Finding lowest frequency in audio file

[bluesman]

9 hours ago, Music Matters said:

Hi,

 

My speakers are Wilson Benesch Vertex, and midwoofer does not have a low pass filter, it's directly connected to the amplifier.

As I have a powerful amplifier, sometimes when listening to a new albums which have very low frequencies the speaker bottoms

out, even on not so extreme sound levels. Does not happen often, but I don't wont to risk damaging the speakers.

 

I would like to know how to analyze FLAC file and to find what is the lowest frequency in that audio track?

The Vertex only has two drivers - it rolls off below 44 Hz per the WB specs. There’s nothing you can do short of adding an external crossover and a sub. It’s hard to imagine that you’re driving a voice coil to its stop at any listenable level, though. They’re substantial speakers that ought to be fabulous with a good high powered amplifier.

 

Have you contacted the manufacturer? 

[bluesman]

9 hours ago, Music Matters said:

I can listed H. Masakela's STIMELA (Live) at 86 without any sign of problem, but than listening L. Cohen (right now don't have time to find what album was) at 67 for a few tracks without problem than on third track POP, and this track I cannot listen more than 63. 

If those numbers are dB SPL, something seems wrong unless your listening room is a huge auditorium. I drove my Rogers LS3/5a speakers with a Hafler 500 for a few years, and I got an occasional coil thump on big band jazz when I pushed it to over 90 dB (on peaks) in my 8’x14’ library with an 8’ ceiling - and that’s loud. Fortunately, I backed off immediately without damage. But I added a powered sub for practicing at live levels.

 

I wonder if you’re getting this from very low frequency transients caused by an intermittent electronic or interconnect problem.  It could even be in your source files - imperfect rips can have some odd noise in them.  I just have a hard time believing that you can’t listen to a Leonard Cohen track above 63 dB at your listening position without bottoming your voice coils in such wonderful speakers.

[bluesman]

Something seems off here. You describe your amplifier as being a high power unit. The scale on your volume control is arbitrary, but listening to anything at 90% of its excursion no matter what its taper seems very high, unless you have attenuation elsewhere in your system &/or your input signal is quite low.

 

This may all be quite fine and normal, but it’s odd to have to limit your program material with a system that good. If all is fine otherwise, you’d probably be fine just adding what we used to call a rumble filter to sharply roll off frequencies below about 30. The one from Harrison goes in the line level input to your amp & seems like what you need - look on Amazon. But it may affect SQ a little.

 

But I still worry that this seems wrong. I’d check everything carefully before just living with it. You could even have a weird problem like DC offset in your amp that’s holding the voice coils closer to one end of their travel.

[bluesman]

4 hours ago, Music Matters said:

I did some measuring, previously mentioned track from Wailin' Jennys, amplifiers setting 65 sound level AVG 95.9 dB (max 100.3 dB) AC/DC - Thunderstruck amp. level 71 - sound level AVG 92.4 dB (MAX 98.3) . 

There’s the answer. You’re pushing your speakers really hard!  I doubt that 1% of AS people ever hit 100 dB SPL (A weighted) at the listening position  - that’s LOUD! It’s a rare symphonic performance (or gas powered lawn mower) that tops 90 dB.  Even loud rock concerts rarely top 105 dB in the venue space.

 

I think you’re asking more than your speakers can deliver. To achieve clean, 20-20k  95+ dB levels at 3+ meters from the speakers, you’ll need more robust units that are built to do that.  

 

There’s probably nothing wrong with your equipment, but (as a physician) I’m a little worried about your hearing from the SPL levels you describe, I practice my playing at  live performance levels that can average 90 on some tunes and when I’m in the mood. But that’s only for a few minutes a few times a month, and it pushed my LS3/5as too hard for my comfort. A few coil thunks stopped me, and I use other speakers for that (currently Focal 726s).

 

Your speakers are wonderful, but they’re not up to sustained full frequency listening levels of 95-98+ dB from 20 Hz.  The bigger WBs probably are, but I’m not even certain of that. Turn ‘em down for your sake and theirs. And you might want to visit an otolaryngologist or audiologist to make sure your hearing is OK. It’s a good idea for any audiophile and music lovers any age.

[bluesman]

I don’t understand how you could be banging your voice coils at 80 dB SPL in a room of typical home size with any program material through a properly functioning system. But I’m not making any progress at figuring out what’s really happening. So here are two answers to your question:

 

You can download any of a dozen+ spectrum analyzers to give you the frequency content of an audio file - just Google it. I’m pretty sure that there’s one in Audacity too.

 

Search Amazon for Harrison Industries crossover high pass inline filters. The one for 30 Hz should do what you think you want. I hope this helps.

[bluesman]

3 hours ago, Music Matters said:

for example L. Cohen's Darkness it is always, but always at the same position, and it is not a CD RIP it's official Qoubuz HiRes download, so I have hard time beliving that cables could be the problem.

If you’ll please post the time point at which it occurs, I’d like to see if I can identify the problem.  Is it only with the Qobuz download or have you experienced this with other sources of the same track (eg CD)?

[bluesman]

I just downloaded a 320 mp3 of Darkness and analyzed it with Friture.  Here are the spectra for lowest levels and highest (red is the peak trace):

 

 

 

I don't see anything in there that would bottom your voice coils.  The general SPL goes up at about 30 seconds, but I didn't see any spikes while watching in real time - and there are no tell-tale peaks recorded in red.  The tiny red traces below 90db weren't even enough to show on the octave spectrum.

 

So I can't identify anything in that file to do what you describe.  You can download Friture and look at your Qobuzfile to see if there';s something in there that's not in the file I have. 

[bluesman]

35 minutes ago, Music Matters said:

 

Just now did the test, same thing with both CD and the HiRes (24-44), the CD is recorded slightly louder so the pop is also louder, the pop is at the 4:20 (21) at 55 amp level everything fine, at 56 (avg 81 dB) starts the pop, as you increase the volume the pop is louder. 

 

That's very interesting!  At exactly 4:20, the beat is discontinuous - count it out and you'll find that the beat is irregular there because a portion is skipped.  I don't see a blip in the spectrum scan and I don't hear any sonic anomalies (e.g. pulse or other extraneous noise) in my 320 mp3 or the Spotify mp3. So I went to YouTube and found it in all of the low res videos, with a slight pop where it skips:

 

 

It is not present in this live version (which is pretty good):

 

 

So it's probably from a glitch in creating the source file, and it's more prominent in the high res file you downloaded from Qobuz than in mp3s or the YouTube versions.  The live versions are clearly different from the others and do not have this anomaly, so I suspect you'll be able to play the live tracks as you wish without fear.

 

As far as I can tell, this is a weird production flaw in the source file.  If I'm correct, your speakers are innocent bystanders.  I wouldn't have expected this to bottom your voice coils at any reasonable level, though, unless you're using EQ/DSP to boost your lows a fair amount. I suspect it's just the pop itself that you're hearing, rather than full voice coil excursion (which is a much more resounding and alarming thump/clunk than what I hear on this recording, even with serious bass boost).

[bluesman]

I just listened to and analyzed the three files you sent me (Darkness Redbook, Darkness hi res, and Fragile). The same discontinuity I heard in the mp3s is in your Darkness files at exactly the same point (4:20) - so they were apparently all derived from the same source.  It's somewhat more prominent in your files, and I can understand how you might have thought you heard a voice coil hitting its stop if you've never actually heard that happen before.  But a bottoming voice coil is a lot more alarming than what we hear on these tracks.

 

It's a little upsetting to discover that a "high resolution" file coming at a premium price has this kind of flaw in it.  As this was Leonard Cohen's last album (as I recall), a proper master should be readily available and should have been used to generate a high res file worth its cost.

 

As for the Jennys' Fragile track, there's nothing except intermodulation and some low level percussive transients louder than 30-40 dB below the track's peak SPLs, except the fundamental frequency (AKA the first harmonic) of the E string on the upright bass (41.2 Hz).  I only listened once so far, but I didn't hear anything wrong with it.  Friture analysis shows nothing below 40 Hz that comes anywhere close to the track's SPL peaks.

 

I'm neither a qualified technician nor an industry professional.  As an experienced amateur audiophile and professional musician, I strongly doubt that there's anything wrong with your equipment.  But that and a dollar will buy you a cheap cup of coffee.  So if you have any doubts at all, you should let your dealer or the manufacturer help you sort it out. You paid a lot of money for your stuff, and I don't want to dissuade you from being sure you got what you paid for - it's great stuff!

[bluesman]

I listened to the Nils Landgren version of Fragile that you sent me, and there's definitely a single loud extraneous pop / crack at the time point you identified.  Friture shows it to be a fairly broad spectral splash, but it has no content below about 63 Hz and it does not sound to me like a bottoming voice coil (although a sound file captured from your speakers with the mic on your phone is far from an accurate replica of the source file). Most importantly, though, this sound is NOT present in what appears to be the same track on YouTube (below) and there's no sonic energy in the spectrogram of the YouTube file to whack your voice coils.  So this also seems to me to be an extraneous noise in your file.  Again, I can't tell you that it's not your speakers because I can't hear it live. 

 

The sound in this YouTube video below is at least passable through HDMI or other digital output into a good DAC & system.  Play the audio below through your system and see if you hear the same noise at the same point.  I don't, and I pushed it to a measured 90 dbA at 1 meter on axis for 2 seconds before and after the point of interest (the practical safe limit of my Rogers monitors, which are the easiest of my speakers to push to coil banging).

 

 

[bluesman]

3 hours ago, Music Matters said:

Thanks, I'll try to find the way to play the YouTube version, and I have to learn to use Friture, seems I have a talent for finding anomalies in songs

I know your intent is humorous, but you've probably unearthed a serious problem. The discontinuity in the Darkness track seems like a production flaw to me.  The timing error had to occur in post production -  the band simply did not play it that way.  The fact that it's present in both the basic mp3 and a supposedly high quality file sold at a premium price is more than a little unsettling.  I expect a premium product to reflect great care in its design, creation, and delivery - and this clearly does not.  Did no one listen to it before adding it to a premium streaming and download service?

 

I don't know for certain what the audible "crack" is in the Fragile track. But I find it very hard to believe that your voice coils are bottoming from what I see and hear in it.  We know that many so-called high resolution audio files are actually not what they're represented to be.  Simply upsampling a Redbook file to 24/192 or better doesn't make it a premium product, and many "premium" files are just hi res versions of mediocre tracks.

 

The moral of the story is an old saw:  caveat emptor!