Regenerating voltage vs. passive smoothing on AC power?

[Boomzilla]

I'm about to test that theory...

[Boomzilla]

Jan 10, 2016 18:51:17 GMT -5 monkumonku said:

Hey but I am using three of my Class D amps in bridged mode so those are monoblock, right? So Class D Monoblocks rule!!!

DDDs?

[audiobill]

Jan 8, 2016 20:42:49 GMT -5 hemster said:

Jan 8, 2016 20:38:41 GMT -5 novisnick said:

Yes,,,,,,,,D's rule!!!

Wait! That's Double-Ds!! 

Stereo?

[mgbpuff]

I think you have a pretty good handle on the problem, Boomzilla. Actually, I think that the Class D amps are less susceptible to the pop you refer to because of the sizable output filters they have. I do use a PS Audio regenerative supply on one system I have, but I only plug source devices into it. My power amps are far too large so I simply use a CMX-2 for them . The transformer is good to eliminate line transient disturbances and ground problems but does nothing for longer term line sags or surges unless it is of the voltage harmonic regulating type such a Sola CVS type.. The best solution would be to have a gigantic uninterruptable supply with a pure sine wave output but the cost would be prohibitive.

[Boomzilla]

It's called a Tesla PowerWall.

[KeithL]

Let me throw in a bit of clarification here... (big surprise, right?)...

The AC power you get from the wall is an AC sine wave (you've all seen a picture of one by now). Since the line frequency here in the US is 60 cycles, that means that the sine wave repeats itself 60 times per second - so, for each sine wave, it goes positive for 1/120 of a second, then negative for 1/120 of a second, and this whole thing repeats sixty times per second. (For folks with 50 cycle power, those are 1/100 second positive, 1/100 second negative, repeated 50 times per second.)

Inside your equipment:

1) A transformer changes the voltage of this AC sine wave if necessary.
2) A rectifier makes the AC into DC; either by simply cutting half of it off, or by "flipping" half of it around so it's all either positive or negative.
    (that rather lyrical description of what happens is actually more or less accurate - although the details are omitted).
3) The output of the rectifier, which now looks like a series of ripples, but all of them either positive or negative, is filtered to make smooth DC.
4) As an extra step, the resulting DC may be further smoothed with a regulator (but usually not for the power rails of the output stage of a power amp - because regulators are inefficient).

Now, there are also a limited number of things that can go wrong with the original sine wave - and each has the potential to produce different problems.....

1) The frequency could vary. You can ignore this one for several reasons. First off, most power supplies simply don't care; even varying the frequency by quite a bit won't actually change the DC coming out after it's filtered to any significant degree. And, even though some things, like some turntable motors, actually do use the frequency of the AC signal for some purpose, and so would be affected if it were to vary, the fact is that the AC line frequency almost never varies significantly anyway (unless you have your own generator - or live on a small island somewhere). The power company is very careful to ensure that the generators powering your neighborhood remain right on speed. (If the generators at different power stations were to drift apart, even a little bit, the result would be severe equipment damage - or worse.)

2) The voltage could vary slowly - this covers anything between variations over a large fraction of a second to variations between time of day or day of week. A slow variation in the overall AC line voltage would simply result in a variation in the unregulated DC voltage coming out of a typical power supply. To an amplifier, this would mean that, instead of the output stage running on +/- 60 volts, it might end up with +/- 57 volts, or +/- 63 volts. With virtually all modern amplifiers, including ours, the only result of this would be a slight variation in the maximum power available. The gain of the amplifier is fixed by the component values in the circuit, so, if the amplifier was running at 10 watts, that wouldn't change at all. However, with the rails at 57 volts, it might start clipping at 480 watts, while, with the rails at 63 volts, it might start clipping at 520 watts. This is insignificant and would not be noticeable. (A power regulator or regenerator would prevent this; a filter would not; and a simple isolation transformer would not.)

(Note that some equipment, especially some tube equipment, and some unusual designs, may in fact be affected by slight changes in line voltage. However, because line voltage does vary from day to day, and from place to place, most well designed equipment - including ours - is designed NOT to be sensitive to changes in line voltage.)

3) The voltage waveform itself could become distorted. It's relatively common for the tips of the sine waves to become squished because many power supplies draw more power at the peaks of the waveforms than during the rest of the cycle. The main result of this would also simply be to reduce the DC voltage coming out of the power supply slightly. Again, the biggest effect this would have on most amplifiers would be to slightly reduce the output level at which they clip. (A power regenerator would prevent this; a regulator might or might not, depending on precisely what type it was; an isolation transformer wouldn't affect it.)

4) There could be sudden sharp variations in the line voltage (which would look like a sharp step in the waveform on an oscilloscope); there could also be various signals and little blips at higher frequencies "riding on" the sine wave. Collectively these are all considered to be "noise". Noise can affect equipment by "sneaking into the circuitry past the filtering" and actually causing audible noises mixed in with the music, or by causing the circuitry to malfunction and produce some type of distortion. The power supply inside most equipment is designed to block most types of noise, but certain types of noise can sneak in, and some power supplies do a better job of this than others. A good quality line FILTER should help to eliminate these types of problems. (Some types of noise respond better to certain types of filters, and there is a huge variation in cost between various filters. However, unless you have a known problem with a certain type of noise, it doesn't make sense to spend a lot of money on an expensive filter to solve a problem you don't have.)

5) Surges. A power surge is a short but very large jump in the line voltage. Surges should be very infrequent, but may well cause serious equipment damage. (A surge suppressor is intended to protect your equipment from destruction. If your equipment is hit by a surge large enough to damage it, then only a heavy duty surge suppressor will stop it. (And you shouldn't be worried about whether you hear a slight pop while it's doing it or not.)

The dips you notice when a heavy piece of equipment switches on generally last for several cycles of the sine wave (a dip lasting for a single cycle, or part of a cycle, probably wouldn't even be noticed). Because the dips you're experiencing last for more than part of a cycle, the only thing that will prevent them would be some sort of regulator/regenerator. (A FILTER won't prevent them. Without being able to regenerate the waveform, a filter can only remove noise that occupies a small part of a single cycle in duration.) A power regenerator, or a regulator with boost capability WOULD prevent them. A simple 1:1 isolation transformer won't have any effect on them. However, a REGULATED transformer, like a Sola transformer, would probably prevent them. The operant issue is that a normal transformer doesn't actually store energy, and so cannot supply energy to "fill in" a dip.) However, since you're not noticing any audible effects from them, that suggests that the power supplies in your equipment (mainly the filter capacitors) are doing their job and supplying enough power to "cover the gap".

In short, your garage sale transformer should provide some benefit as a noise filter (being an inductor); it probably won't do much if anything to eliminate voltage dips and sags, but they're not something that you really need to worry about anyway.

Most low cost surge suppressors act by shunting surge current to ground. Therefore, it makes sense to attach the surge suppressor directly to the outlet (giving it the shortest path to ground), with the transformer after it. (There are a few arguments for doing it the opposite way and, honestly, it probably doesn't matter much, but that's the way I'd do it.

Jan 8, 2016 13:56:29 GMT -5 Boomzilla said:

To the best of my knowledge, my AC power is fairly steady, from a voltage standpoint. The only dips are when a large appliance (central air unit, for example) kicks on / off. I've never heard any artifacts from the momentary dips / peaks.

The local Magnolia store's trying to sell me a Furman power unit that regenerates the input voltage to provide a steady output voltage. In other words, if the line voltage sags or peaks, the Furman unit ensures a steady output voltage via a large capacitor bank that absorbs excess voltage or makes up when needed.

I'm thinking that since my line voltage is relatively steady other than momentary peaks & dips, I'd achieve the same smoothing via a passive 1:1 ballast transformer. Since the inductance of the transformer resists rapid changes in voltage, and since the device is rated for a full 15 amperes, I'm thinking that the $10 garage sale transformer will provide me with 99% the benefits of the $3K+ Furman unit. Additionally, all RF on the AC line will be blocked by the transformer as well.

Are my assumptions technically sound? If so, is there any harm in using BOTH a ballast transformer AND a surge suppression strip? In which order should the devices be wired - Outlet to ballast transformer to surge strip OR outlet to surge strip to ballast transformer? Why?

Thanks - Boomzilla

[Boomzilla]

Thanks, Keith. Good information. Because my house power has internally generated noise on it (Ethernet over AC), I think that I WILL use my yard-sale ballast transformers, but only on the DAC and preamplifier (where signal voltages are lowest, and where the noise is most likely to be at a significant level. I'm suspecting that the internal AC rectification and smoothing of the DAC and preamp power supplies is more than sufficient to suppress the noise, but since I have the transformers anyway...

Because of its higher inline impedance, I see no reason to pass the power amplifiers' AC through the transformers. If anything, the transformers might limit the current needed by the amps.

Thanks again for the info!

Boom