thankful for having access to it. But given the number of people that have questioned this decision it is a little disappointing that user opinion wasn't given more weight. The new LTspice is so clumsy in this respect that I don't use it; I'm still using LTspice IV, even though it is no longer officially supported.
Last edited:
Thanks Ludus for the explanation I never got from Mike.
I agree with Ray, that is is not something I need to get used to, but a very clumsy change that significantly slows down the process.
I've tried what you suggest, but still not getting what I want or need. I can make AC Analysis work, but DC operating point looks odd (contains the command from AC analysis, and I can get Transient to work once and only once. It always now defaults back to AC analysis. I have de-activated everything but transient, and it still tries to apply the settings to AC Analysis.
The Help section is a waste. Is there a tutorial on this?
Quite frankly if I can't make this work quickly AND efficiently it is a deal breaker.
I agree with Ray, that is is not something I need to get used to, but a very clumsy change that significantly slows down the process.
I've tried what you suggest, but still not getting what I want or need. I can make AC Analysis work, but DC operating point looks odd (contains the command from AC analysis, and I can get Transient to work once and only once. It always now defaults back to AC analysis. I have de-activated everything but transient, and it still tries to apply the settings to AC Analysis.
The Help section is a waste. Is there a tutorial on this?
Quite frankly if I can't make this work quickly AND efficiently it is a deal breaker.
Unfortunately, Dan, LTspice XVII is "working as designed" per Mike. When you go back and forth between transient and AC simulations within the same session your earlier variables get overwritten with ones from another tab. Of course, these overwrites make no sense for the simulation you are trying to run.
I'm afraid this is what we are stuck with. This is why I've gone back to LTspice IV. It doesn't have all of the features that LTspice XVII has but I've learned to work within LTspice IV's limitations.
I'm afraid this is what we are stuck with. This is why I've gone back to LTspice IV. It doesn't have all of the features that LTspice XVII has but I've learned to work within LTspice IV's limitations.
I get the impression that since LTspice is Mike's "baby" he is going to make it work the way he likes, and not care too much about what others think. I guess when you are the "master of your domain" you can do that.
Don
Looking at one of the examples that comes with LTspice, a subtlety has become apparent that I forgot to mention above. If you load the example 100W.asc (in the Educational folder) this will become clear.
There is a multi-line Spice directive:
.tran 10m
;.ac oct 50 1M 10Meg ; plot V(out)/V(out1) to see 72° phase margin @ 950kHz
;.dc temp -55 125 1
;.noise V(out) Vin oct 10 10 20K
.options plotwinsize=0 numdgt=15 noopiter
.four 1K V(out)
As it stands the transient simulation runs. Suppose you want to run the ac simulation. Ctrl-right click somewhere in the block to open the text editor. Because it is multi-line, clicking on "Comment" comments out the whole block, which is not particularly helpful. In this instance then, rather than using the Comment button, you need to comment out the .tran line with a semi-colon, and activate the .ac line by deleting the semi-colon.
For better or worse, I prefer to make each command a separate single-line Spice directive, and use the "Comment" and "Spice directive" radio buttons as described in post #1864.
I hope all this makes sense. In my opinion, it's worth switching to XVII, if only because it is regularly updated - IV has not been updated since 2016.
If you need more help, at least to see how to switch between sims in XVII, just post an .asc file...
Looking at one of the examples that comes with LTspice, a subtlety has become apparent that I forgot to mention above. If you load the example 100W.asc (in the Educational folder) this will become clear.
There is a multi-line Spice directive:
.tran 10m
;.ac oct 50 1M 10Meg ; plot V(out)/V(out1) to see 72° phase margin @ 950kHz
;.dc temp -55 125 1
;.noise V(out) Vin oct 10 10 20K
.options plotwinsize=0 numdgt=15 noopiter
.four 1K V(out)
As it stands the transient simulation runs. Suppose you want to run the ac simulation. Ctrl-right click somewhere in the block to open the text editor. Because it is multi-line, clicking on "Comment" comments out the whole block, which is not particularly helpful. In this instance then, rather than using the Comment button, you need to comment out the .tran line with a semi-colon, and activate the .ac line by deleting the semi-colon.
For better or worse, I prefer to make each command a separate single-line Spice directive, and use the "Comment" and "Spice directive" radio buttons as described in post #1864.
I hope all this makes sense. In my opinion, it's worth switching to XVII, if only because it is regularly updated - IV has not been updated since 2016.
If you need more help, at least to see how to switch between sims in XVII, just post an .asc file...
But you can do this with LTspice IV too. So this wasn't a new feature introduced with LTspice XVII. What the new version did was force you to run simulations this way (i.e., with multi-line directives) whether you like it or not. Mike's opinion was that this is the "correct" way.
But it is what it is, as the saying goes. We each have to decide whether the new version's virtues outweigh its downsides. For me, I haven't encountered a situation where LTspice IV couldn't do the job, but I also have LTspice XVII installed just in case I do run into a problem with the older version.
But it is what it is, as the saying goes. We each have to decide whether the new version's virtues outweigh its downsides. For me, I haven't encountered a situation where LTspice IV couldn't do the job, but I also have LTspice XVII installed just in case I do run into a problem with the older version.
Last edited:
Yes there's a new general manager who is looking to integrate all legacy LTC and ADI tools so they work together. I probably shouldn't reveal the name but it's a person I greatly respect whom I'm sure will do a great job.
LTSpice is considered one of the companies crown jewels, it's not going anywhere.
tommost
LTSpice is considered one of the companies crown jewels, it's not going anywhere.
tommost
Updates are the only reason I wanted to switch.
The .asc file I tried was the one of the examples (audioamp). So posting it is irrelevant.
I, like Ray, will stick with IV until such time that this poorly implemented GUI is brought back to what it should be. LTspice is a terrific tool for design that I have been using for many years. I hope this "problem" can be sorted out.
tommost, thanks for letting us know it will continue to be developed. Maybe you culd drop a few hints about the usability.
Thanks for your help guys.