|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
63.230.51.177
Bob, John Curl said in a reponse to this question to ask you. When you design and build a piece of equipment is there a set rule on the polarity of the power transformer? That is on the primary side of the transformer are the leads such that the neutral is always on an identified lead and the hot on an identified lead on the transformer. I ask because I have read old archives on Agon where some have talked about reversing the AC polarity feed to the unit, switch the hot and the neutral. Does it affect the sound as some have said? Also does the fact that the neutral is at ground potential have any bearing on the design of a piece of equipment?
Follow Ups:
...by Neal Levenson and Enid Lumley -- and somewhat later, by myself.See how long it takes to absorb the news?
The same is true with Absolute Polarity.
At least back then we didn't have people like bjh, Austin PhD and Rick W jeering at us. Who knows how long people like that will set back the advancement of audio?
First I do agree that having every component plugged in at the same AC polarity is vital...assuming it is a single phase system. By that I mean there is either 120VAC and 0VAC or 240VAC and 0VAC measured between phase and neutral and ground.But I have recently, due to a move, rewired my listening room so it is 2 x 120 VAC or balanced 240VAC. That is two 120VAC phases comprising 240VAC, which is possible in every household if one wants to. In this configuration I have NOT been able to determine any difference of orientation of the AC plug. Natch, how could it since the balanced lines are the same no matter how the unit under test is plugged in?
I can also testify that I have never had such a low noise and high contrast/dynamic system as now. Everything is the same except how 240VAC is made.
Well, maybe not in rural areas...The important criterion is to make sure all units are plugged *properly* into ONE LEG ONLY of the split. Then you should hear the effect.
I am not using 120VAC derived from one of the two 120V phases of the 240V service entrance. I am using 240V derived from both legs of the service. That is 120V on each leg, there is no 0V neutral leg as in the case of a usual 120V or European 240V service where you phase and neutral, i.e. 0volts.In the case of the "balanced" 240V configuration there is no proper plug orientation.
Hi Garth,Your statements seem to raise more questions than info.
For instance: Your equipment operates at 240 volts?
Do you not realize, that dwellings with a
single phase service, do not possess more
than one phase?
I said I wired a room for 240VAC using two 120VAC phases which nearly every home in NA has. Did you think your oven or AC runs on 120VAC?And of course my sysem is 240VAC, that's why I did it.
The point being in a two phase or balanced system there is no preferred orientation of the power plugs. The potential to ground is alwazs the same.
I think the problem was not that you are using 240V to power your equipment, just in your terminology. Technically it is called "single phase 240 Volt". The utility transformer that feeds your home has a secondary winding with a center tap. The center tap is bonded to ground and becomes the grounded conductor, the neutral. Thus 120/240V single phase.I am curious how this does indeed effect the sound quality of the equipment. As you said with using 240V USA power there is no polarity. The equipment was still designed for 120/240 volt aplications where as one line is at ground potential.
Bob Crump, if you are still following this Thread could you chime in here please...Does it make a difference?
Hi!In Europe we have single phase 240. One "leg" is 240V, called phase and the other is neutral. If you measure between phase and neutral you get 240V. if you measure between neutral and ground you get basically 0V but is some potential of generally a few volts. And If yu measure between phase and ground you also get 240V. That is a single phase system.
In NA where you have 120V in a normal home it is also single phase. When measuring between the legs as above you have 120V and 0V at various places.
I took the other 120V phase in the service box and wired a room with two phase 240V. One leg has 120V and the other has 120V. Measured between them you get 240V. And since they both also measure 120V to ground there are two phase which are basically identical. So now there is no phase orientation because there is no "neutral leg".
There are some commericial power conditioners, Accuphase and Burmester come to mind, that do just this.
I can't say how much better it sounds than if I had just used 120V 'cause my stuff is 240V but it sounds much better than single phase 240V. it's a neat trick that I feel would improve the performance of any NA based system assuming the equipment could be switched to 240VAC which some will.
