Jump to content
IGNORED

Misleading Measurements


Recommended Posts

51 minutes ago, Don Hills said:

1 ohm resistor? Shurely shome mishtake... ☺️

(dBm takes into account the source/load resistance, dBu / dBV doesn't.)

Yeah - I’m doing 3 things at once tonight. But the sentence is correct without the nonsense phrase that I somehow stuck in there for reasons I can’t imagine.  How’s this:

 

“The reference you keep citing as "consumer line level" (0.316 VRMS) is in the middle of the usual range for mean analog line levels when playing music on consumer electronics”

 

?

 

Link to post
Share on other sites
2 hours ago, stereo coffee said:

You only need to attenuate consumer line level,  which is NOT 2v RMS its nominal 0.316v RMS  , You might be reading too many reviews that test equipment at that high level, that is unrelated to what is at a RCA socket when playing CD's in home audio systems. 

 

Why its bad is because your source component contains the very best opportunity of presenting music as it actually can be, adding anything in-between other than resistance to attenuate,  simply detracts from that opportunity.    

Sorry, but you have a fundamental misunderstanding of whats going on.  As explained consumer line level is *not* 0.316 volts rms.  The vast majority of modern digital sources are 2 volts rms at 0dBFS. 

 

The actual required signal voltage for any final acoustic volume level depends upon the pre / power amplifier gain and the speaker sensitivity.  It will not be 0.316V.  0.316 V is a meaningless number.

 

It doesnt follow that other ("reactive") components damage signal quality.  In fact filters are a fundamental and necessary part of audio reproduction.  They do "good", they are not bad.  They are there for a reason.

Link to post
Share on other sites
6 hours ago, Jud said:

 

Your "explanation" of why this is bad is actually just a restatement of your original position.  Would you please try to give an actual explanation as to why it causes deterioration in the resulting sound?

 

Also, you mentioned looking through schematics of good preamplifiers.  Can you please mention which particular preamps you viewed schematics of?  Thanks.

The schematic I looked at was the audio Research reference 1.  With experience of digital pots and how they sound, it won't be on my shopping list - but it certainly is heading in the right direction. The thing I liked was no mechanical switches in the signal path.  I found it here:  https://www.hifiengine.com/manual_library/audio-research/reference-1.shtml

 

To answer your question, it answers itself, if we ask instead, why it might be desirable if we take stance to covey a audio signal differently, knowing one hand it is possible to convey it intact, but we choose to not do so.

 

Why deterioration occurs with reactance... a vast subject for sure,  but this might help. There are many choices to explore reactance and AC analysis, if the one below is not to your liking, please suggest another. But I liked it as it sets out resistance does little if any change to a waveform.

   

 

 

Link to post
Share on other sites
1 hour ago, stereo coffee said:

The schematic I looked at was the audio Research reference 1. 

Perhaps you should read the rest of the specs on the Ref 1.  There are 3 sets of line outputs - 2 main and one "tape" feed.  So if, as you suggest above, "preamp line level outs are adding reactance [and] this is absolutely fine if you also resign to not ever wanting to hear what your source can actually provide", you've ruled out the Ref 1 as a provider of top quaity sound.  

 

The undefined distortion spec ("less than 0.015% distortion") on the main line outputs is measured at 2V RMS single ended and 4V RMS balanced.  But the line level actually used while listening with a Ref 1 can be as little as tens of microvolts or as much as 10+ volts, depending on source material, environment, associated equipment, usage, etc.  If you read a bit further, you'll find that the maximum usable line output is 30 V RMS "at less than 0.5% THD at 1 kHz" - so do you not consider the Ref 1 to be consumer equipment or do you not think it can let you "hear what your source can actually provide"?

 

As for having "no mechanical switches in the signal path", I assume you realize that there are SS relays instead.  Although they are free of the arcing and other ills of mechanical switches with contacts, SSRs are also not perfect conductors.  If you look at a schematic for an SSR-switched circuit, you'll find that almost all have some kind of diode or RC shunting across the SSR's output (often within the SSR).

Link to post
Share on other sites
5 hours ago, stereo coffee said:

Why deterioration occurs with reactance... a vast subject for sure,  but this might help. There are many choices to explore reactance and AC analysis, if the one below is not to your liking, please suggest another. But I liked it as it sets out resistance does little if any change to a waveform.

 

So if one has a preamp designed by a not terribly competent designer, it may not work tremendously well.

 

Thanks.

One never knows, do one? - Fats Waller

The fairest thing we can experience is the mysterious. It is the fundamental emotion which stands at the cradle of true art and true science. - Einstein

Computer, Audirvana -> wi-fi to router -> EtherREGEN -> microRendu -> USPCB -> ISO Regen (powered by LPS-1) -> USPCB -> Pro-Ject Pre Box S2 DAC -> Spectral DMC-12 & DMA-150 -> Vandersteen 3A Signature.

Link to post
Share on other sites
4 hours ago, bluesman said:

Perhaps you should read the rest of the specs on the Ref 1.  There are 3 sets of line outputs - 2 main and one "tape" feed.  So if, as you suggest above, "preamp line level outs are adding reactance [and] this is absolutely fine if you also resign to not ever wanting to hear what your source can actually provide", you've ruled out the Ref 1 as a provider of top quaity sound.  

 

The undefined distortion spec ("less than 0.015% distortion") on the main line outputs is measured at 2V RMS single ended and 4V RMS balanced.  But the line level actually used while listening with a Ref 1 can be as little as tens of microvolts or as much as 10+ volts, depending on source material, environment, associated equipment, usage, etc.  If you read a bit further, you'll find that the maximum usable line output is 30 V RMS "at less than 0.5% THD at 1 kHz" - so do you not consider the Ref 1 to be consumer equipment or do you not think it can let you "hear what your source can actually provide"?

 

As for having "no mechanical switches in the signal path", I assume you realize that there are SS relays instead.  Although they are free of the arcing and other ills of mechanical switches with contacts, SSRs are also not perfect conductors.  If you look at a schematic for an SSR-switched circuit, you'll find that almost all have some kind of diode or RC shunting across the SSR's output (often within the SSR).

I said it was not on my shopping list. I do see what is happening here though. 

Link to post
Share on other sites
35 minutes ago, Jud said:

 

So if one has a preamp designed by a not terribly competent designer, it may not work tremendously well.

 

Thanks.

Unsure what that means.  Back on topic,  a misleading measurement  is IMO consumer line level. 

 

 

 

Link to post
Share on other sites
1 hour ago, stereo coffee said:

I said it was not on my shopping list. I do see what is happening here though. 

Huh?  What's happening here is that you're offering opinions with which several of us disagree.  We're providing objective facts to support our beliefs and would like you to do the same.  Start with anything that supports your belief that consumer line level is a misleading measurement - so far in this thread, I've seen only your opinion.  We need facts.

 

Wheres-th-Beef1.jpg.05d622907979f133bef0aeda91a8a8a5.jpg

 

You started this segment of the thread with this: "I would suggest ignoring all measurements that fail to understand consumer line level, as anything higher is meaningless with respect to equipment you use every day".  Yet almost all relevant analog consumer audio specs for preamps, DACs etc are measured at line output levels of -10 dBV RMS, which seems to be consistent with your beliefs.  Consider the AR Ref 1, which you seem to both love for its lack of mechanical switching in the signal path and hate for its use of digital attenuation.  Its specs include measurements at your preferred reference line level, but it's also spec'ed at levels up to 30V (!) and has 6 sets of line outputs (3 balanced and 3 single-ended unbalanced).

 

If you're trying to say that all consumer power amps should have an input sensitivity that pushes them to their maximum rated output with an input level of -10 dBV, just say it.  This would not be a practical constraint on amp design and no one would adhere to it - but at least you'd be making a coherent argument based on facts.

Link to post
Share on other sites
50 minutes ago, bluesman said:

Huh?  What's happening here is that you're offering opinions with which several of us disagree.  We're providing objective facts to support our beliefs and would like you to do the same.  Start with anything that supports your belief that consumer line level is a misleading measurement - so far in this thread, I've seen only your opinion.  We need facts.

 

Wheres-th-Beef1.jpg.05d622907979f133bef0aeda91a8a8a5.jpg

 

You started this segment of the thread with this: "I would suggest ignoring all measurements that fail to understand consumer line level, as anything higher is meaningless with respect to equipment you use every day".  Yet almost all relevant analog consumer audio specs for preamps, DACs etc are measured at line output levels of -10 dBV RMS, which seems to be consistent with your beliefs.  Consider the AR Ref 1, which you seem to both love for its lack of mechanical switching in the signal path and hate for its use of digital attenuation.  Its specs include measurements at your preferred reference line level, but it's also spec'ed at levels up to 30V (!) and has 6 sets of line outputs (3 balanced and 3 single-ended unbalanced).

