The angle on conduction
Every raw DC supply, be it for a power amp or pre, is suprisingly noisy, a 'scope trace with a bit of ripple really doesn't tell the same story as the FFT analyser. The narrow conduction angle of the rectifiers is the source of this - turning on for small fractions of the mains cycle and passing 10's of amps, generating extended harmonics of the mains input frequency.
Methods to open up conduction angle, i.e. RC filters have all proved sonically detrimental to my ears.
I stand by this one, adding a filter does all you say it does, but for me it's sonically detrimental, compressing dynamics on items powered by the filter, and also severely affecting rythmic performance of other items connected to the same mains spur. The filter appears electrically across the mains, and just plugging it in affects the performance of other items on the same mains supply.
Some may like the effect, but to me it improves the HiFi, but destroys the music. It's the latter bit I want
I strongly suspect your earth current theory may have some validity, although I've noted just the capacitor across Live and Neutral is also detrimental.
Bam,
Agree with everything you've said, I'd not suggest for one second that it's not measurable, but that either methodology or sensitivity of measurement is the problem. Since it's the dynamic performance od the system that suffers, and measuring that is challenging.
The effect of a line filter, as noted above, is audible even to units not powered through it.
Andy.
I cannot believe you have an amplifier with dc supply noisier than the mains supply, if so you should probably be building a new amp
Every raw DC supply, be it for a power amp or pre, is suprisingly noisy, a 'scope trace with a bit of ripple really doesn't tell the same story as the FFT analyser. The narrow conduction angle of the rectifiers is the source of this - turning on for small fractions of the mains cycle and passing 10's of amps, generating extended harmonics of the mains input frequency.
Methods to open up conduction angle, i.e. RC filters have all proved sonically detrimental to my ears.
I stand by this one, adding a filter does all you say it does, but for me it's sonically detrimental, compressing dynamics on items powered by the filter, and also severely affecting rythmic performance of other items connected to the same mains spur. The filter appears electrically across the mains, and just plugging it in affects the performance of other items on the same mains supply.
Some may like the effect, but to me it improves the HiFi, but destroys the music. It's the latter bit I want
I strongly suspect your earth current theory may have some validity, although I've noted just the capacitor across Live and Neutral is also detrimental.
Bam,
Agree with everything you've said, I'd not suggest for one second that it's not measurable, but that either methodology or sensitivity of measurement is the problem. Since it's the dynamic performance od the system that suffers, and measuring that is challenging.
The effect of a line filter, as noted above, is audible even to units not powered through it.
Andy.
Diesel Powered Amplifier?
Two years ago, I headed the project to upgrade our department's power system for emergency standby power.
As the building has been in use for many years, I found that the "what the heck, it'll do" system had been employed over the years: I found two separate 3-phase systems in the building, 120/240 and 110/208!!
It was an interesting project, upgrading the fire dispatch center's main power without ever shutting it down. I learned a lot! Keeping "too many chefs out of the kitchen" was a real task, but it worked.
Man, does a big Onan DGFA (150 kW) put out some clean AC! It's nice being the "only customer" on the mains.
I agree that there could be differences in the supply ending up at the speakers....but the effect of the DC rail filtration is the biggest factor.
Does anyone agree that induced AC hum and noise (at the signal level) is a greater concern?
Bob
Two years ago, I headed the project to upgrade our department's power system for emergency standby power.
As the building has been in use for many years, I found that the "what the heck, it'll do" system had been employed over the years: I found two separate 3-phase systems in the building, 120/240 and 110/208!!
It was an interesting project, upgrading the fire dispatch center's main power without ever shutting it down. I learned a lot! Keeping "too many chefs out of the kitchen" was a real task, but it worked.
Man, does a big Onan DGFA (150 kW) put out some clean AC! It's nice being the "only customer" on the mains.
I agree that there could be differences in the supply ending up at the speakers....but the effect of the DC rail filtration is the biggest factor.
