Erik,

I made many mistakes early on and fixed them without having you to tell me where I'd gone wrong. I could tell when things were not working right because I had a decent picture of how things should behave if working properly, and how the signals should look on the 'scope or logic analyzer. Of course, I also had the means to examine the behavior of the signals I created.

Anyone who attempts to do this sort of thing "blind," i.e. without the instruments that enable him/her to observe the quality of the signals being generated is being foolish. It's been done, often by blindly following an app-note, which are notoriously flawed, but it's been done. Now, I've got two-channel and 4-channel oscilloscopes, and logic analyzers with many more channels. I've concluded that, while you can get by without a logic analyzer most of the time, I don't believe one can do this sort of thing without an oscilloscope of sufficient bandwidth.

Ultimately, if it works as desired, then it's not a mistake. If it doesn't, it probably is, so it's up to the designer to figure out where he's gone wrong. That's not to say it couldn't be done better, cheaper, cleaner, but those are arbitrary terms.

If he asks for help, though, he's asking help in doing what he wants to do, not in doing what YOU want to do. That's not to say you can't come up with a way that's better, cheaper, cleaner, but those are arbitrary terms.

RE