Two AC current carrying conductors is called single phase. You can have 120V single phase, 240V single phase. In commercial and industrial applications, just to give you one example, you can have a panel board that is feed from a power transformer that is 120/208V 3 phase 4 wire wye connected. The center point of the wye being bonded to ground at the first point of attachment or disconnecting means, this being the neutral.. If I have a piece of equipment that needs only 208V "single phase" I install a 2pole breaker in the 3ph 4w panel board, and pull three wires. 2hots, 2 conductors connected to the 2pole breaker, + 1 equipment grd. This application is called single phase 208V not two phase 208V. Granted, the 2 pole breaker is installed across two phases of the 3ph 4w panel panel board. The branch circuit is still called "single phase". Forget about the neutral it has nothing to do with the terminology.Now back to the point.. If the designer of audio equipment takes into account the fact that he has a utility power system with one grounded current carrying conductor, and he does observe the primary AC input power to the equipment power transformer. In your case you do not have one grounded current carrying conductor. If as you say your equipment sounds better then why doesn't Hi-End manufactures redesign their equipment. Two things that come to mind. A 2pole power switch to break both incomming hot AC lines. And fuse protection on each of the 2hot AC power lines.
one voltage line and one return essentially at zero potential, right?I have in this case two 120V lines, therefore two phases.
As far as equipment manufacturers they have no control over how their equipment is going to be connected. Simply putting a fuse on both phase and neutral doesn't quite get you there.
NO actually you have two ungrounded conductors with a difference of potential of 240 volts between them. The utility power transformer has a secondary winding with a center tap. NEC code requires this common center tap entrance conductor to be bonded to grd at the first point of attachment or disconnecting means, that being you main electrical service panel of your house, this grounded conductor is called the neutral. Yes if measured you will have a difference of potential of 120V between L1 to neutral and L2 to neutral. Check the internet, see if you can find a 2phase primary to 2phase secondary transformer. You will find single phase primary/single phase secondary transformers.In the USA you are right the manufacture cannot control how the buyer of his equipment will connect the equipment to a power system. But before he can sell the piece of equipment he will need it UL listed. If his piece of equipment is designed to be connected to a single phase branch circuit source, first the cord cap must meet certain standards depending on the voltage and amperage of the equipment. If the equipment is designed to be connected to 240V he must also break both ungrounded conductors with a 2pole switch simultaneously. If he chooses to add fuse protection he then must install fusing on both ungrounded hot conductors in his equipment.
The effects of "winding polarity" on capacitive coupling between windings and winding to core is discussed in some very old transformer books. However, they use non-audio terms (effective capacitance, capacitive coupling, pulse overshoot etc.) to describe the effects.A case of engineers and audiophiles agreeing, but not realizing they are talking about the same issue. ;-]
Play safe and play longer! Don't be an "OUCH!" casualty.
Unplug it, discharge it and measure it (twice) before you touch it.. . .Oh!. . .Remember: Modifying things voids their warrantee.
a
Its this kind of unintentionally funny post from you that I occasionally comment on. Disagreeing with either the content of one of your posts, the pompous manner you often display, or your penchant for playing the victim is now in your mind tantamount to "jeers" stifling the advancement of audio. I suspect the advancement of audio will continue despite the chuckles some of your posts engender.
Clark I ran across this using Goggle, is this part of what you were saying in your post?
Good thing it wasn't discovered by Austin PhD, bjh, Rick W and their *lk.Power cords and AC polarity for turntables? The jeering would have never stopped.
Hey! You're a good googler.
Thanks, I also found some of your reviews on Google I found them very informative.I have an old AR ES-1 TT with a Rega 300 Arm. I am going to try and see if I can hear a difference in sound reversing the AC polarity. That one does puzzle me. But Hay, It does'nt cost me a dime to try it out...
Jeer, jeer, jeer.Clark..do you have more info on the origional articles?
Yes this is very important or the bass and focus will be affected...........Put your red lead of a multimeter on chassis ground and the black lead on a decent earth ground, set the meter to ACV and measure the ground potential with the unit plugged in one way and then 180 degrees off (cheater plug) and make ACV readings each way......Gear being tested should be out of the system or readings will get confused.......The lowest reading will be the correct way of plugging in the unit.......Most transformers are marked for the hot side, but best to test them........
Bob, then if I understand you correctly the manufacture of power transformers are supposed to make sure the identification of the primary and secondary windings are correct. Is this something that is also checked by the manufacture before he wires the transformer into the audio equipment? Is this the norm for Hi-End manufactures?
Jim, like it or not, we have to test the transformers, as to how to wire the PRIMARY. NOT secondary.
Did my system, great results. And it's free! You should have told us sooner...
If you would stop calling me and others names I certainly would feel like posting more and, yes the things that make the biggest differences are normally free and are things that decent dealers used to do as part of a purchase of a new piece of gear.......If you didn't realize it Curl is my business partner and Johnsen has been one of my friends as well as one of my dealers for decades....
