Measurements of Audio-GD HE-9 Headphone/Pre Amp

[WolfX-700]

Yesterday, SF Express sent me a big box-I was still thinking "I haven't bought anything recently" until I saw the impressive Audio-GD logo on the box. Aha... The Audio-GDHE-9 Headphone/Pre Amp that netizens commissioned me to measure has arrived!

Its huge size forced me to remove it from the test bench to take pictures, and the weight of the device left a deep impression on me... Considering the recent short-term, mid-term and long-term market conditions for non-ferrous metals, this may be An investment method? Okay, stop talking nonsense and start measuring:

Are these measurements good? not good. Bad? Not bad. So why does L7 put almost all the howling wolves and howling wolves? Because...UNBAL:

Measuring such a product, mine is a bit sad. What makes me sadder is that I saw the description of this product:

HE-9 is a true balanced design applying the newest ACSS design which can improve the XLR connection to sound as good as the ACSS connection.
The HE-9 seem complex , but in fact they are the most simple design the world has ever had , their a lot components to process the power supply. The simplest and shortest signal journey brings out the most neutral and least colored sound.

HE-9 applies ultra-low distortion single stage ACSS amplifier , an OPA already had built in two stages amplifier. they are soldered directly on the motherboard.

The ACSS is a non-feedback technology made with discrete amplifiers. Most people know the global feedback design can offer better specs in test measurements, and non-feedback can't do well in test measurements but can offer better sound for the human's ears. Here is a conflict of the classic circuits.

But the ACSS opens a new field, it can offer a least coloration sound which is more neutral with very low distortion and high linearity. So it can retain the dynamics, detail and neutral sound but not sound bright or harsh.


How dare you!


About regenerative power supply:

On Audio-GD’s product introduction page, they said:

While audio is life , the Power supply is water , Pure water create the wonderful life ,clean power supply reproduce the beautiful sound .
The regenerative power supply like a proprietary electric generator built in the unit.
It can block most disturb from the power line , supply the ultra clean power supply for the audio amps , reproduce the most neutral and analog sound like audio signal for the system and the users.
In the regenerative power supply design, the AC power through the regenerative input transformer , conversion to DC power ,through the class A PSUs power supply for the regenerative wave generator and driver stages .
A balance regenerative regular wave generator produce the ultra low distortion 50Hz wave ,through the balance gain amps and the strong output stages , drive the regenerative transformers produce the clean and steady AC power supply feed to the separate class A PSUs power supply for the L and R channel preamps .
The class A parallel connection PSU has very high input impedance to avoid disturb through the PSU from affecting the signal parts and low output impedance with very fast speed and high linear, it is a very clean power supply.

What a pity. Measurement tells us a different story:

We can see that not only the 50HZ noise and its multiplier still exist, but also a series of 52.667HZ multipliers...obviously...this 52.6 frequency is the frequency after "regeneration"...

What’s more serious is that...the unbal output of this product has a dc offset of about 20-30mv...whether it is the headphone output or the preamplifier output is the same

 

[pozz]

Good to see you again

 

[Matias]

To me this is is really bad, considering the size, price (USD 2180) and their intent.

Thanks for the measurements!

 

[Vini darko]

Only good for 10mw clean how have managed that with such a lovely looking build?

 

[Jimbob54]

Only good for 10mw clean how have managed that with such a lovely looking build?

The dark arts of Audio Gd. Part of me thinks they know exactly what they are doing and how it measures.

 

[Veri]

What’s more serious is that...the unbal output of this product has a dc offset of about 20-30mv...whether it is the headphone output or the preamplifier output is the same

Story time.. I once had an amp that had high dc offset. I was using it with a headphone that had a fancy 'occ copper litz' cable that only has plastic sheathing and no further braid or protection. In the span of half a year the cable developed black/dark spots, because of the DC (I believe/assume). Since it was transparent/plastic I saw the cable deteriorate over time. Could be the cable was flimsy though, but imagine what it can do to the headphone drivers!
I stopped using the device and both headphones and cable still work to this day. But DC offset is a silent killer for your (expensive) gear. At the very least a proper amp should have DC offset protection.. or come with a fat warning.

 

[solderdude]

So they built a SS device with performance very similar to that of tube amps and then ruin it by going balanced.

nice:
Shielded R-core transformers.
Relay volume control.
Opamps in sockets (for those thinking there is a huge audible difference).
Balanced and SE out.
Balanced and SE input.
Nice shielded enclosure.
Dual mono construction.
Substantial 'usable' output power: 1W @ 300ohm (250mW SE), 1W @ 68ohm (250mW SE)

but:
300mW @ 32 Ohm (specified at around 8W) but SE will not be much lower. acceptable but not great for inefficient planars as distortion above 0.5W will be reaching potentially audible levels.
Rather high hum despite the complex PS.
Completely unnecessary complex and totally not worth it power supply that doesn't lead to improved performance.
DIY 'look' (some like that)
2400.- Euros !!!
A lot of components thus shorter MTBF with results similar to tube amps (though some may be looking for just that)
avoidable DC offset is a bit on the high side will this become worse when temperature changes ?

Thanks for measurements...

 

[Veri]

Completely unnecessary complex and totally not worth it power supply that doesn't lead to improved performance.

The over-engineered PSU that fixes nothing is the worst thing about this product, IMO

 

[Matias]

The over-engineered PSU that fixes nothing is the worst thing about this product, IMO

Yes, good point. All this over engineering to remove power supply noise, great, and then they keep 2 big harmonic distortions on the signal.

It's like making sure the tires are really clean and polished and leaving the rest of the car dirty.

 

[highpurityusbcable]

Audio-GodDamn

 

[Martin]

It looks like it measures better than most Audio-gd stuff but they set themselves a very low bar and then nicked it on the way over. It never ceases to surprise me at the number of audiophools who gush over this rather expensive junk. There are a lot of much better sounding and measuring headphone amps than this $2,000 over-engineered junk. They must have Rube Goldberg on their engineering staff.

Martin

 

[WolfX-700]

It looks like it measures better than most Audio-gd stuff but they set themselves a very low bar and then nicked it on the way over. It never ceases to surprise me at the number of audiophools who gush over this rather expensive junk. There are a lot of much better sounding and measuring headphone amps than this $2,000 over-engineered junk. They must have Rube Goldberg on their engineering staff.

Martin

I have observed a very interesting phenomenon. Even these manufacturers who claim to be "not paying attention to measurement results" often have better measurement results for their high-end products than low-end products. If they think their so-called "sound characteristics" are their most important part, then why do high-end products have fewer of these?

 

[DSJR]

Can I ask about negative feedback please?

I was told decades ago that *local* feedback loops may be better and supposedly less intrusive than overall global feedback or maybe a sensitive mix of the two. This may be total BS and I apologise if the engineers reading this do their nut, but my feet have been in both camps over the decades and the audiophool side is still there, albeit receding as time passes from when I was fully into it all.

 

[solderdude]

In the end it is all about the waveform shape of the actual music signals (not so much test signals as these test limits that will never be reached by actual recordings)
It really does not matter what circuit topology is used.
As long as the waveform at the output of an amplifier circuit is similar in shape (but different in voltage/current amplitude) under a load then what does the kind of feedback used matter ?
The more accurate the wave form the lower the distortion the higher the signal fidelity. This is all that matters. Crappy circuits make crappy signal fidelity, you can make crappy amps with all kinds of topologies and make great amps with all kinds of topologies.

 

[Tks]

Currawong losing his mind right now.

 

[dfuller]

Can I ask about negative feedback please?

I was told decades ago that *local* feedback loops may be better and supposedly less intrusive than overall global feedback or maybe a sensitive mix of the two. This may be total BS and I apologise if the engineers reading this do their nut, but my feet have been in both camps over the decades and the audiophool side is still there, albeit receding as time passes from when I was fully into it all.

This is complicated. Short version (as I understand it), there's a limit to how much global feedback you can do as it doesn't work near as well at high frequencies as it does low ones. Too much and it can end up causing ultrasonic oscillation and seriously degrading performance. But, you can combine global and local feedback for greater amounts of error correction (read: less distortion) than either alone.

 

[tesla2046]

This is complicated. Short version (as I understand it), there's a limit to how much global feedback you can do as it doesn't work near as well at high frequencies as it does low ones. Too much and it can end up causing ultrasonic oscillation and seriously degrading performance. But, you can combine global and local feedback for greater amounts of error correction (read: less distortion) than either alone.

This is an interesting article talking about feedback and distortion. Probably it explains why many people like the product although the measurement is terrible. They are hearing the distortions for warmth, soundstage...
"The sound of 2nd order type circuits is often praised as “warm” and by comparison 3rd order type circuits are often noted for “dynamic contrast”.
https://www.passlabs.com/technical_article/audio-distortion-and-feedback/

 

[escalibur]

These things look like power plants with a futuristic design from the 80's.

On a more serious note, I really wish they could focus more on sub $500 combos instead.

 

[dfuller]

This is an interesting article talking about feedback and distortion. Probably it explains why many people like the product although the measurement is terrible. They are hearing the distortions for warmth, soundstage...
"The sound of 2nd order type circuits is often praised as “warm” and by comparison 3rd order type circuits are often noted for “dynamic contrast”.
https://www.passlabs.com/technical_article/audio-distortion-and-feedback/

Nonsense, absolute nonsense.

As Mr. Spock said, “Instruments only measure what they were designed to measure.” * Given the complaints of audiophiles over the sound of high-feedback type amplifiers, it is reasonable to examine non-linear distortion in greater depth than is possible with a single number.

Instrumentation today is more sensitive than our ears.

All of these examples have current flowing from pin 1 to pin 2 as a function of the voltage between pin 3 and pin 2.

Nope, no longer taking this article seriously, they screwed up something as basic as the direction current flows. If you don't know that, don't write technical articles.