Subpanel after the 5KVA transformer gives option of two 20A circuit protection, maybe three 15A circuits with a little stretch of imagination. More than that is not advised, even with a 20A 240V breaker from the main panel to the transformer. Safety first.
Generally, everything connected by analog interconnects should be on the same phase. That's not a law.
Plasma can be a large draw in itself, sometimes over 6A for the big ones. With that kind of a SMPS, probably better off not being on the same transformer. Good example of what you're isolating from.
I'm sure the electrician can figure out the subpanel.
Isolated ground? Bad idea and probably not code. |
I think he means Hospital Grade Outlet |
Ngjockey,
Thanks for the reply. Yeah, I'm confused and probably don't know what I'm doing. But I promise I won't begin installing until I figure it out. Safety IS first. I won't be doing any of this myself, and expect my electrician to follow code.
I really thought a 5kva xfmr would be large enough. I recall a thread where Zargon had a 4kva iso before a sub with six dedicated lines. Though I do recall that Lak had two 5kva isos before a sub, but that was to feed three different sound systems. My equipment isn't crazy big: ARC REF210s, a REF 3, and the plasma tv will draw 328 watts.
Regarding the "isolated ground"--I was going to have all the neutrals and ground wires in the dedicated lines run all the way back to the main service grounding point, per suggestion of Gs5556 in another's thread--although I don't know what this does for my sound, since they all hook together at some point.
I've collected a good bit of possible mis-information regarding grounds, and maybe you can help me sort it out: 1)I've heard of using a separate ground rod that is 6 feet from the service ground. 2)I've heard of using a separate ground rod, but bonding it to the service ground rod. 3)I've heard that the iso xfmr is a separate power supply, thus I can ground everything from the subpanel to the nearby water pipe in the utility room. 4)I've heard that isolated ground receptacles are not necessary because I have a wood frame house.
So what would be the best way to provide clean power to my gear? Thanks.
Chris
|
Hi Chris,
I am going through the same process with my AV power.
I have been able to get a great deal of info from not only A-gon threads but also white papers on the Jensen Transformer site (Bill Whitlock, Jensen President, has published a lot of peer-reviewed papers on grounding issues), the Ultra-K white papers from the Controlled Power site, the white papers from the Middle Atlantic Products site, the white papers from the Equi=tech site, white papers from the Ground1 site, and papers from the publications page of the Audio Systems Group site.
There is a difference between Hospital Grade Outlets (greater contact force among other benefits) and outlets with Isolated Grounds. Isolated ground duplexes separate the ground lead from the mounting strap of the duplex -- so the ground wire can be isolated from the building ground (metal boxes in the wall, other outlet grounds, etc) until the isolated ground meets up with the other grounds at the source (main panel or isolation transformer). If you check the Hubbell-Wiring site, you can check their online catalog (Hubbell Wiring Device-Kellems) and will see in section A "straight blade devices" that they make hosptal grade, industrial grade, and commercial spec grade duplexes (and explain the differences); and in section J show isolated ground versions of basic and hospital grade duplexes (plus a good explanation of isolated ground).
I believe the advice you read about isolated grounds not being needed in a wood house is because if you are already running a dedicated line (dedicated hot, neutral, ground) from the panel to your audio gear without making a stop at any other outlet or junction, then the ground is pretty much already isolated because the outlet box is nailed to a wooden stud, not a metal building frame. But I'm not an electrician, nor have I stayed in a Holiday Inn. |
Any time it's going to an outlet, instead of hard-wired, you have to accommodate the POTENTIAL, not the actual load. Who's to say you won't get 600W, class A, OTL monoblocks and then plug in some 11,000 BTU portable air conditioners and a vacuum cleaner. Also have to consider the electrical service, since you're adding another 20A without taking anything else away.
The "isolated ground system" has two ground wires, one insulated and going all the way back to the bus bar. Mixed results, mixed recommendations, and not official NEC code. Romex and armored are available as 3C ... but talk about confusing... Got any green nail polish? ;) Otherwise, conduit, $$$.
1) Wrong
2) Ok, maybe, sort of. Ask Almarg.
