Topic: Clipping Problem

[Mr. Bull]

I just recently taped an exremly loud show.  I was using my Microtrack 24/96 set on L with my 10DB attunuator and my MM-MCSM-5 Microphone Madness mics

Mic Speces

Frequency response: 20 - 20,000 hz
Open curcuit sensitivity: -50dB (5.6 mv) re 1v at 1 pa.
sighal to noise ratio: more than 60 dB, 1Khz at 1 Pa.
Maximum input Sound Level: 120 dB SPL


Previously I have recorded much louder shows but never had any clipping problems.  I just ran it through Audacity and saw all kinds of clipping.  The sound is not bad its just on some deep bass I get a little hint of distortion.  Is this due to my Mics and their Fz response?  Or did I put it on the wrong setting?

[Brian Skalinder]

What do you mean by "clipping"?  Three common ways I've seen that term used:

<1>  overloaded the mics (analog domain)
<2>  overloaded the analog inputs on the preamp or recorder (analog domain)
<3>  more than 3 consecutive samples at 0 dBFS (digital domain)

[DSatz]

Mr. Bull, Audacity (if it's the type of software I think it is) should be able to give you, and us, some useful clues beyond what you've described.

For example, are the waveforms smashed flat up against the absolute limit (digital full scale)? If so, then your recording levels were set too high (assuming that your recorder lets you set the recording level--unfortunately I'm not familiar with the MicroTrack). There may also be other problems, but you definitely need to set levels so that you're not trying to exceed the absolute digital limit ("full scale"). It's even better if you can manage to leave a dB or two of "headroom" when recording.

On the other hand you might see waveforms that repeatedly smash up against some other seeming limit that's lower than digital full scale (0 dB). If so, that's evidence that some piece of your analog gear was very likely being pushed to its maximum voltage (usually the analog inputs of the mike preamp or recorder--though microphones themselves can be overloaded sometimes, too).

If the microphones themselves weren't being overloaded, then you may need more than 10 dB of attenuation going into the recorder's inputs. In fact, if your recording levels were set correctly, then that's the simplest diagnostic test: Use more attenuation, and see whether the problem goes away. If so, problem solved; just keep using the 15 or 18 or 20 dB attenuators instead of 10 dB. (If more attenuation than that is necessary, please consider the likelihood that your hearing will be permanently damaged from the sound levels.)

If increasing the attenuation doesn't make the clipping go away, however, then your microphones may be overloading. Sometimes this is due to an inherent limit in their design, but sometimes it is due to inadequate or incorrect powering and if this is fixed, the same microphones may be able to handle higher sound levels without overload. It depends on what microphones you're using.

Finally, sometimes you look at the waveforms and you see no clipping at all, which implies that something in your playback setup was distorting. That's not the case here, according to what you said. But do you see how you can play detective if you only take a few clues which the editing software can give you?

--best regards

[Mr. Bull]

What do you mean by "clipping"?  Three common ways I've seen that term used:

<1>  overloaded the mics (analog domain)
<2>  overloaded the analog inputs on the preamp or recorder (analog domain)
<3>  more than 3 consecutive samples at 0 dBFS (digital domain)

Well I may be using clipping in the wrong sence.  I thought that when the signal redlined it was clipping.  When I loaded the show onto audacity The wave forms would redline.  Smashed flat to the top of the spectrum in many places.  Like the mics were overloaded or something.   I then normalized the recording, the waves were flat in many places where there should be spikes.  No preamp

[Mr. Bull]

Mr. Bull, Audacity (if it's the type of software I think it is) should be able to give you, and us, some useful clues beyond what you've described.

For example, are the waveforms smashed flat up against the absolute limit (digital full scale)? If so, then your recording levels were set too high (assuming that your recorder lets you set the recording level--unfortunately I'm not familiar with the MicroTrack). There may also be other problems, but you definitely need to set levels so that you're not trying to exceed the absolute digital limit ("full scale"). It's even better if you can manage to leave a dB or two of "headroom" when recording.

On the other hand you might see waveforms that repeatedly smash up against some other seeming limit that's lower than digital full scale (0 dB). If so, that's evidence that some piece of your analog gear was very likely being pushed to its maximum voltage (usually the analog inputs of the mike preamp or recorder--though microphones themselves can be overloaded sometimes, too).

If the microphones themselves weren't being overloaded, then you may need more than 10 dB of attenuation going into the recorder's inputs. In fact, if your recording levels were set correctly, then that's the simplest diagnostic test: Use more attenuation, and see whether the problem goes away. If so, problem solved; just keep using the 15 or 18 or 20 dB attenuators instead of 10 dB. (If more attenuation than that is necessary, please consider the likelihood that your hearing will be permanently damaged from the sound levels.)

If increasing the attenuation doesn't make the clipping go away, however, then your microphones may be overloading. Sometimes this is due to an inherent limit in their design, but sometimes it is due to inadequate or incorrect powering and if this is fixed, the same microphones may be able to handle higher sound levels without overload. It depends on what microphones you're using.

Finally, sometimes you look at the waveforms and you see no clipping at all, which implies that something in your playback setup was distorting. That's not the case here, according to what you said. But do you see how you can play detective if you only take a few clues which the editing software can give you?

--best regards

Very insiteful.  It seems like something failed after looking through what you wrote.  I was thrid row at Roger Waters and did not have any problems.  That show about blew my face off.  No redlining problems there.  This wave form was totaly diffrent more smashed spikes. 

I'm not sure if Microtrack makes 15 db or larger attunator.  This 10 DB was supposed to solve this problem with the MT. 

[Brian Skalinder]

How about a pic of the waveform in Audacity <1> zoomed out, <2> zoomed in on a problem area?

[DSatz]

Mr. Bull, you're using the word "clipping" perfectly. But before we all jump to any conclusions, could you please tell us the status of one crucial "clue" in this mystery:

• The maximum levels in your recording before you "normalized" or did anything else to the file--did they reach digital full scale (a/k/a the absolute maximum level, 0 dB, or whatever else you like to call it)? Or did they appear to "stop" (or clip) at some lower level?

Without a clear, direct answer to that question, we can only speculate as to the cause of the problem.

--best regards

P.S.: When someone writes as long-winded a message as I did before, maybe it's not the best use of bandwidth to repeat the whole thing back! Anyone who's reading the forum can see all the earlier messages.

P.P.S.: Live recording can involve an enormous range of sound levels and of microphone sensitivities. One well-known professional microphone (which happens to be from Neumann) puts out 62 mV at the standard SPL of 1 Pascal while another (which happens to be from Beyer) puts out only 1 mV at the same sound pressure level. Modern transformerless condenser microphones can put out a Volt or even more in some cases. But most recording equipment designed for consumer and semi-pro use--miniature or portable equipment from Far Eastern manufacturers especially--is still built on the assumption that "microphone-level" signals will never be even a tenth as much as that.

If you are going to subject that class of equipment to professional requirements, you usually have to adapt it in one way and/or another; that's just a fact of life. For example, your MicroTrack's "phantom powering" doesn't meet any industry standard whatsoever; some microphones will work with it more or less, but to use condenser microphones from the very best manufacturers with that recorder--or even to get optimal results from the microphones that do work with whatever its cockamamie voltage and supply resistances may be--you'd need to use a professional-quality phantom power supply instead.

Similarly, resistive attenuators ("pads") are very useful tools--in fact I'm constantly surprised that more people don't realize how useful. But if a recorder's microphone inputs are easily overloaded, then especially with LOUD music and sensitive condenser microphones, the incoming voltages may simply be too great sometimes, even with a 10 dB pad in the line. Thus it's wrong of anyone to promise you that a 10 dB pad will always be an adequate solution to the problem of input overload--but realistically, that isn't a promise that anyone ought to believe, either.

[Mr. Bull]

How about a pic of the waveform in Audacity <1> zoomed out, <2> zoomed in on a problem area?

Give me a day or two to get that up.  I live in the stix durring the week and need to transfer the picture by USB stick.

[Mr. Bull]

• The maximum levels in your recording before you "normalized" or did anything else to the file--did they reach digital full scale (a/k/a the absolute maximum level, 0 dB, or whatever else you like to call it)? Or did they appear to "stop" (or clip) at some lower level?

Modern transformerless condenser microphones can put out a Volt or even more in some cases. But most recording equipment designed for consumer and semi-pro use--miniature or portable equipment from Far Eastern manufacturers especially--is still built on the assumption that "microphone-level" signals will never be even a tenth as much as that.

If you are going to subject that class of equipment to professional requirements, you usually have to adapt it in one way and/or another; that's just a fact of life. For example, your MicroTrack's "phantom powering" doesn't meet any industry standard whatsoever; some microphones will work with it more or less, but to use condenser microphones from the very best manufacturers with that recorder--or even to get optimal results from the microphones that do work with whatever its cockamamie voltage and supply resistances may be--you'd need to use a professional-quality phantom power supply instead.

1. I will let you know very soon what the deal is with the signals.

2. If Phantom power makes the mics perform better than and the microtracks phantom power is not standard then its must be Junk.  I realize that this is myfirst setup and I know there is much better out there.  I'm starting to think its my whole set up that is flawed and that I have reached its maximum.  What mics would be better at this point to prevent this from happening again?

[Mr. Bull]

Here is what it looks like