New Balanced Power supply

[Sovereign]

I was going to start this thread tonight, but I'm sat at Gatwick airport waiting for a plane to arrive so I thought I would see what I could do using my phone.
I have been into using balanced power for about six years now and really valued the audio benifits it brings. I have made and sold on maybe six units in all. The last one I just sold was a whopper of a BPS it was 3kVa with a DC blocker fitted and housed in a nice black aluminium case.
I wanted to make another and this time I wanted to make one that was a bit special and was a true reference BPS.
So here she is in her raw, heavy and curvy glory. I've always liked big curvy girls butt this one is off the scale.http://i1138.photobucket.com/albums/n533/SovereignKing/Mobile%20Uploads/image-1.jpg (http://s1138.photobucket.com/user/SovereignKing/media/Mobile%20Uploads/image-1.jpg.html)
http://i1138.photobucket.com/albums/n533/SovereignKing/Mobile%20Uploads/image.jpg (http://s1138.photobucket.com/user/SovereignKing/media/Mobile%20Uploads/image.jpg.html)
I plan to lash it together tonight, but build it properly over the next few days

[Ali Tait]

Linky no worky?

[Sovereign]

I keep on having problems with photobucket on me phone

[Sovereign]

Ali I've just emailed them to you. Not sure if you can do anything your end ;-)

[Ali Tait]

Here you go -

http://imagizer.imageshack.us/v2/1024x768q90/631/VF8IPR.jpg (http://imageshack.com/f/hjVF8IPRj)

[Ali Tait]

Looks meaty.

[BilliumB]

Hi James, looks interesting.

I've been meaning to put together a balanced transformer setup for some time, but have not got round to it yet (currently using a Ben Duncan isolation transformer that I picked up 2nd hand).

James, I know we spoke some time ago - Nick Goram (Lurcher) had suggested to me that an EI transformer might be better suited as the basis for a balanced supply (better rejection of interference possibly) and I've not taken this any further yet. Do you think a toroidal sounds best - it would be really interesting to find out for a 'reference' supply.

I've also been thinking of a 'multiple' balanced transformer setup - a big one for the amp (say 1500VA and a couple of smaller ones say 300VA for the Touch and the DAC) - I wonder how much better this separation of supplies would sound - another thing to consider for a reference setup perhaps?

I also wonder if stepping the voltage down a bit might be a good idea - whenever I measure the mains voltage here it is always up around 248v or so, so most equipment is probably operating close to the edge of any specified mains I/P voltage range.

I really need to get hold of an oscilloscope to see what sort of mess the mains looks like. Are you finding that Nick's DC Blocker sorts out all DC offset and transformer buzz problems?

Will be interesting to see how you get on.

Cheers. Bill

[Sovereign]

Hi Bill, how are you doing mate?
Here is an snippet of a thread that that talks a lot about EI transformers.
With the manufacturer of my balanced torodial we tried many different options and ideas to make the torodial perform at its best, some of them are mentioned in the article below, but much of it had no effect on the performance of the toroidial or the sound of my hifi. For instance we spent much time earthing the massive iron core of the torodial to absolutely no effect. We used a half-wave rectified circuit on the shared supply to generate DC on the mains. As expected earthing the core had absolutely no effect on the noise level, in fact all it provides is a potential risk of transformer failure if there is a breakdown of the core insulation, the long and short of it is don't believe all you read.
One of the problems about EI transformers is finding someone that will make them and to your requirements, I have been shown some shockingly poor quality EI transformers from respected names in the industry and the cost would also be very high indeed.
One of the advantages of my balanced 3kVa toroidial is it uses an oversized core (5kVA) to allow a very low flux density for low mechanical noise and the core will probably never saturate, however it weighs in around 27kg, It is continuously rated at 3kVA – I’m not so sure that many of the other so called 3kVA transformers are.

