Looking for CLASS A power amp (no phase shift). Only Burson?

[Crodo]

Dear users,
I'm new for this forum. I'm from Italy and I need suggestions about amplifiers. I'm not expert of the hi-fi world. I works as recording engineer and I have always reffered at "pro" products. Now I need an ampli for drive a speaker with these caracteristics:

- Wattage: 25 Watt RMS / 50 Watt Peak
- Impedance: 8 OHMS

I looking for a small CLASS A amp (or AB) with two important caracteristics:


- no phase shift deviation
- a good crosstalk between channels
- if possible XLR input

I've found the Burson Bang, rewievd from trl BANG TEST. I don' know the hi-fi market so I ask for suggestion.
Which other brands and models might be right for me?

Thanks in advance!

Crodo.

 

[Hrodulf]

You should look for a Neurochrome amp. Hard to fault them measurement wise and they sound really good. Benchmark's amp also seems to fit the bill.

 

[Crodo]

Thanks Hrodulf, Neurochrome specs are very good but I don't see phase measurement. It's a pity. Also for Benchmark.

 

[MZKM]

I don't see phase measurement.

Because it most is an unnecessary test (and thus increases the workload of the measurement process). I highly doubt any amp in your price range has any phase drift that is even remotely audible.

And, if I recall (as long as the phase is constant between channels), a phase shift is usually a time delay (it is a higher angle of degree with higher frequencies as their waveforms are shorter), so it is the same as all frequencies being delayed equally, thus still in sync; I could be wrong though.

 

[Crodo]

Phase shift is at diffent frequencies, dalay at all freq.

 

[boXem]

Thanks Hrodulf, Neurochrome specs are very good but I don't see phase measurement. It's a pity. Also for Benchmark.

Putting aside discussions about audibility of phase shift, your requirement "no phase shift" is physically impossible.
Any system has a reaction time.
Let's take as an example a reaction time of 500 ns.
For 20 Hz, 500 ns of reaction time represents a phase shift of 0.0000005 * 360 / (1 / 20) = 0.0036 °
For 20 kHz, the same reaction time represents a phase shift 1000 times higher: 3.6 °.

 

[levimax]

Thanks Hrodulf, Neurochrome specs are very good but I don't see phase measurement. It's a pity. Also for Benchmark.

Ask Tom about phase, he will answer e-mails and is a contributor on these boards. I have 4 Mod-86 amps and they work very well.

 

[dwkdnvr]

Phase shift is at diffent frequencies, dalay at all freq.

Phase shift is derived from the frequency response. If an amp is flat, it will have flat phase response within it's operating bandwidth. Since all amps are bandwidth limited, they'll all eventually have some level of phase shift. You should reconsider your requirements and realize that what you're asking for is bandwidth.

The way you phrase this question makes me suspect that you're buying into some Audiophile Marketing hype from somewhere. Even establishing the audibility of phase in the context of speakers seems to indicate that it isn't audible - Linkwitz for example listened over headphones trying to hear the 180-degree phase shift associated with an LR2 crossover, and concluded for himself that it wasn't audible. In this context, a few degrees of phase shift at 20kHz associated with high-end rolloff of an amp is almost certainly inaudible.

But, as always, you're free to spend your time and money where you choose.

 

[Panelhead]

Found your amplifier. Micromega Microamp. 25 watt, Class A, 1 degree phase from 20 Hz to 20 kHz.

 

[Panelhead]

Found you another. Micromega Tempo. Newer than Microamp. 50 watts. Less than 1 degree phase shift from 20 - 20kHz.
Expect some of this is bandwidth related, DC to 500kHz, -3 dB.

 

[trl]

Dear users,
I'm new for this forum. I'm from Italy and I need suggestions about amplifiers. I'm not expert of the hi-fi world. I works as recording engineer and I have always reffered at "pro" products. Now I need an ampli for drive a speaker with these caracteristics:

- Wattage: 25 Watt RMS / 50 Watt Peak
- Impedance: 8 OHMS

I looking for a small CLASS A amp (or AB) with two important caracteristics:


- no phase shift deviation
- a good crosstalk between channels
- if possible XLR input

I've found the Burson Bang, rewievd from trl BANG TEST. I don' know the hi-fi market so I ask for suggestion.
Which other brands and models might be right for me?

Thanks in advance!

Crodo.

Hi, BANG is not Class-A, it's class A/B, which is quite OK. It should drive your 8-Ohms speakers well and if small size and low price is in your mind, then BANG might be a good choice.

