November 2007 - Posts - Andy's Blog
in
Advanced Search
KLIPSCH - The Ultimate Sound Experience

Andy's Blog

November 2007 - Posts

  • Hum and Buzz – Part III

    Ground Loops 101

    Let’s start our discussion of ground loops with two pieces of chassis grounded equipment.  Class II equipment can exhibit hum and buzz too, but the mechanism is a little different, but we’ll talk about that in another installment.  We’ll also start with an “unbalanced” signal connection since it is more prevalent in commercial equipment (RCA-type connections and stereo mini-plug connections are examples of unbalance signal connections).

    In the first example below a ground loop is created when the two pieces of equipment are plugged in and connected.  DUT 1 might be a CD player and DUT 2 might be a Receiver.  The signal from DUT 1 is passed through a signal cable to DUT 2.  The “signal ground” connection is made between points “C” and “D”.  Generally points “C” and “D” are also connected to the metal chassis.  The “chassis ground” connection is made where the “earth ground” wire attached to the chassis.  This happens at points “A” and “B”.  The bare wire connection at points “H” and “J” on the AC outlets are earth grounded by the wire running back to the breaker panel.

    The ground loop is formed, starting at point “H” to “A” to “C” to “D” to “B” to “J” and back to “H”.  For simplicity we’ll just say “HACDBJ”.  Any magnetic flux that passes through the loop will cause an electric current to flow in the loop.  Any time current flows in a wire there will be a voltage impressed on that wire.  Any voltage on the signal ground connection “CD” will become part of the audio signal and will be amplified.  By virtue of the fact that your whole house is wired with AC wiring with currents flowing every which way, the ground loop will be bathed in magnetic flux.  (Cable shielding is only effective for electric fields.)

    Also any current on the earth ground between points “H” and “J”, regardless of where it comes from (like a light dimmer on the same circuit) will create a voltage in proportion to the current on “HJ”.  Since the connection “HACDBJ” is in parallel with “HJ” part of the current will flow in “HACDBJ” and part of the current will flow in “HJ”.  In electronics terms a parallel path is known as a “current divider”.  The current flowing in “CD” will have an associated voltage, and this will be added to the signal from DUT 1.

    Attachment: housewire2.GIF (14606 bytes)

    The second example above is essentially the same as the first.  The difference is that the two pieces of audio equipment are powered by two different circuits on the breaker panel.  DUT 3 sends a signal to DUT 4 along an unbalanced cable with a signal ground “CD”.  The chassis are grounded at “A” and “B” and the outlets are grounded at “H” and “J” just like before.  The big difference here is that now the ground loop goes all the way back to the breaker panel.  The loop is now a very large loop and can have even greater potential to pick up hum from magnetic fields.  Also, now you have ground and neutral currents on the neutral bus bar coming into play.  Ground current and neutral current from all the other circuits in the house can now put voltage on your signal ground “CD”.

    So what can be done?

    1.      Make a block drawing of the components in your system, showing the signal connections (a single connection is fine for stereo connections) and the power cables.  Also show in your sketch which power cables have an earth ground.  Identify the house circuits that you are using, and show any other equipment that is sharing the same circuit (light dimmers, motors – like appliances, things with switch mode power supplies – computers and such, and things that might draw a lot of current or be injecting noise on the ground).

    2.      Study the diagram to see where your trouble might be coming from.  Then try some trouble shooting and make notes of what happens.  Only make one change at a time; do and undo the change so that you can have an A/B comparison.

    3.      Make loops smaller.  Smaller loops have less magnetic flux inside them.  Try running the AC cables right next to each other; use twist ties to hold them together temporarily.  Shorten signal cables and twist tie them up.

    4.      Does your AC cable or signal cable hang in a “half turn” around a power transformer.  (If this needs a sketch please let me know.)  The cable will become part of the transformer.  If your equipment cables hang in loops; tie them up.  Reroute cables to eliminate or shorten loops.

    5.      Move or reroute cables away from cables that carry large currents or are known to have currents with high harmonic distortion (almost every appliance/piece of equipment that is not resistive will have varying levels of high harmonics on the AC power line/cord).  Distance is your friend in these cases.  Try a different cable orientation or a stack your components in a different order.  The sensitive part of a PCB in one component may be right on top of a noisy transformer or rectifier in another.  Maybe it sounds crazy but a sheet of steel between components can help, and in extreme cases it may be necessary to wrap a steel sheet around the offending transformer (a magnetic “short” circuit).  I have measured hum reduction on the order of 25-30dB by doing this.  But it’s better and easier just to move the noisy transformer away from the sensitive equipment.

