2. misc-audio
  3. 1319998342

Balanced vs standard power

Searching for an Isolation Transformer 10 or 15kva. I have the budget for Equitech but I'm sold only on the benefits of isolation and not "yet" on the benefits of balanced power unless one is recording live musicians. My rationale is that most if not all audio components are not designed with balanced power so they may be optimal performers with standard power and lesser performers with balanced power. Until someone does a side by side comparing isolated balanced to isolated standard power we may never know.

I have read the various threads regarding isolation transformers from Equitech 10wq, MGE Topaz etc, and the Euqitech stereophile review. For larger non-balanced options there is Ultra K 600 with K factor correction and triple shielding from the Controlled Power Company. They range from 5 to 25 kva.

I think supersizing Isolation transformers for audio is not well recognized yet because no one has done the necessary review/ comparisons to determine the performance curve of Isolation Transformer size to Audio Performance. Although Isoclean advocates the use of two of their Isolation Transformers for each piece of equipment. Maybe they're trying to tell us something or just sell more transformers. 10 KVA is "plenty" for my system according to Martin at Equitech, but "plenty" is not quantified enough to convince me, so I 'm leaning toward the 15kva on the Ultra K 600 from Controlled Power about $4000 vs Equitech 15 kVA at upwards of $14000.
natan6355

Showing 8 responses by almarg

almarg's avatar
almarg
9,670 posts
 
11-06-11: Dan_ed
I think I get it, but see what you guys think about this. My two amps hum because they reference ground at a slightly different potential, and that creates a small current between them when they are on the same circuit. The balanced connections assure that all components plugged into it reference the same ground so there would be no potential between the amps.
Hi Dan,

See the second paragraph of page 2 of this Jensen paper. A key factor in how much hum reduction would result is the stray capacitance within each component between each of the two ac input lines and chassis, particularly stray capacitance in the power transformers of the components, and how similar or dissimilar those stray capacitances happen to be. So while I would expect there to be SOME improvement, the amount of improvement figures to not have a great deal of predictability.

Best regards,
-- Al
almarg's avatar
almarg
9,670 posts
 
Jim -- Good points, but in no. 2 shouldn't "lead" be "lag"?

Re the 220uf, that corresponds to an impedance (capacitive reactance) of 12 ohms at 60 Hz. That will certainly pull a lot of current through the transformer, but I suppose it's within reason given the 10 kVA rating of the DU-10, and the 3.4 kVA load. Assuming the breakers and wiring can handle it all, of course!

Best regards,
-- Al
almarg's avatar
almarg
9,670 posts
 
Just having one of the other amps plugged in but not turned on or connected in any other way is enough to cause the hum.
Wow! That's different!

When the amp that is not connected to anything except power is turned on, does the hum get better, worse, or stay the same, compared to when it is turned off?

A possible explanation that occurs to me is that, referring to Figure 1 of the Jensen paper that Jim and I both linked to (great minds think alike :-)), when you plug in the unused amp, and it is in the turned off state, you are placing CPS4 (as defined in Figure 1) of the unused amp in parallel with CPS4 of the amp or amps that is/are being used. That would worsen Vy (as shown in Figure 2) for the amp(s) that is/are in use, thereby worsening the ground loop situation between the amp(s) being used and the preamp or whatever is driving it or them.

If the amount of hum decreases when the unused amp is turned on, it would add credibility to that theory, because doing so would place CPS3 of the amps in parallel, as well as CPS4, which would to some degree reduce the effects of the increase in CPS4.

I presume, btw, that the preamp to amp interconnections are unbalanced, as I would expect these kinds of effects to be much less significant with balanced interconnections.

Best regards,
-- Al
almarg's avatar
almarg
9,670 posts
 
Hi Dan,

I read through the Equitech paper, and it strikes me as excellent. However, it also strikes me as not inconsistent with the comments on balanced power in the Jensen paper, which is to say that there would probably be significant hum reduction if you used a balanced power arrangement, but not necessarily a reduction that is fully satisfactory.

From the Jensen paper:
It is also very unlikely that the two capacitances, CPS1 and CPS2 or CPS3 and CPS4, would be exactly matched in any piece of equipment. Mismatch ratios of two to one are common.... Although intuitively attractive, [the balanced power] approach can completely cancel interchassis currents in a system of three or more devices only in the case where each of the devices had such matched capacitances. This would be an extremely rare occurrence.... 10 to 15 dB hum reductions ... would be more routinely achieved.
From the Equitech paper:
If any aspect of the circuit is applied OR LOADED [emphasis added] in an unbalanced manner, noise will appear in the ground.... On the average, 16dB improvement in background noise has been noted.
In other words, any difference between the stray capacitances CPS3 and CPS4 in your particular amplifiers (which is unpredictable) will limit the amount of hum reduction balanced power would provide.

