I'm contemplating if it is worth going down this route. I know Ali and Gazjam are very happy with their units and I know others who swear by balanced mains.
Neal, did you go for it in the end and if so, what do you think?
What do others think on the benefits or otherwise of going for a balanced mains input?
Balanced Power Supply
#63
Neal, I just took delivery this very day of a BPS from Airlink.
Give me some time to install it and spend some time with it and I'll provide another data point...
It vexes me though that with a perfect 230v input it's generating 250v output when it's rated at 230v output.
Doesn't seem right to me, but maybe I'm being too fussy? What do you guys think, should I contact Airlink about it?
On the positive side it does measure exactly 125v on both sides, so it is perfectly balanced.
Give me some time to install it and spend some time with it and I'll provide another data point...
It vexes me though that with a perfect 230v input it's generating 250v output when it's rated at 230v output.
Doesn't seem right to me, but maybe I'm being too fussy? What do you guys think, should I contact Airlink about it?
On the positive side it does measure exactly 125v on both sides, so it is perfectly balanced.
Jason
#64
It will have been wound to give 230v on load, you would expect it to be a bit high off load.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
-
Paul Barker
- Social Sevices have been notified
- Posts: 8997
- Joined: Mon May 21, 2007 9:42 pm
#67
The possible benefits of using the balanced output transformer instead of a CMC in your power supply EMI rejection measures as far as I can work out are few. Airlink have missed out on one two of those by using torroidal transformers which cannot handle DC, and which have good bandwidth.
If you used EI or Core or Shell transformers you could arrange them to to handle levels of DC from the mains. EI transformers in particular of those three would further have the advantage of poorer bandwidth.
Poor bandwidth means less RFI coupling to the secondary. Exactly what we need.
Supplement the primary with voltage clamps and put your usual mains conditioning circuit (of which Swampy's design was perfectly adequate) into the mix. On the secondary side of the balanced transformer, add the final CM rejection caps between the secondary poles and the earth.
Now comes the 3rd and final benefit of the transformer, which a torroid does give just as much as the above. The earth currents generated by the smps's and other solid state elements in your equipement are balanced, so zero out. Whereas in a normal household system such equipmentt as computers televisions and probably a lot of your solid state stuff like AK4396 DAC's the earth leakage current travels predominantly down the phase creating a high earth current which can be sufficient combined to trip your 30mA RCD. there would be non of this after the isolation transformer.
so as long as the core size is of sufficient capacity (above the normal standard which is pathetic for audio, I would say 50% to 100% bigger) you should hear an improvement against such as a Swampy filter. But, like Neal, I haven't got a round Tuit.
But that's how I would do it. EI stacked for DC, wound for diabolicy HF performance.
If you used EI or Core or Shell transformers you could arrange them to to handle levels of DC from the mains. EI transformers in particular of those three would further have the advantage of poorer bandwidth.
Poor bandwidth means less RFI coupling to the secondary. Exactly what we need.
Supplement the primary with voltage clamps and put your usual mains conditioning circuit (of which Swampy's design was perfectly adequate) into the mix. On the secondary side of the balanced transformer, add the final CM rejection caps between the secondary poles and the earth.
Now comes the 3rd and final benefit of the transformer, which a torroid does give just as much as the above. The earth currents generated by the smps's and other solid state elements in your equipement are balanced, so zero out. Whereas in a normal household system such equipmentt as computers televisions and probably a lot of your solid state stuff like AK4396 DAC's the earth leakage current travels predominantly down the phase creating a high earth current which can be sufficient combined to trip your 30mA RCD. there would be non of this after the isolation transformer.
so as long as the core size is of sufficient capacity (above the normal standard which is pathetic for audio, I would say 50% to 100% bigger) you should hear an improvement against such as a Swampy filter. But, like Neal, I haven't got a round Tuit.
But that's how I would do it. EI stacked for DC, wound for diabolicy HF performance.
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein
-
Paul Barker
- Social Sevices have been notified
- Posts: 8997
- Joined: Mon May 21, 2007 9:42 pm
#68
this is where we audio enthusiasts always run into a brick wall with commercial transformer winders.figlet108 wrote:Good point. I'll see what it measures under load.
We want a big transformer to minimise our supply impedance. they only know a world where a customer who specifies a 3kv transformer wants the rated voltage at 3kv load. you might only demand half a Kv from it. So your secondary voltage would be 8% overvoltage.
The larger the transformer the less percentage underwind of the primary to accomodate regulation (the adjustment for core and copper losses). A small transformer the regulation is created by underwinding the primary by as much as 15% this reduces to 10% as you go up in size. Your small current draw may drop the 10% over voltage a few percet, so you will end up over voltage.
In the very olden days they used to give you two or three primary taps relative to 0v which you could fine tune your output voltage with. that is what you need to specify when you don't intend to use the 3kv transformer for 3kv, and so on. For our designs we can't have enough primary taps beyond 0v. You understand each extra tap is in the other direction to zero than the phase tap. Comprenday?
These extra turns also allow you to adjust for your supply voltage. Mine for instance is never less than 245v and often 255v, but most commercial transformers are wound for 230v (the very badly named "harmonized voltage" which has supposed to have given Europe a level playing field for equipment, but in reality has left the UK way over voltage).
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein
-
Paul Barker
- Social Sevices have been notified
- Posts: 8997
- Joined: Mon May 21, 2007 9:42 pm
#69
Air Link have used torroids because it is much more cost effficient to do so.