    6.      Reduce the ground voltage between components.  Plug all your components into the same AC outlet.  Use a power strip if you have to.  If there is a noisy light dimmer on the same circuit and you plug the system into a single outlet you can reduce the ground voltage difference (the voltage on “HJ”) to zero or nearly zero.  Make sure your power strip is rated for 15-20 Amps.

    7.      Plug the offending equipment into a different circuit outlet, so that the noisy ground will not be on the same line as your audio equipment.  Or if there is a permanent fixture/appliance (e.g. a ceiling light on a dimmer), plug your audio equipment into a different circuit.

    8.      If you are building a house, it can’t hurt to run a dedicated AC line to your equipment.  This will generally cost $200-$300.  If you can talk them into it, run a 20A circuit with 12AWG with a corresponding breaker and outlet instead of a 15A circuit.  Keep the run as short as possible and don’t run it parallel with any other runs.  And for a truly esoteric touch (I’m not claiming this will help, but it might) run the line in flexible steel conduit.  If you are in an existing construction, it is not impossible to run a dedicated line, and sometimes there are creative solutions to getting an AC line where you want it, using an existing line that is more noise free.

    9.      Transformer coupling.  Sometimes there is no other way… perhaps that can be its own installment.

    10. The “pin 1 problem” (this is in reference to pro or “prosumer” products using balanced connections).  This may require internal modifications... the manufacturer may already know about if if they have a product with this issue... but that's for balanced connections, and that's another discussion for another time.

    Posted Nov 30 2007, 05:17 PM by Andy W with no comments
    Filed under: , ,
  • Hum and Buzz - Part II

     

    Electrical equipment is grounded for safety reasons.  Grounding can also help prevent RF interference

     

    First… let’s get some agreement on nomenclature… since I’m writing, I get to pick. [:neener:]  There are many different ways to refer to “ground” and much depends on the context.  As with everything else we measure, we need a reference point, and in electronics we call the reference point either “common” or “ground.”

     

    A) Earth Ground:  this refers to the fact that your house wiring is grounded by the Earth (the planet we live on) and we use and we use the terms “earth” and “ground” interchangeably.  Your house wiring is grounded by attaching a large gauge wire from the Neutral wire at your breaker panel to a buried water pipe or a long stake driven into the ground.  If your house were not grounded the voltage on the neutral conductor of your house could be very high compared to the piece of ground you are standing on.  Fortunately with transformers, the primary side of the transformer doesn’t necessarily care what “ground” is on the secondary side.

     

    As shown in last weeks sketch, the neutral coming into your house is earth grounded, this “earth” is distributed around your house along with the “line” and “neutral” to the outlets around your house.

     

    “So if the neutral is grounded at the breaker panel, why do I need another ground wire?” you ask.  A very good question indeed.

     

    Let’s take the chandelier above your dining room table.  You have three wires that come to that round hole in your ceiling: L, N, and E (Line, Neutral, and Earth).  In the US the Line is BLACK, the Neutral is WHITE and the Earth Ground is bare wire.  Now you attach those three wires to corresponding wires in the chandelier.  There is a two wire cable that passes through a metal tube or decorative chain that goes to the light bulbs.  The ground wire is connected to the metal parts or can pass through the tube or chain of the fixture.  (In the sketch I showed the Line and Neutral running along side the chain for clarity.)

     

    Attachment: lightwire.GIF (6892 bytes) 

     

    Now let’s say the person that connected the light decided not to attach the ground wire, everything works and nobody knows any different.  Now imagine that after time the insulator on the LINE connection wears through or has a small cut and is now in contact with the metal parts of the chandelier.  If only the LINE connection is cut, then the light will still work, but the metal parts of the light are now “hot” (energized).  Now if you touch the light fixture you can get zapped.

     

    If the ground wire had been connected correctly, and the insulation on the LINE connection failed, the current would go through the ground wire back to the breaker panel and the circuit breaker would simply open, removing the voltage from that circuit, and no one would be shocked.

     

    B) Chassis Ground:  chassis ground plays the same role as the ground wire in the chandelier in the above example.  A grounded chassis will have the Earth wire entering the chassis and then securely attached to the metal of the chassis.  In case of insulation failure or component failure inside the chassis, the voltage on the chassis will not change; it will remain firmly “grounded.”  If insulation failure occurs inside the chassis, the fault current will be routed to earth ground, and either a fuse or circuit breaker inside the chassis will open, or in extreme cases, the breaker at the house panel will open, eliminating the risk of electric shock.

