Hi,
Can anyone explain the use of a simulator in audio ?
I currently use it for circuit design.
http://www.national.com/models/spice/CL/oa-18.pdf
edit:-I have corrected the thread
thanks.
Can anyone explain the use of a simulator in audio ?
I currently use it for circuit design.
http://www.national.com/models/spice/CL/oa-18.pdf
edit:-I have corrected the thread
thanks.Attachments
Using a simulater is a popular means of analyzing audio circuits without actually building them. A "model" is complex group of equations and coefficients that a simulater uses to represent a particular component.
These models can be fairly simple for components like resistors and capacitors. For semiconductors and integrated circuits (and tubes), the models can become quite complex.
These models can be fairly simple for components like resistors and capacitors. For semiconductors and integrated circuits (and tubes), the models can become quite complex.
Note that op amp models, and IC models in general, are often not very good, although some are surprisingly good.
The best thing with spice is that you can experiment with circuits without using more space than you computer takes, and you probably already have that one for other reasons. It is also much quicker to modify circuits and change measurement setups in most cases.
Many people seem to think that simulation is only useful if it can accurately predict how a real circuit will behave, and since it often cannot do that well enough, or they don't model the circuit well enough, they think it is useless. However, they are then missing many possibilities, although an experienced EE who is used to think and design in the traditional way may not have much use of or need for using it.
You can often ger a fair idea of how a certain circuit will work. Some aspects of it will be closer to reality than others.
It is a good way to learn how circuits work, as a complement to understanding theoretically how they work, not as a substitute. Sometimes when I try to derive the equations for a circuit I don't trust the results I get, or I simply get stuck in the maths. A simulation can then often give a clue if my equations are reasonable or not, or it may be that when I see how the circuit actually behaves in the sim, I realize where I went wrong in my thinking.
You can also do things you cannot do with a real circuit, since you can use the whole spectrum from ideal components to complex almost perfect models. I have often studied the behaviour of certain circuits using ideal, or near ideal, BJTs. Then I see what happens when I add or change a certain parameter. I might try different values for Is, or beta, or Cob etc. You cannot do that with a real circuit, since it is almost impossible to find to transistors that differ in only parameter, or that are ideal except for one or two parameters. So what good is that? Well, it can help to pinpoint which parameters of a transistor are most important for use at a certain position in a circuit.
Many of also don't have access to fancy equipment, such as spectrum analyzers or high-BW scopes, so simulation may be the only way to get some clue at all to what is happening in some cases. I have even found bugs in a circuit i designed which I most probably wouldn't have catched by measurements, since I don't have the necessary equipment and I also probably wouldn't have thought of measuring the right thing, which I got kind of like a bonus in the simulation.
Also don't get stuck in thinking that only absolute figures are useful. If you compare two very similar circuits in simulation you may get a fair idea how they compare relative to each other, although the absolute figures may be off by quite a lot. I often use spectrum analysis of distorsion that way, knowing that the absolute values may not be very reliable, but the relative performance might be reasonably correct, if taken with a pinch of salt.
And there is a lot more one can do. The sky is the limit. Just remember it is a simulation, and you must never assume that a real circuit will behave exactly as the one you simulate. You have to verify your circuit by measurements once you have built it.
The best thing with spice is that you can experiment with circuits without using more space than you computer takes, and you probably already have that one for other reasons. It is also much quicker to modify circuits and change measurement setups in most cases.
Many people seem to think that simulation is only useful if it can accurately predict how a real circuit will behave, and since it often cannot do that well enough, or they don't model the circuit well enough, they think it is useless. However, they are then missing many possibilities, although an experienced EE who is used to think and design in the traditional way may not have much use of or need for using it.
You can often ger a fair idea of how a certain circuit will work. Some aspects of it will be closer to reality than others.
It is a good way to learn how circuits work, as a complement to understanding theoretically how they work, not as a substitute. Sometimes when I try to derive the equations for a circuit I don't trust the results I get, or I simply get stuck in the maths. A simulation can then often give a clue if my equations are reasonable or not, or it may be that when I see how the circuit actually behaves in the sim, I realize where I went wrong in my thinking.
You can also do things you cannot do with a real circuit, since you can use the whole spectrum from ideal components to complex almost perfect models. I have often studied the behaviour of certain circuits using ideal, or near ideal, BJTs. Then I see what happens when I add or change a certain parameter. I might try different values for Is, or beta, or Cob etc. You cannot do that with a real circuit, since it is almost impossible to find to transistors that differ in only parameter, or that are ideal except for one or two parameters. So what good is that? Well, it can help to pinpoint which parameters of a transistor are most important for use at a certain position in a circuit.
Many of also don't have access to fancy equipment, such as spectrum analyzers or high-BW scopes, so simulation may be the only way to get some clue at all to what is happening in some cases. I have even found bugs in a circuit i designed which I most probably wouldn't have catched by measurements, since I don't have the necessary equipment and I also probably wouldn't have thought of measuring the right thing, which I got kind of like a bonus in the simulation.
Also don't get stuck in thinking that only absolute figures are useful. If you compare two very similar circuits in simulation you may get a fair idea how they compare relative to each other, although the absolute figures may be off by quite a lot. I often use spectrum analysis of distorsion that way, knowing that the absolute values may not be very reliable, but the relative performance might be reasonably correct, if taken with a pinch of salt.
And there is a lot more one can do. The sky is the limit. Just remember it is a simulation, and you must never assume that a real circuit will behave exactly as the one you simulate. You have to verify your circuit by measurements once you have built it.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.