ISOTEK EVO3 Aquarius Power Conditioner Review

[amirm]

I'm saying your testing is incomplete and results are therefore unproven scientifically (they might be reasonable theoretically but the experiment design and measurements are incomplete).

You haven't presented any such need or theory of why it would make a difference. By your logic, I could test 100 devices 100 different ways and you would still come back with the same argument. You don't live your everyday life that way, having your water supply tested every hour by a lab to feel comfortable to be consumed. Yet here, you demand the same thing for the same absurd reason.

This class of device has been tested using many methods and a number of devices and tests. If there were any chance of what you saying being true, we should have seen some evidence of it. Or even a hint of it. We don't get that because engineering says there can't be differences. There is only so much we need to do to prove an impossibility.

From here on, the responsibility is on the shoulders of manufacturers of these products. They have all the time in the world to do this testing and show difference. I suggest you write to them and keep arguing with them on why they don't show such measurements. It is their job. It is not mine to exhaustively search for evidence of their product working.

 

[DonR]

Knowing what we know about power supply design and the inherent filtering capacity of its components, it's enough to show an example of how it doesn't affect performance and extrapolate from that.

 

[amirm]

I'm saying your testing is incomplete and results are therefore unproven scientifically (they might be reasonable theoretically but the experiment design and measurements are incomplete).

But you are not understanding the "science." The science part requires you understanding how an audio device power supply works. If you understood that, then you wouldn't be asking me for more tests. That understanding would start with the fact that a power supply's job is to filter AC and create DC. And that audio circuits routinely have PSRR above zero so they reject variations on power supply to boot. And that these devices don't filter down to mains frequency anyway. Scientists don't waste time on topics they understand to be completely untrue. By definition they understand their field.

I on the other hand deal with lay people who hear terms like "AC filtering" and immediately jump to improvements that it must deliver, damn the lack of science and data to back that. So please don't tell me about "science." We are dealing with people with complete lack of. So I show measurements that get any engineer to say, "of course" at my expense hoping you understand this message. To the extent you are dismissive of that as well, then that is that and I suggest moving on than trying to create FUD.

 

[LEFASR160]

It's pretty obvious that power quality doesn't affect sound quality, so what's the point in messing about further?

@TrevC So far, based solely upon the experiments on ASR, it is not obvious that AC quality cannot affect the audio quality of the electronics because so far no one has tried all the different types of power supplies out there. I would design my experiments so as to test the hypothesis that AC quality is irrelevant by testing all known power supply designs (there are only a handful of different designs that are used today) that are variously incorporated in different audio gear and THEN I would make my conclusions. That's the scientific method. Also, the testing should show measurements of the Power Supply DC output in each case: pure low THD sine wave AC Mains vs dirty AC Mains with various levels of THD/noise. And I'd like to see the audio quality test results using test signals of various frequencies from 20 hz to 20khz. I do not believe this was ever done.

 

[LEFASR160]

But you are not understanding the "science." The science part requires you understanding how an audio device power supply works. If you understood that, then you wouldn't be asking me for more tests. That understanding would start with the fact that a power supply's job is to filter AC and create DC. And that audio circuits routinely have PSRR above zero so they reject variations on power supply to boot. And that these devices don't filter down to mains frequency anyway. Scientists don't waste time on topics they understand to be completely untrue. By definition they understand their field.

I on the other hand deal with lay people who hear terms like "AC filtering" and immediately jump to improvements that it must deliver, damn the lack of science and data to back that. So please don't tell me about "science." We are dealing with people with complete lack of. So I show measurements that get any engineer to say, "of course" at my expense hoping you understand this message. To the extent you are dismissive of that as well, then that is that and I suggest moving on than trying to create FUD.

