108dB/W @ 1m translates to approx 100dB/W @ 2.5m listening distance.
If you have two of them in front of you then 1W to each gives an SPL of ~103dB
If your amplifier has a 1Watt Signal to Noise ratio of 103dB, then the noise will just about be inaudible at that listening distance. At half a metre it will be just audible in a very quiet room.
-103dB ref 1W is 0.02mVac (20uVac)
That is a mighty low noise value. Especially from a chipamp using IC technology.
And is equivalent to an input referred noise of 7.1nV per rootHertz, when gain is set to 20times (+26dB).
If you have two of them in front of you then 1W to each gives an SPL of ~103dB
If your amplifier has a 1Watt Signal to Noise ratio of 103dB, then the noise will just about be inaudible at that listening distance. At half a metre it will be just audible in a very quiet room.
-103dB ref 1W is 0.02mVac (20uVac)
That is a mighty low noise value. Especially from a chipamp using IC technology.
And is equivalent to an input referred noise of 7.1nV per rootHertz, when gain is set to 20times (+26dB).
It also means you are continually operating in a range where crossover (zero crossing) distortion is at its worst . . . which is why you will almost never see distortion specs for commercial amps even at 1 Watt, let alone 10 or 100 milliwatts. Most Class AB amps test better at high power than low, but low is where we listen most, and it's especially notable with high sensitivity drivers.
And that's why the improved noise and distortion performance of the Modulus-86 matters.
Conceptually it's a nice idea, but in practice all you see is phase shift, which is difficult to match and null. You end up testing for something that's essentially inaudible, and missing what is . . .
Even with 88dB/Watt sensitivity speakers 1W is barely low enough . . . you really need to see what's happening at 100 milliwatt and below. That's where even "quality" amplifiers start to diverge (and where they actually can sound "different"). Higher sensitivity (horn) drivers push that need an order of magnitude lower . . .
As noted by Andrew (above), with a 108 dB/Watt horn 1 Watt is essentially maximum power (in a home environment), and "normal" listening is a lot lower . . . so testing at 1, 10 and "max" tells you essentially nothing about normal use.
It is possible to create a reference file filtered the same way the amplifier filters the signal to reduce the effects of phase shift. In practice, this would only be an indicator how close the output follows input under real music, then when you start adding realistic load, you will see how it effects the results. This kind of difference testing is just one of the types of test you would use in the development process.
Well, I think horns create distortion that will mask much of the low level amp distortion anyway. Additionally, I would normally sit about a minimum of 4 meters away from a horn, so figure what actual wattage we will be using.
Possible, perhaps, but not easy. There are lots of poles and zeros in amplifiers . . .
You are wrong in that. While all drivers "create distortion", compression drivers included, that does not "mask" either amplifier noise or the particular kind of IM "splatter and grunge" that many Class AB amps produce at low signal levels.
You are able or correlate that through measurements? I have not seen such. Perhaps you can point to some reference.
I'm perfectly capable of using Google, Bing, and other search engines. Thank you. You still haven't listed the reference records you are expecting a reviewer to use. Are you expecting them to use all the ones reported by Google?
That looks roughly like the data sheet performance of an LM3886. That's still significantly worse than the MOD86. I'm not seeing anything noteworthy there. Nice LM3886 build, but that's about it...
Tom
Last edited:
You know that there are plenty of reference recordings, many of which have, especially excellent vinyl rips FLAC 24/96 with very good sound, interpretation and DR too.
I have no doubt that your developments are the best but to take advantage of a DAC is required with balanced outputs, which I have not. The best and the more expensive too!
In no balanced development I think the best is Akitika GT-101 of Dan Joffe (crossed a few emails). I studied it thoroughly, even chose the step potentiometer, a switch audiophile, maybe the op-amp...
I have no doubt that your developments are the best but to take advantage of a DAC is required with balanced outputs, which I have not. The best and the more expensive too!
In no balanced development I think the best is Akitika GT-101 of Dan Joffe (crossed a few emails). I studied it thoroughly, even chose the step potentiometer, a switch audiophile, maybe the op-amp...
Last edited:
You don't need a balanced output to drive a balanced (differential) input.
Now I listen this great vinyl rip 24/96 FLAC.
foobar2000 1.3.8 / Dynamic Range Meter 1.1.1
log date: 2015-07-11 19:06:26
--------------------------------------------------------------------------------
Analyzed: ? / ?
--------------------------------------------------------------------------------
DR Peak RMS Duration Track
--------------------------------------------------------------------------------
DR14 -0.84 dB -22.16 dB 10:54 ?-A1 El salon Mexico
DR15 -1.59 dB -21.55 dB 8:01 ?-A2 Danzon Cubano
DR13 -1.63 dB -20.86 dB 11:12 ?-A3 Symphony No3 - 1, Molto moderato
DR13 -1.75 dB -19.55 dB 8:14 ?-B1 Symphony No3 - 2, Allegro molto
DR16 -7.02 dB -30.01 dB 9:41 ?-B2 Symphony No3 - 3, Andantino quasi allegretto
DR11 -1.32 dB -17.54 dB 12:37 ?-B3 Symphony No3 - 4, Molto deliberato
--------------------------------------------------------------------------------
Number of tracks: 6
Official DR value: DR14
Samplerate: 96000 Hz
Channels: 2
Bits per sample: 24
Bitrate: 2510 kbps
Codec: FLAC
====================================================
An externally hosted image should be here but it was not working when we last tested it.
foobar2000 1.3.8 / Dynamic Range Meter 1.1.1
log date: 2015-07-11 19:06:26
--------------------------------------------------------------------------------
Analyzed: ? / ?
--------------------------------------------------------------------------------
DR Peak RMS Duration Track
--------------------------------------------------------------------------------
DR14 -0.84 dB -22.16 dB 10:54 ?-A1 El salon Mexico
DR15 -1.59 dB -21.55 dB 8:01 ?-A2 Danzon Cubano
DR13 -1.63 dB -20.86 dB 11:12 ?-A3 Symphony No3 - 1, Molto moderato
DR13 -1.75 dB -19.55 dB 8:14 ?-B1 Symphony No3 - 2, Allegro molto
DR16 -7.02 dB -30.01 dB 9:41 ?-B2 Symphony No3 - 3, Andantino quasi allegretto
DR11 -1.32 dB -17.54 dB 12:37 ?-B3 Symphony No3 - 4, Molto deliberato
--------------------------------------------------------------------------------
Number of tracks: 6
Official DR value: DR14
Samplerate: 96000 Hz
Channels: 2
Bits per sample: 24
Bitrate: 2510 kbps
Codec: FLAC
====================================================
Last edited:
Bill--you can turn it down a notch. We DIY for our enjoyment and to appease our many psychoses, so let's let people do that!
The lounge is for Trolls and loonies. There has been some really good discussion on this thread which keeps getting swamped by said T&L turning up. It annoys me. Anyone can start their own thread to discuss their world view on why clockwise wired Thiele network inductors hand painted in rare ungents sound better or other such tripe, but it would be really nice to keep it out of here. Likewise which reference records the designer should use so someone can decide if the snap of a snare drum will be as tight as he wants.
The T&L noise makers have also swamped a more serious discussion on what we can do to prevent the oscillations with our friends Mod-86, which is on topic and useful.
That's nice and off topic. There is a thread 'what are you listening to now' Go post there!
And if you do have a differential input you can improve the output (noise immunity) of an "unbalanced" device with a single resistor (and 3 pin connector). http://www.hypex.nl/docs/papers/The G Word.pdf