[Asterisk-Users] Terrible crackling on analogue line and X100Pcard

[Stuart Ford]

Rich Adamson wrote ...

> What country are you in, and does the chipset on the compat card
> support the telco standards in your country?

I'm in the UK. The card was bought in the UK, but from Ebay, so I suppose it 
could have originated from anywhere. The card dials and answers calls 
without a problem, so it must be doing *something* right.

I didn't *mean* to cheap out over this - I tried to buy a genuine Digium 
part, but they don't seem to do it any more and I can't find it for sale 
anywhere. The Ebay vendor claimed it was 100% compatible.

The card reports itself as:

00:02.0 Communication controller: Individual Computers - Jens Schoenfeld 
Intel 537

When the wcfxo module loads, dmesg reports:

Zapata Telephony Interface Registered on major 196
PCI: Found IRQ 11 for device 0000:00:02.0
Uhhuh. NMI received. Dazed and confused, but trying to continue
You probably have a hardware problem with your RAM chips
wcfxo: DAA mode is 'FCC'
Found a Wildcard FXO: Generic Clone
Registered tone zone 4 (United Kingdom)

The 3rd and 4th lines are suspicous, but I've no idea what they mean. Does 
it refer to the system RAM or some sort of special RAM on the card? What is 
NMI?

> If the chipset doesn't match your telco standards, there is a high
> probability you won't get rid of the echo. If it does match, then try
> echotraining=800
> echocancel=yes

I already use those parameters in zapata.conf, they make no difference :(

> Regarding the crackling noise, have you checked for shared
> interrupts (cat /proc/interrupts)?

This is the output:

           CPU0
  0:  211266080          XT-PIC  timer
  2:          0          XT-PIC  cascade
  7:     488230          XT-PIC  eth0
 10:    2113812          XT-PIC  eth1
 11:  211520617          XT-PIC  aacraid, wcfxo
 14:         11          XT-PIC  ide0
NMI:          1
ERR:         60

It's sharing an interrupt with the RAID controller. I did try to separate 
the interrupts when I installed the card, but any combination other than 
that automatically assigned by the BIOS caused the Linux kernel to fail to 
even uncompress at boot time, much less boot the system, which struck me as 
a pretty alarming failure.

> If you run "cat /proc/interrupts" every ten seconds, do you see
> calculated interrupt values of about 1,000?

I don't know what you mean here.

> Go to /usr/src/zaptel directory and run
> ./zttest
> Do you get something close to 100% over some period of time?

Yep:

# ./zttest
Opened pseudo zap interface, measuring accuracy...
99.987793% 99.987793% 99.987793% 99.987793% 99.987793% 99.987793% 99.987793%
99.987793% 99.987793% 99.987793% 99.987793% 99.987793% 99.987793% 99.987793%

Is that good?

> What version of asterisk are you running?

1.0.7 plus Zaptel of the same version.

Thanks

Stuart