[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: 20 MHz clock oscillator



All Axis products uses the solution suggested in the application
note but it is correct that some customers have had problems that
have been solved by switching to an external oscillator.

The best way to test ethernet is to run lots of traffic and
check ifconfig once in a while. On a half-duplex network you 
would expect some collisions but very few frame or carrier errors. 
On a full duplex network you would expect no collissions.

>RX packets:724 errors:0 dropped:0 overruns:0 frame:2
>TX packets:277 errors:0 dropped:0 overruns:0 carrier:0
>collisions:0 txqueuelen:100 

I am a bit concerned about the frame error counter. Other
customers having this problem have reported that the frame
error counter increases.

/Mikael

-----Original Message-----
From: owner-dev-etrax@xxxxxxx.com">mailto:owner-dev-etrax@xxxxxxx.com] On Behalf Of girerd
Sent: Wednesday, June 30, 2004 7:26 PM
To: dev-etrax
Subject: 20 MHz clock oscillator


Hi,

I read somes messages on the list about 100 MB Ethernet problems. Some guys
recommends to change the discrete oscillator (axis application note) to 
an integrated
clock oscillator.

Is this also recommended and confirmed by Axis ?

Also, I would be very interested to know which kind of test we can run 
to validate the
Ethernet network in a hardware point of view (routing etc ..).
Our design uses a MCM 2+8 and MCM 4+16 soon.

Regards,

Claude
-- 
____________________________________________
ииииииииииииииииииииииииииииииииииииииииииии
 Claude GIRERD - Service Electronique
ииииииииииииииииииииииииииииииииииииииииииии
 INSTITUT  DE  PHYSIQUE  NUCLEAIRE  DE  LYON
 4, rue Enrico Fermi                   
 69622 Villeurbanne cedex 
ииииииииииииииииииииииииииииииииииииииииииии
 tel: 33 (0)4 72 43 11 05
 fax: 33 (0)4 72 44 80 04
____________________________________________                  
ииииииииииииииииииииииииииииииииииииииииииии