Conference napalm::commusic_v1

	Now, before you all rush to reply to this...

	I just got off the phone from a store that told me something that I
hadnt't realized.  The whole idea behind XLR connections is to reduce line noise
in longer cables that would be inherant in unbalanced (1/4") cables.  So by 
trying to run long lengths of unbal cable, I'd be setting myself up for
more noise and less signal.

	Well, now, this brings me to the following questions:

	Is this principally a concern in Mic lines since the signal level is
	lower?  Or would synths and guitars (with pickups) fall into this
	category as well?

	How much is lost in using converters (unbal to bal) on either or both
	ends?  (For instance - would anything be gained by converting an
	instrument line to bal XLR, running a longer cable, and reconverting it
	back to unbal at the amp back?)

	How long is safe for speaker cables?  20ft?  50ft?  100ft?

    Synths generally produce line levels, and can be run reasonable
    distances over high impedance (1/4" unbalanced) cables.  I have
    run stuff 25' with no problems, but I'm not sure I'd want to run
    much further.
    
    Mics and guitars produce lower level, generally unbuffered outputs.
    You can solve the long run problem by using a preamp local to the
    instrument and then running line levels over unbalanced cables, or
    you can go to XLR terminated balanced cables, which you can run almost
    any distance.  You will probably need a direct box to connect a guitar
    to a balanced line so as to not affect your sound by loading the pickup
    wierdly.  The direct box serves as a buffer and level amplifier.
    
    You gain a lot by converting from hi impedance (1/4"; HiZ henceforth)
    to low impedance (balanced XLR; LoZ henceforth), namely relative
    noise immunity on long cable runs.  You lose something in the
    transformer, namely frequency response, but unless they're garbage
    transformers you probably can't hear the difference.
    
    Speaker cables pose a different problem - noise is not the issue
    as much as wire resistance loading the amps output is.  So the heavier
    the wire (and the lower its resistance per foot) the longer you
    can run it without unduly loading the amp (consuming power making
    it hot instead of moving the speaker cones, as well as altering
    the amp's behaviour because of loading effects).  Ordinary 18 gauge
    "lamp cord" is ok for runs of 20 feet or so, but you want heavier
    duty stuff for other reasons (durability, mainly) so you might as
    well get it.
    
    Regarding LoZ vs HiZ, there are really four separate things going
    on here that are related:
    
    	balanced vs unbalanced - a balanced circuit has three wires;
    	two signal (out of phase with respect to one another) and one
    	ground.  At the destination end, the two signals are combined
    	out of phase so the signals reinforce and the noise cancels.
    	An unbalanced circuit has only two wires - one signal and one
    	ground.  No noise cancellation possible.
    
    	1/4" vs XLR - the former mostly used for unbalanced, the latter
    	mostly used for balanced connections.  Just a matter of convention;
    	there are three conductor 1/4" connectors, and you can use XLRs
    	for unbalanced circuits.
    
    	line level vs. mic level - since mic level signals are lower
    	level, and more susceptible to noise, they generally use balanced
    	circuits.  Mic level signals are millivolts to tenths of volts;
    	line level signals are around 1 volt.
    
    	low impedance vs high impedance - the apparent impedance (ac
    	analog of dc resistance; includes inductive and capacitative
    	effects, which are frequency dependent) across the cable/input.
    	Affects noise susceptibility (the higher the impedance, the higher
    	the induced voltage for a given noise source).
    	
        
    The combinations of high impedance, line level, unbalanced 1/4"
    connectors vs. low impedance, mic level, balanced XLR connectors
    reflect sensible, if traditional, choices.
    
    len (who knows just enough EE to be dangerous...)
    

	Thanks, Len, I knew you'd come through.

	So, from the previous reply, can I assume the following:

	If I have balanced (low impedance) cables, I can run them longer without
	much worry.

	If I have 1/4" (high impedance) cables, I should convert them to 
	balanced cable with a transformer if I want to run them any distance
	(say, over 20 ft).

	But, just to clarify any last questions I have, let me pose the 
following:

	My mikes have XLR connections.  Even though mics produce low-level
	signals, can I safely assume that they can be run longer distances
	than 20ft (I have SM58's) because the XLR connector implies balanced
	signal?

