ESD, was : RE: [sdiy] midi optocoupler isolation voltage?

Tue Mar 11 11:18:48 CET 2003

I must admit that I'm only a user of midi gear.
It is very long ago that I looked into the spec.
I'm ignorant.

If the conncetor shield and/or mesh must not be 
connected to chassis ground, where is it connected to?
There would be only a major loop problem, if both devices
have a plug with mains ground. I think only my Atari
computer could have something like that.
The current loop was chosen bcause it has no common mode problem,
I know.

The pin 2/4 seems to be used by some guitar gear people to add supply to
the midi cable. In this case they seem to use a 7 pin plug
and socket.

m.c.

-----Original Message-----
From: harrybissell at prodigy.net [mailto:harrybissell at prodigy.net]
Sent: Montag, 10. März 2003 22:52
To: nej22 at hermes.cam.ac.uk
Cc: synth-diy at dropmix.xs4all.nl
Subject: Re: ESD, was : RE: [sdiy] midi optocoupler isolation voltage?

Neil is correct.

Some manufacturers didn't pay attention to this...

I thought the shield was connected at the Midi output and
floated at the Midi input, so that the shield would be effective without creating a ground loop.

My 360 Systems Pro MIDI Bass... had all the shields tied to chassis...
as well as the tab of the 5V regulator.

MEGA bad karma ;^P

H^) harry

>
>> I *think* that the two wire midi bus is shielded via copper mesh, that
>> should be connected to the "ring" of the midi/DIN cable.
>> Both devices should connect chassis ground to that. So no potential
>> difference can build up between two pieces of midi gear in terms of DC.
>*Ack* no!  From the MIDI 1.0 Specification:
>"The grounding shield connector on the MIDI jacks should not be connected
>to any circuit or chassis ground."
>Also, regarding connection to the shield:
>"Pin 2 of the MIDI In connector should also be left unconnected."
>In essence there must be _no_ electrical connection between the MIDI jacks
>of any two devices, otherwise you'll have an earth loop (Bad Karma).
>Neil
>--
>Neil Johnson :: Computer Laboratory :: University of Cambridge ::
>http://www.njohnson.co.uk          http://www.cl.cam.ac.uk/~nej22
>----  IEE Cambridge Branch: http://www.iee-cambridge.org.uk  ----