Dedicated Power Line Project


I’m going to be adding a dedicated AC line to feed my audio system. Currently, the system is fed by a 75 foot run of 14 gauge romex that is likely daisy chained a few times in the wall between outlets. Needless to say when I turn the volume up, dynamics suffer and the overall sound quality is much less than I know my system is capable of. At louder volumes, bass notes will make the lamps dim and if I go louder, it seems like the system struggles and the bass notes get rounded off or sound “fuzzy.” My calculations tell me that with 75 feet of 14g romex, the voltage at the end of the line could be dropping by 5-10 volts, maybe more, during transient peaks. What I think is happening is that the power amps’ power supply caps are draining and there isn’t enough current to recharge them quickly enough and the amps start to draw directly from the AC line which causes a large voltage drop. 

Initially I was looking at a 10 gauge or 12 gauge line with a 20 amp breaker. After running the numbers, even a 10 gauge run could see some significant voltage sag at 75 feet during transients. At that point I considered 8 gauge, but the problem with that is that no electrical outlets will accept anything larger than 10 gauge, so I’ll have to do something fancy like place a junction box and step down to 10 gauge at the breaker panel and just before the outlet. Then I thought if I’m going to that much trouble with an 8 gauge, why not run 6 gauge? My calcs tell me 6 gauge could have a volt or two less variance than an 8 gauge and it wouldn’t be much more cost. So as of now, the plan is to run a 6/2 solid core romex line and run all the equipment of that. (I could run another line for the front end in the future but that isn’t the plan right now).  I have a few questions if anyone would be so kind to reply:

1) Are there any foreseeable issues with running “too large” a power line? I’m not sure if it’s accurate, but I’ve heard that large gauge is good for bass and small gauge is good for treble. Since I’ll be running everything (including the low current draw components like sources, preamps, etc) off this one line, is there any chance I’ll change the sound signature in a negative way?

2) What is better? In-wall romex or THHN inside a grounded conduit? I’m leaning towards romex because my gut tells me having the wire surrounded by something could hinder transients or hurt clarity but I’m not sure that’s accurate - that may just be from my personal experience with low level signal cables where too much insulation can be a bad thing. 
3) Are there any other “gotchas” that you can think of?

I’ve read the MSB Technology article and I’m subscribing to that theory which is this in a nutshell:

“It is the subject and goal of this paper. The gauge of the wire is FAR MORE IMPORTANT than the fact that the line is “dedicated”. The subject of this paper works on the theory that the varying musical demands of your amplifier are actually modulating the incoming power line, divorced from the utility (power company) by some resistance (12 or 14 gauge wall wiring at some length from the breaker panel has too much resistance for audio purposes).”

Thank you and wish me luck!
128x128mkgus
Been there, done that. Almost exact same distance and scenario. Originally, wired to code, the system circuit daisy chained like yours with my system near the end. This was replaced first with a direct run of 12 ga. Then this was replaced with a direct run of 4 gauge 120v. This was then upgraded with the 4 ga going 240v to a step down transformer just below the listening room. Very high quality audiophile grade silver windings transformer. Then this was all pulled out, cryo'd, and put back in.

A lot of people have ideas, and theories, and write papers and posts pontificating on their ideas and theories. Writing is a whole lot easier than going out and buying wire, running circuits, hooking everything up, playing music, and hearing what actually happens in the real world. As opposed to, you know, in your imagination. Which is what I predict you will get from the other dozen or so responses this will surely generate.

So, setting theory aside for a moment and going with what actually happens in the real world, the main thing you want to do is avoid connections. Your paper I'm sorry to say is BS. The biggest improvement by far is in eliminating connections. Run your big wire if you want, but realize I went a lot bigger than you did and even that didn't make as big a difference as going direct.

My current setup is so direct the "fuse" is 1" of pure silver wire soldered at each end to the 4 ga and bus bar. This is what I would do if I was serious, which I am. But a lot of guys only pretend to be, and almost all the rest are so scary afraid of voltage and understand electricity so poorly they would have you believe I have burned down my whole neighborhood five times over by now. If not brought the whole PNW power grid to oblivion. Instead of everything working just fine for like 15 years now. If you're serious let me know, I can PM you exact details. So as not to shock the Nervous Nancy's, you know.

This is all not really a lot of money, at least not if you DIY, but it is a lot of work. Would be nice to know just what you can expect from such a project. Well, here you go:

You'll get the improvement in speed and slam and pitch definition you are looking for. Most of this however is from the reduced RFI and other noise eliminated by going direct. Based on my ears, and I have no idea what a volt meter will show, but I do know what sounds better. Cryo'd large gauge wire run direct and with a step down transformer is the bomb. HOWEVER, all that is only about as much improvement as one Synergistic Research Blue Quantum Fuse.

Yeah. No kidding. Let that one sink in a while.

And I hear the Orange is a lot better than the Blue.

Of course being serious you want to do both. All. But big wire is a headache. Thick and stiff it has to be wrestled around and is especially hard to connect, unless you either MacGuyver it like me or go with the junction boxes, which loses with connections a lot of what you gained with the gauge. So to me it just makes sense to run direct whatever you're willing to work with, and not worry too much about it. Anything direct, even 10 ga, will be a whole lot better than what you have. 6 or 4 ga will not be a whole lot better- even if you go the Full Monty like I did. 

Especially not when you consider the improvement difference your thicker wire will get you is only a fraction of say one SR fuse. Which I bet will cost a fair bit less- and be a whole lot easier to install.