When I wound a 3kva EI transformer for my milling machine the core cost me £100 and the 2mm winding wire another £100.
When I wound a 3kva EI transformer for my milling machine the core cost me £100 and the 2mm winding wire another £100.
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein
#70
Yep, I built a 1kVA EI transformer for someone just as Paul said, EI, balanced secondary, multiple taps on the primary, and I added a cheap LCD voltage display on the secondary side. The main reason we built it was to lower their mains voltage which kept blowing up their optimistic 1980's valves amps. The transformer from SP Wound cost more than the 3kva from Airlink.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
-
Dave the bass
- Amstrad Tower of Power
- Posts: 12276
- Joined: Tue May 22, 2007 4:36 pm
- Location: NW Kent, Darn Sarf innit.
#71
I had a good nose around inside our-Ali's TX box and IIRC he's blocking DC with one of our-Nicks DC blockers. I think.Paul Barker wrote:The possible benefits of using the balanced output transformer instead of a CMC in your power supply EMI rejection measures as far as I can work out are few. Airlink have missed out on one two of those by using torroidal transformers which cannot handle DC, and which have good bandwidth.
DTB
"The fat bourgeois and his doppelganger"
-
Paul Barker
- Social Sevices have been notified
- Posts: 8997
- Joined: Mon May 21, 2007 9:42 pm
#72
Did you get a chance to find out if it made a significant difference to the sound?Nick wrote:Yep, I built a 1kVA EI transformer for someone just as Paul said, EI, balanced secondary, multiple taps on the primary, and I added a cheap LCD voltage display on the secondary side. The main reason we built it was to lower their mains voltage which kept blowing up their optimistic 1980's valves amps. The transformer from SP Wound cost more than the 3kva from Airlink.
I can say that a multi stage mains EMI filter makes considerable difference. But my mains at home is horrible sounding left alone. I get a lot of noise out of the speakers regardless of how good my amp is, which goes when I condition the mains.
I am not dc blocking right now. Not actually sure anyone is using a hair drier within range when I am listening and if they do it would be brief enough not to matter.
There may be another reason to need dc blocking I am unaware of. I am ready to be educated.
I talked about the dc blocking element above but only because others desire it, and because I understand transformers to a degree worth mentioning.
I will never forget the huge improvement Darren found when I wound him an HT transformer for his B+ supply using a split bobbin so the secondary was perfectly balanced in every respect. That is the same capasitance the same inductance and the same resistance, the same leakage.
Whereas if you balance the resistance only of the two phases of a normally wound B+ transformer, you don't have a balanced secondary so it doesn't operate in the common mode to your advantage as the transformer Ali has does.
in other words the common mode element works. But there are more than one way of doing it.
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein
#73
Its worth it IMHO if you have a toroid, the lack of gap means its prone to saturate. I did this some time ago now.There may be another reason to need dc blocking I am unaware of. I am ready to be educated.
http://www.mains-cables-r-us.co.uk/site ... _Paper.PDF
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
-
Paul Barker
- Social Sevices have been notified
- Posts: 8997
- Joined: Mon May 21, 2007 9:42 pm
#74
Excellent. Definately worth blocking DC then.
I am trying to decide whether there is any mileage in designing a transformer for deliberately poor HF coupling.
torroids by nature have excellent HF coupling. The reason being the way they are wound. The primary layers tend to run right round the core, then the secondaries likewise. The core is therefore double screened by these two sets of windings, and very little gets out. But it we remotely mount our isolation transformer this doesn't matter to us. But the side effect of the winding style is excellent HF coupling between windings. surely we don't want this for our purpose of cleaner mains.
So I am proposing a thesis that we wind an EI transformer capable of DC, though I grant you that your DC blocker may limit the necessity of that.
But wind it rather than on a single bobbin in the double area central protrusion. Wind the primary around the one extreme single area leg and the other winding on the opposite leg. Deliberately liomit coupling to the lower frequencies.
I shall try this when I get organised to wind again, and compare it to the more usual EI power transformer arrangement.
There also goes through my mind the likelyhood that the distortion introduced by an extra transformer may kick us into the diminishing returns law. the better solution to mains noise (without a freshly generated supply of our own) may be a filter and dc blocker conditioner.
We can't just assume a transformer is better than the latter.
I am trying to decide whether there is any mileage in designing a transformer for deliberately poor HF coupling.
torroids by nature have excellent HF coupling. The reason being the way they are wound. The primary layers tend to run right round the core, then the secondaries likewise. The core is therefore double screened by these two sets of windings, and very little gets out. But it we remotely mount our isolation transformer this doesn't matter to us. But the side effect of the winding style is excellent HF coupling between windings. surely we don't want this for our purpose of cleaner mains.
So I am proposing a thesis that we wind an EI transformer capable of DC, though I grant you that your DC blocker may limit the necessity of that.
But wind it rather than on a single bobbin in the double area central protrusion. Wind the primary around the one extreme single area leg and the other winding on the opposite leg. Deliberately liomit coupling to the lower frequencies.
I shall try this when I get organised to wind again, and compare it to the more usual EI power transformer arrangement.
There also goes through my mind the likelyhood that the distortion introduced by an extra transformer may kick us into the diminishing returns law. the better solution to mains noise (without a freshly generated supply of our own) may be a filter and dc blocker conditioner.
We can't just assume a transformer is better than the latter.
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein