Understanding relative volume and 0db on my receiver.

[j0nnyfive]

Howdy.

Well, I've read various articles online, but I find the answers to be confusing/vague/incomplete.

I'm trying to understand how my receiver "knows" it won't clip if I turn it up to 0db (assuming no EQ or calibration. I'm just running things straight).

I don't plan on running it this loud, but it would give me more peace of mind to understand this topic me thinks. For instance, some music is very compressed (runs hot). Movies can have large swings in SPL. How does my receiver (Yamaha Aventage) "know" that it won't clip if I set the volume to 0db?

And, if it is set to +1 db, does this mean it will most likely clip while watching a movie?
If set to -1db, is it possible for this volume setting to clip, depending on the source?

I've looked in the manual, but I couldn't find an explanation of this. Could someone help me understand this? (No hurry though.)

Thanks!

[mikesiskav]

Howdy.

Well, I've read various articles online, but I find the answers to be confusing/vague/incomplete.

I'm trying to understand how my receiver "knows" it won't clip if I turn it up to 0db (assuming no EQ or calibration. I'm just running things straight).

I don't plan on running it this loud, but it would give me more peace of mind to understand this topic me thinks. For instance, some music is very compressed (runs hot). Movies can have large swings in SPL. How does my receiver (Yamaha Aventage) "know" that it won't clip if I set the volume to 0db?

And, if it is set to +1 db, does this mean it will most likely clip while watching a movie?
If set to -1db, is it possible for this volume setting to clip, depending on the source?

I've looked in the manual, but I couldn't find an explanation of this. Could someone help me understand this? (No hurry though.)

Thanks!

Are you sure that your Yamaha has that feature? I haven't noticed any clip limiter function on any receiver I've ever used.

Some QSC amps do have a clip limiter function (The DCA-1622 for example). In the manual it says this:

Clip limiter
Description
When the audio signal drives the amp's output circuit beyond its power capability, it clips, flattening the peaks of the waveform. The clip limiter detects this and quickly reduces the gain to minimize the amount of overdrive. To preserve as much of the program dynamics as possible, limiting occurs only during actual clipping. Each channel has its own clip limiter, and you can switch it on or off independently, as shown at left.

When to use it
When driving full-range speakers, clip limiting reduces high frequency distortion caused by bass overloads. It also protects high frequency drivers from excess overdrive and
harsh clipping harmonics. We recommend using the clip limiters in all cinema applications.

[mikesiskav]

BTW, notice that it says it reduces the gain after clipping occurs. It does not eliminate clipping altogether.

[j0nnyfive]

Hello Mikesiskav!

I'm not thinking of a clip limiting feature per se. I just want to understand what is happening, power-wise, when you turn the receiver up to "0db". In other words... What is it, mechanically, that 0db represents? How much power is this?

The receiver gives you something like -80.... -40..... -20....-10...-5.... (0)....+5....+10, etc.

So, I want to know how loud I can turn the receiver up without worrying about any clipping. Is 0db safe? -5db? And, how does one know this? My understanding is that there is some kind of attenuation process happening when the volume is set at, say, -20 or -10. But, at 0db, supposedly, there is no longer any attenuation taking place. But, I wonder what +5 gives you? Or +10?

I'm really just trying to understand what is happening, "mechanistically", inside the AVR when you set the AVR to 0db. Or +1db. Or -5db. etc.

I know this may not be a very clear question, but I'm no audio expert either, please forgive me about that.

Thanks Mikesiskav!! Maybe you can help clear me up here. :o

[j0nnyfive]

To clarify:

When the volume is set to a negative number, does this mean the input signal is being attenuated digitally?
(I'm curious).

When the volume is set to a positive number, what is happening then?

Is there a "safe" volume level at all? I would assume that at -30, nothing will clip. But what about -5? -10? 0?