 

If you're trying to say that all consumer power amps should have an input sensitivity that pushes them to their maximum rated output with an input level of -10 dBV, just say it.  This would not be a practical constraint on amp design and no one would adhere to it - but at least you'd be making a coherent argument based on facts.

But no they are rarely if ever measured at -10dBV -RMS , and they should be, that is the point. one only has to go back a few posts to find: 

 

16 hours ago, March Audio said:

Sorry, but you have a fundamental misunderstanding of whats going on.  As explained consumer line level is *not* 0.316 volts rms.  The vast majority of modern digital sources are 2 volts rms at 0dBFS. 

 

The actual required signal voltage for any final acoustic volume level depends upon the pre / power amplifier gain and the speaker sensitivity.  It will not be 0.316V.  0.316 V is a meaningless number.

 

It doesnt follow that other ("reactive") components damage signal quality.  In fact filters are a fundamental and necessary part of audio reproduction.  They do "good", they are not bad.  They are there for a reason.

Consumer line level is already adhered to by some manufacturers, Quad as example selling approximately 2500 units, of their 306 power amplifier all with sensitivity at 0.375V RMS 

 

 

Link to post
Share on other sites
17 minutes ago, stereo coffee said:

But no they are rarely if ever measured at -10dBV -RMS , and they should be, that is the point. one only has to go back a few posts to find: 

Both you and MarchAudio appear to be confused about “line level” in consumer audio devices.  It is both -10 dBV and 2 volts, depending on how you spec it. The mean level as measured on an averaging scale is the -10 dBV that you seem to think is the only proper spec, which equals your 0.316 volts.  But consumer line level can also be defined by the maximum peak voltage before clipping and distortion set in, and that (for analog signals) is usually +6 dBV, which equals 2 VRMS.  Not all equipment reaches the upper limit of its distortion spec at exactly the same level, so analog peak line voltage can be spec’ed a bit higher or lower in some equipment. But you’re all talking about the same specification!

 

Some engineers and manufacturers choose to design equipment to a different performance level.  This is irrelevant as long as the line out voltage range is enough (but not too much) to drive the next device(s) in the signal chain to a normal range of listening levels and preserve a reasonable range of gain control.  It is not an indication that the design team failed to understand the concept of line level.

 

So the same consumer line level is correctly stated as -10 dBV average and/or +6 dBV peak.  These are specs for the maximum level at rated distortion that is available to drive the next stage.  Remember too that this is the output voltage from the last line stage, so the associated specs for distortion etc describe the signal as it exits the device through a line output.

Link to post
Share on other sites
11 hours ago, bluesman said:

As for having "no mechanical switches in the signal path", I assume you realize that there are SS relays instead.  Although they are free of the arcing and other ills of mechanical switches with contacts, SSRs are also not perfect conductors.  If you look at a schematic for an SSR-switched circuit, you'll find that almost all have some kind of diode or RC shunting across the SSR's output (often within the SSR).

 

IME, SS relays are good ... if the circuit is well designed, any ills of their non-perfect resistive qualities are far outweighed by getting rid of the evils of contact noise in physical switches - as a general rule, the more ambitious the setup is, the more one has to worry about the influence of mechanically operating parts in the chain.

Frank

 

http://artofaudioconjuring.blogspot.com/

 

 

Over and out.

.

 

Link to post
Share on other sites
4 hours ago, stereo coffee said:

But no they are rarely if ever measured at -10dBV -RMS , and they should be, that is the point. one only has to go back a few posts to find: 

 

Consumer line level is already adhered to by some manufacturers, Quad as example selling approximately 2500 units, of their 306 power amplifier all with sensitivity at 0.375V RMS 

 

 

Can you show me some dacs that don't output 2 volts rms (or very close to) at 0dBFS?

 

Better still can you show me some dacs that output 0.375v at 0dBFS?

 

The Quad 306 is nearly 40 year old design. Modern components do not adhere to this 300mV level.

 

After CD became established domestic line level quickly changed to the 2 volt convention we have now.  It's been thatxway for literally decades.

