3e Audio A5 Stereo Amplifier Review

[pherax]

Thanks for the reply. So I can return the A5 and just use my SR5011.

 

[Panelhead]

I am glad I am not a manufacturer.

 

[Punchy_Power_Cat]

you should not hear much if any difference between properly designed and functioning amplifiers.

Difference between a ~85dB sinad amplifier and a ~100dB sinad amp very hearable. Above 100 I'm not sure, maybe small nuances.

 

[Alannn]

Difference between a ~85dB sinad amplifier and a ~100dB sinad amp very hearable. Above 100 I'm not sure, maybe small nuances.

Do you have a link to double blind testing results, or is this your opinion?

 

[Panelhead]

The published tests have a lot of information. Not just SINAD. I was really interested THD+N at very low power levels. Around 10 - 30mW levels. Also THD+N at different frequency’s.
The really low numbers are generally around 1kHz. Look at the graphs plotted at 15kHz. Way, way different from using a hard number of say a 100 SINAD. May be 80 SINAD at upper frequencies. The SINAD at 5 watts.
This is one way even amps with the same SINAD can sound different. That number is measured at 5 watts. Most listen very little at 5 watts. Normally less than 1 watt.
My speakers are a real 105dB and 8 ohm. Normal listening output is very low, peaks maybe 500mW.
I started looking at the distortion vs power output at only the far left. The distortion vs frequency only at 10 and 15kHz.

 

[Roland68]

I just received my e3 audio A5 amplifier. I intended to use it to upgrade the amplifier bit of my sr5011. Unfortunately I do not hear any audible differences on my MA silver 500 7g speakers. Does this mean the DAC in my sr5011 is the bottleneck? I could also go for another dac but I like the convenience of my sr5011... I stream music via rpi3b (24bit flacs)

I've (unfortunately) had some experience with this topic, built-in DAC chips and their implementation in AVRs, over the past 30 years, and often bad experiences. The problem is the same for all manufacturers in the lower price range (up to €2,000).

I would strongly assume that the limiting factor in the SR5011 is the built-in PCM1690 DAC chip and its very poor implementation. Added to that are HDAM modules and a few other things that are unfortunately typical of AVRs and Marantz.
The PCM1690 is an 8-channel DAC chip for less than €2 (you read that right, I didn't forget a zero). For it to deliver halfway decent performance, it requires a lot of knowledge and attention to implementation, which you can't expect from a device like the Marantz SR5011. Subtract about 20-30% from the values in the datasheet.
You can see that this can actually reach an audible level, for example, with the SR6014 tested by Amir, which is in a higher price range. The result is simply a slap in the face (for the customer).

Very transparent, good, and affordable DACs include the SMSL D1, SU-1, C100, Topping E30 II (Lite), D50 III, or DX5 II (including a very powerful headphone amplifier).
I would recommend you give it a try.
Of course, you shouldn't expect miracles, but for me, the difference was audible in such cases, especially with amplifiers as transparent as the A5.

 

[Punchy_Power_Cat]

Do you have a link to double blind testing results, or is this your opinion?

This is not just opinion but experience. It is not the same. In fact there is a huge difference between the two.

 

[somebodyelse]

This is not just opinion but experience. It is not the same. In fact there is a huge difference between the two.

Was your experience under conditions controlled to avoid known confounding factors? If not then nobody (including yourself) can tell whether the difference you experienced was in the amplifiers or just your brain playing well known tricks.

 

[Alannn]

This is not just opinion but experience. It is not the same. In fact there is a huge difference between the two.

...just opinion???
I would hope your opinions are formed from your experience.

 

[Punchy_Power_Cat]

Was your experience under conditions controlled to avoid known confounding factors?

This is not like "do I hear the ESS imd hump" or anything subtle, but a day and night difference. If you discover new cues and symbols in tracks you have known for 20 years, there is no point discussing if it's just in the brain any further. The sound stage is totally different between the two amps, even vocals sound different. It's silly having to defend that. To me this is like you asking me for a proof that my couch is blue. I can see the couch being blue any day of the week but now you want proof.

The freaking astrophysicists can't even agree on if anything is real or not, but you guys pretend to be robots with "standardized ears". The ugly truth is that human capabilities differ so vastly, that we outperform each other so profoundly that you could put all the math books every written into your neighbors brain and still he wouldn't turn into Einstein!

Same applies here. Not every brain is musical. Not every ear is good. Some are just waaaaaaay better. We have to live with that. Hearing capabilities (as visual capabilities, f.e. edge detection) in anyone can be greatly improved by the use of various drugs entheogens though.

 

[pherax]

Thanks for your expert and experienced comment! Now if I would purchase a topping D50 III, would i connect it to the SR5011 RCA input (to keep the sr5011 functions) or would I better connect it directly to the A5? (Does the sr5011 do anything with analog signals coming in to analog signals going out?)