Enjoy the read:

The High-End Mythology of the Toroidal Power Transformer

Some background about how AC power is delivered to residences

When AC power is transmitted across long distances, high voltages increase transmission efficiency and decrease the size and weight of cables needed to carry the power. However, high voltages in the home are quite dangerous so power companies build sub-stations to reduce the voltage levels for residential and light business use. In the US these substations produce three-phase electrical power. Two of the three phases are what is typically delivered to residences. These electrical phases each carry (nominally) 120v rms AC referenced to the neutral wire… all three phases use the same neutral wire/connection/reference. When two phases are supplied to a home or business, you have three wires coming into the house. One of them is neutral. You can connect the two 120v lines to a single electrical outlet and get 208/240 AC volts which is used to power electric dryers, electric ranges and stoves, electric water heaters and other heavy duty electrical appliances.

The thing to remember about this three-phase and two-phase power delivery system is that the loads on each phase must remain balanced or you can get a DC offset to the AC voltage. If power becomes unbalanced in any one phase, a DC offset appears in your AC power. These DC offsets can be transient or long-term. You still get the 120 (nominal) rms volts, but instead of it being a sine wave with positive peaks of about 170 volts and negative peaks of -170 volts, the sine wave is shifted up or down. A +5-volt DC offset would result in positive peaks of 175 volts and negative peaks of –165 volts (approximate voltages). A –6-volt DC offset would result in peaks of +164 and –176. All because the power becomes unbalanced, i.e. there is more load on one or two phases than on the remaining phase. These DC offsets can very dynamic, changing constantly as residences and businesses consume power and turn machinery on and off. Some of the offsets cancel each other out, resulting in no DC offset, but it is not terribly unusual to have transient DC offsets in power delivered to homes in the US. In Europe, depending on the country, some power grids are owned by the country and are actually running at full capacity or even higher than full capacity. In those situations, the potential for DC offset to exist in the power delivered to residences is quite high.

Two transformer types versus DC offset

What does DC Offset have to do with the AC Power Transformer? Everything! You see, the toroid transformer is very intolerant of DC offset being present at the input (primary winding). In fact, if there is any DC at all in the AC power supplied to the typical AC Power toroid transformer, the laminated metal core of the transformer "saturates" and the transformer no longer operates as designed. The result is often high levels of audible noise from the transformer as well as very high levels of noise getting into the power supply of the component.

When an EI-frame transformer is used, the DC offset problem disappears. The EI-frame transformer continues to supply AC power of the proper voltage with all the noise elimination capabilities designed into the EI-frame transformer still operating perfectly. Industry outside of high-end audio has not widely adopted the AC toroid power transformer because of these very problems. The majority of transformers in general use outside of high-end audio equipment are actually EI-frame transformers.

The main advantages of EI-frame transformers over toroidal transformers are: less susceptibility to core saturation from DC offset; air gaps which act as pressure relief valves for high density magnetic flux in the core. Saturation of the transformer core results in an energy build-up in the core of the transformer. Toroid transformers are wound with the core completely covered by the primary and secondary wires. The toroid core is completely isolated from ground. In the EI-frame transformer, however, a significant portion of the core is external and is attached directly to the chassis of the equipment it is installed in. This gives any energy build-up in the core a direct path to the chassis ground of the component. The other advantage EI-frame transformers have is the small air gaps that are present because of the gaps present when the two parts of the frame/core are put together. These air gaps act as pressure relief valves for the magnetic flux in the core further enhancing the transformer’s resistance to core saturation. You can have an air gap in the core of a toroidal power transformer, but this requires machining a gap perhaps 0.020" wide, about the thickness of a sheet of paper. This is not a trivial task and adds significantly to manufacturing cost of the toroid transformer. It is very rare to encounter toroid transformers with air gaps because of the cost and difficulty of machining such a thin slit across the laminations that make the metal donut. These are two of the reasons EI-frame transformers, when properly designed, can provide better audible performance in your audio components.

What’s the catch?

It seems like an EI-frame transformer ought to be a 100% shoe-in for high-end audio components, but there is a "dark side" to the EI-frame transformer. They radiate a rather significant magnetic field from the exposed windings (which are sometimes covered by rounded metal end bells). Low level audio signals such as those found in preamplifiers and gain stages prior to the output stage of amplifiers are small enough in magnitude, that passing them through a strong magnetic field would induce hum in the audio signal. In a preamplifier, this is easy to avoid. Put the power supply in an external enclosure as many preamp manufacturers do, especially those who use EI-frame power transformers. In an amplifier, the answer is to "aim" the transformer correctly. There is a big "shadow" in the magnetic field where the laminated plates are...the magnetic field is strong only outside the end-bells of the transformer. Aim the EI-frame transformer so that the core points towards the audio signal processing and so that the "bells" point to the sides and the radiated magnetic field won’t bother the audio circuits inside the amp. But putting an EI-frame in a tightly packed multi-channel amp or receiver is just about impossible...there just isn’t enough room to stay away from the radiated magnetic field. In those applications, you will almost always find toroidal power transformers. You can jam a toroid right into the middle of five amplification channels and have no problems from magnetic fields. Magnetic shielding for an EI-frame transformer is also possible, but it tends to be expensive and heavy.

Why lavish so much attention on the transformer?

When you look realistically at what happens in an audio component...you know the answer. The signal leaving the component existed only as AC power from your wall outlet only milliseconds before it is on the way to the next component or to the loudspeakers. AC power is the raw material for the output of the component. Make beer, wine, scotch or bourbon with bad water and you get a bad product. Make an audio component with a generic off-the-shelf toroidal transformer and you are going to get a generic sounding component. It won’t be terrible, but it will fail to scale the heights of what is possible.

So an EI-frame transformer is always better than a toroidal transformer?

Hmmm. Good question. I am tempted to say "yes" because really excellent toroidal transformers are apparently difficult, if not impossible to purchase off-the-shelf. However, finding great EI-transformers is no walk in the park either. But in general, being able to connect the core of the EI-frame transformer electrically to the chassis of the component and the air gaps created by the two-piece laminated core give even off-the-shelf EI-frame transformers some performance capabilities beyond what you can get from off-the-shelf toroidal transformers. If you know your magnetic theory (or your transformer vendor does), there are additional winding tricks that give EI-frame transformers 20dB to 30dB more noise rejection than typical toroidal transformers. This is very significant. 20dB less noise is 1/100th of the noise of the "reference" level. 30dB less noise is 1/1000th of the "reference" noise level. You might be able to find a toroid manufacturer who could do the custom winding, add an air gap and attach a ground wire to the core. You would pay two to three times the cost of an off-the-shelf toroid to get that level of performance. But you sure won’t find a toroid off the shelf that performs as well.

I’ve been lucky enough to speak to a variety of high-end manufacturers about AC power, transformers, and component design. Conversations with Richard Vandersteen, Mike VansEvers and Emil Rotar (Warner Imaging) have been most illuminating. Years ago, I never really gave power transformers much thought. The assumption was that because so many high-end manufacturers were using toroidal power transformers, toroids must be demonstrably superior. Imagine my surprise when I started noticing that many of the best sounding components I’ve experienced in my system have all had EI-frame transformers. Coincidence? I might have thought so before finding out so more about power and transformers. Now I’m convinced that cleverly done EI-frame transformers could/should give us better sounding high-end audio components.

[brian2957]

Is that one of the tyres off the van James .

[Sovereign]

Is that one of the tyres off the van James .
Yup, just driving a reliant robin now :-)

[BilliumB]

Hi James - an interesting read, have you done any comparisons yourself? I guess an EI balanced transformer would have to be located in a box well separated from any sensitive kit.

I understand that SP Wound do a good job on EI transformers, but have no experience myself, as yet!

Cheers. Bill

[Sovereign]

Hi mate, no comparisons at all. I think I'll stick with a torodial, the difference this one has made to the sound of my hifi is fantastic.

[BilliumB]

I guess as you've over specified everything that could explain why it sounds so good.

Have you tried separate balanced transformers to feed separate components - possibly removing any 'cross contamination' of the mains supply? If not what are your thoughts about any possible benefits?

Really interesting to read about what you're putting together.

Cheers. Bill

[Sovereign]

No, never tried separate torodials for each piece of kit, I wonder if anyone has. I need to stress I don't know that much about the subject of balanced power, my knowledge is from others.
I'll try and post some more pictures as I build it over the weekend.

