Any Qucs users here?

[mashaffer]

I just downloaded Qucs for Mac power PC and it looks like it might be a pretty capable program but I am having trouble with my very first test circuit.

I set up a simple LC circuit with AC voltage source and was trying to do an AC analysis on it. The resulting netlist is...

# Qucs 0.0.11 /Applications/qucs0.0.11PowerPC/test.sch

L:L1 _net0 Output L="3.8 mH" I=""
Vac:V1 _net0 gnd U="1 V" f="1 khz" Phase="0" Theta="0"
C:C1 gnd Output C="5 uF" V=""
.AC:AC1 Type="log" Start="10" Stop="100000 Hz" Points="30" Noise="no"

When I run the simulation it gives me the following error message.

line 4: syntax error, unexpected Identifier, expecting '"'

I assume that the line number excludes blank lines and comments so that the last line is the problem.

Any ideas?

[planet10]

Do you have a link? I could download it and see if i can help.

dave

[mashaffer]

ATG Software Download

If you don't already have it you will need to download both Qt/mac3 (near top of page) and Qucs (about 6/10ths of the way down the page). BTW the Engauge program at this site is really cool too.

[mashaffer]

Not sure how but I got it working. Even though the netlist doesn't show it I think I had managed to get a fractional (e.g. 3.8) Steps per decade.

[mashaffer]

Starting to get the hang of it a little bit. Next steps are to figure out how to display results in dB and then how to get vacuum tube models setup.

[mashaffer]

I did find a tutorial that helped to figure out the dB display. This software is quite powerful and as is often the case that makes it a bit hard to get the hang of it. It turns out that the software has an equation calculating facility that adds your equations to the data set. Then you just graph the results of your equation instead of the raw data.

Here is an example of and LC High pass filter.



Note the equation defined at the bottom left uses the built in voltage dB function.

They also claim the ability to import many (most?) spice subcircuits so I suspect that the ability to simulate vacuum tubes most likely lies in that direction.

Another neat feature is that Qucs includes some interesting tools like filter synthesis.

[darenw]

Your original problem is in the line for Vac:V1 where it says f="1 khz". QUCS is really fussy and wants kHz not khz. You can fix this in the GUI popup properties for V1.

(And don't do like me and type in KHz. That's also no-go for QUCS.)

[mashaffer]

Thanks guys. I kind of forgot about this thread for a while (Geez I am only 49 and already...). I played around with a couple more simple sims and am starting to get a little better at it. The next step will be to attempt to get some tube models imported so I can use it for some real work.

[mashaffer]

Getting tube models converted from spice is not working at all. They have a nice feature where it can import random spice models but it won't work with models using the "VALUE" statement (95% of them) and the ones that don't seem to give random errors. The most common is [unexpected identifier, expecting ""].

So far I have had no success. I may have to learn to create models directly in qucs. Egad!

[jessedavis]

I you haven't already figured it out, the error message [unexpected identifier, expecting ""] is a universally meaningless error message. Often it has nothing to do with the current activity. Be VERY careful with units and multipliers. Things like Mhz instead of MHz. The program is case sensitive in a way I do not understand. For tube parameters remember that this program uses the standard spice engine so if a model works elsewhere it will work here, PROVIDED you get the parameter order and multipliers correct. When in doubt revert to the most basic form then add a parameter at a time until you figure it out. It is a powerful and useful program with many nice interface features.