What to use for simulation ?

A

Abbie

Guest
I need to simulate a simple analog circuit which has a led,
an scr, and all other components are passive. I tried Eagle,
but couldn't find any simulation tool there. I tried Circuit
Maker and I couldn't understand how to do it. Are there
any other simulator programs which are free ?
 
"Abbie" <AbbieAbbieAbbie@btinternet.com> schrieb im Newsbeitrag
news:bpoh36$9gn$1@hercules.btinternet.com...
I need to simulate a simple analog circuit which has a led,
an scr, and all other components are passive. I tried Eagle,
but couldn't find any simulation tool there. I tried Circuit
Maker and I couldn't understand how to do it. Are there
any other simulator programs which are free ?
Click to expand...
Hello Abbie,
if you take a look to this news group, you see a lot of discussions
around LTSPICE. LTSPICE is a free and unlimited SPICE simulator from
Linear Technogies. It is also named SwitcherCADIII.
You can download it from www.linear.com/software and there is a
very helpful news group www.groups.yahoo.com/ltspice .

Back to your blinking LED circuit in your other posting. It can never
work this way with a power SCR, because it cannot trigger the SCR with
your low gate current (0.1mA). It is also wrong that you have connected
directly the gate to the anode.
Why do you want a SCR for this purpose? There are lots of better suited
circuits possible, without SCR of course. Maybe it is more your
educational interest which has lead you to this circuit.

PS: It is very helpful if someone knows the basics of "Standard"-SPICE
when using any flavour of SPICE simulators. They all generate an
intermediate SPICE-netlist from your schematic. This netlist is the
common base of all SPICE simulators. Maybe there are as many SPICEs
around as the alphabet has characters.

Best Regards,
Helmut
 
"Abbie" <AbbieAbbieAbbie@btinternet.com> wrote in message
news:bpoq55$ler$1@titan.btinternet.com...
Hi Helmut,
Thanks for all the info that I will check very soon. I am planning to use
either Egale or CircuitMaker to generate a net list. Do you think it
will work on this free version of SPICE?
Click to expand...
Hello Abbie,
it depends on the format of this netlist. A net list for a PCB is very
different from
a SPICE circuit command list. I think there is an EAGLE script to generate
a SPICE netlist. I would never recommend to use a PCB tool for SPICE
schematics.
It's always additional work and loss of convenience. I like to probe
voltages
on the schematic in (LT)SPICE. This is impossible with a text file(net
list).
LTSPICE has its own schematic editor for the SPICE circuit.
The other way to use the SPICE editor for PCB schematic as also useless.
Better you enter your circuit twice. One times for SPICE and the other for
PCB-CAD.

Finally, I think this circuit does work, you may be right about
the resistor, which is why I wanted to do the simulation, but
in principle when the voltage is C becomes sufficently high,
the scr should conduct, dischrging C through the led and emitting
a short light blink. If, however, you do know of a similar circuit
without an scr, I would be interested.
Click to expand...
Take a look here. Maybe it's exactly what you want to have.
http://anatomix.waag.org/wiki/index.php/ElectronicsWorkshop

It is the fourth entry from a search with Google: blinking led 555 circuit
..

Best Regards,
Helmut
 
Hi Helmut,
Thanks for all the info that I will check very soon. I am planning to use
either Egale or CircuitMaker to generate a net list. Do you think it
will work on this free version of SPICE?
Finally, I think this circuit does work, you may be right about
the resistor, which is why I wanted to do the simulation, but
in principle when the voltage is C becomes sufficently high,
the scr should conduct, dischrging C through the led and emitting
a short light blink. If, however, you do know of a similar circuit
without an scr, I would be interested.
Thanks.
"Helmut Sennewald" <HelmutSennewald@t-online.de> wrote in message news:bpoiij$69a$01$1@news.t-online.com...
"Abbie" <AbbieAbbieAbbie@btinternet.com> schrieb im Newsbeitrag
news:bpoh36$9gn$1@hercules.btinternet.com...
I need to simulate a simple analog circuit which has a led,
an scr, and all other components are passive. I tried Eagle,
but couldn't find any simulation tool there. I tried Circuit
Maker and I couldn't understand how to do it. Are there
any other simulator programs which are free ?


Hello Abbie,
if you take a look to this news group, you see a lot of discussions
around LTSPICE. LTSPICE is a free and unlimited SPICE simulator from
Linear Technogies. It is also named SwitcherCADIII.
You can download it from www.linear.com/software and there is a
very helpful news group www.groups.yahoo.com/ltspice .

Back to your blinking LED circuit in your other posting. It can never
work this way with a power SCR, because it cannot trigger the SCR with
your low gate current (0.1mA). It is also wrong that you have connected
directly the gate to the anode.
Why do you want a SCR for this purpose? There are lots of better suited
circuits possible, without SCR of course. Maybe it is more your
educational interest which has lead you to this circuit.

PS: It is very helpful if someone knows the basics of "Standard"-SPICE
when using any flavour of SPICE simulators. They all generate an
intermediate SPICE-netlist from your schematic. This netlist is the
common base of all SPICE simulators. Maybe there are as many SPICEs
around as the alphabet has characters.

Best Regards,
Helmut
Click to expand...
 
Hello again,
This spice program is very nice. It doesn't have an scr
but it has transistors so I'll try to simulate a blinking led
circuit which works with two transistors. I think you are
quite right that it is better to use dedicated programs. It
seems that some people even use graphics/photo
programs to get fancy coloured diagrams, see
http://www.uoguelph.ca/~antoon/circ/2qflash.htm
"Helmut Sennewald" <HelmutSennewald@t-online.de> wrote in message news:bporm2$1q3ub2$1@ID-60775.news.uni-berlin.de...
"Abbie" <AbbieAbbieAbbie@btinternet.com> wrote in message
news:bpoq55$ler$1@titan.btinternet.com...
Hi Helmut,
Thanks for all the info that I will check very soon. I am planning to use
either Egale or CircuitMaker to generate a net list. Do you think it
will work on this free version of SPICE?

Hello Abbie,
it depends on the format of this netlist. A net list for a PCB is very
different from
a SPICE circuit command list. I think there is an EAGLE script to generate
a SPICE netlist. I would never recommend to use a PCB tool for SPICE
schematics.
It's always additional work and loss of convenience. I like to probe
voltages
on the schematic in (LT)SPICE. This is impossible with a text file(net
list).
LTSPICE has its own schematic editor for the SPICE circuit.
The other way to use the SPICE editor for PCB schematic as also useless.
Better you enter your circuit twice. One times for SPICE and the other for
PCB-CAD.

Finally, I think this circuit does work, you may be right about
the resistor, which is why I wanted to do the simulation, but
in principle when the voltage is C becomes sufficently high,
the scr should conduct, dischrging C through the led and emitting
a short light blink. If, however, you do know of a similar circuit
without an scr, I would be interested.

Take a look here. Maybe it's exactly what you want to have.
http://anatomix.waag.org/wiki/index.php/ElectronicsWorkshop

It is the fourth entry from a search with Google: blinking led 555 circuit
.

Best Regards,
Helmut
Click to expand...
 
"Abbie" <AbbieAbbieAbbie@btinternet.com> wrote:

Hi Helmut,
Thanks for all the info that I will check very soon. I am planning to use
either Egale or CircuitMaker to generate a net list. Do you think it
will work on this free version of SPICE?
Finally, I think this circuit does work, you may be right about
the resistor, which is why I wanted to do the simulation, but
in principle when the voltage is C becomes sufficently high,
the scr should conduct, dischrging C through the led and emitting
a short light blink. If, however, you do know of a similar circuit
without an scr, I would be interested.
Thanks.
Click to expand...
Abbie,

First, I agree with Helmut that your circuit should not work anyway.

However, I just tried a similar SCR oscillator circuit that I believe
*should* work, and failed!
http://www.terrypin.dial.pipex.com/Images/scr-nosim.gif

I know I had it working on a breadboard once, and in theory it
should oscillate. When switched on the cap charges via the 220 R, so
voltage across zener rises. When that conducts, gate current reaches
SCR, triggering it. Cap then discharges through SCR and speaker and
voltage falls to zero. So SCR stops conducting. Cycle *should* then
repeat. But best I've seen so far was one darned pulse.

Maybe someone can offer ideas why it fails? Or maybe Helmut can get it
working with LTSpice?

--
Terry Pinnell
Hobbyist, West Sussex, UK
 
On Sat, 22 Nov 2003 23:06:13 +0000 (UTC), "Abbie"
<AbbieAbbieAbbie@btinternet.com> wrote:

Hi Helmut,
Thanks for all the info that I will check very soon. I am planning to use
either Egale or CircuitMaker to generate a net list. Do you think it
will work on this free version of SPICE?
Click to expand...
LTSpice is free, but that doesn't in any way make it less capable than
equivalent simulators that you can pay *lots* of money for. A lot of
us here swear by it and are very grateful to Linear Technology for
making this wonderful tool freely available. Helmut's right. If you
haven't already downloaded a copy of the program, then do so now. You
never know when they might start charging for it!
--

"I expect history will be kind to me, since I intend to write it."
- Winston Churchill
 
Hello Terry,
First of all, regarding my circuit, I have now realised that I forgot
to draw a zener diode between the gate and the cap, I think this
solves the problem, right ?
Now, regarding your circuit, You are shunting part of the current
to ground via the bottom 1K, are you aware of that. Maybe better
to remove this resistor all together. The only other think I can think
of is your R=220. If this value is too low, your scr will never turn
off.


