🚀 go-pugleaf

In article <B2C918319668B610F@cara.demon.co.uk>, peter@cara.demon.co.uk
(Peter Ceresole) wrote:

> In article <916670544.29352.0.nnrp-03.9e981d17@news.demon.co.uk>,
> "SEG" <seg@technologist.com> wrote:
> 
> >I'm no telecoms expert, but this seems to indicate to me that X2/V90 is
> >rather (very?) sensitive to the physical connection/cable conditions.
> >
> >I hope that one of the experts in this group might come up with a better
> >(and preferably on-topic!) answer...
> 
> No; you seem to have got it fairly right.
> 
> One possible bit of help would be to extent the carrier loss wait time.
> It's the time for which the modem hangs on to the connection if it detects
> a loss in carrier. A normal default would be a bit under a second. But that
> means that if the line goes dodgy for even a short burst, the modem drops
> the connection. You can extend the carrier loss wait time, usually by
> changing the value of a register using an "ats(x)=nn" command where (x) is
> the number of the register in question and "nn" is the duration you want.
> Five seconds is a reasonable value. The number of the register you will
> have to find out from the manual or by interrogating the modem to get the
> list of registers (also with an "at" command- it's "ats$" for a Courier but
> probably not for your modem...)

I used S10=30 on my Global Village but the Pace manual doesn't mention S
settings.

> However this may not solve your problem. Some V90 implementations are
> simply dodgy. People are reporting line drop problems which are
> miraculously solved by flashing to a newer and better version of the modem
> firmware. See if there's anything later available for your model.  

Yes, there are lots! So I'll wait a bit for advice.

-- 
Regards Bob,

Bob Kenyon
email: bob@raspbrry.demon.co.uk