@amirm [This is addressed to you too, and a repeat....] So far, based solely upon the experiments on ASR, it is not obvious that AC quality cannot affect the audio quality of the electronics because so far no one has tried all the different types of power supplies out there. I would design my experiments so as to test the hypothesis that AC quality is irrelevant by testing all known power supply designs (there are only a handful of different designs that are used today) that are variously incorporated in different audio gear and THEN I would make my conclusions. That's the scientific method. Also, the testing should show measurements of the Power Supply DC output in each case: pure low THD sine wave AC Mains vs dirty AC Mains with various levels of THD/noise. And I'd like to see the audio quality test results using test signals of various frequencies from 20 hz to 20khz. I do not believe this was ever done.

 

[LEFASR160]

But you are not understanding the "science." The science part requires you understanding how an audio device power supply works. If you understood that, then you wouldn't be asking me for more tests. That understanding would start with the fact that a power supply's job is to filter AC and create DC. And that audio circuits routinely have PSRR above zero so they reject variations on power supply to boot. And that these devices don't filter down to mains frequency anyway. Scientists don't waste time on topics they understand to be completely untrue. By definition they understand their field.

I on the other hand deal with lay people who hear terms like "AC filtering" and immediately jump to improvements that it must deliver, damn the lack of science and data to back that. So please don't tell me about "science." We are dealing with people with complete lack of. So I show measurements that get any engineer to say, "of course" at my expense hoping you understand this message. To the extent you are dismissive of that as well, then that is that and I suggest moving on than trying to create FUD.

@amirm Sometimes experiments lead to surprising results and confound the theorists. (Case in point: Michelson Morley experiment) Limiting your testing to show only results that support what you believe to be true based upon your own depth of understanding is not "science".

 

[LEFASR160]

But you are not understanding the "science." The science part requires you understanding how an audio device power supply works. If you understood that, then you wouldn't be asking me for more tests. That understanding would start with the fact that a power supply's job is to filter AC and create DC. And that audio circuits routinely have PSRR above zero so they reject variations on power supply to boot. And that these devices don't filter down to mains frequency anyway. Scientists don't waste time on topics they understand to be completely untrue. By definition they understand their field.

I on the other hand deal with lay people who hear terms like "AC filtering" and immediately jump to improvements that it must deliver, damn the lack of science and data to back that. So please don't tell me about "science." We are dealing with people with complete lack of. So I show measurements that get any engineer to say, "of course" at my expense hoping you understand this message. To the extent you are dismissive of that as well, then that is that and I suggest moving on than trying to create FUD.

@amirm Different power supplies have different power supply rejection ratios. And even if audio circuits themselves "routinely" have PSRR's that filter out ripple and noise, it does not mean that all of them do, right ? So it is entirely possible for some people with certain power supplies and certain audio circuits and, may I also say, certain dirty AC lines, an AC power source with less THD+N might be beneficial to those people in terms of (1) audio quality; and, (2) component Mean Time to Failure (both within and downstream from the PS). I'm just saying it's possible; I'm not saying it is a foregone conclusion. That is why I suggested that you design your experiments to include different types of power supplies and their respective gear.

 

[amirm]

@amirm Different power supplies have different power supply rejection ratios. And even if audio circuits themselves "routinely" have PSRR's that filter out ripple and noise, it does not mean that all of them do, right ? So it is entirely possible for some people with certain power supplies and certain audio circuits and, may I also say, certain dirty AC lines, an AC power source with less THD+N might be beneficial to those people in terms of (1) audio quality; and, (2) component Mean Time to Failure (both within and downstream from the PS). I'm just saying it's possible; I'm not saying it is a foregone conclusion. That is why I suggested that you design your experiments to include different types of power supplies and their respective gear.

You have shown no evidence of that possibility when the power devices perform no cleaning in audible band.

 

[LEFASR160]

You have shown no evidence of that possibility when the power devices perform no cleaning in audible band.

@amirm Your experiments showed no cleaning in the audio band but for only one or two power supplies in one or two electronic components. You don't need a power conditioner or power regenerator to do the testing if you use your 9801. How do you know that another power supply in another electronic gear, a power supply that has a terrible a PSRR, will do the same ? You haven't shown that. Sorry, no cigar.

