Audiophonics HPA-S400ET Review (Stereo Amplifier)

[nemesis.ie]

Tape and light-dims fixes the bright lights.

 

[Caliban]

The new ADI-2/4 Pro may fit your needs. It has AES-EBU XLR I/O via its breakout cable, top AD conversion, AutoRef volume control for maximum SINAD over wide range, etc.

ADI-2/4 Pro | High-end USB Audio Converter & Headphone Amp - RME Audio Interfaces | Format Converters | Preamps | Network Audio & MADI Solutions

RME Audio Professional Live, Studio, Recording and Broadcast Solutions. Unrivalled Quality, Performance & Stability MADI Interfaces, Converters & Preamps.

rme-audio.com

But the ADI can not display these two..
(Sorry)..

 

[Grael]

Buying experience
I ordered the Audiophonics HPA-S400ET Sparkos Edition a couple of weeks ago. It was out of stock at the time of order, but I was promised a 15 day lead time.

My buying experience with Audiophonics was very good: detailed and instant replies to my questions, and updates on the shipment (1 day UPS shipping from France to Denmark). I was able to purchase B2B with my VAT no.

Setup
B&W 802 Diamond speakers, [insert amp here], Trinnov ST2 Pro correction hardware, Imperium monitor controlller, Lynx HiLo DAC.

Current Classé CA-2300 Delta
My current amp was a Classé CA-2300 Delta, Class A/B. Purchased for approx. 52.000 DKK (EUR €7.000 / USD $7,000) in 2013. The left side blew something recently and I had to get it serviced for a whopping €/$1,345.

The Classé sounds great, lots of power, very clean. It uses quite a bit of electricity in idle mode and has a quiet fan that rarely gets to spin up. I don't put my amps into standby as I use them every day in the studio at random times.

Loaner NAD C298
I got a brand new NAD C298 on loan, but it didn't sound right to me.

Most likely it's simply me being used to different (Class A/B) amp for many years. The low end felt lower and less solid with the C298, the midrange (perhaps as an indrect result) more pronounced compared to the Classé.

Another explanation for my dislike could be a technical mismatch, namely that the still active Trinnov ST2 Pro profile was based on the Classé and it not matching the C298. Or it could be psychoacoustical: the C298 might be more precise; but cleaner also means less perceived sub frequencies due to lower harmonic distortion than what I was used to.

The C298 pulled up a very low level hiss even when turning down the gain to the minimum setting or when using the fixed setting - I wasn't able to find a way to remove the hiss.

The annoying auto-sense can be shut off, fortunately, but I didn't bother for the two weeks I had it on loan.

New HPA-S400 ET Sparkos Edition
With the HPA I was able to set the jumpers to bypass the gain. To match the bypass setting, I changed the jumpers in my Lynx HiLo DAC on the isolated monitor output from -11 dBu to -24 dBu to ramp up the voltage. This allowed me to reduce the hiss to an inaudible level and have plenty of power.

Nice case and neatly done inside. Everything looked stellar. All torques screws had equal pressure, which satisfied my OCD.

Unlike the C298 the HPA-S400 ET sounded right in the low end, which contradicted my inital assumption that the Trinnov profile could be the culprit. Since the HPA-S400ET measures even better than the C298 then psychoacoustics aren't to blame either. So either it's good old placebo or confirmation bias, or the C298 I have on loan is a bit off from the factory.

I decided to hire the same external consultant I used in 2015 to help me make a new Trinnov measurement for the HPA. Same room, same speakers, same monitor controller, same studio console, same measurement mic, same guy(s). The Trinnov 3D measurement mic was placed using a gyro laser level. This allowed us to get the same placement ± a few mm in the sweet spot.

The differences in the measurements between the Classé and the HPA were small or inconclusive, but then again an acoustical measurement is a different beast since there are many variables. The largest difference was around 10 kHz where there was a ±1.25 dB between the two amps in the same setup, withe HPA having more energy there. Obviously I can't tell which is "correct" - only that there's a difference.