[BilliumB]

Look forward to the pics and to see what you think of the results.

Has anyone else tried EI transformers and/or multiple transformers for a balanced supply?

[Sovereign]

I started the build of the case and sockets, hoping to have it finished by tomorrow.
http://i1138.photobucket.com/albums/n533/SovereignKing/imagejpg1-4.jpg (http://s1138.photobucket.com/user/SovereignKing/media/imagejpg1-4.jpg.html)
http://i1138.photobucket.com/albums/n533/SovereignKing/imagejpg2-3.jpg (http://s1138.photobucket.com/user/SovereignKing/media/imagejpg2-3.jpg.html)
http://i1138.photobucket.com/albums/n533/SovereignKing/imagejpg3.jpg (http://s1138.photobucket.com/user/SovereignKing/media/imagejpg3.jpg.html)

[BilliumB]

How did you get on today?

[Sovereign]

Great. The unit itself is finished, well kind of but I still have a little experimenting to do. As this particular toroidal is so quiet I'm going to leave out the DC blocker for now, Then fit it in the unit in a few weeks to see what the difference it makes. For me when I have fitted DC blockers in the past, not only has it quiet ended the toroidal but it makes the music sound better , that's not just because of a quieter noise floor but maybe it enables the toroidal to perform better. This is all circumspect though. However this toroidal does not suffer from loud hum like other large transformers do, so there is very little hum to quieten, But we will see.
I'm using Neutrik power connectors fitted to the unit, but I need a small enough Allen key too fit the female power connectors to my power chords, then We are done, then I can properly report as to the performance of the BPS

[BilliumB]

Great - looking forward to further pictures.

[docfoster]

Hi, Sov.
Nice work. :-)
How do the effects your BPSs compare to the AG1500? Did you ever use both simultaneously?

[Sovereign]

Hey Ben,how are you doing mate?
That's a good question. Yes I've run BPS and a AG 1500 regenerator together, I did so for a couple of years or so. I sold them both to buy my current amp and only now am I re making that kind of stuff again. Talking about these things can be a bit airy fairy, but to get down to the brass tacks for me the AG1500 adds more air and transparency and holographics. However a good balanced power supply brings more weight, meat, better tonal balance, unforced detail etc. if you run them together you get a bit if both of the two different approaches. I have obviously gone for a BPS first and not a regen, and I'll probably just stick with the BPS I have made. There is also a marked difference between previous BPS I have had in the past and what I have now as it really is quite exceptional.
I'm really pleased with the BPS, I gives my Iridium SECA so much more weight, bass, detail, size, and excellent tonal colour.
Hope that helps.

[docfoster]

Hope that helps.

Indeed it does. Thank you.
The fact that our amps are of similar breeding lends weight to your thoughts for me.
Interesting to read that in your experience not all BPS are born equal too. You sound genuinely delighted with your current creation. :-)
I think you mention somewhere above that you have built and sold some BPSs previously. Would you consider making a replica of your current one, and if so, what sort of price would you be looking for? (PM me if you'd prefer.)

[Sovereign]

Hey Ben can you message me your telephone number there are a few options to discuss, cheers big ears

[Sovereign]

Due to recent busy schedule, tonight is the first time that I have been able to sit down and really listen to my hifi, knowing that I can settle in for the night and not have any other agenda but to listen to music and enjoy.
I'm currently listening to Torrd Gustavsen, Being there. I have never heard the music sounding so relaxed and unforced, yet so detailed. The sound stage is huge and full of emotion. What pleases me the most is the bass depth, it sounds like I am listening right down to the very bottom registers of the recorded music bass floor. I have never heard depths like it. Even when I had other balanced power supplies to the exact same hifi set up, I wasn't hearing what I am hearing now, nothing like it really; I'm really surprised.
Nothing sounds forced of recorded, just real, living, breathing music. Well as close as I have ever heard.
This BPS is the first one I've made without DC blocker, as the toroidial is so quiet it just doesn't need it. I plan to let the BPS properly bed in for a few weeks, then I'll fit the little Nick Gorham device to see if it enables the BPS to perform better and improve the sound of the hifi like it has done every time before. My hunch is that it won't, but we will see.

