Here is a circuit to roll your own EFT generator: http://article.sciencepublishinggroup.com/pdf/10.11648.j.jeee.s.2015030201.26.pdf
sorry - did I miss a costed reference schematic that we could all build and try? We do need to be comparing the same thing...
Hard to be sure from the angle of the photo, but not sure I like the look of those crimps. Good crimping tools for that size are not pennies, but worth the money if you do a lot of them.
Oh and at risk of introducing some Engineering into this, went hunting for the reference to Self doing testing injecting noise into power supplies. Couldn't find that, but did find this https://www.edn.com/Home/PrintView?contentItemId=4208718 which is an analysis of PSRR vs frequency for +ve and -ve rails in his preferred amplifier topology.
The thing that worries me still about the brute force big iron power conditioning solutions is that, as least over this side of the pond consumer units have 30mA RCDs. You can't get too agressive with the filtering before you exceed that.
None of this matters if you are building a listening room/studio from scratch and can have a seperate consumer unit with dedicated spurs and maybe commando sockets on the wall, but few of us have that luxury.
The thing that worries me still about the brute force big iron power conditioning solutions is that, as least over this side of the pond consumer units have 30mA RCDs. You can't get too agressive with the filtering before you exceed that.
None of this matters if you are building a listening room/studio from scratch and can have a seperate consumer unit with dedicated spurs and maybe commando sockets on the wall, but few of us have that luxury.
So the wheel turns.....
A prior time RNM mentioned the subject he did identify his current probe. It is one of the most complicated designs used to get the nice flat high bandwidth it provides.
Seems most here don't understand power supply rejection ratio. It is not a single number no matter how often just one number is quoted. The second often ignored parameter is frequency. The assumption that as long as it is respectable at audio frequencies ain't really right. Higher frequencies can shift the operating point of bits of circuitry.
Then there is the issue of how stray electromagnetic interference gets into the circuit. Easy to understand the electro part. As soon as the EMI enters the chassis it can couple to not just internal wiring but into the case and even into the ground plane.
Thus even if you could get PSSR of 100 dB + the interference would still be much greater in the final product.
Yes I do use a buzzer box to create AC power line noise. I also use an audio power amplifier to power devices under test. It is capable of producing more than the current maximum line voltage for the National Electrical Code standards and at better the 25 amps.
Not yet touched on again is that just a power transformer introduces distortion both into the device being powered and back into the AC power line. Large transformers like those in power amplifiers can easily distort the power line.
Just for humor it is not practical to build an AC generator that produces a perfect sine wave. There have to be gaps in the armature to insert the windings. That results in distortion. The next issue about power lines is distribution. The effective impedance of the AC outlet powering your equipment is contràry to popular nonsense quite low. Almost always under an ohm. That is due to the standard of no more than a 5% voltage drop from no load to maximum load at the main power entrance and the same to the outlet. (Main load typical 200 amps and outlet 20 amps.)
The other factor in keeping the impedance low is the the load impedance is in parallel with the source impedance. Unfortunately the load impedance is no longer just incandescent lamps which even they are non-ohmic loads. (Derfy if you do a voltage versus current plot you will learn what non-ohmic means. The plot will not be a straight line very much exponential.)
Now the RNM filter last I looked at behaves in my opinion a bit differently than most and quite possibly than from the design intent. It not only is a filter but it unbalances common mode EMI into differential EMI which is easier to filter effectively.
Just a bit more about dynamic range. Human hearing spans the comfortable range of about 135 dB. Allowing for did fences in the source volume and reproduction control (AKA the volume knob) gives a possible range of 200 db. That in voltage is 100 ppb or in power .01 ppt!
ES
A prior time RNM mentioned the subject he did identify his current probe. It is one of the most complicated designs used to get the nice flat high bandwidth it provides.
Seems most here don't understand power supply rejection ratio. It is not a single number no matter how often just one number is quoted. The second often ignored parameter is frequency. The assumption that as long as it is respectable at audio frequencies ain't really right. Higher frequencies can shift the operating point of bits of circuitry.
Then there is the issue of how stray electromagnetic interference gets into the circuit. Easy to understand the electro part. As soon as the EMI enters the chassis it can couple to not just internal wiring but into the case and even into the ground plane.
Thus even if you could get PSSR of 100 dB + the interference would still be much greater in the final product.
Yes I do use a buzzer box to create AC power line noise. I also use an audio power amplifier to power devices under test. It is capable of producing more than the current maximum line voltage for the National Electrical Code standards and at better the 25 amps.
Not yet touched on again is that just a power transformer introduces distortion both into the device being powered and back into the AC power line. Large transformers like those in power amplifiers can easily distort the power line.