John Marks maybe, but you, never... What's going on here?
have been able to ascertain, I appologize, but another really good reason for you to register...................
nt
I wasn't around to see what happened. When I downloaded the site, I saw what should probbaly be called the minor aftermath. Thank you for being so observant. Reading is a skill...Why I don't register? I don't because I'm presently feeling rebelious. On a daily basis I see our civil rights eroding out of what I consider uncalled for "terror alerts" created to keep us in fear. Fear causes conformity. So, I simply don't feel like being intimidated or conforming when I'm not a part of the problem. Why should I allow some jerk to dictate my actions, especially at a place like this when my decision not to conform will have no disasterous results or cause someone's death?
I guess I'm just not in the mood for any more Big Brother this month...
a dainty man: ""Thank you for being so observant. Reading is a skill...""And, someday perhaps, I will be skilled in reading...:-)
Actually, when it first posted, I used the comment link at the bottom to notify a moderator that it should be removed..it promptly was.
The actual namecalling post was inconsistent with your previous posts, which made me suspicious...so I looked at the two different monikers, and then compared IP's, which were diff..
So, backed you up when Bob mistook the other poster for you..
"Johnsen has been one of my friends as well as one of my dealers for decades..."
You refer to name calling....I think the IP address to the deleted post was different..I don't think it was the same person.
here........
nt
What you are measuring is the coupling between the inductive current in the primary of the transformer and the ground loop that would be formed by external equipment..An excellent thing to do, and I also recommend it. This procedure will reduce the sensitivity of the amp to ground loop hum, to high current draw haversines, rectifier selection..and yes, line cord geometry..
The limitation of your procedure is that you do it only for 60 hz excitation. Since the coupling is proportional to the frequency, and the coupling to the audio signal is proportional to the frequency squared..running a higher freq current through the transformer will be more sensitive..If you do that, you will find quite a few other things you have to do to really minimize the effect.
This of course, only works for the source of the error..much needs to be done to the receiver, which is the amp gain stages..
My offer to you and John still stands..
I've found the "polarity" of transformers to be an issue because of parasitic capacitances. The physical location and mounting rotation of the transformer is usually an issue because of inductive coupling.There is capacitance from each wire on the transformer to every other wire on the transformer. Plus there is capacitance from each wire to both the core and the circuitry around the transformer. Between some leads, the parasitic capacitance is large (10nF in a large power transformer). Between other leads, this capacitance is small (1 pF or less).
What you want to do is to attach your virtual ground (the lead with the lowest voltage noise/ voltage swing) to the lead that has the most capacitance to the other leads and chassis. The other way to say this, is you hook the primary up so the induced voltage on the other leads and core is the smallest.
When testing for polarity issues, test with only one transformer powered at a time. If the transformer isn't in a chassis or circuit, I normally test voltage from both the core to safety ground and from any center taps (CTs) to safety ground.
____Hook the leads to the test points before you apply power.
____Put wire nuts on any disconnected or dangling leads. Shorting the leads of a secondary can when testing could damage the transformer, set your test setup on fire or other nasty things.
____Apply power through a fuse and an outlet with a GFCI.
____Apply power and take the measurement. Turn the power off before touching the circuit under test.When finished testing, hook the primaries up so that all of them generated the lowest voltage measured to safety ground.
If you have many transformers, occasionally it is worthwhile to run a second pass with all transformers powered in the final circuit, but where you change the primary polarity one at a time again to see if there is a reduction in the output noise of the amp. This time listen to the amp to make the decision.
If the transformer lacks end bells or is toroidal, you may have to add electrostatic shielding between the transformer and your circuitry to avoid capacitive coupling. This can be as simple as a pair of L-brackets holding a metal wall between the transformer and the circuit. I've even made this wall from a coffee can. If you can stick a kitchen magnet to your shield, you'll get some magnetic shielding too.
On power transformers, usually only the primary polarity matters. The only exception I can think of is a secondary that drives the filament of a DHT. In this case, the polarity to each filament can sometimes be arranged for hum cancellation.
For single ended output transformers, both the primary polarity and secondary polarity matter. I usually use a 1 kHz test generator to figure this out in a manner similar to what was used for the power transformer.
After the voltage polarities are figured out, then play with physical part placement and physical part rotation to minimize both inductive coupling and occasionally newly found capacitive coupling.
That's all for now.
Play safe and play longer! Don't be an "OUCH!" casualty.
Unplug it, discharge it and measure it (twice) before you touch it.. . .Oh!. . .Remember: Modifying things voids their warrantee.
