Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Tim Campbell]

As Volker has stated B&K has touched on this in some of their literature and shown that temperature and barometric pressure are much more problematic.

[danlavry]

[quote title=Klaus Heyne wrote on Tue, 07 March 2006 02:26]Three considerations to add to the discussion:

Hi Klaus,

"1....Serious capsule distortion sets in very late, at very high volume levels…."

I came here asking what to expect in terms of max and min “rest” capacitance (no sound),
And what to expect the variations in C (capacitance) to be. I do not know what “very late” means (for me, late is a term I use for time related issues). I assume for now that you mean very loud. Yes, the louder, the more non linearity, but I would not “write is off” so fast. Yes, with say 50pf +/-10pf you get huge distortions, but at 50pf+/-1pf you get pretty high distortions also. What is high? Say high compared to mic pre’s power amps, converters… most gear.

"2. If distortion cannot be heard, but measured, is it relevant in microphone applications?"

A distortion will not be heard when it is a very low amplitude distortion. It can not be heard when it is in a frequency range we can not hear.
The distortion mechanism I am addressing is high enough amplitude, and certainly at hearing frequencies (second, third, forth….). Of course “distortion being heard” is all in reference (comparison) to the original sound, just before it came into the microphone.  

"3. Is a distortion-free microphone automatically preferable to the listener to one with, let's say, 0.5% or 1% THD?"

That is a subjective issue. Some people like tube sound because they like that specific distortion sound.
Personally I am into minimal distortions, I like to have the ability to capture the sound (air vibrations) in the performance space as close as possible, and play it exactly the same (as much as possible) in my living room - the exact same air vibrations. That means 0 distortions. Of course I accept the fact that mastering and recording engineers add sonic alterations (EQ, compression reverb....) but such alterations are DELIBERATE ARTISTIC DECISION that vary from song to song... I personally do not like to be "stuck" with some given distortions, because while one can add distortions, taking them out is not even possible.

But the question you raise is regarding distortion that is “preferable to the listener to one with, let's say, 0.5% or 1% THD”. The question is too open ended, because there are INFINITE POSSIBILITIES to have 1% distortions. One can have 1% at 20KHz, or 1% at 3KHz, you will hear the later much more. One can have 1% spared over all the harmonics, with all sorts of “patterns”.

Here I am talking about a specific type of distortion, all the harmonics are there and they decay linearly, a rather unique sonic “signature” of a condenser mic.

At this point, all I am saying is: The condensers, right from the start (the concept, the operating principle) has a specific sonic signature. I most often use condensers for my music band. I do not have better options. I did not analyze dynamics. My guess is that they have their own mechanism yielding a unique sonic signature.

I do not think there is such a thing as “distortion free mic”. I am just talking about a fundamental distortion mechanism that is directly related to condensers. Knowing and understanding the operating principles is the first step towards lowering the distortions.

Do you want lower distortions? Do you like the sonic signature of a condenser? That I can not answer. But is you wish to “check it out”, you could try some comparisons with and without some small fixed capacitor in parallel, a tradeoff between amplitude and linearity. If you do, please let us know your opinion.

If you are really interested in experimenting, I will be glad to contribute a LavryBlue micpre, a very low distortion (and noise) unit, thus ideal for such tests  

Regards
Dan Lavry
www.lavryengineering.com  

[David Bock]

Quote:

3. Is a distortion-free microphone automatically preferable to the listener to one with, let's say, 0.5% or 1% THD?

This is the most important question since it's just assumed to be better without distortion. BUT, most common efforts to remove distortion are done with a complete neglect for the (still) unmeasurable aspects of what makes a great microphone great, often with disappointing results from a sonic point of view. That begs the question, ?are we just conditioned to  distortion? Answer = yes and no, depending upon type, aplication expectation, conditioning and experience, usually. Then there is the problem with the terms- "great mic"? That's totally application driven. THD? I rearely see an attempt to descramble THD into something meaningfull, like composition and dB below signal? And, of course, it's dynamic, so it will vary with stimulus. So THD numbers are only the crudest guidelines. And they are usually not formed from peak readings, which is what most music is comprised of, not the sine waves used for testing.
And here's my favorite part: Most condenser mics are much less distorted than the speakers they will be heard on!
I'll give another example: some tube mics use a shunt from the plate to ground. this increases trhe HF distortion, but it's out of range and no one hears it. But then there's IM distortion that is in range. Well, it's either so far dowm the dB scale no one hears it or.....no one cares!
Hint: even the most experienced engineers sometimes let mic distortion get by them, sometimes they don't. But I don't know any that are 100% satisfied with any loudspeakers......
regards,
David

[Martin Kantola]

danlavry wrote on Tue, 07 March 2006 00:38

You say your mic measures .015% distortions, so how does it gets measured? Does the distortion measurement done withe the capsule removed? I know some measurements are done that way. For example, say you wish to measure the noise of the build in electronics, and wish to take the acoustic noise out. In such case, one could substitute the capsule with a "dummy" cap of an appropriate value. One could do the same (or similar) for measuring the distortions of the ELECTRONICS ONLY.

Dan,

this is exctly what I was trying to say when I wrote "The measurement included only the electronics naturally."

Regarding your example with the whistle and the flute, I did get the part where the whistle is almost a pure sine etc. What was more difficult to understand was how much distortion it would actually take to make it sound like a clarinet? I suspect we need much more than 1% or so.

In answer to Klaus' clever consideration number two; "If distortion cannot be heard, but measured, is it relevant in microphone applications?" I'd like to repeat that the intended use should be setting the goal. Audible, but (with current methods) unmeasurable distortion is therefore also important to me as a recording engineer.

As for the third question, I personally don't believe that either the presence or absence of some lower order harmonics makes a great microphone. The differences must be in other forms of smearing, coloration or perhaps low-level performance. Whatever we call it, we should seek to find ways to measure and hopefully even mathematically describe what we hear.

Sorry if I'm repeating myself here...

Martin    

[danlavry]

Hi Martin,

I know that much: One can easily add distortions, but as a rule, one can not take distortions out. One can add noise, but one can not take it out. What you call "smearing, coloration or perhaps low-level performance" DOES fall into the category of distortion and/or noise.

Say we have an theoretical ideal system. What does it mean? Sound and music has it's "signature", and each channel of audio is made of a time varying amplitude, and nothing else! If one can keep track of how exact way the air moves in time, record it precisely with a mic, convert it to electrical signals...and at some point have the "perfect speaker" vibrate the air in a PERFECTLY IDENTICAL WAY, then we have the ability to move sound from one space to another, and repeat the performance. Ideally you can close your eyes and imagine you are in the performance space.

Of course I am simplifying things, because each mic can only deal with some limited space and directionality. But the special aspects are about how you place mics, combine sounds from different channels. Each mic alone is a device for recording voltage changes in time, and doing the job precisely means no added noise, no coloration, no alteration at high level, low level....

True, we do not always talk about air pressure variations in time, or voltage variation in time. We often talk about frequency response, phase distortions, harmonics... But it is THE SAME THING as voltage in time. It is just a different way of looking at the same thing.

Once again, I understand that very often the sound engineer would wish to alter the sound, and often for very good practical reasons (such as some compression for the CD in the car), or for artistic reasons (such as warmer sound, or special effect?). But I still believe it is best to have such alteration done intentionally, by choice. Such choices can be different for different music. I want a camera to yield the most accurate picture possible. If I choose to highlight a specific picture with say blue tint, that is fine. That is a choice I make. But I would not want a camera to ALWAYS yield all pictures with blue tint. I may want another picture to be ?natural, and another one to have red tint.

The condenser is a device with a fixed ?tint?, a built in distortion that you are stuck with. You end up choosing between all sorts of condensers, all with the same ?generic? coloration, some have more of it, some less.

Do you have a choice? Not much of a choice. I am not here to have you alter your ways ? I would not know what to alter to? I am just taking about an aspect that is built in to one of the tools of the trade. An aspect that is rather ?dramatic? because it is built into the principle of operation. We have to live with it, until such a time that someone would figure a different way to pick sound efficiently.

You said:
"What was more difficult to understand was how much distortion it would actually take to make it sound like a clarinet? I suspect we need much more than 1% or so."

I am sure that a whistle does not sound like a clarinet. All I was saying is that it has some elements of a clarinet sound. If you amplify the distortion harmonics enough, it would sound less like a whistle and more like a clarinet....