 

[LEFASR160]

You are wrong there as well for the same reasons. Full dashboard with detail was provided unlike your claim that it was just "SINAD." In addition, I created highly distorted AC waveforms and still showed no difference: https://www.audiosciencereview.com/...ac-distortion-noise-on-audio-equipment.25501/

I suggest not continuing to shoot from the hip causing me to point out obvious things like this. Please read the reviews carefully to fully understand their nature and not make empty excuses for these products.

No, I'm not wrong. You're setting up a straw man argument here. I think I said that (1) you only tested one signal input: a 1khz signal; and, (2) one or at most two different gear with different power supplies. You did not test all the different power supply designs. You in fact have the equipment necessary so as not to need to test any power conditioner or power regenerator---you know that. In fact, you used your 9801 to simulate near perfect AC and "dimmer switch" AC and used that instead of a passive conditioner or power regenerator. You did not outline the different power supply designs that are available with different gear and different PSRR's and you did not do your experiments on each one in order to prove that AC purity is of no value. Also, you did not address the possibility of a cleaner AC allowing a longer MTTF for components within a given PS and also downstream from it. Your science is incomplete.

 

[LEFASR160]

You have shown no evidence of that possibility when the power devices perform no cleaning in audible band.

Is this evidence ?

The link below shows different engineering design techniques for power supply filtering:

"No matter whether you’re plagued with ripple, noise, or beat frequencies, these techniques will allow you to reduce the noise of your power system to levels low enough for your needs. When you get to 18-and 24-bit measurements and digital-to-analog (DAC) conversions, getting the power-supply rails as clean as possible is essential to obtain the available performance out of the ICs you’re using."

So it appears to me to depend upon what power supply you're testing. Different power supplies and different circuit designs have different PSRR's:

3 Ways to Reduce Power-Supply Noise

Get noise out of your power supply with a multi-prong approach. Filters, bypassing, and post-regulation all can help achieve that goal.

www.electronicdesign.com

 

[DonR]

Is this evidence ?

The link below shows different engineering design techniques for power supply filtering:

"No matter whether you’re plagued with ripple, noise, or beat frequencies, these techniques will allow you to reduce the noise of your power system to levels low enough for your needs. When you get to 18-and 24-bit measurements and digital-to-analog (DAC) conversions, getting the power-supply rails as clean as possible is essential to obtain the available performance out of the ICs you’re using."

So it appears to me to depend upon what power supply you're testing. Different power supplies and different circuit designs have different PSRR's:

3 Ways to Reduce Power-Supply Noise

Get noise out of your power supply with a multi-prong approach. Filters, bypassing, and post-regulation all can help achieve that goal.

www.electronicdesign.com

The vast amount of noise in a power supply is generated by the power supply itself, particularly switching supplies. The signal you feed it with is of absolutely no consequence as it is filtered, deconstructed and then reconstructed.

 

[BlackTalon]

Also, you did not address the possibility of a cleaner AC allowing a longer MTTF for components within a given PS and also downstream from it. Your science is incomplete.

I am curious how you expect him to test for MTTF...

And you do understand, I hope, that Amir tests products people ship to him to test. So if you want some additional testing done by all means send him the equipment. And if you want some tested to component failure let him know that as well.

Be part of the solution. There are already enough armchair quarterbacks out there (myself included, sometimes)

 

[LEFASR160]

@amirm Your experiments showed no cleaning in the audio band but for only one or two power supplies in one or two electronic components. You don't need a power conditioner or power regenerator to do the testing if you use your 9801. How do you know that another power supply in another electronic gear, a power supply that has a terrible a PSRR, will do the same ? You haven't shown that. Sorry, no cigar.

The vast amount of noise in a power supply is generated by the power supply itself, particularly switching supplies. The signal you feed it with is of absolutely no consequence as it is filtered, deconstructed and then reconstructed.

