This is a review and detailed measurements of the ISOTEK EVO3 Aquarius AC Power conditioner and filter. It is on kind loan from a member and costs US $1,999.
From the front, the unit looks fine but on the back, I was disappointed by the thin sheet metal that bent when I tried to insert and remove the included "Premier C19"
AC cable:
Likewise, the sockets feel plasticky and loose as evidenced by them not lining up. On positive front, I like the 20 amp rated socket.
The unit has won a number of accolades:
“In suppressing line noise, IsoTek unlocks low-level transparency in the same way that removing layers of old wax from a fine wood surface allows more of the inherent depth and beauty of the wood-grain patterns to shine forth.”
– Neil Gader, The Absolute Sound
"That the IsoTek Aquarius revealed previously unheard details from my familiar LPs, says it all. An effective improvement to a good hi-fi system, this serious product justifies its price."
– Hi-Fi World
"The lift in performance is quite remarkable."
– Hi-Fi Choice
"[A] mains product really has to do something extremely special to beat the IsoTek Aquarius."
– Hi-Fi+
"The Aquarius is brilliant."
– Record Collector
Let's see if we can objectively verify any of this.
ISOTEK EVO3 Aquarius Measurements
As usual with power devices, I first determine their "transfer function" but treating it as if were a preamplifier. I feed it balanced sine wave and measure what comes out:
While there is some filtering, it doesn't kick in until 1 kHz. You might think this is still good but is not when we look at where the "problem areas" are in AC mains. Here is the measurement and spectrum of my AC feed tonight:
We see that our sinewave is visually distorted. Looking to the right FFT spectrum, we see the desired 60 Hz main in that first tall spike. But then we have a ton of harmonics and noise. The worse part of this is up 600 Hz or so. In order for filtering to make a difference, it needs to go way lower, ideally down to 120 Hz. Let's see if the EVO3 does that:
As we could tell from the transfer function, it does not do that. It attenuated the higher spectrum but that was already way low in level anyway. To wit our SIAND for the AC mains is essentially the same: 34.3 dB vs 34.1 dB. Here is a higher resolution FFT showing the same:
But maybe the device does something with a real audio product connected to it. For that, I grabbed the excellent Topping PA5 stereo amplifier and tested it first with its cord going directly to AC mains:
Now let's power it through the ISOTEK EVO3:
Beyond run to run variation, there is no difference whatsoever.
Let's do a power sweep and see if that shows any difference:
Nothing. No difference at all where the output is noise dominated (e.g. below 60 watts) or at max power.
Conclusions
As I have explained many times, the reason these devices don't do anything for your audio system is twofold:
1. They don't filter high amplitude harmonics and noise in AC mains which is below 500 Hz and most of it below 240 Hz. Such filtering will be quite expensive and large.
2. The first thing our audio devices do is convert AC to DC. That conversion has mandatory filtering which is far easier to implement due to much lower voltages.
So no wonder that in test after test, we find no difference in quality of AC mains, or output of audio device tested.
Why do people attribute better fidelity to these devices? Simple: lay intuition says "filtering" removes noise. So they pay more attention to what is playing and "hear more into the music" with the AC conditioner in the path. And report veils removed, noise floor going down, blacker backgrounds, etc. This happens even if I gave them an empty box but told them it is a power conditioner.
The only time above false observations go away is if the testing only uses one's ear, i.e. blind testing. Then, the listener doesn't know if the device is in the path or not, and will read such differences into either sample showing no statistically valid value to the device. Sadly, people don't want to know the truth here as that would invalidate years of claiming other differences so false statements are made about benefits of such devices. And folks become $2,000 poorer that they could have used on something more useful.
Needless to say, I can't recommend the ISOTEK EVO3 Aquarius.
-----------
As always, questions, comments, recommendations, etc. are welcome.
Any donations are much appreciated using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/
AC cable:
Likewise, the sockets feel plasticky and loose as evidenced by them not lining up. On positive front, I like the 20 amp rated socket.
The unit has won a number of accolades:
“In suppressing line noise, IsoTek unlocks low-level transparency in the same way that removing layers of old wax from a fine wood surface allows more of the inherent depth and beauty of the wood-grain patterns to shine forth.”
– Neil Gader, The Absolute Sound
"That the IsoTek Aquarius revealed previously unheard details from my familiar LPs, says it all. An effective improvement to a good hi-fi system, this serious product justifies its price."
– Hi-Fi World
"The lift in performance is quite remarkable."
– Hi-Fi Choice
"[A] mains product really has to do something extremely special to beat the IsoTek Aquarius."
– Hi-Fi+
"The Aquarius is brilliant."
– Record Collector
Let's see if we can objectively verify any of this.
ISOTEK EVO3 Aquarius Measurements
As usual with power devices, I first determine their "transfer function" but treating it as if were a preamplifier. I feed it balanced sine wave and measure what comes out:
While there is some filtering, it doesn't kick in until 1 kHz. You might think this is still good but is not when we look at where the "problem areas" are in AC mains. Here is the measurement and spectrum of my AC feed tonight:
We see that our sinewave is visually distorted. Looking to the right FFT spectrum, we see the desired 60 Hz main in that first tall spike. But then we have a ton of harmonics and noise. The worse part of this is up 600 Hz or so. In order for filtering to make a difference, it needs to go way lower, ideally down to 120 Hz. Let's see if the EVO3 does that:
As we could tell from the transfer function, it does not do that. It attenuated the higher spectrum but that was already way low in level anyway. To wit our SIAND for the AC mains is essentially the same: 34.3 dB vs 34.1 dB. Here is a higher resolution FFT showing the same:
But maybe the device does something with a real audio product connected to it. For that, I grabbed the excellent Topping PA5 stereo amplifier and tested it first with its cord going directly to AC mains:
Now let's power it through the ISOTEK EVO3:
Beyond run to run variation, there is no difference whatsoever.
Let's do a power sweep and see if that shows any difference:
Nothing. No difference at all where the output is noise dominated (e.g. below 60 watts) or at max power.
Conclusions
As I have explained many times, the reason these devices don't do anything for your audio system is twofold:
1. They don't filter high amplitude harmonics and noise in AC mains which is below 500 Hz and most of it below 240 Hz. Such filtering will be quite expensive and large.
2. The first thing our audio devices do is convert AC to DC. That conversion has mandatory filtering which is far easier to implement due to much lower voltages.
So no wonder that in test after test, we find no difference in quality of AC mains, or output of audio device tested.
Why do people attribute better fidelity to these devices? Simple: lay intuition says "filtering" removes noise. So they pay more attention to what is playing and "hear more into the music" with the AC conditioner in the path. And report veils removed, noise floor going down, blacker backgrounds, etc. This happens even if I gave them an empty box but told them it is a power conditioner.
The only time above false observations go away is if the testing only uses one's ear, i.e. blind testing. Then, the listener doesn't know if the device is in the path or not, and will read such differences into either sample showing no statistically valid value to the device. Sadly, people don't want to know the truth here as that would invalidate years of claiming other differences so false statements are made about benefits of such devices. And folks become $2,000 poorer that they could have used on something more useful.
Needless to say, I can't recommend the ISOTEK EVO3 Aquarius.
-----------
As always, questions, comments, recommendations, etc. are welcome.
Any donations are much appreciated using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/
Last edited:
