Hello All
The utility of power conditioners and cords is well established, particularly for those living in cities or in industrial areas, but it is more difficult, at least for me, to make a case for that utility for those living in rural areas. I am wondering if there are any evaluations a DIYer can conduct that would indicate the need for such equipment?
Cheers,
ceulrich
The utility of power conditioners and cords is well established, particularly for those living in cities or in industrial areas, but it is more difficult, at least for me, to make a case for that utility for those living in rural areas. I am wondering if there are any evaluations a DIYer can conduct that would indicate the need for such equipment?
Cheers,
ceulrich
To reveal distortion on the AC line waveform, you can use a step-down voltage transformer method to display the waveform on an oscilloscope.
See the article here:
https://wiki.analog.com/university/...ltage waveform a,0 to 5 volt range of the M1k.
Also see this diyAudio thread: https://www.diyaudio.com/community/threads/how-to-measure-mains-waveform-safely.143907/
See the article here:
https://wiki.analog.com/university/...ltage waveform a,0 to 5 volt range of the M1k.
Also see this diyAudio thread: https://www.diyaudio.com/community/threads/how-to-measure-mains-waveform-safely.143907/
You can go through the route of testing system-level results as I did with a DIY AC-Filter and various power supplies (including SMPSes vs Linear Regulated) in the chain:
- 1kHz Sine wave generated by Audacity,
- played via Audirvana to my USB DAC,
- captured back via an M-Audio interface,
- recorded into Audacity for FFT analysis
I've read 10 years of reviews of power conditioners and chords and have yet to see a rigorous study which demonstrates that the power supply output impedance or noise is materially affected in any way by a cord or conditioner. I am not a "denier", just want to see the data and await some numbers. (I have a sine "regenerator" which operates at 90kHz to yield 150V, 250V and 350V for measurement of vacuum tube noise.)
I would suggest to John Atkinson (who has re-wired his audio system with a direct line to the breaker-box) that he connect his AP2722 to the power supply rails (of a quality amplifier) and measure the difference in noise before and after.
Even simpler for those without an AP2722. Connect your power amplifier to an outlet which also has a hair dryer on it! Put your scope on the power supply rails and appraise us of the "before and after" results with some pictures.
Hello All, thanks for the responses.
Galu, thanks for the info. I looked over the Wiki Analog document and it is quite interesting, and if I can get some of the required hardware, will be helpful.
YashN, my Keysight DSO has an FFT function, so I can do something similar to your suggestion.
mkane77g, I think a link is broken in your post.
jackinnj, I share your reservations. My thinking goes like this; specialty power cables, and to a lesser degree, power conditioners, have been on the market for a couple of decades, if they did not do something positive, I don’t think the market would keep growing, which it seems to be doing. Similar to the hair dryer idea, I have read that light dimmers put a lot of noise into the power lines. As an educational experiment, I was thinking about making up a test jig with an incandescent lamp and a dimmer that can be used as a noise source.
Cheers,
ceulrich
Galu, thanks for the info. I looked over the Wiki Analog document and it is quite interesting, and if I can get some of the required hardware, will be helpful.
YashN, my Keysight DSO has an FFT function, so I can do something similar to your suggestion.
mkane77g, I think a link is broken in your post.
jackinnj, I share your reservations. My thinking goes like this; specialty power cables, and to a lesser degree, power conditioners, have been on the market for a couple of decades, if they did not do something positive, I don’t think the market would keep growing, which it seems to be doing. Similar to the hair dryer idea, I have read that light dimmers put a lot of noise into the power lines. As an educational experiment, I was thinking about making up a test jig with an incandescent lamp and a dimmer that can be used as a noise source.
Cheers,
ceulrich
So too has the market for multi-kilobuck interconnects and speaker cable - doesn't mean there's anything to it beyond the well-understood strictures of engineering. Never mind the whole range of increasingly esoteric nonsense like audiophile ethernet switches and magic "quantum" doodads. BS "alternative health" products are also more successful now than they've ever been (just look at Paltrow's Goop range of insanity) - popular does not equate to legitimate or effective.
Power conditioners can help if your AC is full of HF hash, especially if your PSU is simple (e.g. CRC). You'd be better off with a straightforward but well-engineered filter than by spending $$$ on an "audiophile" power conditioner. Having said that, a well-designed PSU with the right regulators and noise mitigation should be all but immune to AC line noise.
If you do proceed with your light dimmer experiment (which I encourage you to do), see how much of the introduced noise exists before and after the device's power supply.
Last edited:
He may have put context around that.
Instead of anecdote, try power transformers as audio transformers. EI with primary and secondary on separate legs will fall-off at 400Hz or 2kHz. EI or toroid pri over sec will usually exceed 10kHz. More "hash".
Definitely. I didn't yet have my DSO, which is a Siglent that also has some FFT capabilities, so capturing the audio and analysing in Audacity did the trick at the time.