My current understanding, from what I've read elsewhere: Below 0d, the receiver performs "cuts" to the signal. Above 0d, the receiver "boosts" the signal. I would assume that at 0db, the system is simply amplifying the given signal by some predetermined amount (whatever this is). I could be wrong about all of what I just said. I'm speculating.

My input: HDMI from a PS3 or blue tooth from my phone. Digital input into the receiver.

My output: 2 speakers (L and R) only. No other speakers.

[curtis]

To clarify:

When the volume is set to a negative number, does this mean the input signal is being attenuated digitally?
(I'm curious).

When the volume is set to a positive number, what is happening then?

Is there a "safe" volume level at all? I would assume that at -30, nothing will clip. But what about -5? -10? 0?

My current understanding, from what I've read elsewhere: Below 0d, the receiver performs "cuts" to the signal. Above 0d, the receiver "boosts" the signal. I would assume that at 0db, the system is simply amplifying the given signal by some predetermined amount (whatever this is). I could be wrong about all of what I just said. I'm speculating.

My input: HDMI from a PS3 or blue tooth from my phone. Digital input into the receiver.

My output: 2 speakers (L and R) only. No other speakers.
Jonny,

You got it all wrong. The volume knob is a signal attenuator from beginning to end. You receiver/amp is always "amplifying" the signal. The volume knob controls how much the signal is amplified.

The scale is arbitrary. To make it useful, you should calibrate your system to where 0db is reference level at the listening position. So, when you are at -5db, you are listening at -5db from your reference, or at +5db you are 5db over your reference.

That said, volume levels also change with sources depending on output signal of that source.

Is there a safe level? Depends on the variables: the speakers, the source, the calibration, the material you a listening to.

[j0nnyfive]

Hey Curtis! Thanks! Let me see if I understand what you're saying.

1. Let's say my arbitrary volume range is from -80 all the way up to +15.
So, the attenuation of the signal begins at "+14" and works its way down the scale to -80?
Do you know whether this attenuation takes place digitally? I would assume it would for
accuracy's sake, but I do not know. (It is intuitive to me that a form of amplification is
taking place all along the scale. I suppose I'm trying to figure out how the attenuator and
amplifier interact with each other.)

2. If I calibrate my system such that "0db" now equals a different loudness (db) with a given signal,
I'm afraid this may complicate the issue. Let's assume for simplicity that I have calibrated my
loudness level, and that there was no change from the system's default loudness at 0db. Let's
just assume that the system was already giving me an accurate 75db test signal at 0db. Lucky me. ;)
I think this will help simplify my question. If I'm misunderstanding some aspect of this, please
correct me.

As an aside: I'm not really interested in matching my volume knob's scale to some external reference
such as THX or other, per se. I just turn the volume up until it sounds good. (almost everything I put in the movie player seems to have it's own standard of loudness anyway, so I just compensate the old fashioned
way by using the knob.) :p


3. You make a good point with "it depends". Let me give you the variables.
Source = A highly dynamic blu-ray movie over HDMI. So, the signal is digital.
Speakers = Sierra-2EX (MT 2-way, looks like about 6 ohm minimal?) run full range, stereo pair only.
Calibration = None. I'm just running it straight.
Receiver = Yamaha Aventage 3070 (I think it's rated at 150w 2-channel)


So, to bring it back to my question: Is there a volume level (along the receiver's arbitrary scale) that could be considered a "safe maximum" such that I can be fairly certain there will be no clipping, assuming my source is a highly dynamic blu-ray and the input signal is digital via HDMI into the receiver? I know that we call it "arbitrary" but I'm not convinced that it is COMPLETELY arbitrary. After all, could one not reasonably assume that 2 of these receivers would be roughly the same loudness when compared next to each other, even if not calibrated? (and assuming all other variables are controlled for?) One would assume that, at 0db arbitrary setting, the receiver is delivering a certain percentage of it's full power into a standard "75db" test signal. I think I'm just trying to understand what that percentage might be, and if there are any rules of thumb.

