Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version Tektronix Network News Daemon (B 2.10.2 based); site tektronix.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!mhuxr!mhuxt!houxm!vax135!cornell!uw-beaver!tektronix!larryme
From: larryme@tektronix.UUCP (Larry Meneghin )
Newsgroups: net.dcom,net.lan
Subject: Re: Ethernet compatibility
Message-ID: <5766@tektronix.UUCP>
Date: Fri, 27-Sep-85 14:08:38 EDT
Article-I.D.: tektroni.5766
Posted: Fri Sep 27 14:08:38 1985
Date-Received: Sun, 29-Sep-85 08:07:38 EDT
Organization: Tektronix, Beaverton OR
Lines: 49
Xref: watmath net.dcom:1329 net.lan:1044


>> One interesting thing I discovered when I installed the DELNI:  The
>> ethernet-2 delni is connected to the net with an ethernet-1 TCL
>> transceiver.  We haven't had any problems with that combination, which
>> really surprised me, when I thought about it.
>> -- 
>> Tim (radzy) Radzykewycz
>>      calma!radzy@ucbvax.ARPA
>>  eon more than one DELNI without problems.
>Well almost the same -- the DELNI is connected with an Interlan ethernet-1
>transceiver.
>-- 
>--peter gross
>UUCP:  {hplabs,seismo}!hao!pag
>CSNET: pag@ncar.csnet
>ARPA:  pag%ncar@csnet-relay.arpa
>
In the idle state the differential line drivers in an Ethernet-2 (IEEE 802.3) 
device go to 0 volts, because E-2 is AC (transformer) coupled.  The receivers
in any Ethernet 1 (DC coupled) device however, expect to see a positive 
differential voltage when the driver is idle.  0 volts is an undefined state 
for E-1.  Depending on how the transceiver is implemented, it may work just 
fine.  But!  E-1 equipment is not guaranteed to work with E-2 equipment.  
It may not work.  Worse yet it may work only when you're watching.  (bugs 
have a tendency to do that.)  

Another question is, how do the E-1 drivers react to trying to develop a DC 
voltage across the pulse transformers in the E-2 device.  Well, if they 
current limit, and the transformers don't saturate to the point where they 
cause startup problems at the beginning of each packet, it may work.  
Again, you just can't take that fact to the bank.  (Have you looked at the 
packets you're sending out with an oscilloscope?  Are you sending out all 
the preamble bits you should be?  Do the signals out of your E-1 drivers 
conform to E-1 specifications especially at the beginning of a packet?)

I don't recommend mixing E-1 and E-2 devices on the AUI interface at all.
(It's just fine on the coaxial cable side of course.)



tektronix!larryme

Larry Meneghin
Tektronix Inc.
P.O. Box 500
Beaverton, OR 97077
M.S. 50-761

(503) 627-7479


Brought to you by Super Global Mega Corp .com