Does anyone agree that induced AC hum and noise (at the signal level) is a greater concern?
Bob
i don't like isolation transformers.
most of them are horribly undersized and end up coloring the sound quite a bit. change in tonal balance, compressed dynamics, messed up soundstage, etc. etc. the only one i've found that works well is the Ensemble Isolink, which is made ONLY for digital components. the transformers are massive (the unit, which has 2 xformers, weighs as much as a heavy-duty power amp) and are extremely conservatively rated (around 200VA). the thing also costs $2700, but if i had to pick an isolation transformer for my digital gear i would not look at anything else. i have another good-quality isolation transformer in my system right now supposedly rated at 800VA, but i find it is only good for low-power digital components - for preamps or small amps it is too colored. interestingly, although the total power rating is twice that of the Ensemble, the unit weighs perhaps 1/3rd as much. the Ensemble unit is the real deal, i would not settle for anything less. just goes to show you what a joke those "Power Wedge" units are.
most of them are horribly undersized and end up coloring the sound quite a bit. change in tonal balance, compressed dynamics, messed up soundstage, etc. etc. the only one i've found that works well is the Ensemble Isolink, which is made ONLY for digital components. the transformers are massive (the unit, which has 2 xformers, weighs as much as a heavy-duty power amp) and are extremely conservatively rated (around 200VA). the thing also costs $2700, but if i had to pick an isolation transformer for my digital gear i would not look at anything else. i have another good-quality isolation transformer in my system right now supposedly rated at 800VA, but i find it is only good for low-power digital components - for preamps or small amps it is too colored. interestingly, although the total power rating is twice that of the Ensemble, the unit weighs perhaps 1/3rd as much. the Ensemble unit is the real deal, i would not settle for anything less. just goes to show you what a joke those "Power Wedge" units are.
i don't like isolation transformers
hi dorkus- i know what you mean but at plitron they have medical isolation transformers up 25Kva, which i think should be enough to handle big amps!! just kidding- the mains at 120V and 15A would be 1.85Kva, so a 2Kva iso trans. should be enough. what i was thinking that an iso. trans. has the same windings on pri. and sec. so it would isolate the mains from any "junk". am i thinking correctly or am i just confused? 
hi dorkus- i know what you mean but at plitron they have medical isolation transformers up 25Kva, which i think should be enough to handle big amps!! just kidding- the mains at 120V and 15A would be 1.85Kva, so a 2Kva iso trans. should be enough. what i was thinking that an iso. trans. has the same windings on pri. and sec. so it would isolate the mains from any "junk". am i thinking correctly or am i just confused?
<b>what i was thinking that an iso. trans. has the same windings on pri. and sec. so it would isolate the mains from any "junk". am i thinking correctly or am i just confused?</b>
I haven't looked at the Plitron iso's specifically, but if they're torroids, you're wasting your time. Torroids have a very high bandwidth (smaller ones make nice tube amp output trans) so let in lots of hash and noise that an EI won't. A big EI will have a bandwidth of a few hundred Hz and drop pretty fast above that. ES shields in torroids don't seem to make much difference either.
Some of the other types, like those with completely seperate bobbins for primary and secondary also work and measure well according to reports I've read.
I haven't looked at the Plitron iso's specifically, but if they're torroids, you're wasting your time. Torroids have a very high bandwidth (smaller ones make nice tube amp output trans) so let in lots of hash and noise that an EI won't. A big EI will have a bandwidth of a few hundred Hz and drop pretty fast above that. ES shields in torroids don't seem to make much difference either.
Some of the other types, like those with completely seperate bobbins for primary and secondary also work and measure well according to reports I've read.
well... i think if the transformer were big and well-built enough, it may work ok. i haven't tried such large transformers so i wouldn't know. however, i second Brett - transformers such as toroids generally have pretty wide bandwidth, so most of the noise would make it through anyway. you need a transformer specifically designed for high power and minimal high frequency response. such a device would be huge, heavy, and very expensive at higher power levels, but for lower power levels (e.g. the Ensemble unit) they are workable.
