Conference napalm::commusic_v1

1)  Are MIDI cables supposed to be shielded?  (Are yours?)  A shielded cable
properly grounded at its connector should not radiate no matter what is
going through it.

2)  Is the CZ properly grounded?

    I had the same problem with a CZ230S connected to an MT32. My personal
    guess is that the CZ may have a larger MIDI signal driving the cable.
    The MT32 mostly likely has a bipolar receiver, and any signal that
    goes above it's power supply voltage will leak into the power supply,
    effectly creating noise in the power supply. MIDI is basically RS232
    with a wierdo plug, which means it's not driven at TTL levels, and
    most vendors I assume would take the voltage from the analog supplies.
    Most power supplies are regulated from below, with no correction
    for slight overvoltages, except the tank capacitors.
    
    						John.

But MIDI receivers are optically coupled.

    Oh, no dey ain't! No where is it written that " Thou shalt use optical
    coupling ". Optical couplers cost money. 'nuff said.
    
    							John.

    Uhhh ... yes, I think it *is* written, right in the MIDI 1.0 Spec,
    that the reciever WILL be optically coupled, at least 500 volts
    isolation to the reciever unit, etc.
    	
    I believe the MIDI 1.0 spec also lists an HP part number for an
    optocoupler which is the "recommended implementation" of the Spec.
    
    If anybody has a copy of 1.0 handy, please look at the last two pages.
    I think the optocoupler part number is right there.

    I believe the problem will be found in ground looping.  Though optical
    isolation is used, it may not be electrically isolated if the MIDI
    cable has the ground connected at both ends and both devices use
    the ground connection.
    
    Steve

    Well, I guess I stand corrected. Silly me.
    
    						John.

    ive only gotten to .2 in this note and must reply.
    
    there is NO WAY you can drive onto the power supply of
    a reciever.
    
    Midi is 5 mil current loop with an OPTO isolator specified
    as the reciever. The cables are SPECIFIED to be sheilded
    and DRIVEN ground at the transmitter only.
    
    John, please check a spec before spec-ulating! ;')

    ron
    

    Ok. John recants....
    
    THERE IS an un-spec'ed potential problem that I addressed in
    another note on ground loops.
    
    since there is ground connction ONLY at the transmitter to the
    sheild of the wire, theoretically no loops should form.
    
    BUT. 
    
    The SHELL of the connectors (you know, the silver cylinder protectin
    the prongs) and its conntection to the GROUND of the cable is not
    well spec'd.
    
    SOME cables connect the in-shell to the out-shell. some dont.
    
    NOW. You use one that does. AND your manufacture has connected the
    reciever's shell to ground (not spec'ed precisely).
    
    Boom. Ground loop.
    
    Check your cable. Pin 2 should NOT be connected to the shell. If
    it
    is, try another vendor...
    
    rr
    

    
    Well, I checked the MIDI cords and the connector body is *not* shorted
    to any of the pins of the connector.
    
    Two other things I've noticed:
    
    1.  The problem is much worse with a tube bass amp I plug into often
    than with my mixing board (I've had radiation problems with tube
    amps and digital delays in the past where the clock "bled" into
    the audio).  That suggests a radiation problem.
    
    2.  I was incorrect in my earlier statement that the noise level
    changed with the synth's volume setting.  It doesn't.  Nor does
    it appear to be in the phones output (hard to tell, I can't get
    enough gain to be sure).
    
    Since there is *no* noise until the cable is plugged into the CZ
    and the cable does not have to be connected at the other end to
    cause the problem, I think it's just a radiation problem from the
    *cable* itself.
    
    I'm not about to chop my cable in half, but aren't these things supposed
    to be shielded????
    
    As far as a potential ground loop, I'm still unable to find anything
    with my ohmmeter, but I'll keep looking.

    Thanks for the replies.
    
    

The fault is not necessarily with the MIDI cables.  Home electronic gear
containing audio output amplifiers, such as TV sets and stereos, is prone
to being affected by RF noise.  This is because the designers, to save money,
omit any protective devices on the speaker leads.

To reduce distortion, typical audio amps take a little bit of their output
and feed it back into their inputs inverted.  Any noise introduced by the
amplifier itself thus cancels itself out.

Any RF picked up by the speaker leads can also get carried around this feedback
path to the input of the amplifier stage, which then acts like a small (and
not very efficient) radio detector.  I would not be surprised if tube
amplifiers are more likely to have this happen than transistor amps, due
to the higher impedances that abound in tube circuits.

If you put a small capacitor (.01 ufd? It has been a while since I used this
information) from each speaker lead to chassis ground in the amplifier, any RF
picked up by the speaker wires will be shunted away, while leaving the audio
frequencies unaffected.  The capacitors have to be as close to the amplifier
as possible, so don't put them on at the speaker terminals.

Also, keep your speaker leads and most other audio signal paths away from the
synths and MIDI cables.  Like on the opposite side of the room.

    
    Interesting followup:
    
    Last night I had a CZ101 hooked up to my S80, and the 101 exhibited the
    same behavior, that is, MIDI clocks and controllers caused noise in the
    audio output.  Hmm...still haven't heard this on anything but a CZ. 
    Guess Casio screwed up somewhere in shielding their audio circuit.
    
    							Brian