Re: Trouble receiving faxes from certain sources

Thank you for the detailed explanation. I will need to do some 
experimenting to try to pinpoint this. I probably should have mentioned 
the following points in my initial post:

1) It seems I used to be able to receive from this sender when I was 
still using HylaFAX, before I switched to HylaFAX+. But:
2) In addition to switching to HylaFAX+, I also switched from an x86_64 
to an armhf based device.

I'm suspecting a line quality issue as part of the problem, but it's odd 
that it's only with certain senders, and outgoing faxes seem fine. I 
will continue to tweak and monitor.

Thanks,

-Jason

On Wed, May 08, 2019 at 12:24:26PM -0700, Lee Howard wrote:
> On 05/08/2019 09:59 AM, Jason Rissler wrote:
> > I am having a problem that I can not receive faxes successfully from a 
> > few senders. I'm attaching a log from one of the sessions in hopes that 
> > someone can tell me what setting I could tweak or what the problem might 
> > be. I can receive fine from this sender when using an HP Officejet Pro 
> > multifunction machine but not with HylaFAX+.
> 
> TCF is a signal which is sent by the sender to the receiver and is a
> 1.5-second series of zeroes using the selected carrier modulation.  So,
> here is a list of modulation speeds and the number of bytes that should
> be zeroes in a TCF signal:
> 
> 14400 bps: 2700 bytes
> 12000 bps: 2250 bytes
> 9600 bps: 1800 bytes
> 7200 bps: 1350 byes
> 4800 bps: 900 bytes
> 2400 bps: 450 bytes
> 
> The fax receiver is required to use the reception of the TCF signal to
> determine if its demodulator is properly trained on the signal from the
> sender.  If it is not trained properly then it will not be able to
> demodulate the page image data and will need to be retrained
> successfully before fax protocol can proceed to receive page image data.
> 
> In order to determine if the TCF signal indicates a proper training of
> the demodulator (the modem) HylaFAX looks at two different things in the
> TCF signal:
> 
> 1) the percentage of non-zero data found in the signal (Class1TCFMaxNonZero)
> 2) the length of the largest run of zero data found in the signal
> (Class1TCFMinRun)
> 
> Both of those settings can be tweaked in the modem config file.  (See
> the hylafax-config man page for details.
> https://hylafax.sourceforge.io/man/hylafax-config.php)  By default,
> Class1TCFMaxNonZero is 10 (meaning up to 10% can be non-zero) and
> Class1TCFMinRun is 1000 (meaning 1000 ms of uninterrupted zeroes) which
> is modified by Class1TCFMinRunECMMod to 500 for ECM sessions.
> 
> If you look at your session log, then, you'll see the following results
> from TCF signals in various training attempts:
> 
> 14400 bps: RECV: TCF 2842 bytes, 96% non-zero, 1 zero-run
> 12000 bps: RECV: TCF 2368 bytes, 94% non-zero, 1 zero-run
> 7200 bps: RECV: TCF 1459 bytes, 54% non-zero, 1 zero-run
> 4800 bps: RECV: TCF 865 bytes, 96% non-zero, 1 zero-run
> 2400 bps: RECV: TCF 493 bytes, 60% non-zero, 2 zero-run
> 2400 bps: RECV: TCF 493 bytes, 60% non-zero, 3 zero-run
> 2400 bps: RECV: TCF 493 bytes, 60% non-zero, 3 zero-run
> 
> These are very bad results and indicate that the modem is not training
> practically at all on the signals from the sender.  I don't think there
> is anything you can reasonably do to Class1TCFMaxNonZero or
> Class1TCFMinRun to get things working.  The problem isn't with your
> configuration.
> 
> The problem is that the modem is not training properly on the sender's
> signal.  So, it's a problem with the line audio quality on the call or a
> problem with the modulator/demodulator compatibility between the sender
> and the receiver.
> 
> If you're not having a problem receiving from most other senders, and if
> this sender can send to most other receivers, then it would most likely
> be a problem with the firmware programming on your modem.  The solution
> would be to change modems (preferably to one with a different chipset or
> better firmware).
> 
> Thanks,
> 
> Lee.
> 
> 
> 
> ____________________ HylaFAX(tm) Users Mailing List _______________________
>   To subscribe/unsubscribe, click http://lists.hylafax.org/cgi-bin/lsg2.cgi
>  On UNIX: mail -s unsubscribe [email protected] < /dev/null
>   *To learn about commercial HylaFAX(tm) support, mail [email protected].*


____________________ HylaFAX(tm) Users Mailing List _______________________
  To subscribe/unsubscribe, click http://lists.hylafax.org/cgi-bin/lsg2.cgi
 On UNIX: mail -s unsubscribe [email protected] < /dev/null
  *To learn about commercial HylaFAX(tm) support, mail [email protected].*