looking for simulator

[colin]

Hi,
Im looking for a simulator - mixed mode with schematic capture, ive been
trying to evaluate several freeware/trial ones but when it takes half a day
trying to put a simple sinewave generator and r/c circuit and look at the
output and frequency plot etc i tend to give up with it. ive looked at the
simulation in protel before although its quite good i cant say i was happy
enough with it to pay the very high price for the whole package

Ive just looked at simetrix but although it seems quite powerfull i dont
have any more time to waste trying to figure out how to even do something
trivial. I tried to download ltspice but all i ended up with was some
library files, couldnt seem to find the main program.

To start with I wish to look at the spectrum of a phase modulated carrier
that gets divided with a counter then filtered then multiplied with a
frequency multiplier etc.

Can anyone sugest something that I can just get up and runing in a short
space of time ? ... thanks

Colin =^.^=

 

[James Meyer]

trivial. I tried to download ltspice but all i ended up with was some
library files, couldnt seem to find the main program.

Colin =^.^=

Keep trying to get LTSpice. I think you will find that it's worth it.
Did you get LTSpice directly from the Linear website?

Jim

 

[Tim Hubberstey]

Hi,
Im looking for a simulator - mixed mode with schematic capture, ive been
trying to evaluate several freeware/trial ones but when it takes half a day
trying to put a simple sinewave generator and r/c circuit and look at the
output and frequency plot etc i tend to give up with it. ive looked at the
simulation in protel before although its quite good i cant say i was happy
enough with it to pay the very high price for the whole package

Ive just looked at simetrix but although it seems quite powerfull i dont
have any more time to waste trying to figure out how to even do something
trivial. I tried to download ltspice but all i ended up with was some
library files, couldnt seem to find the main program.

To start with I wish to look at the spectrum of a phase modulated carrier
that gets divided with a counter then filtered then multiplied with a
frequency multiplier etc.

Can anyone sugest something that I can just get up and runing in a short
space of time ? ... thanks

I use LT Spice (aka SwitcherCAD III) when I need analog simulation. I
think you just picked the wrong thing to download. What you want is:

<http://www.linear.com/designtools/softwareRegistration.jsp>

If you don't end up with 'swcadiii.exe', you've done something wrong.

The basic library includes only minimal logic components so you'll need
to build your counter from FFs and gates, create a model from scratch,
or find a more complete digital library.

I don't use it a lot, but I find it to be quite well done and easy to
use, once you get a feel for the GUI.

 

[Genome]

I use LT Spice (aka SwitcherCAD III) when I need analog simulation. I
think you just picked the wrong thing to download. What you want is:

<http://www.linear.com/designtools/softwareRegistration.jsp>

If you don't end up with 'swcadiii.exe', you've done something wrong.

Or try....

http://ltspice.linear.com/software/swcadiii.exe

Which is the 'No Thanks, just download the software'.

Mike Englehart, the author, pops up every so often.

?For Mike, is there an advantage to registering?

The basic library includes only minimal logic components so you'll need
to build your counter from FFs and gates, create a model from scratch,
or find a more complete digital library.

You can get logic models from the LTspice user group at

http://groups.yahoo.com/group/LTspice/

In the files/lib section.

You have to join.

Helmut Sennewald, the moderator, pops up every so often.

Try posting questions in sci.electronics.cad

I've cross posted your query there.

Mike and Helmut are really helpful people. (Sorry if I pisspelt the names)

Tim Hubberstey, P.Eng

DNA

 

[colin]

Keep trying to get LTSpice. I think you will find that it's worth it.
Did you get LTSpice directly from the Linear website?

hi thanks, I had a closer look at the linear webby and downloaded
ltspice/switherchad thinking this just was for the switcher devices before.

after a while i managed to figure out ltspice, had to use a spice cmd cos i
cldnt find a signal source symbol to feed a fm modulated carrier into
several flip flops and now looking at the spectrum.

With I managed to enter several signal sources in simetrix, but i couldnt
seem to get the signal frequency to show up in the frequency plot it was all
flat, tried al the combinations i could think of of generator
options/simulater model/options ac/dc/tf etc ..

maybe im just expecting it to be too user friendly or my idea of what this
is is rather diferent to everyone elses, it just makes me want to write my
own.