Vont second: ""I've found the "polarity" of transformers to be an issue because of parasitic capacitances. The physical location and mounting rotation of the transformer is usually an issue because of inductive coupling.""
It's obvious from you moniker that you are a switchmode guy..(guessing, of course).
For your app, I certainly agree with all you have stated, and it may indeed have a part of what is being discussed..
However, I am confident that this discussion is basically induction..induction from the line cord and the primary feed wires, and how the xfmr inductive current is coupling to the line cord ground.
Note how Bob talks about bass and clarity being affected..this is a result of analog ground bounce, and I don't believe the capacitance of the xfmr is enough to affect the lower octaves...
I do take issue with using the cheater plug, however..while the safety ground can be maintained with an adaptor, I would prefer for safety, the hot always be the switched one, reversal at the outlet will bypas that if the switch is single pole..
Cheers, John
I agree. Just say no to cheater plugs. Fix the wiring inside the electronics after you figure out how it should be wired instead of using a cheater plug.I also agree that the fuse and switch go on the hot side. If the ac hot wiring to the switch is coupling noise into the circuitry, use twisted shielded pair to route the ac hot to the switch and back. Run the shield to safety ground at one end only and put heat shrink over the end of the braid on both ends of the cable.
Ground bounce (from the diodes in the power supply) is only part of the problem. The experimenting I've done indicates that both bass and midrange clarity are affected by high frequency noise. One way to explain the effect of high frequency noise on bass is that the high frequency grundge makes the highs sound artificially louder so that the bass sounds less loud and less clear. The opposite is also true. I've seen 60 Hz hum make the treble sound weak.
Flipping the polarity of the primary of the power transformer doesn't change the magnitude of the external radiated field and internal coupled field, it just affects the phase of the field. However, flipping the polarity of the primary of the power transformer will change the magnitude of both the radiated and coupled electrostatic field. This capacitively coupled field is usually the biggest cause of high frequency noise in audio electronics.
How the transformer and power wiring is dressed and routed is important to control inductive coupling. That's why we moderately twist the leads for an individual winding and some times even shield the twisted leads. (Tight twisting can damage internal connections and cause insulation cut through.)
We should get a first order cancellation of the inductive field from even cheap power cords because the hot and neutral wires are routed close to each other. The radiated field from the current in the hot should be canceled by the radiated field from the current in the neutral. Sometimes, first order cancellation isn't good enough. The better the resolution of the audio equipment you are using, the more likely first order cancellation isn't good enough. So you are right to say the inductive field from power wiring is a potential problem. I've seen the field from the ac hot current flow cause serious problems in even "non-audio" equipment when it wasn't fully "cancelled" or properly shielded.
For now we'll just have to agree to disagree. I still believe "ac polarity" is a capacitive driven thing. Try some experiments and post what you find out. Have fun!
Play safe and play longer! Don't be an "OUCH!" casualty.
Unplug it, discharge it and measure it (twice) before you touch it.. . .Oh!. . .Remember: Modifying things voids their warrantee.
VS: ""For now we'll just have to agree to disagree. I still believe "ac polarity" is a capacitive driven thing. Try some experiments and post what you find out. Have fun!""When I saw that, I said to myself...what are you talking about..I didn't think we disagreed...then you recounted the cap thing...oh yah..
Nice post..Yes, I will be measuring the inductive coupling..I will be including a chassis, IEC, 2 wall outlets wired as two dedicated outlets, a dummy source chassis, IC, and I'll be trying two toroid sizes (200 va and 2kva) and two reglar fmrs, 8 amp 34 vct and 8 amp 80 vct.. First, at 60Hz, then sweep 20 to 20K.
If I can find a cheap or free 100 amp feed panel, I'll also play with wiring techniques there to minimize inductive coupling.
Oh, I'll also measure capacitance, although I'm not sure it will matter much sub 1Khz, but, you present a reasonable enough argument to certainly consider it..
Electrostatically shield your test wires by wrapping copper foil covered with tape around them so they are 100% covered, but the copper foil doesn't make a shorted turn. Ground the copper foil to eliminate most of the electrostatic effects.Have fun and share what you learn.
Play safe and play longer! Don't be an "OUCH!" casualty.
Unplug it, discharge it and measure it (twice) before you touch it.. . .Oh!. . .Remember: Modifying things voids their warrantee.
I had considered the capacitance this way:The procedure has one measuring the voltage across the ground run within the line cord..that ground run has, what, 10 milliohms? And yet, it is possible to measure voltage across that ground, from the earth to the chassis..