Just for humor it is not practical to build an AC generator that produces a perfect sine wave. There have to be gaps in the armature to insert the windings. That results in distortion. The next issue about power lines is distribution. The effective impedance of the AC outlet powering your equipment is contràry to popular nonsense quite low. Almost always under an ohm. That is due to the standard of no more than a 5% voltage drop from no load to maximum load at the main power entrance and the same to the outlet. (Main load typical 200 amps and outlet 20 amps.)
The other factor in keeping the impedance low is the the load impedance is in parallel with the source impedance. Unfortunately the load impedance is no longer just incandescent lamps which even they are non-ohmic loads. (Derfy if you do a voltage versus current plot you will learn what non-ohmic means. The plot will not be a straight line very much exponential.)
Now the RNM filter last I looked at behaves in my opinion a bit differently than most and quite possibly than from the design intent. It not only is a filter but it unbalances common mode EMI into differential EMI which is easier to filter effectively.
Just a bit more about dynamic range. Human hearing spans the comfortable range of about 135 dB. Allowing for did fences in the source volume and reproduction control (AKA the volume knob) gives a possible range of 200 db. That in voltage is 100 ppb or in power .01 ppt!
ES
Last edited:
The main benefit is ... how much it can change the quality of your system, listening to it.
Sometimes, a little improvement in the numbers gives a big one is the perceived result.
A high definition photography of this thread. But look closer, there is only 7 or 8 hot pixels to filter (like all hot pixels they are noisy and we only see them) .
I wondered what pleasure they could find in such a negative attitude: We all have the same school books, so what they say is without any interest. The only plausible answer is that they are just there to pretend they have the biggest one.
On my side, if I stay here, it is because, sometimes, one have an original idée that I find clever or... original ;-).
And some (they are not so many) are so deep in their knowledge that I can add new things to my little knowledge, reading them. (I remember a thread about cables impedances).
The other reason is it is reassuring to see that other designers, from the other part of the world, came, despite a very different life and experience, to the same conclusions than you on something.
... Under the hoot of naysayers, but what does it matter!
All those critics about power conditioner is totally ridiculous. (Scott, you disappointed-me). Everybody knows that an audio gears can only be as good as its power supply. That any parasitic HF added to the signal will add distortion and noise.
There are several ways to reduce them as much as possible. Designing amplifiers with a high PSRR is the beginning. Designing PSU with low noise and high rejection is the following.
When you have done this the best you can, I see two things that can still be done to improve a system further . To reduce the parasitic signals added to the signal references (ground loops and AC likeages) and reduce the parasitic signals that enters in your PSUs.
Those who stop on the way would do better to catch their breath than to whistle those who continue their journey. Everyone is free to put the bar where he wants.
Last edited:
And so the discussion goes round and round.
Previously batteries were mentioned and IME this is a VERY good way to
verify your mains connected power supply. Most SS circuits are easily
powered from +-24V / 18V or 12V.
Just power a particular bit of gear from batteries and listen to it. It will tell
you pretty quickly whether you need to do some work on the power supplies.
That's what we did. They also remain a good reference point to compare as
you try different transformers, regs etc etc.
T
Bill, you have the $20.00 recipe from RNM, if you were to get out of your armchair you could whip one up in next to no time, actually less time than you spend debating this foreign to you concept.
If you hear no difference then so what, you wagered $20.00 or the Pommie equivalent.
If you do hear a difference then you are on your way to experimentation and discovery....hint, it ain't just the electrical filtering properties that changes the sound.
Dan.
Last edited:
Why do you put words in my mouth? I said no proof has been presented that a power conditioner will improve ANY amplifier. That is no one can design a PA with an adequate PS. I think it's time to filter myself out.
Last edited:
I thought the old fashioned way was to measure with your ears. Aren't you worried about your switchers introducing noise (I know they're silent, as is Jan in this thread lately, can't say I blame him)
Yes, there is always the possibility to improve circuits and PS because we are a technical group. However, most people are not able to modify existing gear - The general public. What can you do to make most all of their gear perform better with less intrusions of the smps generated noise etal in their homes on their ac lines? A power conditioner works for them.... the masses/consumers.
The thorough and most complete offering to the public is the -7000. I suggested a cheap and quick way to improve the situation without digging into the gear -- make an inexpensive isolation transformer with filter for the technical types (DIY) to try.
At this point in time it is not necessary to do a R&D program from the basic beginning. I showed the problem exists in all homes and all consumer level gear. So, all that needs to be done is DIY build and listen.
THx-RNMarsh
The thorough and most complete offering to the public is the -7000. I suggested a cheap and quick way to improve the situation without digging into the gear -- make an inexpensive isolation transformer with filter for the technical types (DIY) to try.
At this point in time it is not necessary to do a R&D program from the basic beginning. I showed the problem exists in all homes and all consumer level gear. So, all that needs to be done is DIY build and listen.
THx-RNMarsh
- Status
- Not open for further replies.