Link to post
Share on other sites
3 hours ago, bluesman said:

Both you and MarchAudio appear to be confused about “line level” in consumer audio devices.  It is both -10 dBV and 2 volts, depending on how you spec it. The mean level as measured on an averaging scale is the -10 dBV that you seem to think is the only proper spec, which equals your 0.316 volts.  But consumer line level can also be defined by the maximum peak voltage before clipping and distortion set in, and that (for analog signals) is usually +6 dBV, which equals 2 VRMS.  Not all equipment reaches the upper limit of its distortion spec at exactly the same level, so analog peak line voltage can be spec’ed a bit higher or lower in some equipment. But you’re all talking about the same specification!

 

Some engineers and manufacturers choose to design equipment to a different performance level.  This is irrelevant as long as the line out voltage range is enough (but not too much) to drive the next device(s) in the signal chain to a normal range of listening levels and preserve a reasonable range of gain control.  It is not an indication that the design team failed to understand the concept of line level.

 

So the same consumer line level is correctly stated as -10 dBV average and/or +6 dBV peak.  These are specs for the maximum level at rated distortion that is available to drive the next stage.  Remember too that this is the output voltage from the last line stage, so the associated specs for distortion etc describe the signal as it exits the device through a line output.

Can you show any specs of output voltage of audio equipment where it is measured as "mean level"?

 

I can't say I have ever seen an AC voltage measured in any circumstance in any field as its mean and for good reason.   The mean of the voltage levels in a sinusoidal waveform is zero!

 

This is why rms is used.

Link to post
Share on other sites
1 hour ago, fas42 said:

 

IME, SS relays are good ... if the circuit is well designed, any ills of their non-perfect resistive qualities are far outweighed by getting rid of the evils of contact noise in physical switches - as a general rule, the more ambitious the setup is, the more one has to worry about the influence of mechanically operating parts in the chain.

Absolutely correct.  But mechanical signal switching was only one of the issues stuck in the protestor’s craw. The biggest ones seem to be line level specs and reactance, and R-C shunting in the output path adds the dreaded reactance. I agree that it’s irrelevant and inaudible.  But it’s a bit ironic for the same person to laud SSRs but decry any reactance at all in the signal path.

Link to post
Share on other sites
2 hours ago, March Audio said:

Can you show any specs of output voltage of audio equipment where it is measured as "mean level"?

All such specs are measures of a mean voltage.  Another name for RMS voltage is the quadratic mean voltage.  The -10 dBV RMS value of 0.316V is the quadratic mean of the signal level voltage swing.  If the test signal is a steady sine wave, there are no peaks - the mean voltage over time is the same as the peak voltage.  
 

Measuring line level voltage by applying a pure sine wave to the input is done with a voltmeter or other measuring device across the output.  True RMS measurement is an averaging of the voltage over a certain period of time.  As the test tones used for this are steady, the instantaneous voltage is the same as the mean voltage over time.

 

But for dynamic program material like music, the peak level is the highest instantaneous voltage level in the signal over time, and the (quadratic) mean voltage is the average voltage over time.   Following your logic, a voltmeter set on averaging would always read zero regardless of the AC voltage drop across it - but it doesn’t.

 

Link to post
Share on other sites
22 minutes ago, bluesman said:

All such specs are measures of a mean voltage.  Another name for RMS voltage is the quadratic mean voltage.  The -10 dBV RMS value of 0.316V is the quadratic mean of the signal level voltage swing.  If the test signal is a steady sine wave, there are no peaks - the mean voltage over time is the same as the peak voltage.  
 

Measuring line level voltage by applying a pure sine wave to the input is done with a voltmeter or other measuring device across the output.  True RMS measurement is an averaging of the voltage over a certain period of time.  As the test tones used for this are steady, the instantaneous voltage is the same as the mean voltage over time.

 

But for dynamic program material like music, the peak level is the highest instantaneous RMS voltage level in the signal over time, and the (quadratic) mean voltage is the average RMS voltage over time.   Following your logic, a voltmeter set on averaging would always read zero regardless of the AC voltage drop across it - but it doesn’t.

 

 

No they are not.  RMS is not the same as just mean.

 

A sinusoidal waveform has (identical) positive going values and negative going values.