"Terry Pinnell" <terrypinDELETE@dial.pipexTHIS.com> wrote in message news:9791svkfuhqqn5f3n47atlge4lcbvo5k98@4ax.com...
"Abbie" <AbbieAbbieAbbie@btinternet.com> wrote:

Hi Helmut,
Thanks for all the info that I will check very soon. I am planning to use
either Egale or CircuitMaker to generate a net list. Do you think it
will work on this free version of SPICE?
Finally, I think this circuit does work, you may be right about
the resistor, which is why I wanted to do the simulation, but
in principle when the voltage is C becomes sufficently high,
the scr should conduct, dischrging C through the led and emitting
a short light blink. If, however, you do know of a similar circuit
without an scr, I would be interested.
Thanks.

Abbie,

First, I agree with Helmut that your circuit should not work anyway.

However, I just tried a similar SCR oscillator circuit that I believe
*should* work, and failed!
http://www.terrypin.dial.pipex.com/Images/scr-nosim.gif

I know I had it working on a breadboard once, and in theory it
should oscillate. When switched on the cap charges via the 220 R, so
voltage across zener rises. When that conducts, gate current reaches
SCR, triggering it. Cap then discharges through SCR and speaker and
voltage falls to zero. So SCR stops conducting. Cycle *should* then
repeat. But best I've seen so far was one darned pulse.

Maybe someone can offer ideas why it fails? Or maybe Helmut can get it
working with LTSpice?

--
Terry Pinnell
Hobbyist, West Sussex, UK
Click to expand...
 
"Abbie" <AbbieAbbieAbbie@btinternet.com> wrote:

Hello Terry,
First of all, regarding my circuit, I have now realised that I forgot
to draw a zener diode between the gate and the cap, I think this
solves the problem, right ?
Click to expand...
Yes, that would make a big difference!

Now, regarding your circuit, You are shunting part of the current
to ground via the bottom 1K, are you aware of that.
Click to expand...
Yes.

Maybe better to remove this resistor all together. The only other think I can think
of is your R=220. If this value is too low, your scr will never turn
off.
Click to expand...
No - wide range of choices used with no success. Either SCR never gets
triggered, or stays on permanently.

--
Terry Pinnell
Hobbyist, West Sussex, UK
 
"Terry Pinnell" <terrypinDELETE@dial.pipexTHIS.com> schrieb im Newsbeitrag
news:cgo1svclbmqjt1c678pjaih8a0h14h2l8e@4ax.com...
"Abbie" <AbbieAbbieAbbie@btinternet.com> wrote:

Hello Terry,
First of all, regarding my circuit, I have now realised that I forgot
to draw a zener diode between the gate and the cap, I think this
solves the problem, right ?

Yes, that would make a big difference!

Now, regarding your circuit, You are shunting part of the current
to ground via the bottom 1K, are you aware of that.

Yes.

Maybe better to remove this resistor all together. The only other think I
can think
of is your R=220. If this value is too low, your scr will never turn
off.

No - wide range of choices used with no success. Either SCR never gets
triggered, or stays on permanently.
Click to expand...
Hello Terry,
the chance that this SCR oscillator will be successful is more like a game.

1. The gate trigger current must be low enough.
2. The minimum holding current shouldn't be too low.
3. It should be fast enough for audio frequencies.

Because we can't meet condition 2 normally, we need a resonat
load impedance. Such a load will lead to a zero crossing current
through the SCR and thus switching it off. Your loudspeaker is
an inductive load and together with the capacitor it is a resonant
circuit. Only the right combination of all circuits elements,
including the SCR, make it an oscillator.

I will send you my LTSPICE schematic/circuit file.

Best Regards,
Helmut
 
Did you try to spice it?

"Terry Pinnell" <terrypinDELETE@dial.pipexTHIS.com> wrote in message news:cgo1svclbmqjt1c678pjaih8a0h14h2l8e@4ax.com...
"Abbie" <AbbieAbbieAbbie@btinternet.com> wrote:

Hello Terry,
First of all, regarding my circuit, I have now realised that I forgot
to draw a zener diode between the gate and the cap, I think this
solves the problem, right ?

Yes, that would make a big difference!

Now, regarding your circuit, You are shunting part of the current
to ground via the bottom 1K, are you aware of that.

Yes.

Maybe better to remove this resistor all together. The only other think I can think
of is your R=220. If this value is too low, your scr will never turn
off.

No - wide range of choices used with no success. Either SCR never gets
triggered, or stays on permanently.

--
Terry Pinnell
Hobbyist, West Sussex, UK
Click to expand...
 
"Helmut Sennewald" <HelmutSennewald@t-online.de> wrote:

Hello Terry,
the chance that this SCR oscillator will be successful is more like a game.

1. The gate trigger current must be low enough.
2. The minimum holding current shouldn't be too low.
3. It should be fast enough for audio frequencies.

Because we can't meet condition 2 normally, we need a resonat
load impedance. Such a load will lead to a zero crossing current
through the SCR and thus switching it off. Your loudspeaker is
an inductive load and together with the capacitor it is a resonant
circuit. Only the right combination of all circuits elements,
including the SCR, make it an oscillator.

I will send you my LTSPICE schematic/circuit file.
Click to expand...
Helmut: Thanks for sending the file. Although I have LTSPICE, what I'd
prefer to do is get it converging in CM. If it's not too much trouble,
could you post (or email) me a picture of the schematic please?

--
Terry, West Sussex, UK
 
To close this off, and especially for Rex (Abbie?), Helmut e-mailed
me:
"It's your original circuit. I have changed only a few component
values. There are two circuits in the schematic, because each SCR
needs different component values.
Be aware that my model for the loudspeaker is only a guess.
You could also start with a 1mH+5Ohm coil as your load.
It was also very important to increase the capacitor to 2.2uF."

I also had other input by email from John Durkee. Many thanks to both.
Happily I was then able to get circuit simulated. Here's my summary.

http://www.terrypin.dial.pipex.com/Images/scr-osc-sim.gif

As you see, component values were very critical to get it working at
all. Not a practical circuit for oscillation, whether audio or LED
flashing.

--
Terry Pinnell
Hobbyist, West Sussex, UK
 
In news:fom6svkfkmnllgn7vmogupj9sseud714k1@4ax.com (Terry Pinnell):
To close this off, and especially for Rex (Abbie?), Helmut e-mailed
me:
"It's your original circuit. I have changed only a few component
values. There are two circuits in the schematic, because each SCR
needs different component values.
Be aware that my model for the loudspeaker is only a guess.
You could also start with a 1mH+5Ohm coil as your load.
It was also very important to increase the capacitor to 2.2uF."

I also had other input by email from John Durkee. Many thanks to both.
Happily I was then able to get circuit simulated. Here's my summary.

http://www.terrypin.dial.pipex.com/Images/scr-osc-sim.gif

As you see, component values were very critical to get it working at
all. Not a practical circuit for oscillation, whether audio or LED
flashing.
Click to expand...

What the heck is GMIN stepping really? I get the same kind of results if
simulating any circuit that involves a diode or inductor. (I use Proteus
VSM, based on the Berkeley 3F5 core.) You'd think increasing or decreasing
the GMIN value in the simulation preferences would have something to do with
it, but it doesn't appear to. I've tried tweaking all the simulator options
with no effect either. Is it just a case of N-P junction and inductor
modelling error, or is this problem intrinsic with simulators?
 
"Mark Jones" <127.0.0.1> wrote:

What the heck is GMIN stepping really? I get the same kind of results if
simulating any circuit that involves a diode or inductor. (I use Proteus
VSM, based on the Berkeley 3F5 core.) You'd think increasing or decreasing
the GMIN value in the simulation preferences would have something to do with
it, but it doesn't appear to. I've tried tweaking all the simulator options
with no effect either. Is it just a case of N-P junction and inductor
modelling error, or is this problem intrinsic with simulators?
Click to expand...
I'm hoping the Spice gurus will have some answers. That's the most
common error message I get in CM. Like you, I ring the changes on
several options, rarely with any joy. But at least it often still lets
simulation proceed.

--
Terry Pinnell
Hobbyist, West Sussex, UK
 
Mark,

What the heck is GMIN stepping really? I get the same kind of results if
simulating any circuit that involves a diode or inductor. (I use Proteus
VSM, based on the Berkeley 3F5 core.) You'd think increasing or decreasing
the GMIN value in the simulation preferences would have something to do with
it, but it doesn't appear to. I've tried tweaking all the simulator options
with no effect either. Is it just a case of N-P junction and inductor
modelling error, or is this problem intrinsic with simulators?
Click to expand...
GMIN stepping is method of finding the DC .op point. It
consists of placing a short(of a few milliOhms or so) to
ground from every node in the circuit. Presumably that
circuit can be solved. Then it reduces the value of the
shorts' impedances, and re-solves the new circuit using
the previous solution as an initial guess. The final
step is to remove these shorts entirely and then solve
the real circuit using the previous perturbed solution
as an initial guess.

"Gmin stepping" is something of a misnomer in that it has
nothing to do the the global gmin parameter that is placed
as a conductivity across all pn junctions.

--Mike
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top