Puritan Audio PSM156 Review (AC Filter)

[David_M]

I'm really curious if anyone is willing to send Amir a DC iPurifier2

https://ifi-audio.com/products/dc-ipurifier2/

it's supposed to be exactly what you just mentioned, an active noise cancellation device for DC power.

Bold claims for such a device with 50dB (x316) to 100dB (100,000) attenuation claims, which say best attenuation is in the audio range.

 

[Beershaun]

I gave it one notch higher score because it actually does something as far as filtering. For pure audio use, it would get a headless panther.

Should it rate a piggybank panther since it's audio claims are false even if the device is functional?

 

[pkane]

Yes, the test frequency matters. a 10kHz (or greater) FFT would be more telling since its harmonics are within the attenuation region of the line conditioner. Comparing the 10 kHz FFT before and after the conditioner (while HF AM/FM noise is injected on the mains) would let us know if this device works as designed (though not as claimed).

Do you care if a 20kHz AC frequency harmonic is attenuated by some dB? That was also measured. See the frequency response. So what? Still has no effect on the audible range.

 

[pma]

To both of you: your remarks are thoughtful and make a point.
Just how do you measure it "correctly" and does it "improve" the audio ?
Also:
Why don't the AC filter "box" manufacturers measure it objectively rather than claiming subjective audio heaven ?

A quick immediate test - there is a line amplifier with 30dB gain (called Amp 30dB) on my workbench. Its input is shorted and its SE output is connected via shielded twist-pair cable RCA-XLR to the balanced measuring input. There is no other connection than USB from the soundcard to the PC notebook that is class II instrument.
Now, to the same 230V/50Hz mains line there is a SMPS 24V/5A power supply connected which I switch on and off. The DC output of this power supply goes to AIYIMA A07 amplifier which I also switch on and off. This A07 amplifier is neither connected to the 30dB amp, nor to the soundcard. The only interaction is via mains 230Vac power, which is used to supply the Amp30dB, PC and SMPS.

1) output noise of Amp30dB, SMPS off, A07 off

2) output noise of Amp30dB, SMPS on, A07off

see the spike above 500Hz

3) output noise of Amp30dB, SMPS on, A07 on

Again additional spikes

These interfering spikes are very low in level, however in the audio band. The only coupling is via 230V mains network. The line above 500 Hz is most probably an intermodulation product reflected in the audio band. Additional filter at the SMPS mains line would with highest probability change the measured plots.
Only 50 cm of pseudobalanced RCA-XLR cable was used at the Amp30dB output. Longer cable or 2-wire SE cable would for almost sure worsen the results. The additional spikes have origin in mains interference voltage and also result in cable shield capacitive currents.

 

[Dzhaughn]

I want to say
I'm very embarrassed that ASR makes completely wrong measurements for such products every time.
To evaluate such a product, create a situation that actually causes disturbance such as noise and affects the audio equipment, and measure how the target product is attached and how it works at that time. If you don't, it doesn't mean anything.
It is natural that measuring it in a normal state without any obstacles has no effect. "No effect" is the correct answer. If there is any change in the audio signal, it means a defective product.
Such products are similar to car seat belts and lightning arresters, and only show their effects in the event of an accident. It usually looks like a useless and disturbing entity.
Do not measure such products in the same way as audio equipment. Evaluation of these products requires a completely different and advanced measurement environment.

That is hilarious.

I am working on a 12V DC filter that improves car audio performance during a car crash. Are there any dummies out there who will test it for me?

The manufacture claims these products produces audible differences in "clarity, staging, and dynamics." The manufacturer does not say that is only when one has hooked up a Van de Graf generator and tesla coil on the same AC circuit. Or during a lightning storm. Or anything else.

 

[LTig]

I lived in an apartment where AM radio stations bled through my audio system. FM radio stations also bled through but that due to the long speaker cables that I had to meticulously re-arrange on the floor to minimize the bleedthrough. Frustrating that I'd approach the speaker wire to move it and the station 'disappears', only to re-appear if I moved away.

Now, we know that AC is converted to DC inside the audio device and further filtered to provide clean DC power to the rest of the circuitry. But this DC filtering may not be enough to eliminate AM radio frequencies (~500kHz - 1700kHz) depending on the design of the audio device. These are then rectified to DC or some lower frequency which may be visible on a DC-20kHz FFT spectrum. This is where this Puritan line conditioner could be useful.

No, it won't, if the bleedthrough is captured by the speaker cables (working as antennas) and enters the amplifier through its output.

 

[amirm]

These interfering spikes are very low in level, however in the audio band. The only coupling is via 230V mains network. The line above 500 Hz is most probably an intermodulation product reflected in the audio band. Additional filter at the SMPS mains line would with highest probability change the measured plots.

You haven't demonstrated why any of that is not instrumentation error. Until you do, it doesn't mean anything. It is trivial for measurement gear to pick up all kinds of noise and interference on their sensitive high impedance inputs.

 

[amirm]

No. Typical use is that you connect more components to the filter outlets and you interconnect them with signal cables.

And that is precisely my setup in the tests I showed.

There is one power strip that feeds:
1. A high-powered "gaming" style workstation that is running tons of programs including Audio Precision analyzer software which saturates at least one core. I am also running photoshop, Roon player hooked up to an RME DAC playing music, browser window, etc. This is no idle machine.

2. An LED dimmable light.

3. A computer monitor.

4. Device under test, e.g. the Topping Pre90.

5. Audio Precision Analyzer.

A USB cable then goes between the PC and Audio Precision analyzer for communication. This creates a ground loop between them. Signal cables are of course used all over the place in the process of testing the equipment.

This is as "dirty" as it gets as far as any audiophile is concerned. Yet day in and day out, I measure equipment that produces pristine, instrument grade performance. Spectrum of AC was shown in my measurements which is anything but "clean." Yet the power conditioner had no effect.

 

[Harmonie]

The guillotine got rusty and needs maintenance. There are too many snake oil manufacturers targetting gullible audiophooles, revolutionary Amir cut too many heads these last times... Also, his wife is getting mad at him , his testing lab has become a gory, horror place, full of blood...
.

With the courtesy of @Boris Badinov

https://www.audiosciencereview.com/forum/index.php?threads/a-call-for-humor.9679/post-893551

 

[DanTheMan]

I know with my dimmer switch on, my ADAM ARTist 5 don’t always automagically shutoff while no music is playing, but they do with the lights off. When you turn the lights back on though, they don’t turn back on however… not sure if a device like this would fix that problem or not, but I wouldn’t be willing to spend this kind of money to find out. The thing that really suck though is that you either have to turn off the lights for 10 minutes to get them to shutdown or you have to flip the manual switch on them which means you have to get off the couch. Life is tough ha ha

 

[tmtomh]

Look at the 3 kHz spike off the D90. It is lower with the Puritan audio.
Look at the 5 kHz spike off the D90. It is higher with the Puritan audio.

1) Is this reproducible run to run?
2) What is there a difference if you looked a HD in this region instead of THD 20Hz-20kHz

(I am sure this is below the threshold of audibility but it would be nice to quantify it in the same way we poo-poo gear with 96 dB SINAD versus 120 dB SINAD even though 96 dB is probably good enough.)

Why do you keep asking “what if there is a difference in this region”? There IS a difference in this region, when Amir measures the AC coming from the wall. There is NOT a difference in this region, when Amir measures the performance of both good and mediocre/poor audio components. Both the good and poor components are filtering the junk in the AC sufficiently, so that cleaning it up with an outboard device doesn’t change the audio components performance, at all, in any measured frequency region, be it below 1kHz, above 1kHz, or above the range of human hearing (out to 90kHz, which I believe was the outer limit he tested in this case).

You can keep asking the same question over and over, but you’re not going to get a different response from Amir.

 

[Billy Budapest]

So, the bottom line is that the Puritan box indeed does what they say it does, but what it does doesn’t matter.

 

[Deleted member 30699]

You haven't demonstrated why any of that is not instrumentation error. Until you do, it doesn't mean anything. It is trivial for measurement gear to pick up all kinds of noise and interference on their sensitive high impedance inputs.

@pma measurements are actually possible. While SMPS are good quality, this power supply form factor is not intended for providing the cleanest possible AC.
The frequency residuals origin likely form less-than-ideal filtering of the rectified DC and an under-damped LC(L) filter somewhere in the supply. This can end up in the measurements for various reasons...

This is _not_ an endorsement of mains filtering for audio.
Modern amplifiers would not couple mains harmonics. However, you can couple things into the output if you try hard enough (maybe that is why some hear differences in cables?). Old or ultra-low budget amplifiers may couple some harmonics but @pma did not specify the make/model.
Also, this is >60dB down and only ~6dB above noise of this amp ...

