help about a SPICE model

Hi, I'm an computer science student and I have a problem. I have a picture (and scientific articles) of a 2nd order biquad lowpass filter and I would want (is necessary) write a spice model of this circuit but I'm not able. Is there someone that courteously could convert this circuit design in SPICE code?

thanks

Andrea

I must use SPICE 3f5 Berkeley version on linux system. this is the link of my schematic design

thanx

Andrea

Jim Thompson ha scritto:

[snip]

Your schematic design has no values for the components. What frequency response is it to have?

...Jim Thompson

-- | James E.Thompson, P.E. | mens | | Analog Innovations, Inc. | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | Phoenix, Arizona Voice:(480)460-2350 | | | E-mail Address at Website Fax:(480)460-2142 | Brass Rat | |

| 1962 | I love to cook with wine. Sometimes I even put it in the food.

I believe LTSpice runs under WINE. You can get it here along with SPICE models:

Regards,

Mike Monett

Most of us like to use Spice programs that have a graphical front end and back end. This means we can enter the schematic, check the viability of the models, tell it to run the included Spice 3F5 or whatever, engine, and display the results in graphical plots. Some of these are available in free versions that are sufficiently powerful to run your simple circuit. For example, the Intusoft Spice CD or download that's available free at On Semiconductor.

But it's not hard to simply use the free spice engine by itself and enter your spice program by hand. You must already have the syntax manual for standard spice. If not, it's easy to find.

Number the nodes of your circuit and type in the drawing, line by line for each component's connections. The only thing you'll be missing is an opamp model. A number of possibilities are open to you. Such as, pick some common opamp, and get its model from the manufacturer's website. It'll probably be a variant of the Boyle model. Study the model to see how it works, and modify it for the specs of your problem.

For simple problems I sometimes like to make a basic opamp from a VCCS and a VCVS with a capacitor to ground between them to create the opamp's gain rolloff, plus a series output resistor... it's worth playing with anyway. Go, and simulate your day away. And of course you can come back to us with more questions.

--
 Thanks,
    - Win

Too much classroom, and not enough thinking process ;-)

You can, indeed, parameterize component values. But you must then declare them, and possibly sweep them from the simulator.

The schematic you have shown IS a LINEAR low pass filter, UNLESS over driven.

Are you a PhD candidate ?:-)

...Jim Thompson

--
|  James E.Thompson, P.E.                           |    mens     |
|  Analog Innovations, Inc.                         |     et      |
|  Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
|  Phoenix, Arizona            Voice:(480)460-2350  |             |
|  E-mail Address at Website     Fax:(480)460-2142  |  Brass Rat  |
|       http://www.analog-innovations.com           |    1962     |
             
I love to cook with wine.      Sometimes I even put it in the food.

Jim Thompson ha scritto:

"..The opamp model has an open loop gain of 10000.the output voltage restriction is -1.5V +1.5V and the maximum output current is 80mA. Due to these restrictions the whole system is nonlinear. This circuit has two state variables, namely the capacitor voltages Vc1 and Vc2. Using acharge oriented capacitor model willnormally lead to the two charges as state variables. We changed this, because we found it more convenientto think in voltages than in charges.The corner frequency Wc and the damping factor d ar given by Wc = (Sqrt(R1*R2*C1*C2))^-1 and d =

0.5C1Wc(R1+R2). We use two different value sets for the resistors and capacitors, one with Wc =100s^-1 and d = 0.5 and the second one at the same frequency but with d = 2.Tus, we get twoequivalent lowpass filters differing only by the damping factor. The property to be checked in this example is the occurrence of overschooting in the two filters. Since these properties should be proved for arbitrary input signals, the appropriate input value model si chosen. The input signal range is Vin =[-2..2] so that the nonliearity due to the restricted output voltage will effect the system beahviour. The state space is restricted to Vc1=[-4..4] and Vc2[-2.5 .. 2.5].Vin start from 0." Obviously I will change manually the variables values that I have writen above Thanx Andrea

Thank you, but I'm obliged to use only SPICE 3f5 (3f2 or higher) on linux system. :( A.

Wine runs under Linux:

Mike Monett

LTSpice is SPICE.

It runs under Wine.

Wine runs under Linux.

Therefore, LTSpice runs in Linux.

LTSpice is a very good SPICE program. You can run it in Linux. See above.

Mike Monett

I know, but I must use Spice...

Yes, but I can't use wine because I must open source program and I must throw SPICE by fork system call

jeko ha scritto:

I have downloaded LTSpice and I'm using it under windows. It's cool...but I can't use source code:(

Then you are WAY over your head trying to analyze a circuit.

...Jim Thompson

--
|  James E.Thompson, P.E.                           |    mens     |
|  Analog Innovations, Inc.                         |     et      |
|  Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
|  Phoenix, Arizona            Voice:(480)460-2350  |             |
|  E-mail Address at Website     Fax:(480)460-2142  |  Brass Rat  |
|       http://www.analog-innovations.com           |    1962     |
             
I love to cook with wine.      Sometimes I even put it in the food.

It's not a SPICE simulation that's needed- the problem statement is state-space analysis and system trajectory in a bounded region. The whole circuit thing is a red-herring where outputs are just artificially clamped at boundaries such as 80ma maximum output current and OA output restricted to +/-1.5V with simple circuit equations defining variable trajectories in the linear unbounded region, ideal opamp behavior assumed. Andrea needs to check out the student edition of MatLab.

Jim Thompson ha scritto:

No, I'm making my Computer Science degree thesis (specialist).I'm italian graduand :)