Some of the graphs are on audiophilestyle in a long thread on a DIY AC Filter Box etc...
I should have quoted the reference:
https://www.diyaudio.com/community/...a-thoroughly-modern-tube-phono-preamp.163570/
The raw supply is even more embarrassingly straightforward. It's built into a separate box using only the finest Radio Shack $10 cabinet (figure 8). I used mostly on-hand and surplus parts, but have given part numbers for some currently-available units. The HV power transformer was a lovely surplus item, potted and shielded. The one in the parts list is amazingly well-priced and should work fine. I'd avoid toroids since they are superbly efficient at coupling mains noise into the circuit.
A nice trick- use a separate transformer for the heater supply rather than the usual extra winding on the HV transformer. The reason for this is that, even using high speed soft recovery diodes, rectifier hash from the HV supply will be coupled to the heater winding, providing unwanted rattle. This is a minor effect, but we're handling microvolt signals…
I've been reading about DC offset on the mains voltage supply caused by the likes of electronic dimmers and computer power supplies.
It is claimed that less that 500 mV of DC can be sufficient to cause toroidal transformers to become saturated, which adversely affects sonic performance and may cause audible mechanical vibration.
https://www.isol-8.co.uk/dc-on-the-mains
Is this a real phenomenon or just marketing hype?
It is claimed that less that 500 mV of DC can be sufficient to cause toroidal transformers to become saturated, which adversely affects sonic performance and may cause audible mechanical vibration.
https://www.isol-8.co.uk/dc-on-the-mains
Is this a real phenomenon or just marketing hype?
Like the others, I wish I had some detailed measurements to reveal, but I don't. I didn't believe power cords could make a difference until I built my first Pass amp. I don't know if it matters for Class AB amps or other electronics, but I could clearly hear an improvement going from 18 gauge to 16 gauge. Further improvement at 12 gauge was there, but harder to hear. I don't buy anyone's expensive power cord.
I have also seen some benefit from adding a Furman pro-audio filter and surge protector. Not so much in sound quality, but in reduction in outside interference. For example, I used to be able to listen to the tune played by my paper shredder, but it doesn't bother my sound system at all once the Furman got into the system.
If you are interested in finding out if you have DC on your mains, of course, be very careful as it can kill you. I have a multi-meter rated to 600V AC and DC. The thing is, you can just measure DC directly at your wall outlet, but you won't get the right answer. I tried that and measure between 100 and 200 mV DC. Wrong. Try something like the link below. I found that I had no measurable DC at all.
https://www.diyaudio.com/community/threads/how-to-measure-dc-offset-on-the-mains-voltage.350171/
I have also seen some benefit from adding a Furman pro-audio filter and surge protector. Not so much in sound quality, but in reduction in outside interference. For example, I used to be able to listen to the tune played by my paper shredder, but it doesn't bother my sound system at all once the Furman got into the system.
If you are interested in finding out if you have DC on your mains, of course, be very careful as it can kill you. I have a multi-meter rated to 600V AC and DC. The thing is, you can just measure DC directly at your wall outlet, but you won't get the right answer. I tried that and measure between 100 and 200 mV DC. Wrong. Try something like the link below. I found that I had no measurable DC at all.
https://www.diyaudio.com/community/threads/how-to-measure-dc-offset-on-the-mains-voltage.350171/
A lot seems to depend on specific local conditions. Here where I am its on top of a hill that looks across Sacramento Valley all the way to tops of the Coastal Mountain Range. Around 100mi as the crow files. There is lots of RF coming in from the West. Some of it gets picked up on the AC lines which act like antennas. Doing an FFT of vacuum cleaner motor noise is one thing, using step-down transformers is another. How about EMI/RFI at several hundred MHz or a few GHz?
Based in experimental evidence, I know I need to put prototype dacs in a steel file server chassis as one factor that reduces low level, yet still slightly audible signal-correlated noise from radiated emissions. Also know that a HIGH QUALITY power conditioner helps with conducted EMI/RFI and ground noise. In addition, found that a lesser quality power conditioner did nothing useful at all.
Would only suggest that if someone wants to measure to find out if any change in measurements depending on local EMI/RFI conditions, probably best to reduce other noise sources as much as possible first. Then use a high resolution FFT and look for noise skirts width on spectral peaks that changes depending on DUT presence or absence. Also, sometimes going after low level signal-correlated noise problems can be sort of analogous to peeling away layers of an onion. Often bigger problems are found to have masked lesser problems. Keep optimizing fixes for bigger problems and then recheck for smaller problems, would be one possible strategy.
EDIT: Regarding sensitivity of torrid transforms to AC line DC imbalance, properly designed torroids don't have to be designed to be so close to saturation. Its just cheaper to make them that way, and happens to be possible to make them that way due to the geometry. Magnetic path length around a circle (torroid) tends to be rather short in relation to the winding window area. Standard E-cores, for example, IMHO have rectangular winding windows for a reason. It would be expected to help to keep from being wound in a way that could allow easy saturation. Magnetic path length is relatively long in relation to winding window area.