Colin =^.^=

 

[Terry Pinnell]

Hi,
Im looking for a simulator - mixed mode with schematic capture, ive been
trying to evaluate several freeware/trial ones but when it takes half a day
trying to put a simple sinewave generator and r/c circuit and look at the
output and frequency plot etc i tend to give up with it. ive looked at the
simulation in protel before although its quite good i cant say i was happy
enough with it to pay the very high price for the whole package

Ive just looked at simetrix but although it seems quite powerfull i dont
have any more time to waste trying to figure out how to even do something
trivial. I tried to download ltspice but all i ended up with was some
library files, couldnt seem to find the main program.

To start with I wish to look at the spectrum of a phase modulated carrier
that gets divided with a counter then filtered then multiplied with a
frequency multiplier etc.

Can anyone sugest something that I can just get up and runing in a short
space of time ? ... thanks

Colin =^.^=

See my notes and links to some 60 ECAD programs at
http://www.terrypin.dial.pipex.com/ECADList.html
several of which have simulation facilities.

Terry Pinnell
Hobbyist, West Sussex, UK

 

[colin]

a

Or try....

http://ltspice.linear.com/software/swcadiii.exe

Which is the 'No Thanks, just download the software'.

Mike Englehart, the author, pops up every so often.

?For Mike, is there an advantage to registering?


You can get logic models from the LTspice user group at

http://groups.yahoo.com/group/LTspice/

In the files/lib section.

You have to join.

Helmut Sennewald, the moderator, pops up every so often.

Try posting questions in sci.electronics.cad

I've cross posted your query there.

Mike and Helmut are really helpful people. (Sorry if I pisspelt the names)


DNA

Thanks i got the real ltspice now, just looking at the FFT output now and
trying to figure out all the peaks, with a 1.6khz carrier and 10hz
modulation, the peaks at 10hz from the carrier are preserved but smaller
when divided as i expected, but theres a lot more small peaks close to the
carrier after several stages they get similar size to the expected
sidebands, wich i assume is due to digitisation. not sure if i can redcuce
this yet ...

Colin =^.^=

 

[Genome]

half

Thanks i got the real ltspice now, just looking at the FFT output now and
trying to figure out all the peaks, with a 1.6khz carrier and 10hz
modulation, the peaks at 10hz from the carrier are preserved but smaller
when divided as i expected, but theres a lot more small peaks close to the
carrier after several stages they get similar size to the expected
sidebands, wich i assume is due to digitisation. not sure if i can redcuce
this yet ...

Colin =^.^=

I..... am personally clueless.

The best I can do with an FFT plot is realise that I have to sample between
two points that start and end with the same level and are going in opposite
directions(?)

That's got something to do with FFT's or something. Otherwise I'd have to
figure out the other bits and pieces you can stick in the thing.....

Seee, I am THICK.


However..... in case you didn't know.....

LTspice saves your circuit diagrams as .asc files which are really .txt
files.

If you open up notepad (if you doze) then you can open up the .asc file and
copy then paste it in your newsgroup message so those who know something can
have a look and give better advice.

In reverse, if you select the text in a newsgroup message and paste into
notepad and save it with a .asc extension then you can load it into LTspice
and see what someone else is trying to do.

DNA

 

[Genome]

hi thanks, I had a closer look at the linear webby and downloaded
ltspice/switherchad thinking this just was for the switcher devices before.

after a while i managed to figure out ltspice, had to use a spice cmd cos i
cldnt find a signal source symbol to feed a fm modulated carrier into
several flip flops and now looking at the spectrum.

With I managed to enter several signal sources in simetrix, but i couldnt
seem to get the signal frequency to show up in the frequency plot it was all
flat, tried al the combinations i could think of of generator
options/simulater model/options ac/dc/tf etc ..

maybe im just expecting it to be too user friendly or my idea of what this
is is rather diferent to everyone elses, it just makes me want to write my
own.

Colin =^.^=

Hmmmmm....

Dunno, that's my view of softbollocks. But, if you play about a bit.....

Click on the AND gate in the menu thing to place a component.

Then doubbblelee click on [Special Functions]

Then click on 'modulate'

Have a look in the help file thing.