[brian2957]

Sounds fantastic James . Did you ever get around to making the Klotz /MS Audio interconnects . If not what are you using mate. .

[Sovereign]

Hi Brian,
Yes I did, been running them for a few months now. Really impressed. Slightly bigger soundstage, more solidity and bass is a wee bit deeper with a more distinctive punch to it. Because I'm a wee bit bonkers I have bought another 4m length of the Klotz 500 as I plan to double up or shotgun the interconnects to see if that makes them sound better still, but been too involved in other DIY projects :-)

[brian2957]

Yup they're very good aren't they . Be interesting to see how you get on with shot-gunning the cables as I'm a bit cuckoo myself with things like this :D

[BilliumB]

So James, how's the new balanced transformer sounding now it's had some use?

Cheers. Bill

[Blueflash]

Hi Brian,
Yes I did, been running them for a few months now. Really impressed. Slightly bigger soundstage, more solidity and bass is a wee bit deeper with a more distinctive punch to it. Because I'm a wee bit bonkers I have bought another 4m length of the Klotz 500 as I plan to double up or shotgun the interconnects to see if that makes them sound better still, but been too involved in other DIY projects :-)

I am going to build myself another BPS using a 4KVA toroid from Plitron, but saw your transformer. Are they available in the UK as I would like to send them some specs for a quote.

Many thanks, Alex.

[Sovereign]

Yes mate, PM me

[BilliumB]

Hi Alex

Are you planning a dual transformer setup as before? How much benefit do you think having 2 separate transformers gave you in your current BPS?

Cheers. Bill

[Blueflash]

Hi Alex

Are you planning a dual transformer setup as before? How much benefit do you think having 2 separate transformers gave you in your current BPS?

Cheers. Bill

Hi Bill,

This time around I will use a single 4Kva transformer with as much over winding on the secondary as I can get for the lowest impedance.

I have noticed that the sound quality has been dropping off lately on some days with my first BPS. I am not sure if a dual transformer set up was the way to go. Recording studio`s use a big single BPS 4 or 8 Kva fed from a dedicated consumer unit fed by a 16mm shielded CY cable then from the consumer unit using 10 mm CY cable with a 32A industrial socket for the main BPS input.

I will be redoing my supply as per above then I will use multiple MCB`s on the output of the BPS with captive 2.5 mm shielded cables to the speakers and source. If that still does not work I will have to think about re generation or get my own generator ....

[BilliumB]

So, the thinking is that the lower overall impedance will more than make up for the separate transformers? I was considering possibly 3 balanced transformers - big one for amp and two smaller ones for squeezebox touch and Dac, I'm not sure which is the best way to go. I think some very expensive balanced supplies use separate transformers, so it might be a mistake in going to a single one - perhaps you could utilise the large transformer from your current setup in the new BPS for the source components?

Why do you think the sound quality is dropping off? I've certainly found that a dc blocker in my system has improved system sound quality and also reduced transformer buzzing on my amp, and I've no done anything yet about a new radial or a balanced transformer (but am using a second hand Ben Duncan isolating transformer that I picked up).

Have you considered EI transformers at all? Discussion with Nick Gorham (lurcher) suggested that there might be significant benefits with a good quality EI transformer (reduced sensitivity to dc, higher impedance to higher frequencies so possible filtering benefit), however everyone seems to be using toroids - probably for lower impedance/size issues and reduced external field effects.

As to feed cable, I was recently talking to my electrician who suggested that split concentric cable might be a good idea for a new radial setup to a balanced transformer. This cable has a central live conductor with the neutral and earth conductors 'wrapped around' the live (all in copper) - the benefit over armoured cable would be for an equivalent outside diameter (and hence bending radius and ease of fitting) you could use much larger, lower impedance conductors, the benefit over twin and earth is that the earth conductor is the same size as the live and neutral (whereas with twin and earth, the earth conductor is generally smaller). I have not got around to fitting anything yet, but planning to put something together. How important do you think screening is vs lower impedance on this cable?