Based in experimental evidence, I know I need to put prototype dacs in a steel file server chassis as one factor that reduces low level, yet still slightly audible signal-correlated noise from radiated emissions. Also know that a HIGH QUALITY power conditioner helps with conducted EMI/RFI and ground noise. In addition, found that a lesser quality power conditioner did nothing useful at all.
Would only suggest that if someone wants to measure to find out if any change in measurements depending on local EMI/RFI conditions, probably best to reduce other noise sources as much as possible first. Then use a high resolution FFT and look for noise skirts width on spectral peaks that changes depending on DUT presence or absence. Also, sometimes going after low level signal-correlated noise problems can be sort of analogous to peeling away layers of an onion. Often bigger problems are found to have masked lesser problems. Keep optimizing fixes for bigger problems and then recheck for smaller problems, would be one possible strategy.
EDIT: Regarding sensitivity of torrid transforms to AC line DC imbalance, properly designed torroids don't have to be designed to be so close to saturation. Its just cheaper to make them that way, and happens to be possible to make them that way due to the geometry. Magnetic path length around a circle (torroid) tends to be rather short in relation to the winding window area. Standard E-cores, for example, IMHO have rectangular winding windows for a reason. It would be expected to help to keep from being wound in a way that could allow easy saturation. Magnetic path length is relatively long in relation to winding window area.
Last edited:
I was a skeptic too until I built my own power cable based on a Chris Venhaus design, with some personal modifications.
Big difference with my Panasonic MOSFET amp.
I also tried it with my 4K Blu-Ray player. Slight visual difference if any.
I built my own AC Filter box starting from an old Lukasz Fikus design with a few personal modifications as well. Large difference in the Digital audio setup with my iFi DAC.
On a whim, I also tried the Filter on the TV + Modem + 4K Blu-Ray setup, not prepared to see any difference. I was amazed to see less bleed on the TV (not an OLED), and hence seemingly better contrast perception. In particular, even after calibration, the TV veers towards a slight orange-ish hue on skin colour, as if actors were overly made-up. This artifact is gone when the AC Filter is in.
There's almost nothing you cannot build and verify for yourself nowadays, so people asking for others to provide measurements should instead put on a builder's hat and make their own observations and measurements.
As mentioned, some of my System-level measurements are on audiophilestyle in that AC Filter Box megathread.
I was also a skeptic about different USB cables being able to make an audible difference. Experiments around that and injection of clean linear power vs computer SMPS vs external SMPSes are also in that thread - it's from 2017 or something. Inspiration for the cable initially came from iFi and another company selling USB cables with separated power and data lines... Note here that I am not saying USB cables always make a difference - I think nowadays DACs are better isolated, but at the time, with that particular DAC, it did.
Now, it would be interesting to directly monitor the AC side of things. Dropping down the voltage and probably some additional amplifiers like differential or what not could be interesting.
The internal FFT of my DSO isn't as precise as a Spectral Analyser, but it can provide some inkling about where high-frequency peaks are.
A more precise view of what bands are noisy provide a blueprint for targeting these with several filters.
Just another experience / subjective POV for Sound quality:
Every Power conditioner I have ever tried or assembled (not that many) slug dynamics with the benefit of reduction in some noise induced artefacts (sounds calmer smoother, cleaner) BUT overall I always went back to straight into the wall
For the money to SQ ratio - although at the expense of inconvenience - is a dedicated separate supply. Split the incoming mains after the meter before the normal consumer unit and take a separate spur to a dedicated consumer unit for the music/listening room only - ideally close to where the sockets will be.
I have had good results with a single MCB supplying a single socket on 30A cable to lower impedance. You can then always consider loosing fuses in each plug (not the unit/product) with a 15A non fused plug and MCB to match the connected unit's current consumption and to protect the cable (which is the purpose of the fused plug in the UK as I understand it).
For the mains cable I have always been able to hear an advantage with Belden 83803 as have many others BUT it is a pig to use as it's so damn stiff.
Every Power conditioner I have ever tried or assembled (not that many) slug dynamics with the benefit of reduction in some noise induced artefacts (sounds calmer smoother, cleaner) BUT overall I always went back to straight into the wall
For the money to SQ ratio - although at the expense of inconvenience - is a dedicated separate supply. Split the incoming mains after the meter before the normal consumer unit and take a separate spur to a dedicated consumer unit for the music/listening room only - ideally close to where the sockets will be.
I have had good results with a single MCB supplying a single socket on 30A cable to lower impedance. You can then always consider loosing fuses in each plug (not the unit/product) with a 15A non fused plug and MCB to match the connected unit's current consumption and to protect the cable (which is the purpose of the fused plug in the UK as I understand it).
For the mains cable I have always been able to hear an advantage with Belden 83803 as have many others BUT it is a pig to use as it's so damn stiff.
- Status
- Not open for further replies.