	Conversely, my keyboards have 1/4" (HiZ) jacks.  But since they produce
	line level signals, can they be run longer distances or should they, too
	be converted to balanced for more than 20 ft?

	I read somewhere (AUDIO conference, I think) that 16 guage zip cord 
	(such as can be bought at Spag's) is great for speaker cable.  Is this
	what you meant by better than 'lamp-cord' wire, and can I safely assume
	30 to 50 ft (max) with this?  Or were you talking about even better
	grade than that?  (This is one thing I'd like to know before I leave 
	tonight since I'm going to Spag's).

	Thanks, again for the help.

	Dan

    Let's see if I can get all these right:
    
    1) yes, balanced (low impedance) cables mean longer runs without
       much worry
       
    2) yes, 1/4" means consider conversion to low impedance/balanced
       for runs longer than 20 feet, especially for mic level signals.
    
    3) yes, XLR connectors strongly imply low impedance balanced cables,
       which mean longer runs are ok, even with mic level signals.
    
    4) 1/4" *line* level outputs can be run considerable distances without
       worry.  If you must worry, convert to low impedance at the
       instrument outputs and back to to HiZ at the amp/board inputs.
       Direct boxes may prove helpful/necessary.  Avoid unnecessary
       conversions if possible.
                  
    5) yes, lamp cord is zip cord.  Speaker wire voltage levels are
       considerable (volts) and speakers eat current so a few millivolts
       of noise are not going to drive the speaker very hard, especially
       with no current sourcing capability to speak of.  Hence no shielding
       necessary, but current carrying capacity is important for speaker
       cable.

    Len,
     
           I'm a little confused on the speaker cable topic. If I
    understand the note, ..If I lengthen the speaker cable, its resistance
    goes up. This resistance would appear to be in series with the speaker.
    
    Then wouldn't the impedance of the entire load increase? If it does,
    then the amp should see a higher resistance, or a lesser load than
    it would have with "monster cables" (is an audio amp a voltage 
    source or a current source? I don't know.)
    
    Unless to get the same volume output you crank up the amp a bit
    to compensate for the lossy cables.. therefore dissipating the
    extra power in the cables.

        Hmmm 
    
    Frank

    You got it.  An audio amp is a current source (at least those intended
    to drive "dynamic" speakers, i.e., things with voice coils).  Try
    driving a dynamic speaker with an electrostatic source - lots of
    volts (tens of thousands), only microamps, only microdB.  Assuming
    of course you don't exceed the breakdown voltage of the voice coil
    insulation - you've seen magnet wire, it's insulated with glorified
    varnish.
    
    Even if the amp was a voltage source, the speaker would see less
    of the total output voltage of the amp because of the voltage loss
    across the cable resistance.  This voltage loss makes the cable
    hot, and does not contribute to the speaker's cone movement.
    
    len.
    

	O.K.  I think it's starting to sink in.

	Now a general question on products:  Does anyone have any 
recommendations on any direct boxes?  I know Boss makes one.  Anyone have
oppinions on the ones available, approximate prices, reliability...  I have
used the transformers that are available from Radio Shack with success - (in
fact it was the only product by R.S. that my sound man ever *recommended*).
Are these basically the same thing as a direct box?

	Dan the_tireless_inquisitor

    A direct box can be as simple as a matching transformer, but usually
    there's more to them (e.g., active buffering, ground lifts, etc.).
    
    I'm not sure how a guitar pickup would react to a matching transformer
    without buffering (via an op amp, like in a direct box).
    
    I'm well beyond my "expertise" (i.e., willingness to conjecture)
    now, so I guess I'll just shut up....
    
    len.
    

    I highly recommend checking out the Countryman Direct boxes.
    They're rather expensive as DB's go, about $150.00 last I checked.
    They're totally worth it though...quiet...wicked quiet, and rugged.
    Lots options for different applications, instruments, etc. 
    Check one out, Whirly's or Daddy's should have 'em, any place 
    respectable.
    
    -Bj

    re:.6 
    Actually, a power amp can be correctly modelled as an "ideal" voltage
    source in series with an impedance, OR an "ideal" current source
    in parallel with the same impedance. For the type of amps we're
    talking about, the impedance value is probably a fraction of an
    ohm, i.e. a couple of tenths.
    