Also interested in balanced transformer selection and manufacturer. Have you looked at Toroidy in Poland (I think) who supply what look to be very good quality transformers at a reasonable price. How are you going to decide who to go to?

I'm wondering about additional taps (probably on the primary) to drop the o/p voltage a bit - it's often up around 250v here and I wonder if this is on the limit for a lot of kit. Also a bit concerned about voltage out of a large toroidal transformer under low load conditions - don't you get a higher voltage than at the rated (eg 4kva) loading? What have you found with your current BPS and is this 'low load' voltage (normal operation) an issue?

I'm also looking at earthing issues (I'm on a TT system out in the wilds, so don't have the very serious safety issues associated with PME). Have you done anything interesting in this area? I'm considering a coax earth cable (screen and conductor connected at earth end, conductor only connected at kit end, possibly with a suitable in-line capacitor to limit dc current flow for safety) in addition to normal earth cable, using the coax as a low impedance route to earth for higher frequencies - have not yet decided on best earth termination at the 'earth' end, but wondering, in addition to normal earth spikes, if I could make good use of my old water well and some stainless steel plate! Wondering about WT125 or possibly Belden RG-8/U that some people are now using as speaker cable for the coax. Clearly safety is paramount, then need to be sure on corrosion issues as I don't want to turn the coax earth into some sort of rf detector.

Cheers. Bill

[The Barbarian]

;)

[KONDO]

Hi Bill, how are you doing mate?
Here is an snippet of a thread that that talks a lot about EI transformers.
With the manufacturer of my balanced torodial we tried many different options and ideas to make the torodial perform at its best, some of them are mentioned in the article below, but much of it had no effect on the performance of the toroidial or the sound of my hifi. For instance we spent much time earthing the massive iron core of the torodial to absolutely no effect. We used a half-wave rectified circuit on the shared supply to generate DC on the mains. As expected earthing the core had absolutely no effect on the noise level, in fact all it provides is a potential risk of transformer failure if there is a breakdown of the core insulation, the long and short of it is don't believe all you read.
One of the problems about EI transformers is finding someone that will make them and to your requirements, I have been shown some shockingly poor quality EI transformers from respected names in the industry and the cost would also be very high indeed.
One of the advantages of my balanced 3kVa toroidial is it uses an oversized core (5kVA) to allow a very low flux density for low mechanical noise and the core will probably never saturate, however it weighs in around 27kg, It is continuously rated at 3kVA – I’m not so sure that many of the other so called 3kVA transformers are.

Enjoy the read:

The High-End Mythology of the Toroidal Power Transformer

Some background about how AC power is delivered to residences

When AC power is transmitted across long distances, high voltages increase transmission efficiency and decrease the size and weight of cables needed to carry the power. However, high voltages in the home are quite dangerous so power companies build sub-stations to reduce the voltage levels for residential and light business use. In the US these substations produce three-phase electrical power. Two of the three phases are what is typically delivered to residences. These electrical phases each carry (nominally) 120v rms AC referenced to the neutral wire… all three phases use the same neutral wire/connection/reference. When two phases are supplied to a home or business, you have three wires coming into the house. One of them is neutral. You can connect the two 120v lines to a single electrical outlet and get 208/240 AC volts which is used to power electric dryers, electric ranges and stoves, electric water heaters and other heavy duty electrical appliances.

The thing to remember about this three-phase and two-phase power delivery system is that the loads on each phase must remain balanced or you can get a DC offset to the AC voltage. If power becomes unbalanced in any one phase, a DC offset appears in your AC power. These DC offsets can be transient or long-term. You still get the 120 (nominal) rms volts, but instead of it being a sine wave with positive peaks of about 170 volts and negative peaks of -170 volts, the sine wave is shifted up or down. A +5-volt DC offset would result in positive peaks of 175 volts and negative peaks of –165 volts (approximate voltages). A –6-volt DC offset would result in peaks of +164 and –176. All because the power becomes unbalanced, i.e. there is more load on one or two phases than on the remaining phase. These DC offsets can very dynamic, changing constantly as residences and businesses consume power and turn machinery on and off. Some of the offsets cancel each other out, resulting in no DC offset, but it is not terribly unusual to have transient DC offsets in power delivered to homes in the US. In Europe, depending on the country, some power grids are owned by the country and are actually running at full capacity or even higher than full capacity. In those situations, the potential for DC offset to exist in the power delivered to residences is quite high.

