Erik,

About the spelling ... You need to read your messages before you send 'em. We all mistype things from time to time, and sometimes we hit "send" too soon. However, if you spell the same word two different ways within the same message, (look for the word schematic in an earlier post in this thread) it's confusing to those for whom English isn't their primary language. No post is perfect, but it shows how much you care about the quality of the stuff you send.

Aside from that, I've corrected your spelling of "babbeling" (sic) a time or two already. English was my second language, too, so that's no excuse. You have a spell-checker, don't you? Why not take it out for a test-drive?

In the years I've watched your posts, I've concluded that you enjoy "pimping" for KEIL, SILABS, and PHILIPS. While we all have our favorites, I find it offensive that you almost always recommend them over far-less-costly alternatives without considering the circumstances of the person whom you're addressing. It seems that, much of the time, people are asking for help with something they've already got. Telling them to substitute something else for it often doesn't help.

Now, about the "noise margin" to which you refer. if you go back to where you started attacking someone else's approach, and a reasonably sound one, to receiving signal for the purpose if in-situ-programming of the subject part, because he didn't use a bipolar supply, you'll see where you went wrong. A bipolar supply is needed in order to TRANSMIT RS232 signals. It is not needed in order to receive, because all one has to do is sense whether or not the signal is above 3 volts. Whatever noise margin there is is created by the transmitter. The receiver has a significant noise tolerance, as you pointed out, albeit misinterpreted as a datasheet error, during the discussion of the MC1489 specification. The 6-volt deadband is created by the transmitter. It is not a noise margin.

A receiver doesn't have a 6-volt "noise margin" though the transmitter is supposed to generatea 6-volt deadband. There are good reasons for this. How would you expect the receiver to behave if it receives a signal that momentarily goes below -3 volts, then rings a volt or so, p-p, about that level for half a bit time? A 50-volt noise margin would be better than a 6-volt noise margin, Erik, but RS232 provides neither.

The reason I jumped you is because, in your usual style, you attacked the approach taken by someone else just because you didn't like it and didn't understand it!.

You may be a lot smarter than most of us, and you may know a lot more than most of us, but you don't know everything. Rather than insisting on your approach as the "only" approach, why not attempt to see what the other person sees. It's clear that you often respond to something with which you think you disagree without even reading the entire post. I think that's what you did here. The original poster didn't make the comment that set you off, but Tony said, "A simple communications setup is to use 2 npn transistors as switches and invert the signal from the micro to the PC and the signal from the PC to the micro. For short cable lengths, (several metres), this will work very reliably at 115 KBaud. I have found no need to use a MAX232 for this application." That seems to have been what set you off, though I don't see why. It seems perfectly correct to me.

I suggest you read and consider each post before attacking it.

I suggest you look up the definition of "noise margin" and "deadband" as they relate to RS232. It would be both interesting and informative to see a supporting reference, ...ANY... supporting reference, to the "science" to which you refer.

I also suggest you use your spell-checker.

RE