Path: utzoo!utgpu!news-server.csri.toronto.edu!rutgers!usc!zaphod.mps.ohio-state.edu!caen!uflorida!gatech!usenet.ins.cwru.edu!falstaff.mae.cwru.edu!jb
From: jb@falstaff.mae.cwru.edu (Jim Berilla)
Newsgroups: comp.protocols.tcp-ip
Subject: Re: What is the voltage spec for thinnet?
Summary: What *are* you talking about?
Message-ID: <1991Feb6.033424.21632@usenet.ins.cwru.edu>
Date: 6 Feb 91 03:34:24 GMT
References: <1991Jan30.155606.21529@dartvax.dartmouth.edu> <1991Feb5.131333.2047@isavax.isa.com>
Sender: news@usenet.ins.cwru.edu
Distribution: usa
Organization: Case Western Reserve University
Lines: 73
Nntp-Posting-Host: falstaff.mae.cwru.edu

In article <1991Feb5.131333.2047@isavax.isa.com> cliffb@isavax.isa.com (cliff bedore*) writes:
>In article <1991Jan30.155606.21529@dartvax.dartmouth.edu> wbc@moose.dartmouth.edu (Wayne B. Cripps) writes:
>>
>>
>>What should the voltage be on thinnet? - I get readings of
>>-1.8 to -2.0 volts and .2 to .3 volts - is this in the
>>range?  is 1.2 volts ok?
>>
>>	Wayne
>
>At the risk of stepping into something soft and gooey, I thought I'd put on
>my engineers hat for a while and comment on this.  
>
>First.  It will be very traffic dependent (assuming you're using a voltmeter
>that does some averaging).

True.  Don't use a voltmeter, use an oscilloscope.  You'll see many interesting
things.

>Having stated that and not knowing the details of an ethernet board, but
>knowing something about transmission lines, we can get a wag of ranges for
>the voltages.

Not true.  The voltages on the ethernet are clearly defined.

>The lines are 50 ohms and are terminated in 1/4 or 1/2 watt resistors. (Mine
>is cause I did it myself and haven't had problems).
>
>Power (watts) = voltage ^2 / resistance or
>
>voltage = sqrt( power * resistance)
>
>voltage = sqrt ( 1/4) * 50 ) or 3.5 volts. for 1/4 watt power
>
>voltage = sqrt ( 1/2 * 50 ) or 5 volts for 1/2 watt power

What *are* you talking about?  (And take off that silly hat.)
In the case of ethernet, the voltage depends on the amount of current
pulled out of the tap.  It's independant of the power rating of the
terminating resistors.  Remember that the tap appears as a 25 ohm load,
i.e. it's connected to two 50 ohm transmission lines.

For the AM7996 transceiver (common in a lot of Sun's), the voltages are
specified as follows:  High level is between 0 and -.1 volts, low level
is between -1.625 and -2.2 volts.  As stated above, this voltage is
generated by a current sink (to -9V) by the chip.  An ideal current sink
has infinite impedance, and doesn't load the transmission line.  If two
or more stations transmit at the same time, the voltage on the line goes
below -2.2 volts.  The chip detects collisions on this basis.

>These are for steady state DC levels.  Ethernet is 10mbit/sec square waves
>+- distortion from the coax.  This means the the voltage measures via some
>sort of averaging/rms/other meter would be something less (depending upon
>the actual waveform, activity on the cable etc.  (This is true for thick or
>thinnet. the impedance on the cable is the same)
>
>This the voltages you measure will probably be in the range for correct
>operation, but without looking at the waveform, you could have a short with
>some DC voltage giving you the same result.

You can use a meter for a few tests on the ethernet, but it's very limited.
On an inactive network, an ohmmeter accross the tap will read 25 ohms.  If
there are any stations transmitting, the reading will be jittery.  If the
network is shorted, it will read close to zero ohms.  If a terminator is
bad, or there is a break in the line, it will read 50 ohms.  A voltmeter
might tell you something, but I wouldn't use it if there's a scope around.


-- 
      Jim Berilla / jb@falstaff.cwru.edu / 216-368-6776
"My opinions are my own, except on Wednesday mornings at 9 AM,
           when my opinions are those of my boss."