 

"The mean is a type of average. It is the sum (total) of all the values in a set of data, such as numbers or measurements, divided by the number of values on the list."

 

The mean of a sinus is zero. 

 

image.png.1ec4e8b73b4f37fecd0ffab3db27d93d.png

 

Put as many values as you like into this, put as many cycles as you like into this, the sum is zero, the mean is zero

 

 

As I asked earlier please show me anywhere that ac voltages are represented as just the mean value.

 

Link to post
Share on other sites
3 minutes ago, March Audio said:

 

No they are not.  RMS is not the same as just mean.

 

A sinusoidal waveform has (identical) positive going values and negative going values.

 

"The mean is a type of average. It is the sum (total) of all the values in a set of data, such as numbers or measurements, divided by the number of values on the list."

 

The mean of a sinus is zero.

I’m sorry to seem contrary, but I don’t know how else to say this - you’re simply wrong.  The quadratic mean (also called root mean square and abbreviated as RMS) is another type of average. It measures the absolute magnitude of a set of numbers, not their arithmetic average.  So every RMS voltage measurement is an average by definition.  Not all means are simple mathematical means.

 

This horse has now been pronounced dead.

Link to post
Share on other sites
26 minutes ago, March Audio said:
9 minutes ago, bluesman said:

I’m sorry to seem contrary, but I don’t know how else to say this - you’re simply wrong.  The quadratic mean (also called root mean square and abbreviated as RMS) is another type of average. It measures the absolute magnitude of a set of numbers, not their arithmetic average.  So every RMS voltage measurement is an average by definition.  Not all means are simple mathematical means.

 

This horse has now been pronounced dead.

 

 

With respect the simple example above shows that I am correct.  Its basic maths.

 

As I said RMS is not the same as just mean.  You can look at it as representing the area under the curve or the "heating value".  This is why its important when looking at non sinusoidal signals which may have significant crest factors.

 

To think of RMS an an "average" reading is quite misleading.

 

Anyway, in the context of this discussion there is no way 2 volts rms can be represented as 0.316v mean as you asserted.

Link to post
Share on other sites
25 minutes ago, The Computer Audiophile said:

Thanks for disagreeing politely guys and for getting back to the discussion at hand. 

I think it simply a case of Stereo Coffee thinking that the "consumer level" convention of -10dBv (0.316v rms) - and I say convention because I dont think its been enshrined in any formal standard - is still being used.

 

It may have been prevalent in the 1970s, but no-one uses it now and havent for decades.

Link to post
Share on other sites
1 hour ago, March Audio said:

I think it simply a case of Stereo Coffee thinking that the "consumer level" convention of -10dBv (0.316v rms) - and I say convention because I dont think its been enshrined in any formal standard - is still being used.

 

Correct me if I'm wrong, but there is a big difference between nominal output voltage (e.g. 0.316v rms) and maximum output voltage. The spec sheets for consumer equipment* these days show the maximum output voltage,  somewhere around 2 Vrms (sometimes a little less and sometimes a little more).

 

*My Mytek Brooklyn DAC+  apparently has a maximum output (at 1kHz) of 9.84V from the balanced outputs.

mQa is dead!

Link to post
Share on other sites
1 hour ago, lucretius said:

 

Correct me if I'm wrong, but there is a big difference between nominal output voltage (e.g. 0.316v rms) and maximum output voltage. The spec sheets for consumer equipment* these days show the maximum output voltage,  somewhere around 2 Vrms (sometimes a little less and sometimes a little more).

 

*My Mytek Brooklyn DAC+  apparently has a maximum output (at 1kHz) of 9.84V from the balanced outputs.

The Mytek is a little unusual in that it has output options which make it flexible across professional and domestic applications.  We have a RME ADI2 Pro FS BE which has +4,+13,+19 and +24dBu output options (1.22V, 3.46v, 6.9v and 12.2 volts rms respectively).  However the vast majority of normal domestic dacs will be 2 volts rms from RCA and 4 volts rms from balanced outputs (if they have them) at 0dBFS.

 

This doesnt mean the level is 2 volts all the time, it depends on whats going on in the recording.  Music usually has a low rms level compared to its peak, maybe a ratio of 1:5.  Recordings are very often normalised so that their peak is up to 0dBFS.  This will be a typical peak voltage of around 2.8 volts.

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now


×
×
  • Create New...