![]() |
David Woolley wrote: > > > of things. - When trying to call the fax number using a regular > > telephone one can hear a lot of noise before the normal fax interaction > > appears to start. In other words, the modem starts sounding like a fax > > only after tried to send some other kind of data for a while. > > This is not a valid test. You need to generate fax calling tone from > the calling end for a fax modem to autodetect fax mode; in fact, I'm I think you missed my point. I don't want the modem to autodetect fax mode, I want the modem to stay in fax mode at all times. > pretty sure that that is the only difference between a fax call and a > 300 bps data call. If you don't send calling tone followed by V.21 > originate carrier, the modem will probably go into a V.34 initialisation > sequence, and when that fails, eventually fall back to V.21 data mode. > > Roughly the protocol is something like: > > caller sends fax calling tone > callee send 2100Hz echo suppressor/answer tone, followed by V.21 (300bps) > carrier > conversation at 300 bps > caller sends high speed training sequence (e.g. V.29, 9600 bps half duplex) > somehow callee confirms successful train (may be by secondary channel) > caller sends fax data using V.29 etc. > caller and callee switch back to V.21 for end of page > cycle repeats > > The first V.21 session is the infamous Phase B and the second is Phase D. > The V.29 portion is Phase C. > > As a result, you will get rather a lot of different tone types in a valid > fax call. -- - T. Lund