out of balance

I would like to know if the unevenness is evidence of tracking distortion, something else, or is totally normal. If from mis-tracking, is it possible to evaluate whether it is permanent damage to the vinyl or evidence of current playback difficulties? I can provide a longer sample if this is not enough to be useful.

The only thing wrong with that clip is its polarity is reversed.  The peaks should be positive, not negative.

This is the classic close-miked trumpet/trombone thing where an asymmetrical waveform is totally natural.  Just envision those little puffs of air originating at the mouthpiece and still remaining a bit "puffy" by the time they reach the instrument's bell.

Thanks, but it is the unevenness of the sound over time, not the asymmetry of the wave form, that interest me. The visual correlation of this is easily seen in Spectral View.

OK, sorry, I did misunderstand you.  But having another look in spectral view again looks totally normal to me, both for the music and that fact that this is a vinyl transfer.  So perhaps you'd be willing to elaborate on the "unevenness" or perhaps put up a graphic so this gets into my dense head correctly.

The only reference I know is the tracks for testing cartridge alignment/tonearm adjustment on some test LPs. They are a single frequency tone (e.g. 300Hz). There are a number of them cut at progressively higher levels, the highest of which (suggest the literature with the LP) probably cannot be perfectly tracked by any tonearm-cartridge setup.

Going from track to track in order, and making fine adjustments to correct tracking, I finally come to the limit of my ability to make a fix. On the highest level track, and on some lesser ones before the adjustment is optimized, there is a sort of buzzing sound added to the basic 300Hz tone. It is not constant like the test tone but comes and goes. Adjusting away that extra sound as the evidence of improved tracking that comes with more correct tonearm settings.

On the LP from which this sample came, and some others with trumpet, I hear something that sounds to me rather like that buzzing. I have not found any evidence of it on any trumpet on any commercial CD. However, I don't have many such commercial CDs, and none have quite the same kind of music.

In the sample are three ‘puffs' on a trumpet. On none of them is the sound constant for the entire duration, and there is a sound that reminds me of the buzzing. Possibly this is the proper sound for trumpet played in this style, I have no way to audition live performances. Or, maybe it is evidence of my poor system performance, maybe it is evidence that the disk was damaged by the poor equipment of previous owners. Its similarity to the sound on the test LPs makes me think it is distortion.

When you observe in Spectral View, you can readily see that the intensity varies. The first and third occurrence get much brighter towards the end. If you zoom in close, you can see that there is moment to moment variation within the brighter regions. This, like the sound, is somewhat similar to what I see in recordings of those harder to track sections on the test LP.

Sorry for the delay, I've been away and forgot about this.  My take is that this has nothing to do with tracking problems at all, and what you've got here is a good representation of the style and audio quality, given the source (which I presume is pre-1950).  Have a look at the processed version of your file attached to this post.  What I did was
1) de-click it; they're typical small noises but don't help the situation
2) ran a phase-extraction to preserve only "center-channel" sound, since this is a mono record
3) summed the remainder to mono

The audio bandwidth pretty much dies above 5Khz.

Anyway, any unevenness now presents itself as the "pulse" at the start and stop of each note, and that is just how the instrument is.  Or am I still missing it?  If so, you'll have to draw on a picture of the spectral view to get it into my brain...

This may be correct, but I don't understand some things. Recently I open a thread to ask about signal and noise in mono recordings.
http://audiomastersforum.net/amforum/index.php/topic,6390.0.html
Perhaps I also do not understand something important in that thread, but I do see not anything that leads me to believe there is a significant deficiency in the way I convert to mono, which is usually with the Channel Mixer's LR to Mid-Side preset. The other two methods I listed give identical results.

I don't believe my version of the CoolEdit  has any ‘phase-extraction' facility. Do you say that your step 2 phase-extraction process is necessary to successfully remove the stereo cartridge artifacts, that the summing to mono I do is incorrect or insufficient?

Is it in this phase-extraction process that you removed (essentially) everything above 5kHz?

Opening the unprocessed sample, if you zoom in close, say putting the cursor at 0:00.190 and zooming in until full screen is about 25 milliseconds, you see many recurring waveforms. At 0:00.186 is a small click, but most of the cycles in that vicinity appear to be free of clicks. Switching to Spectral View, you see various colors corresponding to frequencies in those waveforms, going all the way to the Nyquist limit at the top of the screen.

I believe, for the most part, those frequencies much above 1kHz are harmonics of the trumpet's fundamentals and/or harmonic distortion, not evidence of clicks or crackle. (You disagree? based on what?) Above 4kHz to 5kHz, most of the energy may be distortion, but it would probably be similar in a true stereo recording (from LP) i.e. it has nothing in particular to do with this being a mono LP. It would be nice to have something similar for comparison from a commercial CD, but I do not.

For evidence of my beliefs, I look at the test LP 300Hz tracking tracks I mentioned in the earlier post. That may have some some important difference vis a vis results from this mono recordings because, I believe, the test LP is true stereo, simply with identical information on both channels. However, as I don't know any result of that, I plow onward.

Zoomed in, I see vertical color bands in Spectral View, corresponding to the 300Hz waveforms visible in Waveform View, similar to what I see in the trumpet recording. On the lowest level 300Hz track there is almost none of this higher frequency color, on the highest level 300Hz track the color handily reaches 22kHz.

On the highest level 300Hz track, a Frequency Analysis graph shows the 300 Hz fundamental, plus strong harmonic peaks every 300 Hz, gradually decreasing in amplitude, all the way to the right hand side of the display. There are only a few distinct harmonic peaks in the Frequency Analysis graph of the lowest level 300Hz track.

The recording was made from a new LP, thus there is no reason for me to believe it has many clicks or any crackle to contribute to these frequencies higher than 300Hz.

Zoomed in, the waveform view of the highest level 300Hz track has very obvious distortion on the left channel troughs and the right channel peaks. The lowest level 300Hz track just looks like a smooth sine wave.

If I run an FFT low pass filter on the highest level 300Hz track, set at 525Hz (where the 600Hz peak starts to rise in the Frequency Analysis graph), all of the harmonics and the Spectral View color above that cutoff frequency are removed. The Waveform View now show a very smooth sine wave; all the visible distortion is gone.

Since the only difference there is supposed to be between these tracks is their level (+12dB vs. +18dB, according to the LP jacket), the higher order harmonics and distortion in the highest level 300Hz track almost has to come from poorer tracking due to its high level. (Is there a more reasonable explanation?) The buzzing in its sound is consistent with the instruction's description of mis-tracking. That buzzing is also gone after applying the low pass filter.

In the version attached to this post, I have also done declicking. The declicking is much more aggressive than I would use on all but the worse condition LPs (which this was not). That was followed by two different parameter decrackling runs. While this might not have eliminated everything, it certainly seems to me to have gone beyond anything audible. I then converted to mono by my normal method.

As you can see in Spectral View, there is still a fair amount of energy above 5kHz. My result is easily discriminated from yours in blind ABX tests. The primary difference seems to be those higher frequencies, especially at the ends of the first and third "puffs" although I am not certain this is what I hear. As I said in the earlier post, the sound reminds me of the buzzing on the higher level 300Hz tracks, although it is hardly identical, which brings us back to the beginning, and the reason I asked about mis-tracking distortion.

I guess I now want to know is this is somehow a phase difference effect rather than tracking distortion (or anything else) and that is why it was removed by your processing. I've never done any kind of "fixing" that has had such an extreme effect on the higher frequencies, except a few deliberate low-pass filters, applied only to limited sections, in order to get ride of some bad high frequency noise. If your rendition is truer to the original recording, then I probably need to add something to my LP clean-up regime. I don't have any idea of what however.

I have had similar symptoms on various transfers of stereo LPs, always, or at least mainly, on very high level bursts. That is why I did not make a point of this being a mono recording, nor provide a sample of only one channel. When something similar occurs on a real stereo LP, is it likely a phase difference/conflict problem? Could it be solved there without destroying the rest of the recording?