3) Very wrong. Can't count on copper service pipe. Removed or prohibited from code decades ago. Transformer/shield has to be grounded too.
4) When are they necessary? Google.
Off to a good start. The transformer, though unusual for residential, will provide some common mode noise reduction. Multiple transformers allows isolation between components. Naturally, "the best way" is completely impractical.
Where is this transformer going to be located? They can hum. |
DMatt,
Thanks for the info regarding the hospital grade recepticles. I have seen a number of favorite ports mentioned in threads, most often Hubbell 5362s.
Based on what I've gleaned from other threads there is still a benefit to using recepticles with an isolated ground and running the dedicated lines in armored cable. I'm undecided on which to use here. #10 BX has been recommended. I've also seen recommendations for the Belden 8300 series. Also for twisting the wires. Do they need to be twisted if they are running through metal? You'd think someone on Audiogon would have created a cookbook for all this by now.
Ngjockey,
Thanks, I get what you mean by potential draw on the circuits. And I'll simply let the electrician worry about grounding things properly.
You brought up another point though regarding the service amperage that'll go to the sub. I'm trying to get a handle on what's possible. I guess I would like a minimum of four circuits off the subpanel: two for the monos, one analog, and one digital. I've seen recommendations for a minimum 60amp breaker at the main panel to protect what is going to the sub. 1)How do I know what amperage is available to lend the sub? 2)If I can use a 60amp breaker does this mean that I can't exceed a total of 60amps in breakers on the sub?
Chris |
Sorry I forgot to add that the transformer and subpanel would be in the furnace room, which unfortunately is right behind where I'd be listening. There is a door. It shouldn't be any more audible than the furnace, should it? Anyway, it'll have to do.
Chris |
Chris,
The Controlled Power Ultra-K isolation transformers show an audible noise spec in their website brochure of 45-55dBA at 1 meter.
Take a boombox and your Radio Shack sound meter out to the furnace room and play white noise at 60 dBA from three feet back to see what it will sound like (put the tuner between stations for white noise).
My guess is it will be close to the furnace in loudness, but will be making noise all the time.
David |
RE twisting:
Last week I read a paper that favored twisting the hot and neutral together and leaving the ground separate. Pull the three wires through conduit. This has been touted as the way to reduce the induction effect of the H and N on the ground, reducing induced ground voltage that causes noise.
Bill Whitlock (Jensen Transformers) and Jamie Fox presented the paper at last November's AES convention that supported this approach with test data. Titled: "Ground Loops: The Rest of the Story", the twisted configuration offered lower ground voltage induction than Romex (where the ground is between the other conductors) and aluminum MC (metal clad with the three wires configured symmetrically inside).
The synopsis from the AES library website: "The mechanisms that enable so-called ground loops to cause well-known hum, buzz, and other audio system noise problems are well known. But what causes power-line related currents to flow in signal cables in the first place? This paper explains how magnetic induction in ordinary premises AC wiring creates the small voltage differences normally found among system ground connections, even if “isolated” or “technical” grounding is used. The theoretical basis is explored, experimental data shown, and an actual case history related. Little has been written about this “elephant in the room” topic in engineering literature and apparently none in the context of audio or video systems. It is shown that simply twisting L-N pairs in the premises wiring can profoundly reduce system noise problems."
Hope that helps. I am taking this approach on my dedicated lines (with isolated grounds) from my sub panel.
David |
David,
I'll head to Radio Shack tonight--snowstorm be damned, and give the sound meter a try. Should be interesting.
So what sort of wire are going you going to twist? Are you aware of any shielded cables that come in this configuration already? What sort of shielding is best?
Thanks again.
Chris |
I've seen recommendations for a minimum 60amp breaker at the main panel to protect what is going to the sub. 1)How do I know what amperage is available to lend the sub? 2)If I can use a 60amp breaker does this mean that I can't exceed a total of 60amps in breakers on the sub?
01-31-11: Iabirder
Iabirder, Sub panel? If you are still talking about using a 5 KVA ISO transformer the panel connected to the output of the transformer will not be a sub panel. The transformer will create a new derived AC power system. The new electrical panel becomes a main panel. Per NEC the new derived power system must become an AC grounded system. If you configure the secondary of the transformer for 120V out only, (parallel the two secondary windings), then one leg of the parallel must be connected to earth. This grounded conductor becomes the neutral conductor. The connection to earth can be done at the transformer or in the new electrical panel. I would do it at the new panel. What you will end up with is more than likely what your existing main electrical panel looks like. All of the branch circuit neutral conductors and equipment grounding conductor terminated on the same neutral/ground bar. (STAR grounding)... A bonding screw will connect the neutral/ground bar to the metal electrical panel enclosure. From this neutral/ground bar the earth ground wire connects, (called the grounding electrode conductor). Per NEC this wire Shall, must, connect to the existing grounding electrode system of the main electrical service of your house. The primary side of the ISO 5 KVA transformer and branch circuit wire feeding the transformer must be protected by an overcurrent device, breaker. 5 KVA / 240V = 20.8 FLA amps X 125% = 26 amps. Minimum primary wire size #10 awg. Breaker size, 2 pole 30 amp. If it were me the minimum wire size I would use would be #8 awg copper, breaker size 2 pole 40 amp. Secondary rating of a 5 KVA transformer. 5 KVA / 120V = 41.67 amps @ 120V. Main breaker size 45 amps. Minimum wire size from transformer to new panel. 41.67 X 125% = 52 amps. #6 awg copper minimum. I would probably use #4 awg copper. With your ARC equipment, JMHO, a 5 KVA xfmr is too small. A 7.5 KVA would be more appropriate imo..... The above info is just a brief not complete method of an electrical wiring installation. Your electrician must follow the local electrical codes of your area...... Even when no load is connected to the output of the XFMR the primary still draws idle current.... The electric meter wheel still turns 24/7..... Are you sure you want/need an ISO XFMR? Is your house power that dirty? How big is your existing house service. 100 amp? 200 amp? 400 amp? Are there spare spaces in the existing panel for new dedicated branch circuits? If not you could have a sub panel installed. If you have a sub panel installed over kill the feeder wire size feeding the panel... #4awg copper minimum. # 3 or #2 copper better yet. Jim |
Jim did the math, but if you want four circuits...
4 X 15A X 120V = 7200 VA
or
(2 X 20A X 120V) + (2 X 15A X 120V) = 8400 VA
You never know how noisy a transformer will be till it's installed and the same one could be quiet in one nearly identical setup to where it was noisy elsewhere or vice versa. I have a 5KVA and a 2KVA and both are inaudible to within inches. Work with and around a lot of xformers and, except for massive (15 KVA+) three-phase, unusual noise seems to be exception, not the rule. I've had open and regulated xformers that have been noisy. Both mine at home are shielded , encapsulated (NEMA 3R) and potted in epoxy, as typical for the type. Both cost $100 off Ebay (industrial recyclers) and cost more to ship to Canada. Both times the mounting brackets got bent in shipping but casing remained solid as a rock. A 7.5KVA of this type can weigh 140 lbs and put you on a UPS hate list. |
Ngjockey,
Just so Iabirder understands the max a 5 KVA xfmr can be loaded is 41.67 amps @ 120V (5000 va).
Technically the continuous connected load @ 120V = 80% of 41.67 amps = 33.3 amps (3996 va).
Most would not use the 80% rule for calculating the size of a xfmr for a connected load of a power amp.....
Jim |
Jim, agreed.
In my setup with 2 Plinius SA100 and the transformer derated by half for the center tap...
(2 X 500W) / 2500 VA = 40%
Don't know the max draw on Iabirder's amps but if they're A/B, it could be much more and a different characteristic.
BTW, the xformer's idle (exciting) current drops quite a bit if there is a minimum load, like "standby", as opposed to no load. |
Ngjockey,
I've been abusing the FedEx guys for some time now. I think its only fair I spread the love.
David,
60dB is what the fan on my computer case generates from an inch away, which is fairly unobtrusive. But as you suggested I did set a boombox at the same level from a meter away and shut the furnace room door. It was quite audible--possibly my hearing is better at these frequencies. And then the burner on the water heater kicked in. I will definitely be putting some sort of sound insulation on that room, transformer or not.
Jim,
Thanks greatly for all of that. You've clarified a number of items. But there are a few things I'm not sure about yet. If I understand you correctly there will be three wires running from the secondary winding to the isolated main panel. One leg will be the hot at 120V to feed the breakers. The second (center-tap?) will go to the neutral bar. The third at the other end of the winding will be attached to ground on the panel. And this will necessarily put all circuits in the isolated panel on the same phase. And the neutral bar, ground bar, and panel will all be connected and grounded back at the main service ground. Is this correct?
And if my math is right, a 7.5kva transformer would accomodate a load of 62.5 amps@120V, just enough for four 15 amp circuits. And I would need a 10kva transformer to accomodate a load of 83.3 amps@120V, enough for four 20 amp circuits. Is there any wiggle room here?
You raised the issue of cost. Is having a transformer hooked up 24/7 like leaving my electric range on all the time? Well probably not that expensive. But 10kw x 24hours x 365days x 8.38cents/kwh=$7,340.88 per year to keep it hooked up. Is this the way it works? That would be prohibitive.
As far as whether I need a transformer in the first place, I honestly don't know. How would I know whether my power is dirty? I thought that perhaps I should also be concerned with noise in my ciruits from everything else on my service. I've got fluorescent lights, dimmer switches, computers, a plasma, etc.
I've got a 200amp service to the house and three unused spaces on the panel. Could you please clarify your last comment regarding overkilling the feeder wire size--the feeder going to the subpanel?
Chris |
If I understand you correctly there will be three wires running from the secondary winding to the isolated main panel. One leg will be the hot at 120V to feed the breakers. The second (center-tap?) will go to the neutral bar. The third at the other end of the winding will be attached to ground on the panel.
If you configure the secondary of the xfmr 120/240V out and only use one leg to neutral you will only be using one of the two secondary windings of the xfmr.... Usable va rating is cut in half. If you are only needing 120V then configure the output for 120V out only.... The two secondary windings are wired in parallel.... 2500 va x 2 = 5000 va. Note: Winding polarity is very important. If wired incorrectly the two windings will buck one another. Also polarity should be observed for the paralleled leg that will be grounded. So the two secondary windings will be paralleled and (2) lines, legs, from the xfmr will extend to the new electrical panel. One leg will terminate on the neutral/ground bar, (becomes the neutral the grounded conductor). The other leg terminates on a single pole 45 amp main breaker, (the hot ungrounded conductor). The electrical panel is 120V only. And the neutral bar, ground bar, and panel will all be connected and grounded back at the main service ground. Is this correct?
Correct. And if my math is right, a 7.5kva transformer would accommodate a load of 62.5 amps@120V,
Correct x 80% just enough for four 15 amp circuits. And I would need a 10kva transformer to accomodate a load of 83.3 amps@120V, enough for four 20 amp circuits. Is there any wiggle room here?
The main breaker on the new panel would limit the amount of load connected. You could fill the panel with 15 or 20 amp breakers. The main breaker could care less... Add up all the "breaker handle" breakers in your main house panel. You will find they may well add up to over 200 amps. You raised the issue of cost. Is having a transformer hooked up 24/7 like leaving my electric range on all the time? Well probably not that expensive. But 10kw x 24hours x 365days x 8.38cents/kwh=$7,340.88 per year to keep it hooked up. Is this the way it works? That would be prohibitive.
LOL, I coresponded by email with a guy that hooked up a 7.5 KVA xfmr for his audio equipment. After a couple of electric bills his wife made him shut the xfmr off when not in use. I've got a 200amp service to the house and three unused spaces on the panel. Could you please clarify your last comment regarding overkilling the feeder wire size--the feeder going to the subpanel?
Is your audio equipment now fed from any dedicated circuits? If not I would try that first. Have a sub panel installed in the electrical/mechanical room. That is where you were going to install the iso xfmr and new panel. Feed the sub panel from your existing main house panel. The electrician can move a couple of breakers near the top just below the 200 amp main breaker to accomidate a new 2 pole breaker that will feed the new sub panel. Wire size to feed the sub panel? At least #4 copper. Feed breaker 2 pole 70 amp. Better yet #2 copper. Feed breaker could still be a 70 or up to a 90 amp. Note: I would wire the sub panel 120/240V for future and house resale. It also allows you to use the Line, leg, (L1 or L2) that yeilds the best sound from your audio system. No main breaker will be needed in the sub panel. Copper busing only for the panel.... stay away from aluminum. Example: Square D QO is plated copper... Square D "Home Line" is aluminum. |
Jim,
Actually I haven't been able to hook up any of my gear. We moved into a new house and after 18 months of remodeling, we're just now to the point when I can begin to put "my space" together.
My wife will certainly appreciate your subpanel suggestions, especially the resale benefit. She doesn't understand any of this and I think a transformer might scare her--even more than all the non-integrated components. And that wouldn't do anything for the music.
I'm not going to pretend to understand everything about the transformer hookup. I find all this difficult without diagrams. I am going to look for some and review them along with your explanation. My primary reason for starting this thread is that there are so many kinds of transformers, that I didn't really know what I should be shopping for. I figured if I understood how it all needed to be hooked up, then I would be able to shop halfway intelligently.
I can relate to your guy with the high electric bill, but I'm not sure I figured the cost correctly. I did some reading today and I came across a formula for excitation current in the primary winding, which would depend on the inductance of the transformer. I=E/(2x3.14xL) The current in the primary winding would then increase only in response to a load placed on the secondary winding. So it wouldn't be pulling current full bore anytime it was hooked up. Without a draw on the secondary, only an excitation current would be drawn on the primary. Does this sound correct? What would be a more realistic cost estimate for allowing a transformer to hum away?
I've got just a few more questions regarding breaker sizes. My understanding is that if I'm needing 20 amps on a circuit, then I will need a 20 amp breaker at the subpanel, because a 15 amp breaker would trip. And a circuit protected by a 20 amp breaker would require #12AWG or larger, because 20 amps would overheat anything smaller. Is this correct?
If I have 80 amps worth of breakers on the subpanel, would I need to have at least an 80 amp breaker protecting the feeder on the main panel? And then the feeder size would be based on whatever wire can handle 80 amps? Can I have an even larger breaker on the main panel as long as I increase the size of my feeder appropriately?
If I were to have an isolation transformer with a new main panel, what would determine the size of the new main breaker? If the rating of the isolation transformer wasn't large, realisticly wouldn't the limiting factor for the amps on the new main panel be the breaker protecting the isolation transformer?
Thanks again.
Chris |
To give you an idea of the running cost.
On my 5KVA, in sequence of testing tonight...
With amps in A mode, no music playing
primary (240V) 3.3-3.5A, secondary (120V) 6.3-6.7A (approx 95% eff)
With amps in A/B mode, no music playing
primary 0.9A , secondary 0.9A
amps switched off
primary 0.6A (144W - got light bulbs more than that)
So, either I was wrong with my earlier statement or conditions were different on previous test. Your results may vary.
Bigger transformers have less of an exciting current as a % of the total KVA but will always be more than a smaller xformer.
==========
Are you ready for this? Does your electrician have a heart condition?
http://mysite.verizon.net/vze22yzp/id10.html
Some of the links are broken or misplaced but not completely dead yet.
http://www.audioasylum.com/audio/tweaks/messages/55.html
http://www.contractorsav.com/article/2006/11/07/ac-power
Maybe slightly biased
My experience has been that balanced AC has little benefit for power amps but has been occasionally dramatic with sources and preamps. |
My understanding is that if I'm needing 20 amps on a circuit, then I will need a 20 amp breaker at the subpanel, because a 15 amp breaker would trip. And a circuit protected by a 20 amp breaker would require #12AWG or larger, because 20 amps would overheat anything smaller. Is this correct?
02-01-11: Iabirder
More than likely your ARC power amp/s requires it to be connected to a 20 amp branch circuit. Not because it draws anywhere near 2400 va (20 amps x 120V = 2400 va), but rather because of the amount of inrush current the amp draws when it is first turned on. Read the owners manual for the Amp. So why use 20 amp branch circuits for other equipment instead of 15 amp? 15 amp would be more than enough. If VD, (voltage drop), is an issue #12 or #10 wire could be used and terminated a on 15 amp breaker. The breaker determines the size of the branch circuit. First a little back ground info. In most cases, if not all today, the guts inside a good quality 15 amp receptacle are the same as a 20 amp receptacle. The only difference is the face plate. A 15 amp recept will only accept a 15 amp male plug. It's a safety thing... Most 15 amp branch circuits use #14 awg wire, bare minimum per NEC, which is rated for 15 amps. NEMA /UL /CSA requires manufactures that use a cord and plug for an appliance/equipment cannot exceed 12 amps FLA continuous load if a 15 amp plug is used. Per NEC code 2 or more 15 amp receptacles can be installed on a 20 amp branch circuit. ( a duplex is two recepts.) Per NEC code a 20 amp receptacle can only be installed on a 20 amp branch circuit. So if you want to install Oyaide R1 recepts, per NEC code, the branch circuit must be a 20 amp.... Per code it cannot be installed on a 15 amp branch circuit. If a manufacture builds a piece of equipment with a cord and plug that draws more than 12 amps continuous FLA then he must use a NEMA 5-20P plug. A 20 amp plug will not plug into a 15 amp recept. The max allowable continuous FLA for a 5-20P plug is 16 amps, 1920 va. If I have 80 amps worth of breakers on the subpanel, would I need to have at least an 80 amp breaker protecting the feeder on the main panel? And then the feeder size would be based on whatever wire can handle 80 amps? Can I have an even larger breaker on the main panel as long as I increase the size of my feeder appropriately?
You could have 200 amps worth of breakers in a panel. The main breaker a head of the branch circuit breakers is the gate keeper.... If the main breaker has a handle rating of 60 amps any continuous load of 60 amps and over will/should trip the breaker open. Technically, per NEC, the max continuous load placed on the breaker should not exceed 80%. Same for a branch circuit breaker, 80%. In a residential environment most convenience outlet branch circuits have very little load if any on them. In a bedroom maybe a clock radio, table lamps, TV.... A 15 amp branch circuit with maybe an amp or two of load at any given time. Family room, living room, den, 15 amp branch circuit/s with hardly any load. Several 15 amp branch circuits breakers in the panel but hardly any load at any given time of the day. Night time add a few lights. 20 amp branch circuits? Kitchen has a few required by NEC. Microwave dish washer refrigerator garbage disposal GFI recepts above the counter tops for what ever.... Most of the time during the day they just sit idle. But when needed the branch circuit must be able to handle the load placed upon it. Laundry room. 20 amp branch circuit for the washer 30 amp 240V branch circuit for the cloths dryer. The list goes on and on. Lots of branch circuits and breakers. At any given time of the day hardly any load if any. I believe you were talking about installing 4 new dedicated circuits for your audio/video equipment. Other than the inrush current of the ARC power amp/s a 15 amp circuit would more than likely handle all the connected load. More than likely a 20 amp branch circuit would handle the inrush current of the ARC amp/s plus the rest of the equipment. So why do you want 4 dedicated circuits? Why do some guys have 10? Here is just one post on the subject. There are many out there... http://www.audioasylum.com/audio/tweaks/messages/16/167770.html |
Ngjockey,
It sounds like maybe having the transformer on isn't as costly as we feared. Also that formula I posted for excitation current wasn't correct--I forgot to include the frequency in Hz. so I=E/(2 x 3.14 x Hz x L) where I is the excitation current in Amps, E is the voltage, 3.14 is pi, and L is the inductance of the transformer in Henries.
Thanks for those links. I hadn't yet looked into balanced power. Looks like I'll be busy.
Jim,
I don't have any problem with two of the circuits being 15 amp. I think four 20 amps circuits just made discussing the hypotheticals easier. Thanks for the info on the plugs and outlets. That would have probably tripped me up at some point. I read the thread you linked. And I printed off the Hansen paper--can't wait to read it.
In case anyone else is interested, here is the link for it: The AC Power Line and Audio Equipment, by Charles Hansen
http://www.conceptorg.com/techlibrary/PowerTechnology/AC_and_Audio.pdf
I appreciate greatly all the help you've both provided. I'll check back in to see if anyone adds anything more. But I'm now going to retreat to my reading and my drawing board.
Chris |
