These things are rather confusing, originating in early teletype days - if you expect to receive exactly what you type, you first need to tell the device on the other end to send it. If you see what you type on your hyperterminal screen, it doesn't necessarily mean the device is sending it. Most teletype devices can be configured to display, or not, what you type, no matter if there's anything reading the line on the other end, or if it is configured correctly. This is local echo. Neat thing unless you're sending files etc. There's another option, remote echo - anything you send to the device, is being sent back immediately. But if the remote device is not receiving your commands, it won't send them back so you will see them on your screen with remote echo on, only if the connection is okay. (moreover, if the connection exists but is bad, e.g. long serial line, too high speed or wrong serial timing, the received data will be garbled). What happens when you turn both on? TThhiiss hhaappeennss. If you turn both off, you see nothing. If in response to AT<cr><lf> you expect AT<cr><lf>OK<cr><lf>, then enable remote echo by sending ATE 1<cr><lf> before anything else - probably Hyperterminal does this before displaying anything, or displays local echo. Otherwise, just expect OK<cr><lf> after the last character of AT<cr><lf>, which is the normal course of things if there's no screen attached to the devices communicating.