You might have to put in a behavioural voltage source and tell it to be the
absolute value of the output of the modulator plus something.

DNA

 

[colin]

looked powerfull with

I..... am personally clueless.

The best I can do with an FFT plot is realise that I have to sample between
two points that start and end with the same level and are going in opposite
directions(?)

That's got something to do with FFT's or something. Otherwise I'd have to
figure out the other bits and pieces you can stick in the thing.....

Seee, I am THICK.


However..... in case you didn't know.....

LTspice saves your circuit diagrams as .asc files which are really .txt
files.

If you open up notepad (if you doze) then you can open up the .asc file and
copy then paste it in your newsgroup message so those who know something can
have a look and give better advice.

In reverse, if you select the text in a newsgroup message and paste into
notepad and save it with a .asc extension then you can load it into LTspice
and see what someone else is trying to do.

DNA

Without much thinking I just ran the simulation in time domain mode, and if
u click on the graph u get an FFT option, i found u can select the
'alternate' simulator and the millions of spikes in the fft droped by about
40db.

I tend to run out of brain power these days so thats usefull after a cat
nap, although i just open up the software and select open last file used.

Just need to decide if its worth modeling a frequency multiplier.

Colin =^.^=

 

[colin]

hi thanks, I had a closer look at the linear webby and downloaded
ltspice/switherchad thinking this just was for the switcher devices before.

after a while i managed to figure out ltspice, had to use a spice cmd

cos

i
cldnt find a signal source symbol to feed a fm modulated carrier into
several flip flops and now looking at the spectrum.

With I managed to enter several signal sources in simetrix, but i couldnt
seem to get the signal frequency to show up in the frequency plot it was all
flat, tried al the combinations i could think of of generator
options/simulater model/options ac/dc/tf etc ..

maybe im just expecting it to be too user friendly or my idea of what this
is is rather diferent to everyone elses, it just makes me want to write my
own.

Colin =^.^=

Hmmmmm....

Dunno, that's my view of softbollocks. But, if you play about a bit.....

Click on the AND gate in the menu thing to place a component.

Then doubbblelee click on [Special Functions]

Then click on 'modulate'

Have a look in the help file thing.

You might have to put in a behavioural voltage source and tell it to be the
absolute value of the output of the modulator plus something.

DNA

the and gate icon just has simple logic gates, flip flop and a PD etc, it
doesnt seem to have a [special function] in the list, i asume the libraries
add more here ?

conveniently there is a spice directive (SFFM) for a FM modulated signal
wich made things easier.

Colin =^.^=

 

[Helmut Sennewald]

Without much thinking I just ran the simulation in time domain mode, and
if
u click on the graph u get an FFT option, i found u can select the
'alternate' simulator and the millions of spikes in the fft droped by
about
40db.

I tend to run out of brain power these days so thats usefull after a cat
nap, although i just open up the software and select open last file used.

Just need to decide if its worth modeling a frequency multiplier.

Colin =^.^=

Hello Colin,

Maybe choosing the "Alternate"-solver has simply changed
the automatic timestep control. I recommend to force a useful
timestep and to switch off any data compression. Data compression
is useful for SMPS simulations but not for low distortion application.
Watch how the spurious signals drop from -80dBc to -130dBc just
by reducing the timestep from 1u to 0.1u.

I have attached an example schematic file. Copy the whole following
text in a file "myfm1.asc" . Only the ".asc" counts. Let's hope
that no longer line will be split by any mail server.