In considering the 10 to 16 db numbers, keep in mind the rule of thumb that a reduction of 10db is subjectively "half as loud."

Best regards,
-- Al
almarg's avatar
almarg
9,670 posts
 
Perhaps Jim or Al can comment on the issue of UL/CSA standards and shunting/grounding as compared to IEC. I gotta step back and listen when those guys are talking.
Thanks for the compliment, NG, but those kinds of standards are not among my areas of expertise. Jim?

Best regards,
-- Al
almarg's avatar
almarg
9,670 posts
 
For the record, I just noticed what appears to be a boo boo in the paper by Mr. Whitlock of Jensen Transformers.

In figure 2, which I referred to earlier, the formulas for Vx and Vy should have CPS1 and CPS3 in their numerators, respectively, rather than CPS2 and CPS4. That follows from the fact that the voltage divider effect between the two capacitances in each component will be dependent on the reactance of the capacitors (1/(2piFC)), not on the capacitance values themselves.

Therefore I retract my previous theory about CPS4 of the unused amplifier causing Vy, and hence the hum level, to worsen.

Best regards,
-- Al
almarg's avatar
almarg
9,670 posts
 
03-11-12: Jea48
Norm has the transformer configured for balanced power. That cuts the KVA rating of the xfmr in half. The xfmr maximum continuous FLA rating is 5 KVA.
Good point, Jim. Agreed.
03-11-12: Jea48
Just wonder if the poor PF, caused by too much capacitance on the AC line, is doing anything to the power xfmrs and switching power supplies of his audio equipment. You are the EE here.... What say you?
Good question, which I had been wondering myself. I don't know the answer.
03-11-12: Norm
Would you please elaborate how increasing the capacitance to the secondary side of the transformer may cause voltage to "lag"?
The voltage across a capacitor cannot change instantly. It changes in response to the accumulation or depletion of the charge it is storing, which in turn is proportional to the integral of current, over time. As a consequence of that, for a sinusoidal AC waveform of a given frequency, and assuming an idealized capacitor model, and since the integral of a sine wave is an inverted cosine wave, the voltage across a capacitor will lag the current by 90 degrees, or 1/4 cycle.
03-11-12: Norm
Also, how much additional current is the transformer "pulling" by adding 220uf.
10 amps, based on the assumption that 120 volts is placed across the 12 ohm impedance which the capacitor has at 60 Hz.
03-11-12: Norm
Jim, are you saying that by adding 220uf to the circuit I have effectively lowered my 10kVA xfmr rating to 5kVA?
No, the capacitor has nothing to do with that, although the capacitor significantly increases the amount of current the transformer has to supply, and significantly increases how much of the 5 kVA capability is being utilized. Referring to the data sheet for the transformer, each of the two secondary windings is rated to handle 41 amps. Presumably you have them connected in series, with nominally 60 volts appearing across each winding, and 120 volts across the series combination. 120 x 41 = 4920 VA, or a little under 5 kVA.

Also, to make sure I'm envisioning your setup correctly, I'm assuming that the two primary windings are connected in series, and are fed by a single-phase 120V line and breaker rated at 40 amps or more. Is that correct?

Best regards,
-- Al
almarg's avatar
almarg
9,670 posts
 
03-11-12: Norm
Would you care to take a guess what is happening to the electronics(tube gear) with voltage and amperage being 1/4 cycle out? The overall results with the additional capacitance is a big improvement, especially LF. Am I doing any damage to my xfer and/or electronics by the additional capacitance?
Hi Norm,

As I indicated earlier I don't really know, as my technical background is not specifically relevant. I will say, though, that it does strike me as plausible that there could be SOME risk over the long term, but I don't have much if any feel for how significant it might be.

One mechanism that occurs to me by which a problem could result is if a resonant frequency that is formed by the interaction of the capacitor and the inductance of the wiring and/or the transformer were to happen to coincide with a frequency component of either a transient on the power line or a harmonic distortion component that may be present on it (perhaps created by the amplifier's power supply, or by something external to the system), the resonance could conceivably boost the magnitude of that transient or distortion component to very high and potentially damaging levels.

As I say, though, I don't have any particular instinct for how significant that or other risks may be.

Best regards,
-- Al
More to discover