Dedicated power

If one has to choose between getting a dedicated line or a good power conditioner, which one provides the biggest bang for the buck.

First determine if you have a problem. The best answer maybe neither.

We use conduit hard pipe (emt) in Chicago.

Romex has better rejection than loose wires in EMT.
Twisted L-N in conduit it best, but it's a bitch to pull.
See pp 31ff here  Microsoft PowerPoint - Indy AES 2012 Seminar w-Notes v1-0.ppt (wordpress.com)

Mr Carbon is a president of TAMC [Tweak A Minute Club]

Headroom loss for 14ga vs 10ga is less than ¼db @ ≈14A / ≈1600W
See  ieLogical CableSnakeOil AC Wiring

Music power is a fraction of that. Head banger rock @ 120BPM uses about 2 cycles peak energy of the 3600 power cycles in a minute.

Clean screwed connections, not push terminals, make more difference than doubling the wire gauge.

What most of you big gauge gurus fail to realize is that musical transients are asynchronous to the power line frequency. Low E on a bass is 41.2 Hz. The lowest common multiple with 60Hz is 6180 or every 103 line periods for bipolar supplies and every 51.5 for unipolar.

@ 120 BPM there are 30 line cycles per beat. The power supply is topped up on every line peak which almost never coincide with an audio transient.

Clean connections and no computer junk on the audio circuit solve the vast majority of power problems. Note that many wall-warts for 'audio' devices are SMPS and should not be used.

You don’t want the AC line mains voltage bouncing around in beat with the music when pushing the amp hard. An amplifier’s power supply like/needs the AC line voltage steady state.

The power line does not bounce around in tempo with the music. The filter caps supply DC current modulated by the output devices. The line charges the caps @ A/C line frequency. see ieLogical CableSnakeOil and http://192.168.1.160/assets/CblSnkOil/Signal_vs_ACLine.png and Linear Power Supply Design (sound-au.com)

For the umpty-billionth time, PS capacitors charge in sync with the line frequency, NOT THE MUSIC!!! A 15amp circuit with clean, screwed terminals can supply all the current necessary to vaporize any voice coil. The only time the full line current is used continuously is when testing on a load resistor.

If the sound collapses, the PSU is inadequate to supply the required current. Either the caps have high ESR or their capacity is too low.

@OP - If you plan to purchase dinosaur amps, plan to recap them. Otherwise your fat wire install will be for naught. Capacitor technology has advanced considerably in the last three decades.

in May 2020 invalid wrote:

I just installed a sub panel in my dedicated audio room which is a separate building from my house. I had to isolate the neutral wires in the sub panel and install a grounding rod, should I connect the grounding rod to the grounding rod on the main panel.

Audiogon Discussion Forum

Question was answered then by jea48.

Starting with the 2008 National Electric Code, the only acceptable way to wire a subpanel is with a four-wire feed. Two hots, one ground, and one neutral wire. The grounds and neutrals must be isolated.

These images show post 2008 NEC main and sub connections
https://structuretech1.com/wp-content/uploads/2019/07/E15-Service_Panel-445x500.jpg
https://structuretech1.com/wp-content/uploads/2019/07/E16-Subpanel-446x500.jpg

The purpose of a ground rod is to provide a low impedance path to earth in the event a power line is hit by lightning. All the rods in the area provide an earth mesh.

The Earth Safety is to carry no current EXCEPT IN THE EVENT OF AN ELECTRICAL FAILURE! The fault current is returned to the neutral, NOT the ground rod.

Adding a ground rod will do nothing to cure ground loop noise because the loop still exists and depending on the installation with a different, possibly higher earth impedance.

Theoretically, one ground rod with a 1-inch diameter driven in homogeneous 1,000-ohm per meter (ohm/meter) soil for one meter would yield 765 ohms. Driving it two meters into the soil would give 437 ohms. Going to three meters, however, does not give as great a change (309 ohms). One would get faster ohmic reduction and easier installation by using three rods, each one meter long, giving 230 ohms compared to that of one rod three meters long. This assumes they are spaced “greater than the sum of their lengths apart”.
see https://www.polyphaser.com/News/DownloadFile?downloadGuid=25400030-29fb-4981-862f-c24c4118b556

Adding additional rods can be detrimental depending on rod material, soil moisture and pH. Improperly done, a battery is created which will destroy one rod. Too close and the rod effectiveness is diminished. 