    An "electrostatic" source would actually be a large voltage source
    in series with a large impedance. Due to the impedance mis-match,
    an "electrostatic" source would tranfer very little power to a speaker.
    Len, you had the right result for the wrong reason.
    
    Regarding speaker cabling, what I recommend is getting heavy-guage
    AC power cable of the type that is "round", rather than any kind
    of zip-cord. The round stuff is more durable, easier to roll up,
    and is probably available in heavier guages. Go with 14 or 16 guage.
    The price difference is negligible, and 14 guage is good for running
    lots of power into a long cable run.
    
    As someone hinted, using wire that is too small means that your
    cable will dissipate a PERCENTAGE of the amp's power, meaning that
    you will have to drive the amp slightly harder to get the same power
    at the speakers.
    
    Peter
    

    re: direct boxes
    Before spending REAL money on a DB, if I were you I would try just
    using a 1/4 - XLR convertor on each end a standard mic cable.
    When wiring it up, connect the "third" wire to ground so that it
    doesn't float and pick up extra noise.
    
    I suspect that in many (most?) applications, a DB is expensive
    overkill. In my PA setup, I always used adaptor connectors (home
    made) plugged into my snake. Noise from those low-level signals
    was never a problem.
    Bad AC power distribution/grounding in clubs WAS a problem, however.
    Don't always assume that 3-prong outlets are properly grounded.
    
    Peter
    

RE .10

	What I actually bought was just what you described - 14 guage round
power cable.  It was a bit difficult finding the right 1/4" plug for the ends
since the size of the cord was too big for most common plugs.  I finally
uncovered some plugs I had bought months ago from Radio Shack that had 1/2"
diameter barrel.  Even with those I had to drill out the hole to fit the wire 
in, but they worked out fine after that.  Plastic barrel, though, makes me 
concerned about someone stepping on it and breaking it.  Anyone know where to 
find a good LARGE metal barrel plug for such large cable?

	Just an aside - when I brought the 14 guage cord home I thought I had
succombed to major overkill and had wasted $20 (you can't return wire once it's 
cut).  It was encouraging to come in this morning and find it recommended.

RE .11   -< direct boxes >-

	Making converters

	Can anyone provide a diagram of the wiring scheme on XLR connectors?
Or is it quite flexible or apparent when you open one up to wire it?  I came
across someone selling a six-channel (home-brew) snake for $50.  Even though
it doesn't have returns and I really wanted more than 6 sends (since I want to
expand later, and since I can put more than one input into each channel of my 
mixer) I think I'll buy it.  I go to look at it next week.  It may be obvious 
when I see it that it's a shoddy job.  We'll see.  Any suggestions of what to 
watch out for apart from bad wiring/soldering?


	Dan

    Dan,
    I've got those large plastic Radio Shack plugs, too. At least half
    of them are now broken, but after the plastic breaks, just wrap
    them up with whole bunch of vinyl tape and hope for the best.
    I suppose it might work to make your own metal housing for them
    by using some kind of inside-threaded pipe - maybe from a plumbing
    supply?
    
    I'll check one of my XLR - 1/4 converters tonight for the wiring.
    I found that it was NOT intuitively obvious how to hook it up. As
    a guide, I used the schematic from my Allen-Heath mixer schematic.
    
    re: the $50 snake. The single most important thing is to check the
    quality and condition of the cable used. The cable should also be
    of a type that has VERY fine strands in the conductors - otherwise
    they will eventually break. Much of the Radio-Shack type cable is
    not intended for repeated flexing. The kind used in professional
    XLR mic cords is best - Whirlwind is probably the best brand (i.e.
    most expensive).
    
    Peter
    

    re .10 - the "electrostatic" source I had in mind was in fact
    electrostatic in nature, not meant to drive electrostatic speakers.
    Consider a van de Graaf generator, rather than a high voltage power
    supply.  A charged capacitor (capable of storing very high voltages)
    has an effectively infinite source resistance.  I know about impedance
    mismatching, I was making a point about current sourcing capability.
    Ideal sources don't have such limitations.
    
    len (who turned into a computer type before his EE education was
    completed).
    
