Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site redwood.UUCP
Path: utzoo!watmath!clyde!bonnie!akgua!sdcsvax!sdcrdcf!hplabs!hpda!fortune!redwood!rpw3
From: rpw3@redwood.UUCP (Rob Warnock)
Newsgroups: net.music.synth
Subject: Re: Re: MIDI
Message-ID: <182@redwood.UUCP>
Date: Sun, 3-Mar-85 06:46:22 EST
Article-I.D.: redwood.182
Posted: Sun Mar  3 06:46:22 1985
Date-Received: Thu, 7-Mar-85 04:01:47 EST
References: <3416@alice.UUCP> <8617@brl-tgr.ARPA> <102@cvl.UUCP>
Organization: [Consultant], Foster City, CA
Lines: 59

+---------------
| > > the MIDI standard requires a current loop interface, hence an opto-
| > > isolator.
| > Eh?  The schematics for the MPU show the lines to be balanced pairs.
| > They are driven by convetional drivers, no optoisolaters are involved.
| > -Ron
| Refer to David Dronan's Exploring MIDI: the Musical Instrument Digital
| Interface.  The spec calls for a 5ma current loop interface.  How the
| Macintosh can accomplish an interface with NO MORE THAN A CABLE would be
| my next question.  How is the Mac capable of handling current loops?
| -dave
+---------------

Tutorial time on current loops, folks! (Having spent some years building
current-loop communications equipment...)

In any current-loop system, there must be a battery or "loop supply".

The loop battery can be external to the all of the controllers (or
interfaces) in which case all controllers must be floating (that is,
opto-isolated), or it may be embedded in exactly ONE of the controllers,
in which case that controller ONLY need NOT be "current loop", per se.

Likewise, exactly ONE of the controllers can be connected to earth
ground without causing "ground loops" and other hazards. Usually,
that will be the loop battery or the controller which is acting as
the "battery". (For example, your telephone is a "current loop",
device with the battery at the central office, and grounded there.)

The -422 interface of a Mac is capable of being the "battery", and
can support a limited number of opto loop devices (given certain
resistors, etc., in the cable), as long as the sum of the voltage
drops (each opto receiver is about 1.5 volts and each transmitter
about 0.2 volt) does not exceed the voltage drive of the -422 drivers
in the Mac. Therefore, if the Mac is the "battery", IT need not have
any opto isolators. (Clearly, TWO Macs may not be connected in this
way to the same MIDI current loop.)

Given that the -422 drivers do not break down, extra voltage could
be supplied to the "battery" (say, with a resistor to +12) to extend
the number of current-loop devices supported.  There will be a finite
(and probably small) limit on the number of controllers that can be
in the loop. Teletype interfaces used to use quite high voltages, such
as 60 to 100 volts, to extend the compliance of the loop. The target
loop current was set with a large, high-power resister. These days,
one uses "constant-current" FET devices, or op-amp circuits. Still,
there must be enough voltage in (around) the loop to turn on all the
opto receivers. (Note that the loop voltage must also not be so high
as to break down the opto drivers in the other controllers!)

I hope this has helped a little to explain the apparent contradiction.


Rob Warnock
Systems Architecture Consultant

UUCP:	{ihnp4,ucbvax!dual}!fortune!redwood!rpw3
DDD:	(415)572-2607
USPS:	510 Trinidad Lane, Foster City, CA  94404