     

    WARNING:  If you are thinking about using a “cheater” plug on your audio equipment to eliminate a ground loop, re-read the above paragraph, and then don’t do it, or fault current could be routed through YOU.

     

    The chassis ground wire according to UL, CSA, and EU standards must be able to carry the full fault current without fusing (melting the wire).  This is required since if the chassis ground wire fuses before the internal fuse or panel breaker opens, the chassis will remain “hot” and you can still be shocked.

     

    Electrical equipment that does not utilize a three-prong AC cord does not have an earth grounded chassis.  In this case UL, CSA, and EU standards require that hazardous live voltages must have double or reinforced insulation (two layers of adequate insulation).  This is also referred to as Class II insulation.  Transformers of Class II products also have to meet the insulation requirements.  Parts that are not insulated must meet specific “creepage and clearance” requirements (which is to say there must be adequate space between conductive parts and hazardous live voltages – the air is the insulation).

     

    Some equipment will also have a polarized (wide blade) AC plug.  This ensures that the fuse or switch inside the equipment is always on the “hot” side of the AC line.  If you defeat the polarized plug by replacing it or cutting it, it is possible to flip the plug over and you are putting the fuse on the neutral.  Now if you have an insulation failure the chassis could still be “hot” even if the fuse blows!

     

    WARNING:  If you can’t plug in a polarized AC plug, get an electrician to change your obsolete AC outlet.  Defeating the polarized plug can be very dangerous.

     

    C)  Signal Ground:  This is the reference voltage for the audio (or video) signal in your system.  Each component in your audio rack will have a signal ground.  The signal ground connection must be “clean,” in other words, any voltage on this ground will be treated just like any other voltage on the signal line; it will be amplified and become part of the sound coming out of your speakers, or if in the video, can cause a noisy picture.  Sometimes this is called “common” as in a “common emitter” amplifier.

     

    D)  Digital Ground:  This is the ground reference for digital signals.  As with analog signals, poor ground on digital signals will have an equally degrading effect.  One common problem is “ground bounce” this occurs when there is noise on the digital ground and the digital circuit misbehaves because the voltage of the digital bits is no longer valid.  Ground bounce can also cause many digital circuits to reset without explanation, as it would if there were a power loss or a “brownout” (a partial loss of voltage).

     

    E)  Power ground:  The power supply in your audio equipment needs a return path too.  We call this the “power ground,” and sometimes this is also called “common.”  Electrical current needed to power the circuit components returns to the power source on the power ground.

     

     

    Next Week: Ground Loops 101

    Posted Nov 16 2007, 01:57 PM by Andy W with 1 comment(s)
    Filed under: ,
  • Hum and Buzz – Part I

     

    It is a fact of life.  Ground current happens.  And where it flows makes all the difference.

     

    In a perfect world there would be no ground current on the AC power line.  AC would enter on the LINE, and return on the NEUTRAL, and the earth ground would just sit there.  In the real world there is inductive and capacitive coupling that creates ground current.  Ground current can also be created by various electrical and electronic equipment, or it can be created by the pick up RF energy, especially when the ground forms loops (loops can increase the amount of RF energy that is picked up).

     

    Before we get too deep into ground current let’s take a look at how a typical house is wired.

     

    Attachment: housewire.GIF (8161 bytes)

     

    Power comes from an electric substation to the pole outside your house.  The distribution voltage is approximately 13kV, but higher voltages are used.  The power then goes to a transformer which takes the high voltage AC and steps it down to 120V for your house.  Usually a few neighbors share a transformer.  The center tapped secondary of the transformer gives you two “hot” wires, each with 120V, and a neutral wire.  The voltage between the two “hot” wires is 240V which is the sum of 120V+120V, which are 180° out of phase.  Even though there are two “hot” leads entering your house, and they are 180° out of phase, technically this is still a single phase system.

     

    SIDEBAR:  13,000 Volts eh?  Yes, so don’t be tempted to touch it, not even with a 10 foot pole.  The moisture in a wooden broom handle will conduct enough current to light a 100W light bulb (meaning it could kill you).  Linesmen use fiberglass poles.

     

    Victims of electrocution at this level will have a burn where they contacted the power line and their feet are blown off (or whatever part of their body was touching the ground at the time).  Not cool.