    
    

    as long as we're sort of on the subject.....does anyone have a simple
    schematic for 10db pads? I used to have a resistor divider schematic
    that used two resistors to get a 10db drop....now I find I need
    ti and it's gone or misplaced...thanks
    
    dave

    Don't know what you had before, but assuming you want to drive about
    a 10K high impedance input, how about a 2K series resistor shunted
    by a 1K resistor.  So long as your source has an impedance of around 
    300 ohms or less this ought to work, I think.  I'm rusty on what
    kind of values to use, so pardon me if I have the source and load
    impedances off.  This should give you about a 10 dB drop.
    
    Steve
        

    re:.14
    Gee, Len, I hate to be a pain but ANY and ALL electrical sources
    are modelled by a combination of an IDEAL V or I source and an
    impedance. The purpose of the impedance is to make the model work
    in the real world. I don't believe that an "electrostatic" source
    uses a different model - it's simply a very high voltage source
    (ideal) in series with a VERY large impedance. The key (related
    to driving speakers) is that a source such as a Van de Graaf generator
    does not produce enough electrical ENERGY to make much noise, not
    to mention that the impedance match is so far off that there is
    almost zero power transfer.
    
    NOW, does anybody remember what we were originally arguing about?
    8^)
    
    Peter
    

    Oh yeah, I remember now: my original beef was Len's statements in
    .6 that said that a power amp was a current source, not a voltage
    source; when it fact ANY amplifier is BOTH.
    
    re: -.2
    Do you need a 10db pad that is EXACTLY 10db? If so, you will have
    to customize the resistor values to work with the source and load
    impedance. OR, you can use active electronics in your pad, so that
    your pad has (effectively) infinite input impedance and 0 output
    impedance.
    If ABOUT 10db is OK, then a couple of off-the-shelf resistors is
    good enough.
    
    Peter
    

    Sorry, me again...
    The answer to how to wire up a 1/4-to-XLR adapter is:
    I used XLR pin 3 as the signal wire, with pins 1 and 2 and the
    connector shell all connected to the cable's shield. Pin 3 is the
    "center" pin in an XLR plug.
    
    Peter
    

    
    Dan,
    
    For the speaker wires in my sound reinforcement system, I also went
    to Spags. Their prices aren't the lowest but they're certainly
    convenient. I use the 16G round rubber cord for the mids and highs,
    the 14G for the low mids, and I use Monster Cable for the subwoofers.
    I'm pushing 1000 watts into the subs, and I like the added damping
    you get from the Monster Cable (10 gauge). I can hear the difference
    at high levels between it and the 14G. For your purposes, though,
    the 14G cord should do fine.
    
    Switchcraft makes a shielded metal 1/4" connector with a large barrel
    that should accept the rubber cord. I get around this problem by
    using dual bananna plugs for my speaker wires. They work great.
    
    CdH
    

< Note 772.20 by MELODY::DEHAHN >

>    Switchcraft makes a shielded metal 1/4" connector with a large barrel
>    that should accept the rubber cord. I get around this problem by
>    using dual bananna plugs for my speaker wires. They work great.
    
	Where can I find Switchcraft plugs around here?

	And, just curious, how do you use dual banana plugs (sounds like I'm
setting myself up for a gag...).

	Dan

    
    Dan,
    
    If you still live in Clinton (I bought your CB remember?) then you're
    fairly close to Newark on 12 Harvard St in Worcester or Industrial
    Components Corp. on Route 20 in Wilberham. I've only bought things
    from ICC because they take a Master Charge, Newark at one point would 
    only allow retail purchase by account (this might have changed).
    
    Newark is 757-4515, ICC I don't have handy but it's a 413 area code.
    The Switchcraft part number for the plug is #170, they're 11/16ths
    diameter. As long as you're not using them in a patch bay or other
    tight quarters they work great.
    
    Bananna plugs (I'll let some of the more humorous folks like Mr.
    Ross or Mr. Cote provide the jokes) are easy to use for speaker
    cables. You strip the rubber (uh oh) insulation off the cable and
    expose (double uh oh) about an inch of the conductors. Strip each
    conductor about 3/8 inch back. Twist the conductors real tight and
    twist (ooooh) them into the receptacle on the bananna plug. Then
    you take a small tie wrap (kinky) and tie the rest of the conductor
    onto the strain relief on the plug. Done (what a disappointment)
    This even works for Monster Cable. I use Pomona plugs, model number
    MDP, and H. Smith single bananna panel jacks because they're better
    quality. Avoid Rat Shack bananna connectors as they're made of cheap
    plastic, not phenolic like the good ones, and will break easily
    if stepped on. The Pomonas withstand stepping on all night long
    on my cables and I haven't had to replace one yet.
    