Reagrds
Dan Lavry
www.lavryengineering.com

[Klaus Heyne]

danlavry wrote on Tue, 07 March 2006 10:35

Klaus Heyne wrote on Tue, 07 March 2006 02:26 Three considerations to add to the discussion: Hi Klaus, "1....Serious capsule distortion sets in very late, at very high volume levels?." I came here asking what to expect in terms of max and min ?rest? capacitance (no sound)  As a forum leader yourself, you must know that sometimes discussions get broader from post to post, and a thread may not stay limited to the original quest. My remark was in response to observations of others regarding capsule vs. amp distortion. Quote:  I do not know what ?very late? means (for me, late is a term I use for time related issues).  Great, and I thought I was the King of Anal. Quote:  "2. If distortion cannot be heard, but measured, is it relevant in microphone applications?" A distortion will not be heard when it is a very low amplitude distortion. It can not be heard when it is in a frequency range we can not hear. The distortion mechanism I am addressing is high enough amplitude, and certainly at hearing frequencies (second, third, forth?.). Of course ?distortion being heard? is all in reference (comparison) to the original sound, just before it came into the microphone.   I am confused: Do you mind distortion that cannot be heard, or not? Quote:  "3. Is a distortion-free microphone automatically preferable to the listener to one with, let's say, 0.5% or 1% THD?" That is a subjective issue. Some people like tube sound because they like that specific distortion sound. Personally I am into minimal distortions, I like to have the ability to capture the sound (air vibrations) in the performance space as close as possible, and play it exactly the same (as much as possible) in my living room - the exact same air vibrations. That means 0 distortions. Dan, I cannot reflexively support your assumptions, though they sound perfectly logical.   I've been listening to very good microphones for a long time now, and the more I listen, the less I am convinced of what you assume to be self-evident: less measurable artifact equals better rendition of reality. I have come to the (preliminary) conclusion that the best mics, in terms of delivering music, are not necessarily those with the best performance on those meagerly few parameters we put all our hopes and aspirtations on: linearity of frequency response, THD-free processors, high headroom. I believe that our hearing and mental processing of sound is so unique and idiosyncratic that our attempt in duplicating this process with transducers and processors is at best a passable simulation, but nothing close to the real thing. So I have come to believe that certain types of artifacts in the process of micing a musical event are beneficial to a more realistic recreation of the musical event, rather than detrimental. As an example: Somewhere else I reported on my findings of (wrongly) believing that a certain type of theoretically superior tube biasing would yield a superior sounding mic. Boy, was I wrong when I listened to it. I have come to the conclusion that better mics are made when the listening dictates the direction of the research, rather than theory, or the constraints theories can put on free exploration. I will gladly employ the theorists to explain my empirical conclusions, never the other way around. Quote: I personally do not like to be "stuck" with some given distortions, because while one can add distortions, taking them out is not even possible. Again, I find this statement logical but only on the surface: How much artifacts (distortion) are present in every mic?  What tradeoffs is one willing to make in order to achieve "good" specs? High headroom vs. phase shift? Is that a good tradeoff? High amounts of processing to achieve a flat frequency response? We know so little, and it always takes years to get back to square one after each attempt by a manufacturer  to sell us lousy mics with good specs. quote] I am just talking about a fundamental distortion mechanism that is directly related to condensers. Knowing and understanding the operating principles is the first step towards lowering the distortions.

There are analogies of mic designers having gotten rid of all kinds of distortions, only to attract new kinds of distortions, and often worse than the ones they eliminated. (Neumann's TLMs, for example)

In your quest to eliminate the particular kind of distortion you are after, you too may find that there might be an unpleasant trade off.

Quote:

If you are really interested in experimenting, I will be glad to contribute a LavryBlue micpre, a very low distortion (and noise) unit, thus ideal for such tests.

Thanks for the offer. But again, maybe the simulation we are all after is not best served by devices that are as linear as can be?

I don't want to sound like a Luddite, and I confess that I stand at the beginning of understanding a few of the ingredients of  what makes a microphone a good witness of reality, but I cannot deny that I have lost some respect for the basic approach audio engineering has taken in its attempt to designing better microphones in the last twenty years.

I am inviting you and others to question every assumption about good audio until and unless it is confirmed by your ears. Nothing else matters.

Best regards,

[Martin Kantola]

Dan,

yes I agree, it can be expressed as voltage in time. (or velocity/pressure before it even becomes voltage)

danlavry wrote on Tue, 07 March 2006 23:17

Once again, I understand that very often the sound engineer would wish to alter the sound, and often for very good practical reasons (such as some compression for the CD in the car), or for artistic reasons (such as warmer sound, or special effect…). But I still believe it is best to have such alteration done intentionally, by choice. Such choices can be different for different music.

Right. In most cases I actually do prefer to record a very true representation of the sound. But the big question is, true in what sense?

Quote:

I want a camera to yield the most accurate picture possible. If I choose to highlight a specific picture with say blue tint, that is fine. That is a choice I make. But I would not want a camera to ALWAYS yield all pictures with blue tint. I may want another picture to be “natural, and another one to have red tint.

Good way of expressing it. How about UV filters then, which are a much more subtle thing, but in fact used to give a better result? And to me, good photography is much more than capturing an accurate image. In an excellent photo, there is not only a sense of depth and presence, but it can also accurately picture something of the personality and emotion of a person.

Not an expert, but I think a photographer picks the different lenses (which all distort the natural perspective more or less) to best capture a true representation of the subject. It is seldom acheived without the artistic and creative input of the photographer, and has sometimes very little to do with the actual optical degree of realism or even resolution.

We need to remember that our hearing mechanism consists not only of our ears, but also of the advanced processing unit our brain is. For this reason, a simple microphone can never "hear" things as we do. Lack of harmonic distortion doesn't change that fact.  

Quote:

The condenser is a device with a fixed “tint”, a built in distortion that you are stuck with. You end up choosing between all sorts of condensers, all with the same “generic” coloration, some have more of it, some less.

Working with condensers, dynamic and ribbon microphones on a daily basis I find it hard to fully agree with you. All these microphones have colorations, but if I would use our small diaphragm measurement microphone for all my recordings to minimize the "generic coloration" you are talking about, I don't think it would yield better results. Not even when I'm doing a minimalistic, "natural" recording without any added EQ, effects or dynamics processing.

Sometimes a small diaphragm omniphonic condenser does give the most natural impression. But not always. Remember the cocktail effect?

So, to sum up; do we need tools that are true to the music, to the impression or to the waveform (in one specific point in the acoustic space)?

Martin

PS. Klaus, I was writing my post while you posted yours... You said many things much better than I did!

[maxdimario]

The most compelling sounds often don't come from the most linear devices.
this doesn't mean that linearity is a bad thing, or that distortion is good.
as far as I am concerned linearity IS good -- and distortion is a bad thing.

But then one has to place themselves the universe of proportion and everyday reality.

when is too much, too much?

it is human nature to try and make our creations as 'perfect' as possible.
as people It's too easy to get caught up in what 'should be' and not what is the naked truth.
there is always a point of diminishing returns, and most of the time those who love technology and perfection above nature and it's inherent imperfection, step over that line in an effort to achieve the seemingly un-achievable.

if you get .0001 total distortion out of a circuit or total system which is made of components which DON'T have the inherent capacity for that degree of accuracy you have taken a large element of the aforementioned component's natural behaviour away.

whenever you correct the behaviour of an analog system which has its inherent limitations BEYOND a certain point, you alter the behaviour of the system. When you alter the natural behaviour of the components of an analog system you lose or alter some of the information which is transmitted through the system.

IF you are NOT aware of all the information which is passing through a system, and the system is left to behave 'naturally' with minimal correction (such as extending freq. response too much  through resonance, or various corrective electronic processes.).. THEN the analog system is likely to leave most info. untouched.

on the other hand if you FOCUS on only the type of distortions that you as a designer know and understand, it is very likely that in an effort to improve the distortion figures that you understand, you make the 'unknown' elements distort.

[danlavry]

Klaus Heyne wrote on Wed,

> If you are really interested in experimenting, I will be glad to contribute a LavryBlue micpre, a very low distortion (and noise) unit, thus ideal for such tests.< Thanks for the offer. But again, maybe the simulation we are all after is not best served by devices that are as linear as can be? Best regards,

Hi Klaus,

I really did not understand your comment. A simulation is in my view an attempt to model a real life case as closly as possible. If you want to study a distortion of a mic, you will be best served by use of as linear a device as possible. If you take a non linear device and look at it with another non linear device, you will have a difficult (and often impossible) task of seperating where the distortion originated.

But, fine, I will keep my very high end micpre.

Yes maybe we all want things to sound different then in the performance space. Maybe we want tube sound. Maybe we really like the specific signature of the condenser non linearity.
I have no problem with any of it.

But lets put the "maybe" aside for a minute.

Knowledge is good, and understanding what is going on is a good thing. This is a mic forum, so I wanted to share an aspect regarding mics with the group. I did that. My point was technical, so while some care about it, others will not care.
No surprises here.

Thank you all for the hospitality.