 

[trl]

Phase shift is at diffent frequencies, dalay at all freq.

There is no phase shift on LM3886 amps, trust me. If not, then let's @tomchr jump in.
There's also no pre and no post ringing too with square-waves @20Hz and @20KHz, again Tom will need to confirm this, but I am 110% I'm right.

 

[andreasmaaan]

Unless the amp has infinite bandwidth (which is impossible), it will have non-constant (almost certainly inaudible) group delay.

 

[trl]

We're speaking about phase shift across audible bandwidth, there should be no phase shift.

 

[RayDunzl]

no phase shift deviation

Here's my fifteen year old amp, at 5W, driving 4 ohm speakers:

Frequency response +/- .06dB

Phase nearly flat except rising a little in the bass, to 7 degrees.

Group Delay - 1ms at 20Hz

You will find the phase shift from conventional speakers to be far far more. and even more if measured at your listening position as reflections from the room creates chaos..

Disclaimer: These are not "lab grade" measurements, just what I can throw together with what I have.

 

[tomchr]

We're speaking about phase shift across audible bandwidth, there should be no phase shift.

It depends on what you mean by "no phase shift". From a theoretical point of view, having zero phase shift is a physical impossibility for any real circuit. The circuit would need to have infinite bandwidth. Similarly, it is a mathematical (and statistical!) impossibility to have a circuit that has exactly 0.000º phase shift from 20.000 Hz to 20.000 kHz and phase shift elsewhere.

The LM3886 has limited bandwidth. Even if you scrap the "not-so-optional" stability components needed to make the amp work well, you won't get more than about 1 MHz bandwidth out of it. So you'll have some fraction of a degree of phase shift at the upper end of the audio band. A well-designed LM3886 amp, such as my LM3886DR, typically has a bandwidth of around 80-100 kHz, so you'll get a few degrees of phase shift at 20 kHz.

Similarly, unless you implement a DC offset adjustment, you'll likely place a pole around 0.2-2 Hz to prevent the LM3886 from putting out significant DC offset. That gives you a few degrees of phase shift at 20 Hz.

So the LM3886 will have some phase shift. Physics made it so!

The question is whether that phase shift is audible or whether phase shift more broadly is audible. I'm sure the latter is a barrel of worms that has been discussed ad nauseam elsewhere. I highly doubt the few degrees of phase shift caused by any reasonably well-designed amp is audible.

As for your question about ringing on the square wave: That depends entirely on the design of the amp. I design mine to have a clean square wave with a reasonable capacitive load. Usually "reasonable" exceeds 100 nF. I also design my amps to be stable with significant capacitive load (> 1 uF). This is a rather tall order, by the way. Other designers may make different design decisions for a variety of reasons.
I've attached the transient response plots of the LM3886DR for 100 nF || 8 Ω load and 1.0 uF || 8 Ω load. The response at 100 nF load is clean. The ringing that you see on the 1.0 uF plot is caused by the resonance of the output inductor in the amp and the load cap (darn physics again!). The amp itself is stable and rock solid.

Tom

 

[tomchr]

Thanks Hrodulf, Neurochrome specs are very good but I don't see phase measurement. It's a pity. Also for Benchmark.

If there's enough market demand for such measurements, I can certainly add them. Unfortunately the AP doesn't automatically give me the phase with the amplitude response. That would have been nice...

Tom

 

[andreasmaaan]

We're speaking about phase shift across audible bandwidth, there should be no phase shift.

Also impossible without infinite bandwidth

Sorry, I'm not meaning to troll the thread, I just think it would make more sense to specify some allowable degree of (inevitable) phase shift instead of demanding the impossible.

And to be honest, I'd also be curious as to what the rationale is?

 

[tomchr]

- no phase shift deviation

As outlined above, that's unfortunately a physical impossibility.

- a good crosstalk between channels
- if possible XLR input

My Modulus-86, Modulus-186, or Modulus-286 would fit the bill nicely.

The Modulus-86 is all DIY. The 186 and 286 come as fully assembled modules. The 286 is the most versatile of the amps. It'll deliver 60-70 W into 8 Ω and about 125 W into 4 Ω. The 186 is more like 40-45 W into 8 Ω and 65 W into 4 Ω. I go into more detail in my Amplifier Selection Guide.

Tom

 

[Thomas savage]

Btw this guys not been back since April, the day he posted this .