I think these simulation programs were written by computer scientists who never built a circuit.
Spice eats text listings (ascii). What I know, Windows user interface is just a tool to make it easier. Silly errors, like pins do not connect and so on, are easy to make.
Nothing to apologise about. If you donât like it, donât use it.
Life is really that simple.
Iâm an engineer and I know/calculate what to expect before I do a sim.
Not true at all. The snag with simulation is how detailed to make the device models. Too simple and strange things can happen. Too complex and the simulation performance chokes.
Then you get into the model accuracy. Many vendor supplied models are very flawed.
An example of this is audio power transistors, where the standard model doesnât model the beta drop at high currents and the Ft drop at very low currents. So Bob Cordell published his own models that match reality a lot better.
I was not particularly recommending the ADCMP609. It is a good comparator but it is not cheap. I was recommending an LM339. I showed the ADCMP609 only because that is a drawing which I had at hand.
The true advantage of simulation is that results are not true! They are perfectly repeatable. Not like in real life with noise and jitter.
Another huge advantage is that it takes just a few seconds to modify a circuit. Change a resistor or capacitor, or the the configuration of a whole filter, and then run the simulation again. You can also quickly isolate a particular section of your schematic and simulate only that.
A disadvantage of simulation is that it can not simulate the PCB itself. Things like track resistance and capacitive coupling.
And real life circuit probing is also not trivial, especially for high impedance or high frequency circuits. If I measure the frequency on the crystal oscillator of my uC with a scope, then I get a different measurement for each of the pins. Large parts of radio receivers can also simply not be measured at all in real life because the measurement would disturb the circuit.
In the end, a spice simulation is just a tool and it has itâs own advantages and limitations and it takes time and effort to learn those.
1 Like
LTSpice has a Forum LTspice@groups.io | Messages and over 140 000 messages there.
A user should have some idea of what a circuit does. Often it is not enough to just drop a circuit in the SIM, and wish good results.
And of course, forums are better at solving simulation issues than PC board issues.
I had a friend who also taught some undergraduate circuit design classes at a local university. He would assign a problem early in the semester where he instructed the class to analyze a simple circuit, and then simulate it in SPICE. He had a collection of circuits that would converge to the wrong answer in SPICE from which he chose the circuit.
Most student turned in the homework with the wrong answer, because they did the simulation and then tried to make their analysis match the wrong answer. Only a few would say they could not get them to match. It was his lesson on the pitfalls of simulation. They still used SPICE in the rest of the class, but it was a good lesson to learn.
John
2 Likes
If simulating were always easy, they wouldnât need 150 000 posts in the LTSpice forum.
1 Like