The two hypothesized mechanisms, inductive coupling, and capacitive coupling, are significantly different.
Inductively, it is a trivial matter to generate hundreds of millivolts across that ground path, by trapping 60 hz mag flux within the loop formed by the ground wire and the meter leads.. This was shown by test by Dave Kingsland, albeit a different setup in tweaks.. one of separate outlets and a 1.5kw space heater..however, he measured 100 to 200 mv ground to ground, but yet no current was being drawn through the grounds, meaning it was inductive coupling..and, it wasn't capacitive coupling, as there was no voltage in the loop with no current being drawn..
Capacitively, how much is required to force a ground current..remember, milliohms, and, what capacitance..a uf, a nf...
Differentiating the two should be rather easy...given a setup of a xfmr on a chassis, test the polarity with the core attached to ground, and then by isolating it by lifting it say 100 mils..bring the capacitance down to the pico range..if the polarity distinction goes away, it indeed is capacitance..if it does not, it is inductive..
Test will be done...tis in the queue, gotta get my quad wire spiraller finished soon though, I need the 150 foot run in a coupla weeks..
Cheers, John
I've seen nearly 80V ac drop to 30V ac on the chassis by swapping the polarity of the transformer leads.100 feet of romex from the outlet to the breaker box at 10 amps inducing 200 mV on the safety ground doesn't surprise me, but sounds high. With safety ground centered between the Hot and Neutral, the induced current should cancel if the insulation thickness and spacing is symmetrical around the safety ground. I'll have to play with the toaster oven sometime.
Play safe and play longer! Don't be an "OUCH!" casualty.
Unplug it, discharge it and measure it (twice) before you touch it.. . .Oh!. . .Remember: Modifying things voids their warrantee.
Did you load the chassis to ground before measuring that, or are you relying on the dvm? If unloaded, what would the AC current be withn the ground loop, the driving impedance of the transformer capacitance within the audio band relative to the ground loop impedance is rather low..Your desc of ground in the middle/cancellation is correct..I am talking about a ground to ground potential, with the loop being formed by either two cords to the same outlet, or two dedicated runs.
I usually use a 1 megohm DVM. I've seen up to 7 mA of chassis leakage current. A dangerous amount. I seem to remember the chassis voltage was 60V in that case. The Thevenin equivalent would be -j8.5K (0.31 uF) and 60V at 60 Hz. This will drop to -j85 ohms at 6 kHz.My bet is two dedicated runs will generate more safety ground noise than a pair of runs to the same duplex. Consider a preamp on one dedicated run and a huge class A power amp on the other. The preamp won't have much current flowing so over 100 feet so on an non-matched romex line it won't make much safety ground noise. The power amp will generate 20 times the amount of safety ground noise.
The questions will become what is the source impedance for the noisy safety ground and what is the voltage on the safety ground in the breaker box.
The next thing that will comes up is power cord design. From this conversation, it looks like we want to decouple the safety ground from the hot and neutral. May be do a four braid or a twist of the hot and neutral, cover that with an insulator and add the ground wire and cover with one more insulator.
Side bar: I've made a few power cords with built in common mode chokes on the hot and neutral (nothing on safety ground). I like them.
Play safe and play longer! Don't be an "OUCH!" casualty.
Unplug it, discharge it and measure it (twice) before you touch it.. . .Oh!. . .Remember: Modifying things voids their warrantee.
< < measure the ground potential with the unit plugged in one way and then 180 degrees off (cheater plug) > >Replying on the powercord & cheater plug to orientate the polarity will we be able to find out if only 1 out of 2 trannies is inverted?
If you have 2 power transformers in the one chassis and you DON'T notice a difference in the chassis-ground potential when you invert the mains, then the transformers are oriented "oppositely"!!You need to choose one transformer to reverse the primary leads on - then repeat the test. This time you will notice a chassis-ground voltage difference. Hopefully, you will have chosen the correct transformer to have reversed - and they're both now oriented the right way! If not, you'll have to reverse them both.
Regards,
I unhook one tranny while testing the other one and vice versa until I ascertain the hot side of each tranny, hook them both back up to the proper side and go on down the road........
z
Actually Clark, that is what Bob does in order to get things right. This is one of Bob's suggestions to Parasound in order to get a Class A rating on their products.
a
JC: "" This is one of Bob's suggestions to Parasound in order to get a Class A rating on their products. ""But, do they actually do it?
I would certainly hope so..as there are valid reasons..
Cheers, John
Good point, we find this all the time.
This post is made possible by the generous support of people like you and our sponsors: