What happens when an amplifier's noise floor stops rising in loudness? - AVS Forum | Home Theater Discussions And Reviews
Forum Jump: 
 
Thread Tools
post #1 of 9 Old 07-05-2019, 12:38 PM - Thread Starter
Senior Member
 
Join Date: Jan 2015
Posts: 264
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 178 Post(s)
Liked: 34
What happens when an amplifier's noise floor stops rising in loudness?

The amp in question is actually a receiver: Yamaha RX-V381.

The test signal is 10 minutes of mute content (-infinite dB True Peak) in a 24 bit 48 KHz 5.1 audio track.

When playing back the track, if I leave all channels active and I max out all channels trims at +10 dB and the input signal trim at +6 dB, the noise floor (the hiss coming from the speakers) stops rising at +14 dB (I mean above 0.0 dB) out of a max MV of +16.5 dB.

If I disable (downmix) the center channel and/or the surrounds, the noise floor stops rising at +8 dB of the MV.

If I disable the subwoofer, the noise floor stops rising at -8.5 dB.

Also, with the MV left untouched, the hiss (noise floor) from the front speakers gets louder when disabling the center/surrounds and/or the sub, but playback level of actual content, like a movie, remains the same.

Of course, I also tried with channels and input trims at the neutral 0 dB, but, then, the MV levels at which the noise floor stops rising are simply an equal amount higher (16 dB), thus, with the sub disabled, the noise floor stops rising at +7.5 dB of the MV, but with all channels enabled, the noise floor never stops rising, because I can't go beyond +16.5 dB.

Lastly, if I hit the pause button, the hiss disappears.

My question is, technically, what is happening when I reach the point where the noise floor stops rising in loudness?

Last edited by Optimus_Fine; 07-05-2019 at 01:05 PM.
Optimus_Fine is offline  
Sponsored Links
Advertisement
 
post #2 of 9 Old 07-06-2019, 09:58 AM
AVS Forum Addicted Member
 
BassThatHz's Avatar
 
Join Date: Apr 2008
Location: Northern Okan range (NW Cascades region)
Posts: 10,632
Mentioned: 221 Post(s)
Tagged: 0 Thread(s)
Quoted: 3678 Post(s)
Liked: 4169
I wonder if the muted content isn't actually muted... have you verified this?
Perhaps it is re-sampling and applying dither.

Entry-level receivers are known for being cheap and having cut corners. Made by robots 24/7/365 in china by the millions of units.

If you have high efficiency speakers then you may have to upgrade to something of higher quality, perhaps a more expensive receiver or even separates.

I'd be curious to see what would happen if you brought this disc into Magnolia (or any other hifi place) and tried the same thing on one of their more expensive products (try several different brands). See if they are better, the same or worse.

Then you'll at least know the problem is solvable and what product to save up for.
If you have the money you could even bring it home for a demo with a no-questions return policy I'm sure...

I get it that you are trying to understand the why of it, but it could be a million different reasons of why it does this and we can only make guesses. (and those guesses won't fix it either.)
You may just want to take the Yamaha RX-V381 back and get a refund, while you still can/if possible.

One unanswered question I have is:
Why do you need a million db of gain to start with, why cranked so loud?
Generally that indicates the speakers are too inefficient and the unit lacks the power to make up the difference. Just wondering what the backstory of your inquiry is...
Generally it is unusual to NEED to crank a device to a high setting and then listening for amplified quantum-noise.

Last edited by BassThatHz; 07-06-2019 at 10:12 AM.
BassThatHz is offline  
post #3 of 9 Old 07-07-2019, 04:27 AM - Thread Starter
Senior Member
 
Join Date: Jan 2015
Posts: 264
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 178 Post(s)
Liked: 34
There's been a misunderstanding.

Those levels are not the ones I use for playback of movie soundtracks; they're not even necessary to reach Reference Level at my listening distance of 3 meters (-20 dBFS RMS at 85 dBC measured with my external SPL meter and tone calibrator), as I can get that at over 30 dB below the level at which the SNR stops increasing.

I simply wanted to know what was happening when I pushed the RX-V381 to the max and with different channels configurations.

Anyway, I made another test: this time I mastered and encoded in Dolby D+ a 1 KHz tone at -90 dBFS RMS in the center channel, in order to be tolerable to hear.

As I predicted, the tone stops increasing in loudness at exactly the same level where the SNR stops increasing too, which could mean that that's the RX-V381 maximum output, which shifts depending on channels trims and input signal trim levels and/or channels configurations.

As soon as I can get a good AC Voltage meter with a great price/accuracy ratio, I'll measure the sensitivity of my speakers.
Optimus_Fine is offline  
Sponsored Links
Advertisement
 
post #4 of 9 Old 07-08-2019, 03:30 AM - Thread Starter
Senior Member
 
Join Date: Jan 2015
Posts: 264
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 178 Post(s)
Liked: 34
So, in the meantime, I've objectioned to myself that this AVR cannot have so much audio power and there must be something else going on.

I've read somewhere that, when an AVR is set with MV, channels trims and input trim at 0 dB, it is outputting a source signal at 1 V RMS, thus with no gain.

Is that true?

If so, would a full scale sine wave at 2.828 V RMS be obtained at +9 dB on the MV?

Last edited by Optimus_Fine; 07-08-2019 at 10:39 AM.
Optimus_Fine is offline  
post #5 of 9 Old 07-08-2019, 10:46 AM
AVS Forum Special Member
 
bigguyca's Avatar
 
Join Date: Oct 2007
Location: East Turkestan
Posts: 1,543
Mentioned: 27 Post(s)
Tagged: 0 Thread(s)
Quoted: 955 Post(s)
Liked: 776
Quote:
Originally Posted by Optimus_Fine View Post
The amp in question is actually a receiver: Yamaha RX-V381.

The test signal is 10 minutes of mute content (-infinite dB True Peak) in a 24 bit 48 KHz 5.1 audio track.

When playing back the track, if I leave all channels active and I max out all channels trims at +10 dB and the input signal trim at +6 dB, the noise floor (the hiss coming from the speakers) stops rising at +14 dB (I mean above 0.0 dB) out of a max MV of +16.5 dB.

If I disable (downmix) the center channel and/or the surrounds, the noise floor stops rising at +8 dB of the MV.

If I disable the subwoofer, the noise floor stops rising at -8.5 dB.

Also, with the MV left untouched, the hiss (noise floor) from the front speakers gets louder when disabling the center/surrounds and/or the sub, but playback level of actual content, like a movie, remains the same.

Of course, I also tried with channels and input trims at the neutral 0 dB, but, then, the MV levels at which the noise floor stops rising are simply an equal amount higher (16 dB), thus, with the sub disabled, the noise floor stops rising at +7.5 dB of the MV, but with all channels enabled, the noise floor never stops rising, because I can't go beyond +16.5 dB.

Lastly, if I hit the pause button, the hiss disappears.

My question is, technically, what is happening when I reach the point where the noise floor stops rising in loudness?
Quote:
Originally Posted by Optimus_Fine View Post
So, in the meantime, I've got into questioning to myself that this AVR cannot have so much audio power and there must be something else going on.

I've read somewhere that, when an AVR is set with MV, channels trims and input trim at 0 dB, it is outputting a source signal at 1 V RMS, thus with no gain.

Is that true?

If so, would a full scale sine wave at 2.828 V RMS be obtained at +9 dB on the MV?

Because that would give a total headroom of +16.5 dB of MV -9 dB at 2.83 V +10 dB of channels trims + 6 dB of input trim = 23.5 dB of gain, which amounts to 224 V, which is within the 220-240 V electric tension of the wall plug and the AVR transformer.

But, that would make my speakers insanely efficient: considering that, at 1 m, I measure a single speaker at 85 dBC, with my calibrated IEC 61672 SPL meter, at -15 dB MV with a -20 dBFS RMS narrow band pink noise encoded in Dolby Digital with DialNorm -31, that would, theoretically, equal to measure, with a full scale sine wave, 85 + 20 + 15 + 9 = 129 dB SPL at +9 dB MV (if that was a 2.83 V output level).

Now, 100% efficiency equals 112 dB SPL and my speakers, even accounting for a +18 dB of 6 boundaries gain, would have a free field sensitivity of 111 dB and a 79% efficiency, with just a 13 liters volume, 5.25" midrange and 2.5" tweeter.

You have based your posts on the an estimate of the output of the internal preamp stage, likely the output of the volume control. A typical output at 0 dBFS would be 1V to 2V RMS. The only RCA output on your AVR are for the subwoofer so an assumption can be made that you are connecting your speakers to the power amplifier outputs on the back of the unit, not to the internal preamp stage.

A typical AVR power amplifier has a gain of about 28x, that is, about 29 dB. If the output of the preamp stage is 1V RMS at 0 dBFS, then the output of the amplifier section, to which you have connected your speakers, can be estimated at 1 x 28 = 28V RMS.

The are a lot of approximations in your last post, but using the 111 dB number, something like 111 - 29 = 82 dB is more in the ballpark, but that could vary a lot, depending on the validity of you other a assumptions. Grossly, 82 dB SPL gives a power efficiency of something like .1%. 90dB SPL would be around .6%. Typical home loudspeakers have a low power efficiency.
bigguyca is offline  
post #6 of 9 Old 07-08-2019, 02:17 PM - Thread Starter
Senior Member
 
Join Date: Jan 2015
Posts: 264
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 178 Post(s)
Liked: 34
Quote:
Originally Posted by bigguyca View Post
A typical AVR power amplifier has a gain of about 28x, that is, about 29 dB. If the output of the preamp stage is 1V RMS at 0 dBFS, then the output of the amplifier section, to which you have connected your speakers, can be estimated at 1 x 28 = 28V RMS.
Yeah, I've kind of gone too much far off with those calculations.

So, assuming my AVR outputs 28.28 V RMS at 0 dB MV/0 dB channels trims/0 dB input trim, it should go like the following:

-29 dB MV = 1 V RMS at 0 dBFS RMS
-20 dB MV = 2.828 V RMS at 0 dBFS RMS
-9 dB MV = 1 V RMS at -20 dBFS RMS
0 dB MV = 2.828 V RMS at -20 dBFS RMS

Here's what I can't wrap my head around with the data above: each of my speakers at 1 m measures 85 dBC at -15 MV with Dolby + DialNorm -31 and at -11 dB MV with DTS + DialNorm -31.

Even with the worst case scenario of -11 dB MV, if I add 85 dB and the 20 dB of signal headroom, I end up with 116 dB SPL of sensitivity including room gain.

Even if I take off 18 dB of 6 boundaries gain, which is a worst case scenario in on itself, the net result is still a whopping 98 dB SPL of free field sensitivity.

But with an enclosure of 13 liters, 5.25" + 1" drivers and a rated 4 Ohms impedance, is that even physically possible?
Optimus_Fine is offline  
post #7 of 9 Old 07-08-2019, 03:10 PM
AVS Forum Special Member
 
bigguyca's Avatar
 
Join Date: Oct 2007
Location: East Turkestan
Posts: 1,543
Mentioned: 27 Post(s)
Tagged: 0 Thread(s)
Quoted: 955 Post(s)
Liked: 776
Quote:
Originally Posted by bigguyca View Post
You have based your posts on the an estimate of the output of the internal preamp stage, likely the output of the volume control. A typical output at 0 dBFS would be 1V to 2V RMS. The only RCA output on your AVR are for the subwoofer so an assumption can be made that you are connecting your speakers to the power amplifier outputs on the back of the unit, not to the internal preamp stage.

A typical AVR power amplifier has a gain of about 28x, that is, about 29 dB. If the output of the preamp stage is 1V RMS at 0 dBFS, then the output of the amplifier section, to which you have connected your speakers, can be estimated at 1 x 28 = 28V RMS.

The are a lot of approximations in your last post, but using the 111 dB number, something like 111 - 29 = 82 dB is more in the ballpark, but that could vary a lot, depending on the validity of you other a assumptions. Grossly, 82 dB SPL gives a power efficiency of something like .1%. 90dB SPL would be around .6%. Typical home loudspeakers have a low power efficiency.
Quote:
Originally Posted by Optimus_Fine View Post
Yeah, I've kind of gone too much far off with those calculations.

So, assuming my AVR outputs 28.28 V RMS at 0 dB MV/0 dB channels trims/0 dB input trim, it should go like the following:

-29 dB MV = 1 V RMS at 0 dBFS RMS
-20 dB MV = 2.828 V RMS at 0 dBFS RMS
-9 dB MV = 1 V RMS at -20 dBFS RMS
0 dB MV = 2.828 V RMS at -20 dBFS RMS

Here's what I can't wrap my head around with the data above: each of my speakers at 1 m measures 85 dBC at -15 MV with Dolby + DialNorm -31 and at -11 dB MV with DTS + DialNorm -31.

Even with the worst case scenario of -11 dB MV, if I add 85 dB and the 20 dB of signal headroom, I end up with 116 dB SPL of sensitivity including room gain.

Even if I take off 18 dB of 6 boundaries gain, which is a worst case scenario in on itself, the net result is still a whopping 98 dB SPL of free field sensitivity.

But with an enclosure of 13 liters, 5.25" + 1" drivers and a rated 4 Ohms impedance, is that even physically possible?
There is no reason to believe you have 20dB of headroom. Calibration programs don't check to see if the amplifier/speaker combination is capable of the implied output. For example, many people run say Audyssey, which runs at 75dB SPL per channel and adjusts to 85dB SPL with 20dB of implied headroom to 105dB maximum at the listening position, and believe they are getting all that. This Audyssey calibration in no way means that the AVR or amplifier in use can cleanly drive the loudspeaker to 105dB SPL or that their speaker can perform at the level. Audyssey didn't measure for that, it didn't even measure to 85dB. The calibration program did its room calibration, but really didn't find out if the system can perform at any specific SPL or acceptably at a specific SPL. Few systems can do 105dB SPL per speaker at the listening position, without compression or horrible distortion and overloads, or can do it at all. One of the reasons many people listen at -10dB or lower levels IMNVHO, is that the system is unpleasant, due to being overloaded, above that level.

In the -11dB case above, adding 11dB of gain evidently gets you to 0 dBFS which in the calculations above (for ease of calculations, 28V RMS say into 8 ohms, your AVR doesn't seem to be rated into 4 ohms) gets you to the limit of your AVR of about 100 watts per channel. (Assuming I've understood your post, which at this point is becoming less likely!) That puts you at 85 + 11 = 96 dB SPL. How linear one of your speakers at that level at your listening position - unknown.

And no, your speakers as you have briefly described them, likely can't each do a clean 105dB SPL at your listening position.

Last edited by bigguyca; 07-08-2019 at 03:19 PM.
bigguyca is offline  
post #8 of 9 Old 07-12-2019, 02:18 AM - Thread Starter
Senior Member
 
Join Date: Jan 2015
Posts: 264
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 178 Post(s)
Liked: 34
So, I decided to buy a Mercury MTB01 multimeter.

It has a 45-450 Hz frequency response and a 2% accuracy, which is less than the RX-V381 master volume precision of 0.5 dB.

First test was the AVR AC Volt output with a 200 Hz tone at -0.01 dBFS RMS encoded in the left channel of a Dolby Digital Plus 7.1 track with DialNorm -31.

The AVR was setup with all channels active and with trims at 0 dB and input trim at 0 dB.

1.0 V at -32 dB MV
2.8 V at -25 dB MV
9.8 V at -15 dB MV
28.6 V (+29.1 dB) at -6 dB MV
35.2 V (+31 dB) at -4.0 dB MV (beyond this level, with Dolby, linearity drops, thus, with LPCM and DTS with DialNorm -31, that will occur beyond 0 dB MV)
37.6 V (+31.5 dB) at -3.0 dB MV
39.7 V (+32 dB) at -2.0 dB MV
41.1 V (+32.25 dB) at -1.0 dB MV
42.5 V (+32.5 dB) at 0 dB MV
50.8 V (+34.1 dB) at +16.5 dB MV (max volume)

Second batch was done with a 200 Hz tone at -20.00 dBFS RMS, still encoded in the left channel, but, this time, of a Dolby Digital 5.1 640 kbps track with DialNorm -31.

1 V at -12 dB MV
2.8 V at -5 dB MV
10 V at +5 dB MV
27.7 V at +13.5 dB MV
29.4 V at +14 dB MV

After that, I measured the in room sensitivity of my left channel Pioneer loudspeaker, with a calibrated Sinometer SL812 and with the same 500-2000 Hz pink noise at -20.00 dBFS RMS at 2.8 V output (-5 dB MV), at 1 m on axis: the result was 95.4 dB LAeq.

To obtain 85 dB SPL, though, I had to drop the MV at -17 dB, in order to measure 84.9 dB LAeq, which suggests that at 2.8 V the speaker was compressing by 1.5 dB SPL.

Last edited by Optimus_Fine; 07-12-2019 at 02:36 AM.
Optimus_Fine is offline  
post #9 of 9 Old 07-12-2019, 02:47 AM
AVS Forum Special Member
 
torii's Avatar
 
Join Date: Mar 2015
Posts: 6,689
Mentioned: 47 Post(s)
Tagged: 0 Thread(s)
Quoted: 3318 Post(s)
Liked: 1871
I hate to say this, but most people arent good at math. I would start with something simple like how loud is your system at what volume on avr and then work backwards. most math nerds like myself failed a 5 page calculus answer by missing 1 step/1 detail. whats a noise floor? the terms you are using might lead to confusion.

Power: Marantz sr7008, NAD C 275Bee x 2, Video: Oppo 103, Samsung 75un6300
Speakers: Focal aria 948, Focal cc900, Klipsch synergy KSF 10.5, Magnepan LRS
Subs: Rythmik FV25HP, Rythmik FV15HP
torii is offline  
Sponsored Links
Advertisement
 
Reply Receivers, Amps, and Processors

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off