Banned
Joined 2002
Anyone thought about Balanced Mains Isolation or technical mains as I have heard it said once before.
It maybe what some of you mean by Isolation transformers.
But the method is quite simple. Unbalanced mains voltage on the i/p or primary and on the secondary windings remains floating except for a centre tap to mains ground.
The idea is that any noise that appears on either the Active or Neutral line gets cancelled out by differentation. Just as in Balanced Audio transmission lines...
Just a thought
Anthony Holton
It maybe what some of you mean by Isolation transformers.
But the method is quite simple. Unbalanced mains voltage on the i/p or primary and on the secondary windings remains floating except for a centre tap to mains ground.
The idea is that any noise that appears on either the Active or Neutral line gets cancelled out by differentation. Just as in Balanced Audio transmission lines...
Just a thought
Anthony Holton
I deliver balanced AC to my equipment through a 2x60VAC 2KVA Plitron transformer. As each 60VAC Secondary is delivered 180 Deg out of phase to each other the reactive cuurents sum to Zero and the CT of the transformer (Ground). Their has been a lot of lively discussion about this technology by me and several members on this site. If you search for "Balanced AC" you will find a lot of perspectives in discussion.
As already indicated North American power is split from the power company lines as single phase AC. This is one Hot line with an unbalanced Neutral (essentially a ground return). For safety reasons a local ground is introduced to protect circuits and people from a direct short on the Hot side by tripping the breaker. The problem is it only takes 20 mA to kill a person at 120 VAC. So GFI's and GFCI's were introduced to detect and respond faster and at minute current flows to provide better protection.
Recently AFCI or Arc Fault Current Indicators were introduce to protect where HF current is present due to arcing. Anyway the addition of a local common ground means that all equipment sharing the ground also share AC noise artifacts. The most insidious is usually AC Hum, but much higher frequencies can wreak havoc with Digital circuits.
When you convert your AC supply to two phase through a line transformer your equipment will pass most of the noise picked up through ground back down the 2 phase AC line to ground out of phase where they some to zero. This effect is very similar to a line level balanced input. Balancing you AC supply can reduce your noise floor by up to 12 dB.
The downside for this technology is the bottleneck it will invariable introduce when driving high current loads. It will dampen bass response due to the inherent characteristics of a transformer to be slow to react to changes in it's magnetic flux. That is it's ability to supply rapid changes in current on demand.
If you were using this technology on a low power device this would not be an issue, but an Aelph Class A Amp would give it kyniption fits. A standard residential circuit delivers 15 A, a 2KVA transformer would be the minimum required size to operate at say 50% capacity. That means 7 to 8 Amps max if you want any headroom at all.
If you can afford or are so inclined, the best way to deliver balanced power is right from the AC Mains entering your home. Here you can hardwire a 10KVA system that should run the larger demands of a serious Audio system. Such a system can be $5,000 to $10,000 dollars.
For about $700 you can start to build a small unit for your digital sources and your TV. Truth be told your better off trying to clean source noise than the DC rails of your amps anyway. Video is especially agreeable to balanced power as the effects can be quite noticeable.
Equi-Tech are one of the leadres in this technology and are supprisingly open about supplying information on Balanced power, see their link below. If you wish to build your own Balanced supply start with their white papers and one of the new Plitron IMIN transformers, 2KVA is about $500 and a GFI and a Corcom filter and your all set.
Anthony
http://www.equitech.com/
http://www.equitech.com/articles/articles.html
As already indicated North American power is split from the power company lines as single phase AC. This is one Hot line with an unbalanced Neutral (essentially a ground return). For safety reasons a local ground is introduced to protect circuits and people from a direct short on the Hot side by tripping the breaker. The problem is it only takes 20 mA to kill a person at 120 VAC. So GFI's and GFCI's were introduced to detect and respond faster and at minute current flows to provide better protection.