Regards
Dan Lavry
www.lavryengineering.com

[Martin Kantola]

danlavry wrote on Wed, 08 March 2006 19:25

Maybe we want tube sound.

What is "tube sound"? Sounds like a very non-technical generalisation to me...

Quote:

Maybe we really like the specific signature of the condenser non linearity.

Sorry if I'm slow here, but I have yet to understand both the magnitude and importance of this non-linearity you've described.

Quote:

My point was technical, so while some care about it, others will not care. No surprises here.

Are you saying that some of us participating in this discussion are less technical and therefore not interested?

BTW, are you familiar with this technology, maybe this could be the ultimate microphone since it's not based on a capacitor?

http://www.microflown.com/

Martin

[danlavry]

Martin,

You said:
?Sorry if I'm slow here, but I have yet to understand both the magnitude and importance of this non-linearity you've described.?

I say:
Well good luck in understanding it. I do have an FFT on my PSN forum, in response to some values suggested by a poster. But I do not have a goal to convince you of anything.

You said:
?Are you saying that some of us participating in this discussion are less technical and therefore not interested??

Certainly some are more technical, others are less. I am suggesting exactly what I said, no more no less. I would not be surprised if those that do not understand details would lose interest or have the focus change towards the areas they feel more comfortable with. There a requirement to get into the engineering aspects, and I do not give grades

You said:
?BTW, are you familiar with this technology, maybe this could be the ultimate microphone since it's not based on a capacitor?

I say:
I truly do not understand what you said. The statement is confusing to me. I thought we are talking about a condenser mic, one that IS based on a charged capacitor. So what are you referring to?

Dan Lavry
www.lavryengineering.com

[danlavry]

Well, as I stated earlier, I am not a mic guru. I am aware of only 3 principles of operation - the condenser, the dynamic and the ribbon. Just looking at the very basic "theory of operation" they all have thier strong points and also their weak points. I would be very interested to know about other type devices that do not rely on one of the above 3 ways.  

Regards
Dan Lavry
www.lavryengineering.com

[Tim Campbell]

Capacitor microphone capsule design is so different from one type to another that in order to have a realistic discussion you
need to specify a particular capsule. Are we discussing stainless steel, nickel, aluminum, mylar or pvc membranes? at what tuning frequency? At what polarization voltage?  At what spacing from the backplate? What design of backplate? Omni? Cardiod? Figure 8? Sealed? vented?

When looking for improvements in capactor capsule design, distortion, although it exists is a miniscule problem when compared with non-linearities caused by resistive networks, backplate design, imprecise dampning, variations in tuning frequencies, the effect of aging on the capsule and that's before we begin to explore the problems caused by the head grille or poor implementation of the amplifier design.

Only measurement microphones try to address all these problems.
Even they must be calibrated to compensate for their shortcomings. I wouldn't really want to make a recording only using measurement mics even if they could be built distortion free.

[danlavry]

OK. I understand that there are a lot of variables. The one I pointed out is only one, but it is a fundamental one, because it is not about a specific design or material. It covers the basic concept of condenser mic thus it contributes to a sonic signature that is fundamentally built it to all condenser mics, ON TOP of all the other issues you and others raised.

I keep seeing statements about other mechanisms. For me, a designer of gear, a mechanism that is fundamental to the basic concept of operation, is to be treated differently then mechanisms that have to do with specific implementation.

For example, it is good to know that a gasoline engine (Carno cycle) has some theoretical limit as to the expected efficiency. A diesel engine has a different maximum possible limits. We never get near those limits due to many practical factors, but we do design engines with full understanding of those limits and with comprehension of the operating principles.

So here I bring out a limitation that is built in to the very basic operation of a condenser. Why? Because knowledge is a good thing. That does not mean that I do not see or acknowledge that there are many practical problems. It also does not mean that I offer a better way to do things. I do not know of a better way.

Regards
Dan Lavry
www.lavryengineering

[Tim Campbell]

Dan,
Your point is well taken. All microphone designs have their limitations. I'm sure some better way will eventually be found. Up until now condensers have proved the least problematic. That's why they've been used to set standards almost from Wente's first design. There may be dynamic or ribbon measurement microphones but I'm unaware of any.

Reading through all these posts it's easy to see the great divide between members who look at recorded performances as emotional or empirical. I'm afraid I fall into the emotional category. Building capsules for me is much closer to cultivating orchids than carrying out experiments in search of truth.
No one would deny the beauty inherent in truth, but the truth I find in beauty seems to have a much more lasting effect on me.