For example: One "rule of thumb" floating around the internets is "Just don't turn it past 0db and you'll be fine". But, I don't know how grounded this little rule is. I don't know if the receiver would really be "fine" no matter what. This is what I'm trying to figure out. Is there really a good "rule of thumb" that makes sense, in general, for receivers that use this type of arbitrary scale? (-80 to +15, etc).

So, my good Curtis, if you read all of this... I'm so sorry. lol
Oh, and BTW, I didn't get it ALL wrong. I DO have 2 speakers. I counted them. :cool:
Thanks again, everyone!

[curtis]

Hey Curtis! Thanks! Let me see if I understand what you're saying.

1. Let's say my arbitrary volume range is from -80 all the way up to +15.
So, the attenuation of the signal begins at "+14" and works its way down the scale to -80?
Do you know whether this attenuation takes place digitally? I would assume it would for
accuracy's sake, but I do not know. (It is intuitive to me that a form of amplification is
taking place all along the scale. I suppose I'm trying to figure out how the attenuator and
amplifier interact with each other.)

Not necessarily. The range is just a range. It may go beyond +15, it is just the available range set by the manufacturer.
The attenuator controls how much the amplification is applied to the signal.


Hey Curtis! Thanks! Let me see if I understand what you're saying.
2. If I calibrate my system such that "0db" now equals a different loudness (db) with a given signal,
I'm afraid this may complicate the issue. Let's assume for simplicity that I have calibrated my
loudness level, and that there was no change from the system's default loudness at 0db. Let's
just assume that the system was already giving me an accurate 75db test signal at 0db. Lucky me. ;)
I think this will help simplify my question. If I'm misunderstanding some aspect of this, please
correct me.

As an aside: I'm not really interested in matching my volume knob's scale to some external reference
such as THX or other, per se. I just turn the volume up until it sounds good. (almost everything I put in the movie player seems to have it's own standard of loudness anyway, so I just compensate the old fashioned
way by using the knob.) :p

Not sure what to tell you here other than I think you are way over thinking the use the volume knob.


Hey Curtis! Thanks! Let me see if I understand what you're saying.

3. You make a good point with "it depends". Let me give you the variables.
Source = A highly dynamic blu-ray movie over HDMI. So, the signal is digital.
Speakers = Sierra-2EX (MT 2-way, looks like about 6 ohm minimal?) run full range, stereo pair only.
Calibration = None. I'm just running it straight.
Receiver = Yamaha Aventage 3070 (I think it's rated at 150w 2-channel)


So, to bring it back to my question: Is there a volume level (along the receiver's arbitrary scale) that could be considered a "safe maximum" such that I can be fairly certain there will be no clipping, assuming my source is a highly dynamic blu-ray and the input signal is digital via HDMI into the receiver? I know that we call it "arbitrary" but I'm not convinced that it is COMPLETELY arbitrary. After all, could one not reasonably assume that 2 of these receivers would be roughly the same loudness when compared next to each other, even if not calibrated? (and assuming all other variables are controlled for?) One would assume that, at 0db arbitrary setting, the receiver is delivering a certain percentage of it's full power into a standard "75db" test signal. I think I'm just trying to understand what that percentage might be, and if there are any rules of thumb.

For example: One "rule of thumb" floating around the internets is "Just don't turn it past 0db and you'll be fine". But, I don't know how grounded this little rule is. I don't know if the receiver would really be "fine" no matter what. This is what I'm trying to figure out. Is there really a good "rule of thumb" that makes sense, in general, for receivers that use this type of arbitrary scale? (-80 to +15, etc).

It is absolutely arbitrary. You can set 0db to whatever you want.

What is safe is the volume/spl that doesn't blow your speakers or hurt your hearing. It isn't complicated.

There is no rule of thumb since there are a lot of variables as for the setting on the receiver.

On my pre/pro, I can set the maximum volume on the scale, and the pre/pro will not go beyond that. I know how loud I can stand, I set it there. So, if for some reason the volume gets cranked up, it will not go past that setting. All is safe.

[j0nnyfive]

Hey Curtis!

Okay, I think I see where you are coming from in answering this question. I need to be more specific about my potential uses of these speakers.

I may use them in scenarios where I need to turn up the volume as loud as it can safely go for limited periods of time. This is probably why you're under the impression that I'm over-thinking the volume control. Haha! I can see that now. You're probably like "what is this dudes deal? It's just a volume knob, man, geeze." I see. lol That's funny!

Loud as it (AVR) can safely go. It seems to me that this AVR (assuming it isn't clipping) won't surpass the Ascends' safe limits. Again, assuming it isn't clipping. That's what I'm looking for. The highest volume setting where clipping is still very unlikely to occur assuming that my source is digital (HDMI/blu-ray/etc).

Okay, I'm thinking about what all you've said here... yep. You can set 0db to what you want, blowing speakers and eardrums is bad, attenuators attenuate, and the answer probably won't be very complicated. Amen brother. I agree with you on all points.

Here is where I think I need to focus: " It may go beyond +15, it is just the available range set by the manufacturer."

So, what you're telling me here is... call Yamaha and report back? :p

What do you mean by "It may go beyond +15"? I'm not understanding here. Thanks again Curtis! I appreciate your input, no pun.

[jimb]

I think you are asking, what the maximum non-clipped volume setting would be from a digital source (Bluray disc). If your receiver power rating level is 150W/channel for the nominal impedance of the S-2EX, that is safely under the permitted peak power for the speakers. In that case, it is all about the exact specs for your receiver: at what volume knob setting does it drive its power amp stage to clipping, from a "full" digital signal. I'm not sure anyone here has those specs for your receiver.

It IS possible that "0" does correspond to that (but not assured).

It that were true, it would mean that right at "0", a maximum digital signal peak could just reach to the start of clipping.

If that were true, it would be 'safe' to run right up to "0" without expecting to damage anything.

If that is true, then why would a manufacturer provide a "+" range to the control?

If I was the designer, I would do so to permit getting more volume on signals that are recorded at less than digital maximum (or with analog inputs at lower voltages).

That's all I have.

[curtis]

+15 is an arbitrary limit...as you set what what SPL 0db on the volume knob.

I get it, with all the variables involved, you are looking for the safe volume limit. The AVR isn't going to do that for you. You will not get a definitive answer for that. Nobody knows how loud you like to listen. You have to be smart about how you use the system and how loud you play it.

[j0nnyfive]

Hey Jimb! Welcome!

Thank you for your insights! Yes, maximum non-clipped volume setting from a digital source. Bingo.

Yep, from your input and Curtis's and Mike's, I see that I need to speak with these AVR manufacturers directly. Hopefully I can get some kind of answer for this. I will report back with what I learn.
Thanks, guys!!

[davef]

j0nnyfive,

I think you are misunderstanding volume control and how it relates to a receiver's or amplifier's "power output"....

First off, as others have mentioned - "0dB" is just an arbitrary number - it is meaningless and has no relationship to the output volume of your speakers or how much wattage your receiver or amplifier is producing.

That said, if you calibrate your receiver using its built in auto speaker setup it takes measurements of your speakers and will then set "0dB" to typically refer to 75 or 85dB spl output from one speaker at the microphone position. (there is no standard, some receivers calibrate to 75, others to 85, others to whatever the user wishes). This then translates such that +3dB on your volume control would then mean 78dB from one speaker at the same mic position (if the auto level calibrated to 75db). However, this is not precisely accurate.

The receiver has no idea how much power it is sending to the speakers at any volume control level. Volume controls control "gain". As such, the receiver does not know when it is clipping or when it has reached maximum output. It is complicated because every speaker model has a unique sensitivity rating and unique impedance profile. At what output volume level from the speakers your receiver will start to clip will dramatically vary from one model speaker to another.

There are some receivers/amplifiers out there that do measure ripple current from the power supply and thus initiate some form of limiter or compression circuit to reduce gain in order to reduce clipping - but this is quite rare and these limiting circuits often come at a cost of a reduction in dynamics.

While we understand what you want to accomplish -- the only way to do this is as follows.

Your receiver must be connected to the speakers and you must check your AC line voltage to confirm proper voltage.

With the receiver connected to the speakers, connect an oscilloscope across the receiver's speaker output (+ and -) so that you can view the waveform.

Play some music or pink noise, and gradually increase the volume level of the receiver while monitoring the waveform on the scope. At this first hint of "clipping" - (a squaring off of the crest and trough) - using a voltage meter (or the scope) - jot down the peak voltage level and there you have it, your maximum safe peak voltage output from the receiver connected to this particular set of speakers. Or, but somewhat less accurate, using an SPL meter take a peak spl measurement at the first sign of clipping -- this would then be the maximum safe volume level you can reach. If your receiver is calibrated, note the dB level of the volume control and simply do not exceed this....

I'm afraid that if you call your receiver manufacturer and ask to speak to a tech, they are not going to even understand what you are asking or even worse, provide you with wrong information.

[j0nnyfive]

Hey Dave!

Thank you for your technical insight! (And thank all of you for yours!) I appreciate it. Unfortunately... (drum roll)... I have more questions. lol :o

"0dB" is just an arbitrary number - it is meaningless and has no relationship to the output volume of your speakers or how much wattage your receiver or amplifier is producing."

My impression of the volume knob control system is that they try to make it smooth and predictable, roughly correlating to 1db change per "tick" of the knob. It seems to me that, given a certain impedance profile, they (engineers) should be able to roughly predict how much electricity can be delivered to each speaker at each "tick" of the volume knob. Is this not possible?

And, since the sources are decoded inside the AVR itself, this would seem to give us even more predictability.

I'm not an audio expert, but surely.... surely!... there must be some system at play here (even if only the engineers know) that would at least give us a rough estimate of how loud the volume knob can be turned for most 8-ohm speakers, most of the time? Maybe? Maybe just a rough guess? :eek: I don't know. I don't need a super exact number. I'm just looking for "rough" guess. Also, does anyone know why they make the positive numbers go up to +15? +16.5? +18? Is there a technical reason for this?

" At what output volume level from the speakers your receiver will start to clip will dramatically vary from one model speaker to another."


How dramatically are we talking here? Could 2 different 8 ohm bookshelf speakers (MT, 2-way) clip at 2 dramatically different points on the volume knob? Like "-10" on one speaker and "+5" on another speaker? Is it this dramatic? I'm just trying to get a sense of scale here. And maybe hold on to irrational optimism. lol :o

Thanks again!

P.S. Thank you for your advice about the oscilloscope and SPL meter. Also, I'm listening to the Sierra-2EX right now! I'm loving them! My ears have to adjust to the difference for now, but I'm loving them. Man, you weren't kidding about the bass. lol Talk about PUNCH! I'm hearing some singers breathing in the background when they're not singing (jazz). Awesome!

[davef]

Hey Dave!

Thank you for your technical insight! (And thank all of you for yours!) I appreciate it. Unfortunately... (drum roll)... I have more questions. lol :o

"0dB" is just an arbitrary number - it is meaningless and has no relationship to the output volume of your speakers or how much wattage your receiver or amplifier is producing."

My impression of the volume knob control system is that they try to make it smooth and predictable, roughly correlating to 1db change per "tick" of the knob. It seems to me that, given a certain impedance profile, they (engineers) should be able to roughly predict how much electricity can be delivered to each speaker at each "tick" of the volume knob. Is this not possible?

And, since the sources are decoded inside the AVR itself, this would seem to give us even more predictability.


Perhaps this explanation will make more sense. Basically, there are 3 main sections to a modern AVR. section 1 is where all the DSP processing occurs. Section 2 is the pre-amp. Section 3 is the amplifier.