 

[Beave]

This is not like "do I hear the ESS imd hump" or anything subtle, but a day and night difference. If you discover new cues and symbols in tracks you have known for 20 years, there is no point discussing if it's just in the brain any further. The sound stage is totally different between the two amps, even vocals sound different. It's silly having to defend that. To me this is like you asking me for a proof that my couch is blue. I can see the couch being blue any day of the week but now you want proof.

The freaking astrophysicists can't even agree on if anything is real or not, but you guys pretend to be robots with "standardized ears". The ugly truth is that human capabilities differ so vastly, that we outperform each other so profoundly that you could put all the math books every written into your neighbors brain and still he wouldn't turn into Einstein!

Same applies here. Not every brain is musical. Not every ear is good. Some are just waaaaaaay better. We have to live with that. Hearing capabilities (as visual capabilities, f.e. edge detection) in anyone can be greatly improved by the use of various drugs entheogens though.

Uh huh.

 

[Roland68]

Thanks for your expert and experienced comment! Now if I would purchase a topping D50 III, would i connect it to the SR5011 RCA input (to keep the sr5011 functions) or would I better connect it directly to the A5? (Does the sr5011 do anything with analog signals coming in to analog signals going out?)

You should connect the D50 III directly to the A5; the SR5011 makes no sense in this chain and configuration. It's best to use a balanced connection.

With the SR5011, connect the preamp outputs of the main speakers to the RCA inputs of the A5. You then need to adjust and mark the volume on the A5 to match the surround playback of the SR5011.
This will give you two completely separate systems, one for surround and one for stereo music.
I've been running my system this way for 30 years.

 

[harkpabst]

Playing "praise the lord -asap" bass heavy track with full range signal ~85db (very loud)
2x12070 PBTL = strong visible woofer excursion
2xTPA3251 PBTL = no woofer excursion at all

Either something's seriously broken, or you're not comparing them like for like. This kind of difference between amplifiers is impossible without EQ of filtering. That's as indisputable as the color of my sofa.

 

[Punchy_Power_Cat]

This kind of difference between amplifiers is impossible without EQ of filtering.

I can get the woofers to bounce on that track with the A5 too but I will have to tune the volume up much further, up to a point at which it will get painful to listen to at my distance. There is for sure a difference. The base overall is more "loose", not muddy or unprecise, but clearly not as full and powerful sounding. You will have to consider multiple factors that may be at play here. Let's start by comparing some crucial factors:

2x12070 PBTL
1.6 Ohm max./ ~460W theoretical max.
26V/17A/400W PSU
Gain 26db

2xTPA3251 PBTL
2.0 Ohm max./ ~720W theoretical max.
38V/5A/190W PSU
Gain 23db

I don't know but even though they claim that it can reach much higher spike loads, the GaN PSU seems to be a bottleneck. Both these ICs take measures of their connected power supplies and adjust themselfes to it. If the IC finds that the PSU is lacking, it probably doesn't utilize it's full capabilities (f.e. for safety reasons). We don't know how these methods operate exactly, so it is down to guess, but there is something here.

 

[Roland68]

I can get the woofers to bounce on that track with the A5 too but I will have to tune the volume up much further, up to a point at which it will get painful to listen to at my distance. There is for sure a difference. The base overall is more "loose", not muddy or unprecise, but clearly not as full and powerful sounding. You will have to consider multiple factors that may be at play here. Let's start by comparing some crucial factors:

2x12070 PBTL
1.6 Ohm max./ ~460W theoretical max.
26V/17A/400W PSU
Gain 26db

2xTPA3251 PBTL
2.0 Ohm max./ ~720W theoretical max.
38V/5A/190W PSU
Gain 23db

I don't know but even though they claim that it can reach much higher spike loads, the GaN PSU seems to be a bottleneck. Both these ICs take measures of their connected power supplies and adjust themselfes to it. If the IC finds that the PSU is lacking, it probably doesn't utilize it's full capabilities (f.e. for safety reasons). We don't know how these methods operate exactly, so it is down to guess, but there is something here.

Are you talking about the old Sabaj A20a in the black case, slim and tall?
Bass and treble adjustment set to 0?
EQ menu set to E1 or E8?
I always had the misconception with the old A20a that Sabaj also added some bass or loudness to E1/E8.

By the way, 38 volts is the maximum continuous operating voltage for the TPA3251. The absolute maximum ratings are 50 volts, which is well within the safe range even for transformer-based power supplies. You don't have to worry about 38 volts.

 

[harkpabst]

I can get the woofers to bounce on that track with the A5 too but I will have to tune the volume up much further, up to a point at which it will get painful to listen to at my distance. There is for sure a difference. The base overall is more "loose", not muddy or unprecise, but clearly not as full and powerful sounding. You will have to consider multiple factors that may be at play here. Let's start by comparing some crucial factors:

2x12070 PBTL
1.6 Ohm max./ ~460W theoretical max.
26V/17A/400W PSU
Gain 26db

2xTPA3251 PBTL
2.0 Ohm max./ ~720W theoretical max.
38V/5A/190W PSU
Gain 23db

I don't know but even though they claim that it can reach much higher spike loads, the GaN PSU seems to be a bottleneck. Both these ICs take measures of their connected power supplies and adjust themselfes to it. If the IC finds that the PSU is lacking, it probably doesn't utilize it's full capabilities. We don't know how these methods operate exactly, so it is down to guess, but there is something here.

Max power output has been measured. There really shouldn't be any surprise. The LS50 Meta is less hard to drive than many assume. See here:

KEF LS50 Meta loudspeaker Measurements | Stereophile.com

Sidebar 3: Measurements I used DRA Labs' MLSSA system and a calibrated DPA 4006 microphone to measure the KEF LS50's frequency response in the farfield, and an Earthworks QTC-40 for the nearfield and spatially averaged room responses. My estimate of the LS50 Meta's voltage sensitivity was...

www.stereophile.com

EPDR is low in certain frequency regions, that's true, and sensitivity is lowish. But the amount of air they can move is very limited and so is (deep) bass output. You will hear very ugly noises from the voice coils hitting the pole plates long before any of these amplifiers is maxed out.

When I said something is broken or you're not comparing them like for like I literally meant it. If nothing is broken, something must be enhancing the bass with the other amp. EQ would perfectly explain what's happening.

 

[Punchy_Power_Cat]

Are you talking about the old Sabaj A20a in the black case, slim and tall?
Bass and treble adjustment set to 0?
EQ menu set to E1 or E8?
I always had the misconception with the old A20a that Sabaj also added some bass or loudness to E1/E8.

Very interesting. I totally forgot they had EQ and tone controls. Never used any of it but there is a good chance you are correct with the assumption. I will have it checked.

Anyways, the GaN PSU is still not out of questioning. I have read reports of high frequency noise and there are safety concerns with such low cost PSU in any case. Certificates seem fake too. The manufacturer is blacklisted by german TUV for fake certificates on other PSU units. I am contemplating a Mean Well LOP-400-36 to run the A5. It has ~95% efficiency with loads below 10%. Probably the most efficient PSU for amps there is. Rise time is about 6ms and ripple is very low @1Khz as it is @100Hz. It even has 150% peak ability.

 

[Roland68]

Very interesting. I totally forgot they had EQ and tone controls. Never used any of it but there is a good chance you are correct with the assumption. I will have it checked.

Anyways, the GaN PSU is still not out of questioning. I have read reports of high frequency noise and there are safety concerns with such low cost PSU in any case. Certificates seem fake too. The manufacturer is blacklisted by german TUV for fake certificates on other PSU units. I am contemplating a Mean Well LOP-400-36 to run the A5. It has ~95% efficiency with loads below 10%. Probably the most efficient PSU for amps there is. Rise time is about 6ms and ripple is very low @1Khz as it is @100Hz. It even has 150% peak ability.

I would recommend the Meanwell HRP-150N3-36 or HRP-300N3-36. You definitely won't need more than 300 watts with the A5, and it can be set to 38 volts.
In my opinion, the HRP N3 series is the best power supply for audio amplifiers, as it's designed to handle rapid load changes and delivers 350% power for 5 seconds. At 300 watts, that would be over 1000 watts, or over 30 A.

 

[Punchy_Power_Cat]

I would recommend the Meanwell HRP-150N3-36 or HRP-300N3-36

Yes, I have been looking closely into these.

Meanwell HRP-150N3-36
TRANSIENT RECOVERY TIME: 169mVp-p
DYNAMIC LOAD 120Hz: 196mVp-p | 1Khz: 195mVp-p
Leakage Current: 0.833mA
Rise Time: 26ms

Meanwell HRP-300N3-36
TRANSIENT RECOVERY TIME: 441mVp-p
DYNAMIC LOAD 120Hz: 687mVp-p | 1Khz: 420mVp-p
Leakage Current: 1.477mA
Rise Time: 6.5ms

Meanwell LOP-400-36
TRANSIENT RECOVERY TIME: 380mVp-p
DYNAMIC LOAD 120Hz: 680mVp-p | 1Khz: 600mVp-p
Leakage Current: 322uA
Rise Time: 6.4ms

I'm not an expert but I suppose leakage current can under some circumstances reach your speakers, hence people like to buy medical PSUs for they have near to none. However, this might be a factor with internally installed PSUs not with external units. The HRP300 has better ripple@1Khz which is the most important figure I suppose, but this is measured by a 50% duty-load and the LOP is more powerful (peak ability not considered) so it is actually probably a little better in terms of ripple. Both have equally very fast rising times so they should be both good for audio. The HPR150 has by far the lowest ripple but it less than half the power. The rise time might be problematicly high tho. I don't know if this would affect/delay the sound. With those 350% peak ability I'm not sure if those aren't achieved by the use of extraordinarily high ripple/high frequency noise. There are no figures for the ripple at those peaks.