 

[GXAlan]

Why do you keep asking “what if there is a difference in this region”? There IS a difference in this region, when Amir measures the AC coming from the wall. There is NOT a difference in this region, when Amir measures the performance of both good and mediocre/poor audio components.

Look at the graphs to the right. SINAD is worse by 0.2dB *with* the filter. But if you ONLY looked at 3 kHz, you'd expect the THD% from 1-5k for example to be much higher?

 

[Yevhen]

Hi Amir, the filtering of such devices depends on the load. I believe these are optimized for low-ohmic AC loads with 5-15A currents as you normally have with the power amplifier. With almost zero loads (as your AP of topping DAC), it won't do itś filtering function properly.

On top of it, the noise could be balanced (on both F and N wires) and unbalanced (for example F + GND).

https://asset.conrad.com/media10/ad...-mh-l-x-b-x-h-1135-x-575-x-454-mm-1-stuks.pdf (see measurement conditions, the bottom of page 2) They test with 50 Ohm, 100 Ohm and 0.1 Ohm loads at 220V.

I saw these filters to be useful when my power amp Audiolab 8300A with toroidal transformer picked up RF noise from the LAN via 220V TP-Link TL-PA4010P. Of course, the filter would be redundant / absolute waste in the system with a switching mode power supply (SMPS).

 

[bidn]

Yes, the test frequency matters. a 10kHz (or greater) FFT would be more telling since its harmonics are within the attenuation region of the line conditioner. Comparing the 10 kHz FFT before and after the conditioner (while HF AM/FM noise is injected on the mains) would let us know if this device works as designed (though not as claimed).

In your posts in this thread, you seem to systematically confuse completely different things:

- the signal fed, in these test cases usually a simple waveform oscillating with a single frequency
(e.g.1000 times the same pattern per second = with a parameter of 1 kHz)

- the FFT, a which is not a signal but an operation, not characterised by a frequency. Its maths will convert the data from the time domain to the frequency domain, revealing the frequencies of the input , signal.

Before criticizing Amir, one should at least understand the very basics ...
The famous philosopher Wittgenstein said something like:
Wovon man nicht weiss, darüber soll man schweigen...
.

 

[sarumbear]

A quick immediate test - there is a line amplifier with 30dB gain (called Amp 30dB) on my workbench. Its input is shorted and its SE output is connected via shielded twist-pair cable RCA-XLR to the balanced measuring input. There is no other connection than USB from the soundcard to the PC notebook that is class II instrument.
Now, to the same 230V/50Hz mains line there is a SMPS 24V/5A power supply connected which I switch on and off. The DC output of this power supply goes to AIYIMA A07 amplifier which I also switch on and off. This A07 amplifier is neither connected to the 30dB amp, nor to the soundcard. The only interaction is via mains 230Vac power, which is used to supply the Amp30dB, PC and SMPS.

1) output noise of Amp30dB, SMPS off, A07 off
View attachment 150116

2) output noise of Amp30dB, SMPS on, A07off
View attachment 150117
see the spike above 500Hz

3) output noise of Amp30dB, SMPS on, A07 on
View attachment 150118
Again additional spikes

These interfering spikes are very low in level, however in the audio band. The only coupling is via 230V mains network. The line above 500 Hz is most probably an intermodulation product reflected in the audio band. Additional filter at the SMPS mains line would with highest probability change the measured plots.
Only 50 cm of pseudobalanced RCA-XLR cable was used at the Amp30dB output. Longer cable or 2-wire SE cable would for almost sure worsen the results. The additional spikes have origin in mains interference voltage and also result in cable shield capacitive currents.

Am I correct to understand that those spikes are at -140dBV?

 

[amirm]

Look at the graphs to the right. SINAD is worse by 0.2dB *with* the filter. But if you ONLY looked at 3 kHz, you'd expect the THD% from 1-5k for example to be much higher?

View attachment 150156

Those are run to run variations with amplitudes down to -145 dB.

 

[amirm]

Hi Amir, the filtering of such devices depends on the load.

It doesn't really. I showed you the filtering when feeding it with my AP and capturing it with the same with very low currents involved. Passive circuits like that don't care about current as far as filtering. Current specifications are there with respect to voltage drops and safety which is independent of performance.

 

[sarumbear]

So, the bottom line is that the Puritan box indeed does what they say it does, but what it does doesn’t matter.

Have you read what the manufacturer says the box does?