VT220

SP wrote: Login will usually respond to any kind of garbage sent through the serial port. Ok if everything is already set up, but useless for real debuging. Make sure you are NOT using a DEC 9-pin to 25-pin cable when plugging into the PC. DEC's 9-pin layout is different than IBM's. Do you need a null modem? I've had PCs with opposite "polaritry" before. Those cheap RS232 testers available from RadioShack (and others) are useful here (i.e. a block with two DB25's and a lot of leds); if both 2-3 light up, cabling is probably ok and you don't need to add a null modem; if only one does then you probably need one; if none light, you have another problem. Test that the terminal works. Tie pins 2 and 3 together (on the VT220's 25-pin connector) to make a loopback connector (paper clips often come in handy here), and try typing on its keyboard. It should echo. If you don't get echo, then the line drivers are blown. Also double check baud rates and bit-sizes (Try 8bit-no parity-one stop bit). Another thing to try is to write a program that sends a continuous stream of characters to the serial port. Make sure it doesn't hang while outputing (indicates it's a flow control problem). Then you can play with the VT220 settings without needing login working properly. Or you can use kermit (or another terminal program) from the PC. Type on the VT220 keyboard, and you should see what you typed on the PC. Also vice versa. you can see if characters are flowing on one direction or the other properly that way. If you get garbled characters, you have [baud-rate, char width, stop bit, cabling, etc] problems. Try slower speeds (300 baud). Could be a time-base problem in one or the other devices. Slower speeds may not be as badly affected.