Two transformer types versus DC offset

What does DC Offset have to do with the AC Power Transformer? Everything! You see, the toroid transformer is very intolerant of DC offset being present at the input (primary winding). In fact, if there is any DC at all in the AC power supplied to the typical AC Power toroid transformer, the laminated metal core of the transformer "saturates" and the transformer no longer operates as designed. The result is often high levels of audible noise from the transformer as well as very high levels of noise getting into the power supply of the component.

When an EI-frame transformer is used, the DC offset problem disappears. The EI-frame transformer continues to supply AC power of the proper voltage with all the noise elimination capabilities designed into the EI-frame transformer still operating perfectly. Industry outside of high-end audio has not widely adopted the AC toroid power transformer because of these very problems. The majority of transformers in general use outside of high-end audio equipment are actually EI-frame transformers.

The main advantages of EI-frame transformers over toroidal transformers are: less susceptibility to core saturation from DC offset; air gaps which act as pressure relief valves for high density magnetic flux in the core. Saturation of the transformer core results in an energy build-up in the core of the transformer. Toroid transformers are wound with the core completely covered by the primary and secondary wires. The toroid core is completely isolated from ground. In the EI-frame transformer, however, a significant portion of the core is external and is attached directly to the chassis of the equipment it is installed in. This gives any energy build-up in the core a direct path to the chassis ground of the component. The other advantage EI-frame transformers have is the small air gaps that are present because of the gaps present when the two parts of the frame/core are put together. These air gaps act as pressure relief valves for the magnetic flux in the core further enhancing the transformer’s resistance to core saturation. You can have an air gap in the core of a toroidal power transformer, but this requires machining a gap perhaps 0.020" wide, about the thickness of a sheet of paper. This is not a trivial task and adds significantly to manufacturing cost of the toroid transformer. It is very rare to encounter toroid transformers with air gaps because of the cost and difficulty of machining such a thin slit across the laminations that make the metal donut. These are two of the reasons EI-frame transformers, when properly designed, can provide better audible performance in your audio components.

What’s the catch?

It seems like an EI-frame transformer ought to be a 100% shoe-in for high-end audio components, but there is a "dark side" to the EI-frame transformer. They radiate a rather significant magnetic field from the exposed windings (which are sometimes covered by rounded metal end bells). Low level audio signals such as those found in preamplifiers and gain stages prior to the output stage of amplifiers are small enough in magnitude, that passing them through a strong magnetic field would induce hum in the audio signal. In a preamplifier, this is easy to avoid. Put the power supply in an external enclosure as many preamp manufacturers do, especially those who use EI-frame power transformers. In an amplifier, the answer is to "aim" the transformer correctly. There is a big "shadow" in the magnetic field where the laminated plates are...the magnetic field is strong only outside the end-bells of the transformer. Aim the EI-frame transformer so that the core points towards the audio signal processing and so that the "bells" point to the sides and the radiated magnetic field won’t bother the audio circuits inside the amp. But putting an EI-frame in a tightly packed multi-channel amp or receiver is just about impossible...there just isn’t enough room to stay away from the radiated magnetic field. In those applications, you will almost always find toroidal power transformers. You can jam a toroid right into the middle of five amplification channels and have no problems from magnetic fields. Magnetic shielding for an EI-frame transformer is also possible, but it tends to be expensive and heavy.

Why lavish so much attention on the transformer?

When you look realistically at what happens in an audio component...you know the answer. The signal leaving the component existed only as AC power from your wall outlet only milliseconds before it is on the way to the next component or to the loudspeakers. AC power is the raw material for the output of the component. Make beer, wine, scotch or bourbon with bad water and you get a bad product. Make an audio component with a generic off-the-shelf toroidal transformer and you are going to get a generic sounding component. It won’t be terrible, but it will fail to scale the heights of what is possible.