@DonR

Is the 120 hz ripple, mentioned in the article, generated by the PS itself ? Answer: NO...it’s due to the 60 hz sine wave fed to it.

The link below shows different engineering design techniques for power supply filtering:

"No matter whether you’re plagued with ripple, noise, or beat frequencies, these techniques will allow you to reduce the noise of your power system to levels low enough for your needs. When you get to 18-and 24-bit measurements and digital-to-analog (DAC) conversions, getting the power-supply rails as clean as possible is essential to obtain the available performance out of the ICs you’re using."

So it appears to me to depend upon what power supply you're testing. Different power supplies and different circuit designs have different PSRR's:

3 Ways to Reduce Power-Supply Noise

Get noise out of your power supply with a multi-prong approach. Filters, bypassing, and post-regulation all can help achieve that goal.

www.electronicdesign.com

 

[LEFASR160]

No, I'm not wrong. You're setting up a straw man argument here. I think I said that (1) you only tested one signal input: a 1khz signal; and, (2) one or at most two different gear with different power supplies. You did not test all the different power supply designs. You in fact have the equipment necessary so as not to need to test any power conditioner or power regenerator---you know that. In fact, you used your 9801 to simulate near perfect AC and "dimmer switch" AC and used that instead of a passive conditioner or power regenerator. You did not outline the different power supply designs that are available with different gear and different PSRR's and you did not do your experiments on each one in order to prove that AC purity is of no value. Also, you did not address the possibility of a cleaner AC allowing a longer MTTF for components within a given PS and also downstream from it. Your science is incomplete.

@amirm I’m not excusing any product. You need to re-read what I wrote to you. I read what you did carefully before and I understood it before. You showed that 3 different AC qualities (generated by your 9801and the Mains) did not affect the audio output OF ONE DEVICE with ONE POWER SUPPLY. Now test different devices with different PS designs.

 

[DonR]

@DonR

Is the 120 hz ripple, mentioned in the article, generated by the PS itself ? Answer: NO...it’s due to the 60 hz sine wave fed to it.

The link below shows different engineering design techniques for power supply filtering:

"No matter whether you’re plagued with ripple, noise, or beat frequencies, these techniques will allow you to reduce the noise of your power system to levels low enough for your needs. When you get to 18-and 24-bit measurements and digital-to-analog (DAC) conversions, getting the power-supply rails as clean as possible is essential to obtain the available performance out of the ICs you’re using."

So it appears to me to depend upon what power supply you're testing. Different power supplies and different circuit designs have different PSRR's:

3 Ways to Reduce Power-Supply Noise

Get noise out of your power supply with a multi-prong approach. Filters, bypassing, and post-regulation all can help achieve that goal.

www.electronicdesign.com

The 120Hz ripple is indeed an artifact of the power supply ITSELF!!!! That is how linear supplies work.

 

[LEFASR160]

The 120Hz ripple is indeed an artifact of the power supply ITSELF!!!! That is how linear supplies work.

@DonR It would be 100 hz ripple if the AC line going in was 50 hz. So120 hz ripple is there only because of the incoming AC which is 60 hz as well as the PS's inability to smooth it completely to DC.

 

[LEFASR160]

The 120Hz ripple is indeed an artifact of the power supply ITSELF!!!! That is how linear supplies work.

@DonR Unfortunately, words are ambiguous. What I meant was that the value of 120 hz was due to the 60hz AC. And you are correct too. Ripple is a function of at least two variables: AC coming in and PS itself

 

[DonR]

@DonR It would be 100 hz ripple if the AC line going in was 50 hz. So120 hz ripple is there only because of the incoming AC which is 60 hz as well as the PS's inability to smooth it completely to DC.

Indeed it would be. But it is an artifact of rectification and would be present whether the incoming signal had zero distortion or not.

 

[LEFASR160]

Indeed it would be. But it is an artifact of rectification and would be present whether the incoming signal had zero distortion or not.

@DonR I totally agree.