Balanced Power

[breeze]

I've been reading a lot about balanced power units, such as Equitech, to reduce or eliminate noise. Instead of 120VAC, it provides -60V and +60V from a torroidal transformer's secondaries, with a center tap that goes to ground. When I found this schematic of a unit made by Transcendant, I decided to build it. I have all the parts, and I'm just about to begin, but I just found this schematic by (or named) Felicia. I see the Transcendant has caps from the hot secondaries to ground, while the Felicicia has caps running between the hots. The Felicia also has a fuse on each side of the secondaries. I'm concerned about shock and other hazards. Any thoughts?

[wizard333]

Would a normal PT even run on balanced DC?

[Albert Kreuzer]

Would a normal PT even run on balanced DC?

Uh, no...
You'd need a special DC transformer for that

But I think he's talking about balanced AC. Interesting - has anyone tried this?

[wizard333]

Instead of 120VAC, it provides -60V and +60V from a torroidal transformer's secondaries

If you have a constant +60v on one secondary output, and a constant -60v on the other secondary output, is that not DC? Sounds like what he's talking about is what you'd see on the B+ and B- rails in a solid state design.

AC is alternating current; whats alternating?

[Albert Kreuzer]

I think he means "in phase" and "out of phase", not + and - like with DC.
The whole thing looks like a balanced line out, but for mains voltage...

Again: Has anyone tried that?

[wizard333]

Do you mean this:

Instead of one line that operates at a 120v rms sine wave relative to ground, two lines that operate at 60v rms sine wave relative to ground but phased such that there is a 120vrms potential between the 2?

Thats not exactly what he said; he said +60v/-60v which usually indicates a DC voltage relative to ground. Otherwise the +/- are not really necessary are they. It could be he was just using +/- to indicate that they are out of phase? If he meant two lines of 60vAC out of phase so there is a 120vrms potential between the two, thats different.

[Albert Kreuzer]

It could be he was just using +/- to indicate that they are out of phase? If he meant two lines of 60vAC out of phase so there is a 120vrms potential between the two, thats different.

Yes, I think that's what he meant.

[breeze]

This is a schematic for an AC power supply, to plug your gear into. Sorry, my mistake for posting in "guitar amp" theory and design. It's just theory and design.

It uses an Avel torroidal 60+60 transformer.The secondaries are (+60VAC), and (-60VAC), and the center tap is ground. You still get 120VAC at between the hots, but like a humbucking pickup, or noise cancelling headphones maybe, any noise that is present in the line, is cancelled by the secong leg (-60VAC). THe trannys are only about $100, but complete units are much more, so I thought I'd build one.

My concern, as I said, is with the caps on the secondaries, going to ground. Would that present a shock hazard? Could I build the unit without the caps? I guess they are for filtering(?). Thanks for your replies.

For more info on the principle, Google "balanced power", Equitech, Transcendantsound,B-P-T, Tortech, etc.

[breeze]

.... Yeah, they are AC voltages that are out of phase with each other.

[Merlinb]

My concern, as I said, is with the caps on the secondaries, going to ground. Would that present a shock hazard? Could I build the unit without the caps? I guess they are for filtering(?). Thanks for your replies.

Using caps like that is very common in mains filters, and not considered a risk. You must use X or y class capacitors though. You can read about them here: Line-Filter

You don't have to use the caps, but it won't be as effective as a filter without them.

[breeze]

Thanks Merlinb. I think class Y would be a wise choice. I found these Schurter filters 5500-2042, and their spec sheet shows Cx and Cy values. I don't know if they just used x and y as variables (and could just as well have used a and b), or if they are class x and y (probably). I'm going to hook this up without any filter, first, then try the Schurter filter, just to compare. I have the trans in the chassis, and just need a lazy afternoon (hard to come by) to drill some holes, and hook everything else up.

Someone corrected me on this. This is not a power supply, but a power conditioner.

[dai h.]

has anyone ever tried one of the (commercial) units from equi tech, furman, etc. on a guitar amp? I'd be interested to hear how they work (how effective). If the object is to reduce gtr. amp noise though, IMO good layout, grounding, power supply design should in most cases (unless it's some insanely high gain amp or something) prove adequate to deal with noise. Also, I remember reading (on the Equi Tech site IIRC) that theirs was a specially wound iso. PT (for better symmetry than usual?) so I wonder if trying to reproduce the (apparently positive noise-reducing) results if something special is required. On another site, there was someone reporting they did this with a 100/50-0-50V PT (100V here in Japan) and it worked (according to the guy) but haven't tried it myself. Also from what I understand (one way) AC power can come into a building as 100-0-100 (and I think also the same in the States but 120-0-120 and travels to various AC outlets as 0-100/120) and since that appears to be "balanced" if that would work (stepped down to half).

[Redhat]

Equi Tech 3RQ recommended!

Hey,

I have an Equi Tech 3RQ 30 Amp model for my home studio and it was totally worth the bucks.

60 Hz hum is virtually gone except for some that's noticable in my Marshall JVM410 when in OD1 Yellow + and OD2 Green +.

An iso xfrmr here and there fixes that.

I don't know if I'd want to try building one since it is somewhat more than just a huge toroidal transformer. It has a digital electronics section for computers etc...

One major thing to beware of is using a UPS power supply or connecting any unfiltered circuits to it since its ground reference is 60 Vac away from the ground of something just plugged into the wall

[dai h.]

cool. thx for the report. I've read on a Japanese studio website where the engineer was using one of these types of balanced power iso transformers (different maker) and the guy claimed that it gave more consistency to the audio over the day (which if true might be more/less noise on the AC as more/less equipment gets turned on and off?). Typically they seem to be pricey, though one thing I remember reading (couple years ago) was that Equi Tech sold extras/surplus units for less than usual on their site, so possibly a source for anyone wanting one (assuming they still do).

[Redhat]

geek spk

According to my calculations, the orginal schematic shows a 4ea .47uF caps. They combine to form an impedance of 5643 ohms across 120 Vrms. So you "burn" (21 mA rms)^2 times whatever the resistance of the cap's leads plus the cap's ESR.

Essentially the caps slightly increase the output's power factor angle (kind of like the Furman PF linear conditioner). This has the advantage of "correcting" for highly inductive loads such as power amps' transformers.

My Equi Tech works by the method suggested by the above post. Its input is 240 VAC (from the clothes dryer circuit). Each side of the toroidal transformer then steps down the voltage "exactly" half. That's what costs so much. To get truly balanced power, you must use one transfomer and not two so that the flux is common to both sides. With two separate flux paths, the impedance balance is gone.


And its the impedance to gnd that you really want to balance, not just the voltage in order to increase the CMRR.