If lights are 'blinking' they are on the same circuit. Sockets and lamps should be wired on separate circuits. Likely there are other high current loads as well.

I've been doing this a very long time. Including professional recording studios. 12ga wire and 20A breakers for walls of amplifiers. Main monitors typically 1 x 20A circuit for 4 x 500 watts [250wpc]. Playback on STUN didn't dim the lights [on a separate circuit] or jiggle a peak responding line voltage monitor.

On a 14ga 15A circuit with 10 foot run to the socket, I currently have tri-amped speakers - total 315WPC - plus 2 subs 300W each. Speakers are ≈84dB/w/m and peak levels are +100dB. The REAL cannon in Telarc's 1812 barely jiggle the line voltage when measured with a DSO. Amplifier outputs scale to input levels, i.e. NO compression.

a forceful foot pedal hit on a bass drum.

Exactly what is shown here http://ielogical.com/assets/CblSnkOil/Signal_vs_ACLine.png
The drum transient occurs right in the middle of the negative A/C signal and the wire is supplying ZERO current to the +ve PSU filter bank and will continue to supply zero current until the line voltage rises above the filter bank voltage X the power transformer ratio. At that time a large current spike will charge the caps. However that spike is well below the line capacity.

Examine the "White Rabbit" track. The signal oscillates either side of 0v through the entire A/C line wave cycle. The filter caps are only charged when it is their turn on their half of the A/C cycle.

Rock music has about a 20db dynamic range or 100:1 power ratio. So a 500W amp is putting out a fraction of its total capability. The ONLY time anywhere the line ampacity is required is when driving a load resistor.

 Voltage, length of the wiring X 2 and the gauge of the wire used.

I covered that in Headroom Loss for 1600w on 14ga/120v | Audiogon Discussion Forum

Upping the gauge from 14 to 10 results in a 170mΩ decrease in wire resistance over 54 feet. The voltage ratio is  20 * log ( 0.987 / 0.968 ) =  0.169db relative to the unloaded line voltage!  see  ieLogical CableSnakeOil

It's my contention that improvements heard after a 10ga rewire are largely due to direct clean connections and not the wire gauge. If rewiring for audio, go direct and only use screw terminals on quality outlets.

My media room has a dozen outlets which were daisy chained with push terminal quick connections. This circuit is used for 2x 600w electric heaters. I'd estimate the length at about 80 feet. I had a 5-6v drop from the first outlet to the last when loaded by both heaters. Removing the quick connects, cleaning the wire and using the screw terminals resulted in under a volt drop when loaded.

Sunfire XTEQ:

• Input Power100-120 / 220-240 V~, 50/60Hz, 900W
• Output Power   3,000 Watt

¿Que?

2800W is way overload for a 15A breaker, moderate overload for a 20A.  It's enough to cause a breaker grief, particularly an aged one.

One thought, what are the street input connections like? I've seen some really horrible stuff in the meter and into the box.

Thinking back, these are some of the problems that caused power problems:

• squirrels nearly chewed through the bare aluminum neutral. Apparently, they love it for sharpening their teeth.
• both a recording studio and a mastering lab had issues from underground transformers that got wet over several years and eventually became unbalanced.
• pole transformer failure severely unbalanced 120v legs.
• this was a toughy: one of the load balancing transformers the DWP switched in at times was defective in combination with some others. I called DWP, but of course It’s all good here. Must be your end. I logged power and voltage for about a month and gave them specific times, loads and voltages. When they compared with their switching logs, they were able to identify the faulty unit. Once replaced, no more issues.
• etc.

Where all the services are underground, there are vaults with transformers.

In my current location we have small above ground transformers every few houses and a large vault with a larger transformer. It also got rather wet and shorted in our 101 inch rain in 2017

what you find acceptable electrical service would be considered of "third world" quality in many parts of the world

The City of Los Angeles extracts $¼ BILLION annually from DWP. Hence ZERO maintenance. The pole in my yard was installed in 1962. Ditto the phone wiring. Crackles every time it rains. Pac Bell says the problem is inside the house because there is no crackling when they show up in dry, sunny weather.

Sheesh!