     

    Really super cool:  These are linesmen working on LIVE high tension wires (100kV-500kV).  Yes those are electrically conductive suits.  Talk about a “bird on a wire!”  http://www.chopperworx.co.za/powerline.html  A helicopter drops them off and picks them up.  http://infao5501.ag5.mpi-sb.mpg.de:8080/topx/archive?link=Wikipedia-Lip6-2/1536956.xml&style#3 for more info on live line work.

     

     

    OK, back to the blog…  In your house the power goes to your breaker panel and to the “main breaker” (the main breaker is omitted in my sketch for clarity).  The main breaker feeds the bus bars.  Opening the main breaker disconnects the electricity to the entire house.  Each bus bar carries one of the two “phases,” and where you install the individual circuit breakers determines which “phase” you get.  Installing a 240V breaker connects to both “phases” (the breaker is a “double” wide).  These circuits are used for HVAC, dryer, and range.  There is also a neutral bus bar.  The house earth ground is attached here also.  (Notice the earth ground symbol is a capital “E” turned on its side)  If you cut open an AC wire you will see three wires as shown.  All of the neutral wires (WHITE insulation) from all of the house circuits are attached to the neutral bus bar, as are all of the circuit ground wires (BARE copper wire, can also be GREEN for 240V).  The “hot” wires (BLACK for 120V, BLACK and RED for 240V) from each circuit are connected to the circuit breaker.

     

    Did you notice in the diagram how both the “N” and the “G” go back to the same bus bar in the breaker panel?  The electric current could care less whether it gets back to the bus bar on the “N” wire or the “G” wire.  The insulation prevents the current from flowing on the “G”  wire… of course it’s this ground current that we’ll focus on here.

     

    More next week…

    Posted Nov 09 2007, 04:12 PM by Andy W with 5 comment(s)
    Filed under: ,
  • AES 2007 – DSP-Based RIAA Filters (Attn: Vinyl Fans) - Part II

    For those of you waiting with bated breath (LOL), here is part II, finally…

     

    The author presents an argument that if instead of normalization after truncation (see part I of this blog for an example of truncation), the digital samples are processed by a digitally implemented inverse-RIAA filter, and then carefully matching the gain, the residual error is greatly reduced.  In fact, he claims that the magnitude of error is only on the order a few LSBs out of 24 bits, and he claims that this proves to be over four bits less than expected 6 or 7 bit loss.

     

    I have no way of knowing what filtering or algorithm is applied in DSP so I can’t comment on his findings, however today’s floating point DSPs have incredible amounts of processing power, and the accuracy of the computations involved in not in question.

     

    The other argument presented was based on the review of perhaps thousands of available vinyl recording, his own and others.  He selected around 40 recordings that were singled out based on the amount of high frequency content.  His theory is that if a recording has a high level of high frequency or treble content, the resolution of the bass frequencies would be the most impaired.

     

    This stems from the fact that when doing a digital recording (or an analog to digital conversion), you will eventually run out of headroom, and any transient that exceeds 0dBFS (full-scale) will be clipped.

     

    When doing an inverse-RIAA filter digitally it is imperative that the maximum level on the recording is known, otherwise you will either 1) clip the signal on the peaks if the level is too high or, 2) underutilize the available resolution if the level is too low.

     

    By sampling these specially chosen recordings, and using peak responding, peak-hold audio analyzer software, he discovered that the peaks were lower that what would be expected based on the RIAA curve alone (which would imply that you would have several of resolution for free).

     

    That is not strictly the case however, since the analog gain level must be set based on the peak value of the coherently summed signal of bass, mids, and treble, not just the peak value at discreet frequencies across the spectrum.  But remembering that the mids and especially the bass are de-emphasized, and by analysis of the program peak level on the selected recodings, validated the use of the peak responding, peak-hold analysis, and that with LP recordings this is unlikely to exceed 1 bit (6dB).

     

    His argument continues that RIAA equalization in the analog domain followed by digital conversion would be a reduction in fidelity since 1) the program peak level would now be defined by the recording’s bass level, 2) there would now be less resolution available for the treble, so the treble will be truncated instead of the bass, 3) humans have better hearing in the treble range, 4) treble doesn’t benefit from the reconstruction effects of the filter.

     

    So there you go…  If there are errors in getting his point across they are probably mine.

    Posted Nov 02 2007, 06:24 PM by Andy W with 2 comment(s)
    Filed under:
©2008 Klipsch Audio Technologies. All rights reserved.