    I had to replace the bananna jacks on the back of one of my amps
    because it was too worn out to make a tight connection with the
    plug. That's the drawback with bananna connectors, they're not captive
    like XLR's or 1/4" connectors. If you're careful with them and treat
    them like you treat your performing equipment they'll do fine.
    
    CdH
    
    
       

    I have often thought about using heavy duty twist lock connectors
    like the ones used for 3 phase wiring.  I seem to remember seeing
    large sound reinforcement companies using something like that.
    
    The one problem might be people plugging 220V into your speakers.
    Oh well, you can see how well they reproduce 60 Hz for about 5 msec...
    
    Also, ITT/Canon are proposing a new standard for speaker
    interconnection like the XLR connector for balanced signals.  When
    that comes about, should help standardize some of this stuff.  The
    controversy now is over what color should represent left and what
    color should represent right.
    
    		Derek

    
    Hubbell TwistLok connectors are no doubt the best for high current,
    large diameter cabling. But I priced them out....$6-$7 for the cable
    connectors or receptacle.
    
    I have 48 speaker connectors in my system. That's a lotta bucks
    for connectors. Bananna plugs were $1.75-$2.50 ea.
    
    CdH
    

    Dan,
    
    The only thing in those Radio Shack "transformers" is the connector
    on each end and an actual transformer (one part). So for your box,
    you just need a transformer for each line. I'm not sure if you want
    low impedance in/low impedance out, or if you want to go from low
    to high impedance. You'd need a different transformer for the two 
    different situations.
    
    One place that has an excellent reputaion in audio transformers
    is:
           Jensen Transformers
    	   10735 Burbank Bl.
    	   N. Hollywood, CA 91601
    	   213-876-0059
    
    I'm sure they'd have the ones you need. I imagine they have a catalog,
    not sure on this though.
    
    When you get your transformers, each one would simply be wired up
    like this:
    
                            +--------+
    		2 o---------|        |----------o
                            | xfrmr  |             1/4" out
           XLR              |        |            
          side  3 o---------|        |----+-----o
                            +--------+    |
                                          |
    		1 o----+                  |
    		       |		  |
    		       V                  V
    		    ground (case)      ground (case)
    
    
    You'd want to be consistent and wire them all up the same (i.e.
    xlr pin two should go to the same transformer pin on all of them).
    Hmm, right at the moment I don't remember if pin 2 or pin 3 is the
    hot lead on an xlr connector. If you connected pins 2 and 3 backwards,
    it would reverse the phase. That might or might not matter, at worst
    it would degrade the sound.
    
    Hope I've helped.
         -Keith

    
    Industry convention is pin 2 hot.
    
    Jensen transformers are nice but very expensive. The cheapest ones are
    $18 each. How many lines are we talking here? Transformers also limit
    your headroom. Even the best ones (around $100 each) still saturate at
    +22 dBm. But if you need real isolation, they're the best way to go.
    Why are you running balanced lines? Which way are they running? Where
    is the snake terminating?
    
    CdH
    
    

    
    Sounds like you want to build a *direct box*, Dan.  I'd stop by
    Union and ask to look at what types they stock.  They run from about
    $20 to over $150.  Some direct boxes use active circuitry rather than 
    transformers, but typically these cost more (plus they use a battery).  
    
    I have a schematic for an active direct box in an old Guitar Player
    issue if you might be interested.
    
    One thing commonly found on direct boxes which you have left out
    is a ground lift switch for the XLR.  This is used to disconnect
    the XLR shield from the 1/4" shield to correct hum from ground
    loops.

    
    re: .27
    
    +22dBm is almost 10v across a 600 ohm load.. I'd think that most
    line level type applications would be far lower than this, so headroom
    shouldn't be too difficult to deal with.
    
    re: .28
    
    I was under the impression that the Rat Shack thingies cost about
    that much anyway. If you'd only be saving a few $$, you'd be better
    off getting some high quality ones. Don't forget, your sound will
    only be as good as the weakest link. I have no idea what the frequency
    response is of the ones from Rat Shack either, it may not be good.
    
