Mathematical Model for a project

I'm not sure where you're hearing that. A more general statement is that you should know what you're doing before starting an electronics project, and unless you know exactly how they're going to behave any analog circuits in your project should be modeled mathematically. You shouldn't automatically make a mathematical model of a circuit -- making a complete and detailed mathematical model of a microprocessor circuit would be a waste of time unless you were the chip manufacturer trying to make sure that a new chip will function correctly. On the other hand, if it's a very fast microprocessor it would be a good idea to do some form of preliminary validation that you are going to achieve the speeds you want on your various buses, and know what you need to do to get there.

A mathematical model of a circuit (or anything else) is a description, using math, of that circuit which is sufficient for predicting its behavior. Sometimes you'll be interested in different aspects of a circuit's behavior and you'll develop a model for each -- for example a power amplifier may have a behavioral model that tells you what it does assuming the finals don't get too hot, and a thermal model that tells you how hot the finals do get.

So you develop a mathematical model by asking questions about your circuit that you need answers to, then figuring out how to model the circuit to answer those questions.

The model itself doesn't end up being incorporated into the actual product, but any good engineering project should be accompanied with good documentation; your models should be implemented in some manner where they can stand or their own, or they should be incorporated into a circuit description that anyone who works on your circuit in the future will be able to read so they understand what the heck you were trying to do.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

well Mr. Wescott have give a link to its web page after checking that link i could tell u that i have studied these stuff at the 3 rd yr of my engineering degree ( i just want to say that these things are complicated if ur not used to them...) u could start by something like the RLC circuit modeling whish could give u a much more easier differential equation.... Best Regards, Eng. Roy.

No, that is one of the steps in doing an electronics project not something you do before it. The steps in doing a project are:

(1) Determine if this is a really stupid idea that can be rejected before we even start. This step involves checking that the spec. doesn't suggest that you know the future or produce more energy than you start with.

(2) Guess at a few ways that the problem can be solved. If its audio guess that it will be good op-amps if it is image processing guess that it will involve a CPU of some sort. .. etc .. Either on paper or in your head make a list of methods.

(3) Now you make a very simple model in a form that works well for thinking about how you are going to solve this problem. This model may well show that you picked the wrong way to attack the problem. If it does, change your mind[1]. The sooner you get rid of the dumb ideas, the less time you'll spend going down the "garden path".

[1] When I change my mind, I always ask "will this new one work better".

--
--
kensmith@rahul.net   forging knowledge

In article , Tim Wescott wrote: [...]

Stop one step short of "anyone who". There are some real morons out there that won't get it if you spoon feed them. Assume the other guy isn't as smart as you but that he does know how to tie his shoes.

A couple do and don'ts to demonstrate the point:

If you are writing software make your comments state why you do something:

Don't: ADD A,#209 ; Add 209

Do: ADD A,#209 ; (0-backslash) for compare

If you are writing about op-amp circuits:

Don't: U1 is an op-amp that makes a gain of 7 and inverts the signal

Do: U1's offset current forced me to use a low value resistor for R17

--
--
kensmith@rahul.net   forging knowledge

1) You use differential equations and algebra, and knowledge of the behavior of the components of the system. Your project's requirements and specifications should be able to be used to mathematically describe how the circuit or system should behave, and how it needs to interface with the rest of the world. And with a BSEE degree or equivalent knowledge you should be able to create rudimentary mathematical models of most electronic circuits, and the parts of the rest of the world to which your circuits will be interfaced. You should be able to end up with a set of equations, either in the time domain or the frequency domain, that behave almost (within the limits of your modeling techniques and knowledge) as your system behaves. i.e. If you solve the set of time-domain differential equations that represent your circuit, then, the algebraic equations that result, when plotted, should show all of the various voltages or currents, at every point in your circuit, as functions of time. Similarly, if you use an equivalent set of equations, but in the frequency domain, the solution should show the voltages or currents as functions of frequency.

for that matter

2) A mathematical model is one tool that can be used to help analyze or synthesize a system. Other helpful tools might include imagination, reasoning, logic, cleverness, research, and experience. Mathematical models can be used in several ways. Two typical ways might be: a) put pencil to paper, or, b) simulate on a computer. In both cases, you should be able to learn how the properties of the components of a system affect its internal and external behaviors. And you should also be able to solve the equations, to arrive at the needed properties (values) of components, in order to produce the desired responses of the system.

All of the above can range from (at least) almost trivial to almost impossible, depending on your project's requirements and complexity.

After you know how to theoretically do most of the modeling and solving with pencil and paper, you can probably then safely use computer simulation software, such as the freely-downloadable LT-SPICE, from

. If you need more component models, one of the many places to look, for help with finding them, is: . For help with using the LT-SPICE software and the models, go to the LT-SPICE discussion group, at .

Good luck!

Tom Gootee

Entire projects/products are seldom modeled. Sometimes one will model parts of a project, like one specific circuit, usually using Spice but occasionally other tools, including specifically coded programs.

The first thing I do for a new project is write the manual.

John