We did several additional measurements between 0 and 40 cms on the Z-axis and they all confirmed a variation in this frequency area, so it's not acoustical, i.e. not due to a difference in mic placement between the two amps. It will make the HPA sound a bit more sparkly compared to the Classé or vice versa, if you will.

I'm saving on electricty with the HPA, about €/$399 a year (Danish energy prices are crazy), so I'm probably buying another HPA (non-Sparkos) to replace another old Class A/B amp. That's a big bonus you can tell yourself (or your wife) when switching out ye olde Class A/B for a brand new Class D.

One criticism is that the HPA doesn't come with any kind of manual or even a piece of instruction paper, but that's what Google is for. Also, the heat sinks on the sides of the unit can resonate at high frequencies. I'm not sure if that's a real life issue at 3 meters listening distance, but I'm thinking about dampening them somehow. Last thing: the power on LED is quite bright and can't easily be dimmed.

I'm sorry to ask, I hope it won't hurt your feelings, but why getting the Sparkos edition if the expensive Sparkos opamps are bypassed ?

For the Sparkos regulators ? (And are they in use when the buffer is bypassed?)

 

[eduardw]

Great read, NAD C298 is off the list then...I will certainly hear the hiss too, having my speakers close to me, no need to even try it..
Nice to see a Hilo user, are you satisfied with it ?

Maybe a bit hastily to remove the c298 from you list. I have two and I very happy with them cannot hear any Hiss. I use it with
the topping D90 dac and topping pre90.

 

[UNICRON-WMD]

Will the HPA-S400ET go in stand-by or is it always in idle?

Thx,
WMD

 

[nemesis.ie]

Will the HPA-S400ET go in stand-by or is it always in idle?

Thx,
WMD

 

[nemesis.ie]

They don't have auto-on with input/standby after none for a time. The models with a trigger go into standby via that or attach a power trigger like the one they sell which I linked earlier.

 

[Lagerfeldt]

I'm sorry to ask, I hope it won't hurt your feelings, but why getting the Sparkos edition if the expensive Sparkos opamps are bypassed ?

For the Sparkos regulators ? (And are they in use when the buffer is bypassed?)

As a long time member of Gearslutz (now p.c. named Gearspace) I assure you nothing can hurt my feelings ;-)

I'll give you two explanations and you can choose the most likely or both:

1) I wasn't sure whether I was going to use the on-board gain or bypass it, so I got the Sparkos just in case
2) I've got too much money and I'm a bit dumb

Also, I'm not sure the Sparkos actually measures that much better than the regular version. Don't know about the regulators, someone here might know.

PS. Getting the non-Sparkos edition of the next HPA I'm buying for a set of Minipods.

 

[Lagerfeldt]

Maybe a bit hastily to remove the c298 from you list. I have two and I very happy with them cannot hear any Hiss. I use it with
the topping D90 dac and topping pre90.

The hiss isn't produced by the C298. It's a result of something earlier in the chain, but I couldn't get it low enough with the C298 at zero/fixed gain compared to the HPA with the jumpers set to bypass.

In my dead silent studio it was barely audible at 3 meters distance with the C298, but I prefer it as quiet as possible.

I can still technically hear it with the HPA, but then we're talking max output from my monitor controller and my ear pressed against the treble on the B&W 802 Diamond. Not a moment to have your dog step on your keyboard...

 

[Lagerfeldt]

Will the HPA-S400ET go in stand-by or is it always in idle?

Thx,
WMD

Always idle. I measured ≈16W which is acceptable to me.

 

[Caliban]

The hiss isn't produced by the C298. It's a result of something earlier in the chain, but I couldn't get it low enough with the C298 at zero/fixed gain compared to the HPA with the jumpers set to bypass.

In my dead silent studio it was barely audible at 3 meters distance with the C298, but I prefer it as quiet as possible.

I can still technically hear it with the HPA, but then we're talking max output from my monitor controller and my ear pressed against the treble on the B&W 802 Diamond. Not a moment to have your dog step on your keyboard...

About the gain - zero setting on C298 is 8.5 dB according to the manual, the passthrough gain on the HPA is 12.8 dB.
For comparison - that is 4.3 dB less gain for input noise to be amplified.
But there were more noise (set at zero gain) when the NAD was in the chain vs the HPA.. And you say it did not come from the NAD itself ?
Or am I missing something ?
Just curious

 

[Lagerfeldt]

Good point. I'm not sure why it was more audible with the NAD when you look at the nominal values. I still have the NAD on loan, so I could test it again.

Yes, the very low hiss close to the hearing threshold is a byproduct of combining HiLo's monitor DAC and Trinnov's AD/DA, which is the only way to have an analog monitor controller with sum/difference aka mid/side soloing in the chain. Of course M/S monitoring could be done digitally via a plug-in post mastering but pre-DAC... but call me old skool.

Alternatively you can loop the Trinnov via AES/EBU on the HiLo only using the HiLo's monitor out for DAC or you can use a separate DAC (such as the Topping D90SE) after the Trinnov and supply the Trinnov with AES/EBU from the HiLo.

However, neither of these solutions will work with an analog monitor controller with sum/difference because the Trinnov's processing will change the sum/difference, and especially with the latter that would negate the whole concept as it would introduce new material into the difference channel.

Another aspect is my preference for 96 kHz processing in the Trinnov as opposed to 44k1Hz because Trinnov not only locks to the source SR but doesn't upsample or oversample internally, i.e. all processing happens at the nominal SR. Now, I don't give a rat's ass about sample rates most of the time, i.e. I'm perfectly fine with 44k1Hz, but I hear a difference - as well as the expected latency difference - when Trinnov is using 44k1Hz.

Despite what some think, most music is still done at 44k1Hz, even if oversampling is standard during processing. There's a move towards 48 kHz because of Opus encoding, I suspect, but most of the approximately thousand songs I master a year are still at 44k1Hz. So that's the detailed explanation.

Basically a non-issue (probably) for most hi-fi buffs, but a bit of a puzzle and trade-off for a sound engineer using Trinnov. Still, I can't recommend Trinnov enough, it's a game changer.

 

[Caliban]

Good point. I'm not sure why it was more audible with the NAD when you look at the nominal values. I still have the NAD on loan, so I could test it again.

Yes, the very low hiss close to the hearing threshold is a byproduct of combining HiLo's monitor DAC and Trinnov's AD/DA, which is the only way to have an analog monitor controller with sum/difference aka mid/side soloing in the chain. Of course M/S monitoring could be done digitally via a plug-in post mastering but pre-DAC... but call me old skool.

Alternatively you can loop the Trinnov via AES/EBU on the HiLo only using the HiLo's monitor out for DAC or you can use a separate DAC (such as the Topping D90SE) after the Trinnov and supply the Trinnov with AES/EBU from the HiLo.

However, neither of these solutions will work with an analog monitor controller with sum/difference because the Trinnov's processing will change the sum/difference, and especially with the latter that would negate the whole concept as it would introduce new material into the difference channel.

Another aspect is my preference for 96 kHz processing in the Trinnov as opposed to 44k1Hz because Trinnov not only locks to the source SR but doesn't upsample or oversample internally, i.e. all processing happens at the nominal SR. Now, I don't give a rat's ass about sample rates most of the time, i.e. I'm perfectly fine with 44k1Hz, but I hear a difference - as well as the expected latency difference - when Trinnov is using 44k1Hz.

Despite what some think, most music is still done at 44k1Hz, even if oversampling is standard during processing. There's a move towards 48 kHz because of Opus encoding, I suspect, but most of the approximately thousand songs I master a year are still at 44k1Hz. So that's the detailed explanation.

Basically a non-issue (probably) for most hi-fi buffs, but a bit of a puzzle and trade-off for a sound engineer using Trinnov. Still, I can't recommend Trinnov enough, it's a game changer.

Look who I found ! (hope you do not mind)

Forgive my ignorance, I had to google some bits and pieces of your setup, some really cool stuff you have there !
You are a professional I guess, and this gear are your work tools ? I am just a audio/electronics hobbyist that once discovered the beauty of room correction software.
I run Dirac live on a PC, that was really a game changer for me, I have been using it for almost ten years now.
It has been with me since windows 7 and it is still working perfectly in windows 10, latest update, and I am waiting for the windows update when it will finally crash / BS on me..
Maybe a Trinnov SR2 or ST2 could step in then
Or perhaps a newer Dirac version, we will see. It depends on how rich and stupid I am at that time.

Oh, well, this could have its own thread, I better stop here..

 

[Satyadhara]

Before I got the HPA, I had a Hegel H300 with sometimes quite audible transistor hum. I got a DC filter, which reduced this hum considerably. Not knowing enough about Class D amps, I am wondering if the HPA could generally profit from a DC or AC mains filter. Any thoughts on that?

 

[Rottmannash]

Re: Sparkos- when @amirm measured the VTV Purifi with the Sparkos vs OG buffer the Sparkos measured worse, did it not?

 

[tmtomh]

Before I got the HPA, I had a Hegel H300 with sometimes quite audible transistor hum. I got a DC filter, which reduced this hum considerably. Not knowing enough about Class D amps, I am wondering if the HPA could generally profit from a DC or AC mains filter. Any thoughts on that?

The PSU inside a Class D amp like the HPA-S400ET is going to be less susceptible to DC-induced transistor hum than the large toroidal linear PSU inside the Hegel H300. Large toroidals are the most susceptible to such hum.

I had an Adcom GFA-5400 with a large toroidal PSU and had to use a DC blocker with it in two different homes. I got the Audiophonics HPA-S400ET and the DC blocker is now in my spare parts closet.

 

[Lagerfeldt]

Re: Sparkos- when @amirm measured the VTV Purifi with the Sparkos vs OG buffer the Sparkos measured worse, did it not?

I seem to remember reading this somewhere or at least it didn't measure better.

 

[nemesis.ie]

@Caliban you don't have to be stupid to get a Trinnov - or for those that upgrade processors every few years for new features, extra rich.
Something like an Altitude 16 (20 channels) is game changing and at 2 to 3 times the cost of a mostly "fixed function" processor is good value over 10 years.
I was nervous about dropping the $$$$$ on the Alt32 but it has been fantastic and gets better over time. Now with 4 more channels than it started with at no extra cost and more cool features to come.

 

[DubbyMcDubs]

Audiophonics HPA-S400ET Amplifier Class D Stereo Purifi 2x400W 4 Ohm - Audiophonics

Class D stereo power amplifier based on two PURIFI 1ET400A modules and a HYPEX SMPS1200A400 power supply.

www.audiophonics.fr

vs

AUDIOPHONICS LPA-S400ET Class D Stereo Amplifier Purifi 2x400W 4 Ohm - Audiophonics

We are proud to offer you the Audiophonics LPA-S400ET, a Class D stereo power amplifier based on two Purifi 1ET400A modules. This ready-built and ready-to-use version is for audiophiles who want to equip themselves with a foolproof amplification

www.audiophonics.fr

The only difference is the case. Would anyone predict that the close proximity of the power supply to the modules would create any measurable decrease in performance?

I am looking at pairing one of these with my Benchmark HPA4 to drive a set of Revel Studio2's.

 

[nemesis.ie]

I have a couple of the 4ch units, everything is jammed in there with not even enough room left for a trigger module and they sound superb.

Even if there is a small bit more noise creeping in from the PSU I suspect even if it were measured it would be inaudible.

So far the audiphonics amps are the only thing that has cured my tweeter hiss (which some folks can't even hear).

Using another amp (Emotiva HC-1, 29 dB gain, vs the 400ET, 12.8 dB, bypass mode) with the Trinnov DACs and also via external DACs (Schiit Modius and Gustard X16 (at any setting on the voume control, including -90 dB) still has audible hiss.

So given I can hear ZERO backround hiss with my ear right on the AMT driver with the 4ch Audiophics says to me that either of the two channel models will also have inaudible noise - I also have one of the two channel models in the smaller case and it too is silent.

If you plan to drive them really hard, the larger case might possibly provide more cooling ...