    Are you sure the the output of your submix board is high impedance?
    The 1/4" connector doesn't necessarily mean that, it could be an
    unbalanced low impedance output. If the R.S. transformers are
    low-to-high (or high-to-low, depending on which side your looking
    at), they wouldn't be correct. You'd want low/low transformers.
    
    	-Keith

RE < Note 772.29 by AQUA::ROST "She's looking better every beer" >

	I'm not sure if a direct box is what I want or not.  If it can be built
at a low cost and will give better performance than the transformers (which cost
about $11 now and seem to work just fine for me), I'd be interested.  My main 
purpose is to do two things;  1)  Cut down the need to fumble around inside the 
rack, and 2)  Eliminate the need to carry around those RatShack 
transformer/adapters.

	I'd like to see that schematic on an active direct box, though.

	Dan
    

RE < Note 772.30 by AQUA::KANOUN >

	RatShack adapters cost around $11.  The cost difference wouldn't be all 
that significant except that I already have the RatShack adapters now.  I have
not noticed anything bad about their frequency response.  I'd just like to make
a permanent wiring arrangement instead of carrying them around.

	Perhaps I said it wrong about hi-Z output on my board.  What I mean is
that I want to send a BALNCED signal (which I thought was the same as saying
I wanted to send a lo-Z signal).  My snake as well as my board uses balanced 
lines.  I just figured I'd use them if they are better than unbalanced lines.
My submix board does not send out balanced lines.  That is why I'm asking about 
transformers.

	Dan

    
    +22dBm is EASY to generate. The 'weakest link' analogy is right. If
    whatever's driving it is capable of +27dBm and your transformer starts
    to generate high THD at +22dBm then you know what you're throwing away.
    
    $18 is the CHEAPEST Jensen transformer. It might not even be the right
    one for your needs. What you are building is definitely a direct box.
    If you can't hear the sound degradation in the Rat Shack adapters, then
    you shouldn't be worried about Jensen transformers. Cheaper ones like
    Stancor or Quam or Soundolier would be fine for your needs. They are
    more like $5-6 each.
    
    CdH
    

RE < Note 772.33 by STROKR::DEHAHN >

	Well, Chris, we've had conversations like this before.  I honestly
can't hear a lot of things most others can when they listen to a sound system.
'Skinda funny when you think about the fact that I'm so heavily involved in
sound as a hobby.  That's why I take my wife along whenever critical listening
is necessary.  We just recently bought a tape deck for the car.  While I could
hear the differences in quality between an $80 deck and a $300 deck, it took
my wife to distinguish between two $300 decks.  (And she has to listen to four
screaming kids all day! &^)

	I have a NEWARK catalog here and I found STANCOR transformers.  Can you
help me out by being more specific as to what kinds of transformers I'll need?
Also, was the wiring diagram in an earlier reply correct?

	Dan

    
    Dan,
    
    Yes, we've had that conversation before, and I agree with you. There's
    no sense spending $$ on something you can't appreciate.
    
    I have an old (#108) newark in the office, they don't really have what
    you want. Newark isn't really an audio jobber. I'll check what I have
    at home and post it here.
    
    CdH
    

    Hi Dan,
    What make/model mixer are you using? Does it have a "Sum" output
    as well as Sub A and Sub B outputs? Are you sure these aren't
    balanced outputs? Are you terminating these balanced lines at a
    balanced input? 
    	Low impedance doesn't necessarily mean balanced and you may
    not want to go balanced especially if sound quality suffers
    because of cheap transformers. Technically, you don't need to run
    balanced lines if the signal is line level. Cheap transformers
    will limit your bandwidth and IMO active balancing (op-amps) is
    cheaper and better for your application.
    
    Neal
    
    
    	

    
    If there is room inside the Kawai, why not do the conversion there?
    
    I seem to recall that you found that the XLR jacks were simply
    in parallel with the 1/4" jacks; so just cut those wires, take the
    transformers out of your RS adapters and wire 'em up, or maybe
    use an active circuit (where you could steal the DC you need from
    the Kawai supply).  The parts for an op-amp driven balanced cirucit
    would certainly be under $10.  Time to dredge up those schematics....