Oh, one more thing- ground. You can try a dedicated ground rod if you want. But ground is tricky. You may be better off with existing normal to code ground. Or not. So many wrinkles its hard to say for sure. Something to keep in mind, as they say, just in case.






mkgus
... I was looking at a 10 gauge or 12 gauge line with a 20 amp breaker. After running the numbers, even a 10 gauge run could see some significant voltage sag at 75 feet during transients. At that point I considered 8 gauge, but the problem with that is that no electrical outlets will accept anything larger than 10 gauge, so I’ll have to do something fancy like place a junction box and step down to 10 gauge at the breaker panel and just before the outlet ...
That is one solution. Another is to use a pigtail inside the outlet box. Either will work. I have found great benefit from using derated lines. Given that you now see your lights dimming on transient peaks, I’m pretty sure you will, too.

While you’re at it, I also suggest you consider multiple dedicated lines, unless you have a very simple low power system. There’s no better time to do it and you’ll never regret having the capacity.

millercarbon
You can try a dedicated ground rod if you want. But ground is tricky. You may be better off with existing normal to code ground.
It’s likely that you’ll hear from others about the potential benefits of "dedicated" grounds. It’s best to ignore them. Per NEC, all grounds must but be bonded together at the service panel neutral. No exceptions. Where you might get improvement - depending on what you have now - is checking the connections on your existing ground rods and consider adding new rods to the system.

On that topic, I also suggest that you get a permit to install these lines, unless they are absolutely not required in your jurisdiction. Your electrical inspector is there to protect safety and if you ever sell your house, the buyer will likely want to know that the work was done to code.

To satisfy @millercarbon’s concerns, I’ve actually already done this work myself, DIY. I have a biamplified system with separate lines for each amp, then also separate lines for analog and digital. It made a huge difference in my system.
Thank you for your help, Miller and Cleeds! I’ve had the same thought about eliminating as many connections as possible and trying to avoid any junction boxes and pigtails. Maybe I can find a specialty outlet or make an electrically safe modification so I can avoid connections. Surely copper atoms chemically bonded to each other (pure wire) is better than any mechanical connection. I will likely start with one dedicated line and go from there with additional lines in the future if that makes sense. Fortunately the path I’ll take is mostly unfinished or in-wall spaces so I can add future lines easily. I’m curious to know the benefits of dedicated lines for each piece of equipment assuming the gauge is large enough so voltage drop is irrelevant. I’m thinking it has to do with each piece of equipment injecting noise into the AC wires so that if they’re on their own lines they won’t “talk” to each other.  
You mean dimming the lights is not a feature? :-)

Your main feed (200A?) is likely 2/0, so even if it is 500 feet to the pole, that 75 feet of 14A is doing way more damage. Some amplifiers have excellent power supply rejection and can tolerate this, some don’t. 6V or so at 15A, but that is only the fuse rating. Short term peaks could be higher depending on how far the capacitors have drooped and the saturation limit of the power supply transformer.

There are no issues with pigtails (connections always made in electrical boxes to meet code), as long as the connection is made properly. If you want a perfect connection, use a proper compression splice with a proper crimp tool (and make sure the conductors are clean of corrosion). Don’t use screw lugs and definitely don’t solder them together. A properly implemented compression crimp connection is pretty much like a continuous wire. Any impact of this connection will be orders of magnitude smaller than the connection of the power cord to the AC socket and to your equipment. The ground should connect direct to the junction box and not pigtailed to a thinner gauge. That is frowned upon.

If you do run multiple AC lines, I would encouraging terminating them at the receptacles in the same junction electrical box so that all the receptacles have the same ground connection and you don’t end up with any unintentional ground loops.  As Cleed pointed out, the easiest time to do it is now and it sounds like you already have the 14 awg run, so unless you were planning to use that as a snake, just add the additional wire, even if you leave the 14 unconnected for now.

Running multiple AC lines is akin to bi-amping (not biwiring) speakers. Bi-amping allows you to prevent stresses in one range of frequencies (bass) from generating distortion in the higher frequencies, most prevalent obviously when run hard. Bi-wiring has limited (if any) value as the cross-over splits up the frequencies anyway. Running a separate AC line for power amps can prevent modulation of the AC line from impacting the more sensitive low power equipment. You are already accomplishing much of that by running heavy gauge and most low power equipment has better power supply "noise" rejection.

Interesting comment about large gauge being good for bass, and small gauge for treble. There is, unfortunately, with the way most amplifiers are designed, the potential for this. Almost as a rule amplifiers have good power supply noise rejection at low frequency (within limits as you have found out), and poor(er) power supply noise rejection at higher frequencies. Some low feedback designs can have pretty poor rejection at bass frequencies too. Reducing the resistance in the AC line can improve your bass, but it also means faster transitioning of the diodes (at higher peak currents) in the power supply which means more high frequency noise, something the amplifier doesn’t deal with as well.


Thank you for your reply. Very helpful!

Today I flipped off the breaker and started testing outlets to figure out everything that’s on the existing line. It’s worse than I thought. Included is at least 2 bedrooms and 1 bathroom which includes roughly 11 outlets, 3 light fixtures, a fan, and a bunch of switches. And who knows how many low quality 14 gauge daisy chain connections the power goes through before reaching the amps and other gear. No wonder I’m not ecstatic with my sound!

The plan as of now is to run 2 dedicated lines, a 6 gauge and a 10 gauge, and terminate them in the same junction box so they can share a ground. The 6 gauge will be for the power amps and the 10 gauge for the front end. 
Millercarbon, if you would be ever so kind, please PM me the details of your power supply system. I probably won’t forgo a breaker at the moment, but I’m sure I can learn a lot from your large gauge setup if you’re . Thank you!