Best regards,
Helmut





Version 4
SHEET 1 1244 680
WIRE -48 224 -48 160
WIRE -48 368 -48 304
WIRE 32 160 -48 160
FLAG -48 368 0
FLAG 32 160 fm
IOPIN 32 160 Out
SYMBOL voltage -48 208 R0
WINDOW 123 0 0 Left 0
WINDOW 39 0 0 Left 0
SYMATTR InstName V1
SYMATTR Value SFFM(0 1 10k 10 200)
TEXT -104 72 Left 0 !.tran 0 0.1 0 1u
TEXT -104 104 Left 0 !.options plotwinsize=0
TEXT 160 64 Left 0 ;.options reltol=1e-4
TEXT 384 64 Left 0 ;.options trtol=1e-4
TEXT -104 -32 Left 0 ;between 0.1u and 1u.
TEXT -104 0 Left 0 ;Watch the difference in the FFT.
TEXT -104 -64 Left 0 ;Try different timesteps in .tran
TEXT -104 -264 Left 0 ;F0=10kHZ Fm=0.2KHz dF=2kHz
TEXT -104 -304 Left 0 ;FM-Source
TEXT -104 -224 Left 0 ;Modulation Index = dF/Fm = 10
TEXT -104 -184 Left 0 ;Voltage Source: SFFM(0 1 10k 10 200)
TEXT -104 -144 Left 0 ;Checked with FFT of size 131072 points.
TEXT 392 -264 Left 0 ;Other FM and AM sources
TEXT 392 -224 Left 0 ;Components: Special Functions -> Modulate
TEXT 160 104 Left 0 ;Always use ".options plotwinsize=0"

 

[colin]

Hello Colin,

Maybe choosing the "Alternate"-solver has simply changed
the automatic timestep control. I recommend to force a useful
timestep and to switch off any data compression. Data compression
is useful for SMPS simulations but not for low distortion application.
Watch how the spurious signals drop from -80dBc to -130dBc just
by reducing the timestep from 1u to 0.1u.

I have attached an example schematic file. Copy the whole following
text in a file "myfm1.asc" . Only the ".asc" counts. Let's hope
that no longer line will be split by any mail server.

Best regards,
Helmut

thanks il try that, i reduced the max time step to 1ns, its crucnhing away
the numvbers as i write this, i increased the carier frequency to 32khz but
can no longer see the 10hz sidebands so trying to increase the data points
for the fft etc but i think i just reached the limit as it said it cldnt get
a contigous block of 131000000, although realisticaly id be surprised if it
can cope.

Colin =^.^=

 

[Helmut Sennewald]

thanks il try that, i reduced the max time step to 1ns, its crucnhing away
the numvbers as i write this, i increased the carier frequency to 32khz
but
can no longer see the 10hz sidebands so trying to increase the data points
for the fft etc but i think i just reached the limit as it said it cldnt
get
a contigous block of 131000000, although realisticaly id be surprised if
it
can cope.

Colin =^.^=

Hello Colin,

The resolution of a DFT or FFT is always 1/Time_span.
If you want see something at 10Hz, then you need a simulation
time which is a multiple of 1/10Hz, e.g. 0.1s, 0.2s, ...

Maybe you see now the drawback of such simulations in Spice if
you have a high carrier frequency. The simulation can take endless.

Best regards,
Helmut

 

[Jim Thompson]

Hello Colin,

The resolution of a DFT or FFT is always 1/Time_span.
If you want see something at 10Hz, then you need a simulation
time which is a multiple of 1/10Hz, e.g. 0.1s, 0.2s, ...

Maybe you see now the drawback of such simulations in Spice if
you have a high carrier frequency. The simulation can take endless.

Best regards,
Helmut

AND you need a small timestep, otherwise the "nose" sharpness is
meaningless.

...Jim Thompson

 

[Kevin Aylward]

Hi,
Im looking for a simulator - mixed mode with schematic capture, ive
been trying to evaluate several freeware/trial ones but when it takes
half a day trying to put a simple sinewave generator and r/c circuit
and look at the output and frequency plot etc i tend to give up with
it. ive looked at the simulation in protel before although its quite
good i cant say i was happy enough with it to pay the very high price
for the whole package

Ive just looked at simetrix but although it seems quite powerfull i
dont have any more time to waste trying to figure out how to even do
something trivial. I tried to download ltspice but all i ended up
with was some library files, couldnt seem to find the main program.

To start with I wish to look at the spectrum of a phase modulated
carrier that gets divided with a counter then filtered then
multiplied with a frequency multiplier etc.

Can anyone sugest something that I can just get up and runing in a
short space of time ? ... thanks

Yes. My opinion is that SuperSpice is the easiest simulator to use in
the known 3 universes.

Kevin Aylward
[email protected]
http://www.anasoft.co.uk
SuperSpice, a very affordable Mixed-Mode
Windows Simulator with Schematic Capture,
Waveform Display, FFT's and Filter Design.