On Mon, Apr 19, 2010 at 1:09 PM, Olin Lathrop w= rote: > Electron wrote: >> PS: sorry, I forgot: one feature I'd appreciate much is a huge >> database of real components. MultiSim 11 has a good one, but a lot of >> real world components are still missing. > > Maybe you should step back and ask why you need all this. =A0In my experi= ence, > if you can't solve your circuit with a few minutes and a calculator, you > don't know what you're doing. =A0Personally I haven't used a circuit simu= lator > yet. =A0That's because I haven't felt the need for one. =A0I agree that > simulation can be a useful tool to see some details or get verification o= nce > you already know what your circuit is doing. =A0I have occasionally seen > others use it for that, and that's fine. =A0To some extent that depends on > what you're comfortable with. > > However, simulation is not a substitute for understanding a circuit, and = is > not for synthesizing a design. =A0It also has the problem of appearing to= give > a precise answer when things are +- 10% in the real circuit, making it too > easy to design a one-off that works only with components exactly like the > simulator thinks they are. > > While I agree that it is very dangerous to use simulation as a core part of circuit design, sometimes it is unavoidable. For example, when designing circuits involving feedback loops, I often use simulation as a sort of calculator/Bode plot generator. I break open the loop and feed signals in and see the relative gain and phase at the output versus frequency. I can then check for gain and phase margin. For things like simple op-amp circuits, it is better to do this using back of the envelope calcs, but if your system is more complex, the calculations get very tedious on paper and yield really no more understanding than a simulation-computed Bode plot. That said, if one were to just close the loop, simulate it, and then declare "yup, it works in sim so it must be stable in real life" that would be a mistake. Another handy use of simulation is to provide a check step between design and prototyping. Instead of drawing up a schematic and doing hand calcs and then going directly to breadboarding, I often catch additional mistakes by doing a sim run inbetween. Sean > ******************************************************************** > Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products > (978) 742-9014. =A0Gold level PIC consultants since 2000. > -- > http://www.piclist.com PIC/SX FAQ & list archive > View/change your membership options at > http://mailman.mit.edu/mailman/listinfo/piclist > -- = http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist