• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

Thoughts on power conditioners?

Joined
Aug 2, 2018
Messages
64
Likes
68
Location
Campbell, CA
I worked as an engineer at Panamax for 16 years, and performed quite a few leakage tests. The limit for plug-in residential equipment is 0.5 mA (thought to be the limit of "sensibility") through a simple "body model" (1500 Ohm resistor || 0.22 µF cap). Leakage can also become audible as mains hum if it causes a voltage difference between grounds in earth-referenced audio equipment. That would be a LOT of leakage.

It's more likely that mains interference in audio signals is happening inside of the equipment, due to stray ac fields. Sometimes there can be electrostatic interference outside of the equipment if low-voltage (line-level) analog audio wiring is run closely adjacent to mains wiring in the wall; I've seen installers co-locate A/V cables with the ac when running in-wall wiring to video monitors, and still wondered why they could see "hum bars". In the digital world, though, the signals are more tolerant of this.

The NEC has a recommendation for <5% drop in a circuit branch, but it actually isn't a requirement. There are conductor sizes that are required per the outlet ampacity (say, 14 AWG for 15 A), but if the run is long enough, you'll need a heavier gauge to achieve the NEC recommendation. Not all electricians take voltage drop into account when speccing wire.

This is especially evident with large copy/imaging machines in the far reaches of office environments. The fuser inrush current can cause a temporary undervoltage on the line, and sometimes microcontrollers in the office machines will go into shutdown if the supply voltage sags enough.

And yes, the earth conductor is only allowed to carry "temporary" currents; its primary intention is to trip a breaker in the circuit branch if a metal enclosure is accidentally energized.
 

Speedskater

Active Member
Joined
Mar 5, 2016
Messages
282
Likes
132
Location
Cleveland, Ohio USA
The only grounding is on the chassis and is to be connected to Audio_Ground and not to Earth.
Without knowing where you live, the meaning of those words could be different, but:
a] what is the "Audio_Ground" ?
b] by "Earth" do you mean "Planet Earth" ?

The chassis of a three wire component, needs to be connected to the Safety Ground/Protective Earth. While these is continuity between the SG/PE and Planet Earth, the chassis is not connected to the Planet Earth system.

A modern two wire component (that is one with a double insulated enclosure and has the square within a square safety symbol) is not connected to the SG/PE.
 

nick-v

New Member
Joined
Sep 14, 2018
Messages
4
Likes
1
I find it quite difficult to differentiate between the different types of power conditioners.

I use a Sound Application CF-X in my main system, a Furman IT Reference 7i in my bedroom system, and a Yulong P18 in my desktop system.

I can't find a picture of the inside of the Sound Application unit,

But, here's the Furman IT Reference 7i:



and the Yulong P-18:



They're clearly very different designs. Of the two with pictures of the internals, it looks like the Furman is probably the superior design, but it would be nice to cut through the marketing babble and have a real understanding of what the specific type of unit is doing.

The Sound Application unit retailed for around $5K, but I'm not really sure exactly what it does (I got it slightly used for much less $$)
 

RayDunzl

Major Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
7,067
Likes
3,391
Location
Riverview, FL
it would be nice to cut through the marketing babble and have a real understanding of what the specific type of unit is doing.
The Sound Application CF-X seems to be a line filter... surge protection/noise reduction, similar to the Yulong P-18.

The Furman appears to be an isolation transformer that outputs balanced power, along with filtering and surge suppression, but is limited to 7 amp putput. That's around 840 watts.

I have a device similar to the Furman in my system, but with a higher power rating.

1537935506637.png
 

trl

Addicted to Fun and Learning
King of Mods
Joined
Feb 28, 2018
Messages
672
Likes
435
Location
Iasi, RO
Without knowing where you live, the meaning of those words could be different, but:
a] what is the "Audio_Ground" ?
b] by "Earth" do you mean "Planet Earth" ?

The chassis of a three wire component, needs to be connected to the Safety Ground/Protective Earth. While these is continuity between the SG/PE and Planet Earth, the chassis is not connected to the Planet Earth system.

A modern two wire component (that is one with a double insulated enclosure and has the square within a square safety symbol) is not connected to the SG/PE.
1) Grounding - inside a Hi-Fi equipment we might find digital ground, analogue round, case ground and earth ground (the actual ground find in our mains outlet). Some manufacturers are interconnecting analogue with digital grounds and case ground is directly connected to Earth ground, some others are floating analogue and digital grounds and others are interconnecting all of them etc. It's about safety, but also about hum noise that can get injected from mains ground...it all depends on the design. Most speakers amps I've seen were using 2-pole mains plugs and "audio ground" screw connected to the chassis on the backside of amplifier; that means it should be only connected to your Hi-Fi chain's Ground only, not to 120/230V Earth/Grounding.
2) Earthing - Earth protection circuit coming from your mains, usually <3-4 Ohms.

P.S: https://www.analogictips.com/faq-ground-rules-earth-chassis-signal-ground-faq/

P.P.S.: My ASR profile has been filled in from the very first day: I live in Iasi, RO, EU.
 
Last edited:

trl

Addicted to Fun and Learning
King of Mods
Joined
Feb 28, 2018
Messages
672
Likes
435
Location
Iasi, RO
The Sound Application CF-X seems to be a line filter... surge protection/noise reduction, similar to the Yulong P-18.

The Furman appears to be an isolation transformer that outputs balanced power, along with filtering and surge suppression, but is limited to 7 amp putput. That's around 840 watts.

I have a device similar to the Furman in my system, but with a higher power rating.

View attachment 15906

Could you please get some graphs from mains-input, then mains output? Also, can you use ARTA and RMAA (or similar) to get some FFT, SNR, THD with and without the Furman device? i did that myself in my home, but like I said before, my mains sine is looking too good to see any differences.

Thank you,
Raul.
 

RayDunzl

Major Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
7,067
Likes
3,391
Location
Riverview, FL
I don't have the quiet accessories to make fine electrical measurements (nor do I have the Furman).

My mains sine is rather flat-topped, with a little bit of HF spikes and fuzz, and the output of the Equitech is similarly flawed, when observing on an oscilloscope.
 

RayDunzl

Major Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
7,067
Likes
3,391
Location
Riverview, FL
Here's a mains picture from a while back, voltage on Hot and Neutral:

170Vpk flat-top sine on the Hot wire and about 4Vpk on the Neutral

1537939957721.png
 

RayDunzl

Major Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
7,067
Likes
3,391
Location
Riverview, FL

nick-v

New Member
Joined
Sep 14, 2018
Messages
4
Likes
1
The Sound Application CF-X seems to be a line filter... surge protection/noise reduction, similar to the Yulong P-18.

The Furman appears to be an isolation transformer that outputs balanced power, along with filtering and surge suppression, but is limited to 7 amp putput. That's around 840 watts.

I have a device similar to the Furman in my system, but with a higher power rating.

View attachment 15906
I found this on the Sound Application unit:

"
This filter uses no transformers or other current limiting devices. It is designed to provide an extremely low impedance capacitive shunt to ground for EM and RF energy over a bandwidth from 10 kHz to 2.5 GHz. That's an upper limit of 2.5 Gigahertz - clear out to the microwave range! I'm not aware of any competing models that can claim a bandwidth beyond the megahertz range. Noise reduction of up to -60 dB is claimed across this bandwidth. Designer Jim Weil, to put it bluntly, is fanatical about construction detail and parts quality. He's the kind of guy that has apparently never read the book "Don't Sweat the Small Stuff," because he constantly worries over small details. The CF-X represents the culmination of 12 years worth of development effort and tweaking. It's interesting to note that Jim uses the video quality of his Sony XBR TV to gauge the effect of various part substitutions. He maintains that visual sensory input in this case is more refined and reliable in making fine judgments. In what follows, I can only give you a glimpse of his attention to detail that is essential in making a parallel filter work effectively over a bandwidth extended beyond 2 GHz.
All of the circuitry is housed in a machined aluminum case. The internal power distribution is via 4N (99.99% pure) 1/8-inch copper bus bars and 6N (99.9999% pure) copper wire. All internal wiring is point-to-point, and is rated at 100 amps. The unit can safely handle 6 kilowatts of power, well beyond the rating of a standard 15-amp circuit breaker. The filter stages are preceded by spike and surge protection circuitry. A custom high-speed magnetic circuit breaker provides an interrupt time of less than one millisecond for high-current transients - much faster than the thermal breakers commonly used in competing products. A bank of eight varistors follows for voltage spike suppression and protection. A 23-step stage filter bank of high-quality capacitors, mainly film and foil types, is used to provide a near-zero impedance pathway for both transverse and common mode noise. The primary emphasis, however, is on transverse mode EMI and RFI rejection. For effective filtering, it is necessary to cover the desired bandwidth with multiple capacitor sizes, as each capacitor's high-frequency self-resonance limits its upper range of applicability. Did I mention the several pounds of Cascade Audio dampening compound used internally, or that all soldering is performed in an inert argon gas environment to eliminate the formation of surface oxides?
The CF-X features a total of six Hubbell duplex outlets, and is shipped with the MAC Delta power cord ($5,000 retail), or optionally without a cord ($4,200 retail). A long and hard break-in protocol is specified in the installation instructions. The initial break in period is about 300 hours and can be expedited by plugging high-current devices such as vacuum cleaners into the unit. The break-in of the high-voltage caps will take much longer. Expect slight sonic improvement for many months to come."

The Furman is good enough for my bedroom system. I'm only running a 50" TV, a Dali Kubik One Active Soundbar, a Chromecast Audio and my HDTV cable box through it. My Martin Logan subwoofer plugs directly into the wall, so 7A is lots for that system I think.

I was considering picking up the 15 or 20 amp version for my main system (Furman IT Reference 15i or 20i), and was wondering, would I just plug it into the Sound Application unit?

I also run a couple Bybee AC Quantum Purifiers in my system (which I'm sure is another can of worms).
 
Joined
Aug 2, 2018
Messages
64
Likes
68
Location
Campbell, CA
Hmmm...sounds good to test a power conditioner. :)
To have an effective instrumented test, you'll need a consistent ac power source, conducted-emissions power noise source, and audio load.

The first item (consistent, clean power source) is very difficult to achieve, so I'd recommend that the power noise source be extreme enough to swamp ambient EMI. A bank of SMPS (for PC applications) with some large power resistors works well. A power resistor in series with the mains can also simulate the resistance of a longer circuit branch, and will increase the influence of the conducted emissions. Also, the mains circuit branch should not have any other loads on it (such as a refrigerator).

[Edit]: A vacuum cleaner would also be a valid data point as a noise source.
 
Last edited:

trl

Addicted to Fun and Learning
King of Mods
Joined
Feb 28, 2018
Messages
672
Likes
435
Location
Iasi, RO
Joined
Aug 2, 2018
Messages
64
Likes
68
Location
Campbell, CA
Well, there's that. :)
That's why we used SMPS for "real-world" load testing. Rectified incandescent lamp loads were also fun for making toroids buzz from DC offset.
That said, we never settled on a protocol to directly measure the ac noise influence on audio signals.

Our sales guys did run around with AudioPrism Noise Sniffers, which basically were non-linear RFI demo boxes that would show stark differences when using ac line filters (even a simple pi filter with 18 µH series inductance would easily kill the detected "noise" from an ac outlet), while using an unloaded cell phone charger as a noise source.
 

Blumlein 88

Major Contributor
Joined
Feb 23, 2016
Messages
6,499
Likes
6,233
I used pi filters on my phone line back when internet was dialup. My older neighborhood at the time had old loading transformers in the phone lines. Various noise issues made my dialup bad. I eventually made some headway getting better speed. I could have tried it with streaming audio, except you can't stream much when your good speeds are 40-56 kbps instead of only 28.8 kbps.
 

SIY

Technical Expert
Technical Expert
Joined
Apr 6, 2018
Messages
2,876
Likes
4,975
Location
Phoenix, AZ
That said, we never settled on a protocol to directly measure the ac noise influence on audio signals.
There's the key, right there.

Look at the output of an audio device plugged into a normal outlet. You know, like equipment reviewers do in their measurements. When testing electronics, I don't plug them into power conditioners. If memory serves, neither does Amir. Nor Atkinson.
 
Joined
Aug 2, 2018
Messages
64
Likes
68
Location
Campbell, CA
There's the key, right there.

Look at the output of an audio device plugged into a normal outlet. You know, like equipment reviewers do in their measurements. When testing electronics, I don't plug them into power conditioners. If memory serves, neither does Amir. Nor Atkinson.
I’m open to suggestions. Note that I’m not advocating any end-use solutions here, other than stout copper mains. I don’t have any skin in this game. As for test protocol, a “normal outlet” should be characterized before its alternatives are dismissed. Establish baseline environmental criteria, and maintain it for comparative tests, or overwhelm it to an acceptable degree of certainty with controlled stimulus. What is the intent of this thread?
 
Top Bottom