Recently AFCI or Arc Fault Current Indicators were introduce to protect where HF current is present due to arcing. Anyway the addition of a local common ground means that all equipment sharing the ground also share AC noise artifacts. The most insidious is usually AC Hum, but much higher frequencies can wreak havoc with Digital circuits.
When you convert your AC supply to two phase through a line transformer your equipment will pass most of the noise picked up through ground back down the 2 phase AC line to ground out of phase where they some to zero. This effect is very similar to a line level balanced input. Balancing you AC supply can reduce your noise floor by up to 12 dB.
The downside for this technology is the bottleneck it will invariable introduce when driving high current loads. It will dampen bass response due to the inherent characteristics of a transformer to be slow to react to changes in it's magnetic flux. That is it's ability to supply rapid changes in current on demand.
If you were using this technology on a low power device this would not be an issue, but an Aelph Class A Amp would give it kyniption fits. A standard residential circuit delivers 15 A, a 2KVA transformer would be the minimum required size to operate at say 50% capacity. That means 7 to 8 Amps max if you want any headroom at all.
If you can afford or are so inclined, the best way to deliver balanced power is right from the AC Mains entering your home. Here you can hardwire a 10KVA system that should run the larger demands of a serious Audio system. Such a system can be $5,000 to $10,000 dollars.
For about $700 you can start to build a small unit for your digital sources and your TV. Truth be told your better off trying to clean source noise than the DC rails of your amps anyway. Video is especially agreeable to balanced power as the effects can be quite noticeable.
Equi-Tech are one of the leadres in this technology and are supprisingly open about supplying information on Balanced power, see their link below. If you wish to build your own Balanced supply start with their white papers and one of the new Plitron IMIN transformers, 2KVA is about $500 and a GFI and a Corcom filter and your all set.
Anthony
http://www.equitech.com/
http://www.equitech.com/articles/articles.html
hey coulomb- your thinking is correct. i am an electrician by trade and what you say is correct. your idea about balanced power is interesting, very good point!! can you elaborate on this alittle further. also just like a power supply, couldn't you just add some electrolitic caps on the sec. side to " stiffen" the load so there wouldn't be any "sag" on the current demand? the reason i ask is i want to build an iso. trans. and maybe line filter for my audio system. if this works, i will build them for all my electronics!!! also why would a 2Kva trans. only give you 7 to 8 amps max. how does current affect headroom. greg
The lower frequencies require ecponentially more current due to load of the speakers. My reference to Headroom is a misnomer as I am refering to the Amps ability to draw sudden current reserves from the ISO transformer when it needs it. If you run the Iso at full load for nominal listening where is the additional power going to come from when required. The same theory applies to your branch circuits, but your mains run from a 13.6KVA distribution transformer on the pole. Due to the extremely high power output this transformer appears transparent to a high current load such as a class A Amp. The much smaller 2KVA transformer (about max on 115VAC, 15A branch) will appear as a tremondous bottle neck.
My next experiment is going to be utilizing a dryer or stove branch into a 4 KVA transfomer. Not your typical DIY project if you want to hardwire to the panel, but I used to be an apprentice journeyman so I will manage.
Anthony
ok i'm still confused!! if i balance the line as per coulomb with 60V sec. and CT to ground- there is still another trans. in all electronic equipment, right? what is the purpose (read benefit) of having a balanced line? the second trans. wouldn't "see" a difference, would it??? also i thought having two trans. in the line would have detremental impact on the sound!! also if i used 240V as per nelson's statement- how would i convert it back to 120V. would you use another trans.? also having the same results. i like the idea of having "balanced" lines (i read the link to equitech from coulomb and it was very interesting) but still don't understand why. is this just another sheep in king's clothing?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.