An audio amplifier has a fixed gain, typically a gain of 28. An amp does not have variable gain. The pre-amp section sends a variable low voltage signal to the amp section, and the amp amplifies this analog signal. For example, put in 1 volt, get 28 volts out.

The DSP section does the various processing and in the last stage of this section, the digital signal is converted to analog (an audio signal - AC current). This is sent to the pre-amp section. The signal source and type, or where it is decoded makes no difference at all. If decoded outside, then generally the signal bypasses the DSP section and just hits the pre-amp section.

Speakers have varying impedances and efficiencies. An 8 ohm speaker might dip as low as 3 ohms at say 100Hz, while another speaker might have a 30ohm impedance at 100Hz. Impedance rating of a speaker is a bit like an overall average.




I'm not an audio expert, but surely.... surely!... there must be some system at play here (even if only the engineers know) that would at least give us a rough estimate of how loud the volume knob can be turned for most 8-ohm speakers, most of the time? Maybe? Maybe just a rough guess? :eek: I don't know. I don't need a super exact number. I'm just looking for "rough" guess. Also, does anyone know why they make the positive numbers go up to +15? +16.5? +18? Is there a technical reason for this?

There is no special system at play here, if you want to properly estimate how loud you can play before clipping for any speaker, I gave you the procedure. Or, you can simply do the math and get a decent estimate using a calibrated SPL meter at your listening position. Those dB increments on your volume control are purely arbitrary. If you calibrate, then you set 0dB to = a specific SPL at the mic position from 1 front speaker.

With a speaker rated at a sensitivity of 85 and an 80 watt receiver, and with a listening distance of ~30 feet - calibrate the receiver such that 0dB = 85dB, and you will be clipping your receiver at the 0dB volume control level. Your receiver does not know this...



P.S. Thank you for your advice about the oscilloscope and SPL meter. Also, I'm listening to the Sierra-2EX right now! I'm loving them! My ears have to adjust to the difference for now, but I'm loving them. Man, you weren't kidding about the bass. lol Talk about PUNCH! I'm hearing some singers breathing in the background when they're not singing (jazz). Awesome!

Happy to hear you are enjoying the Sierra-2EX. I can't continue answering your questions regarding the 0dB setting though as I and others here have done our best.

[j0nnyfive]

I want to thank you guys for your help. I was going to say to please ignore my last post because I know you guys are busy doing what you do, and I appreciate it. You even responded faster than I could stop you (which is awesome)! Thank you for the oscilloscope and SPL meter method advice. I know you have much more understanding than I do regarding these things, and I will take your advice. Thank you also for your explanation of how the receiver works. You guys aren't even in the AVR business and here I am asking AVR questions!

Thank you for the GREAT speakers. I love them. Thank you also to Curtis, Jimb, and Mike for your contributions. I have some learning to do.

[j0nnyfive]

Okay, I just want to thank you guys again and say sorry for bugging ya'll so much. lol And...

I think I've figured out what I'm going to do! I feel compelled to at least tell you guys my plans, for good or bad. You don't have to respond, but if you feel it's a bad idea, please don't hesitate to let me know!


Here's my logic:

IF dynamic range peak loudness = max wattage,
THEN proportionally lower dynamic range loudness = proportionally lower wattage.
SET speaker output @ 1 meter to equivalent of proportionally lower wattage.

The resulting volume setting will be what I'm looking for. This represents the "theoretical danger zone". I will stay well south of this volume setting.


What does this accomplish for me? A little more piece of mind. It's not super accurate like the Oscilloscope method, but it will satisfy me enough. (This is my creative interpretation of Dave's "do the math")


Again, thanks guys for your help! You all got me thinking in a better direction. And, if this is a dumb idea and my logic is completely convoluted or just plain stupid, please let me know!! lol
Thanks again. :)

Edit: If you have any concerns or additional ideas to add to this approach, please feel free!