So an EI-frame transformer is always better than a toroidal transformer?

Hmmm. Good question. I am tempted to say "yes" because really excellent toroidal transformers are apparently difficult, if not impossible to purchase off-the-shelf. However, finding great EI-transformers is no walk in the park either. But in general, being able to connect the core of the EI-frame transformer electrically to the chassis of the component and the air gaps created by the two-piece laminated core give even off-the-shelf EI-frame transformers some performance capabilities beyond what you can get from off-the-shelf toroidal transformers. If you know your magnetic theory (or your transformer vendor does), there are additional winding tricks that give EI-frame transformers 20dB to 30dB more noise rejection than typical toroidal transformers. This is very significant. 20dB less noise is 1/100th of the noise of the "reference" level. 30dB less noise is 1/1000th of the "reference" noise level. You might be able to find a toroid manufacturer who could do the custom winding, add an air gap and attach a ground wire to the core. You would pay two to three times the cost of an off-the-shelf toroid to get that level of performance. But you sure won’t find a toroid off the shelf that performs as well.

I’ve been lucky enough to speak to a variety of high-end manufacturers about AC power, transformers, and component design. Conversations with Richard Vandersteen, Mike VansEvers and Emil Rotar (Warner Imaging) have been most illuminating. Years ago, I never really gave power transformers much thought. The assumption was that because so many high-end manufacturers were using toroidal power transformers, toroids must be demonstrably superior. Imagine my surprise when I started noticing that many of the best sounding components I’ve experienced in my system have all had EI-frame transformers. Coincidence? I might have thought so before finding out so more about power and transformers. Now I’m convinced that cleverly done EI-frame transformers could/should give us better sounding high-end audio components.hi,enjoyed that,noel keywood done a feature in hi fi world,about the benefits,great stuff

[docfoster]

I recently asked Sov' to build me one of his MTBPS ("Mother Trucker Balanced Power Supply" as he Christened it).
I know that Sov' has similar tastes to me (NVA, Colin Wonfor class A power amps, speakers with large bass drivers, etc.,...) and that he, like me, had found benefits from using an AG1500 regenerator. Also, he seemed like a thoroughly good human being when I met him at a get together at RFC towers a year or so ago, so I felt confident to ask him to build me a MTBPS. It seems that my confidence was not misplaced!
The Sovereign MTBPS arrived a couple of days ago, and is really making a difference.
I haven't chopped and change much yet. Initially just plugged in the TOCA SECA power amp and left the TFS media PC feeding of a AG1500 regenerator. Now I have both coming off the MTBPS and things are even better.
The presentation is just so smooth and sweet across the frequency range - high frequencies (eg cymbals) are just sublime, and bass seems to have more depth and definition and clear texture (kick drums sound and feel even more like kick drums!). Vocals on hi-res tracks are being better exploited. Quite tear jerking in fact. Amazing and wonderful how extra degrees of realism bring additional emotional impact...

http://i1003.photobucket.com/albums/af151/docfoster/MTBPS_zpss4vhajhh.jpg

(It's the big black unit sitting on the floor.)

The BPS has surpassed my expectations. I'd hoped it would bring what the AG1500 did, but without the slight compressive effect on the bass - something that I only notice when using the AG1500 with the TOCA; with previous amps (NVA TSS and Inca Tech ID25) I could not detect any negative effects. In actuality Sov's MTBPS has improved on everything, simply made things yet more realistic.

Good work sir! :-)

[Ali Tait]

Are you still using the AG1500?

If so, try taking it out a feed everything from the BPS.

[docfoster]

No indeed Ali, the AG1500 is taking a break.
I may try it with the projector tomorrow.

[Ali Tait]

Yes, I found it better feeding everything direct from the BPS.

[Sovereign]

Really glad your happy with it Ben, it made a significant difference in my system.
I've been messing around with balanced power for about six years and this toroial that I now use is significantly better then anything I have heard before; and dead quiet in it's own right.
I really enjoyed making it and quite happy to make more.

[Ali Tait]

Yes, can be one of the most cost effective upgrades to make IMHO.