Head Azimuth Alignment

[john]

Just doing an alignment on my Revox C270 and I am having trouble getting a stable pattern on the oscilloscope at 10Khz and 15Khz. I get a stable reading at 1Khz this I understand is par for the course but just what causes the lissajous pattern to open and close  at 10 and 15Khz is it mechanical, electrical or both !

[docb]

It's possible that the tape has been slightly stretched. Watch to see if the pattern is repeatable over a given section of tape.

[ironbut]

Hi John,

If the variations (what you're calling stability) seem to occur throughout the 10 or 15kHz tones, it could be more of a matter of perception (although all machine's performance degrades as you climb in frequency).

Most of the time I use a software scope to do my alignment. I used to wonder about what I was seeing during the upper freq. test tones (perhaps the same as you?).
If I'm not mistaken, with most digital scopes with auto ranging, the ellipse that we see during the 10k tone is 1/10th the size of the ellipse that we see during the 1k tone so effectively it's a 10x zoom to fill the same number of hash marks on the screen. That would lead one to believe that these variations are larger than the actually are.
If you have one of these scopes, try turning off the auto ranging feature and I think you'll see less variation between frequencies (although it's harder to see and use).

The source of these variations are probably a combinations of mechanical and electrical. I think that imperfections of bearings and scrap flutter caused by the fixed elements are a large contributor. But, the higher you move up the "food chain" to something like a Studer 820, the percentage of these mechanical distortions become less and less.
At that level, issues like the balance of rotating parts (unsprung weight) becomes an issue.

[john]

Thanks Doc, but I don't think its tape stretch as it happened with a new alignment tape. Steve I do zoom in to make the ellipse the same size at all frequencies, so I suppose I am magnifying any movement by doing this but i just expected it to be more stable. My Technics 1506 is the same and thats a new machine (new old stock) I tried recording a 10KHz tone on some new tape and playing it back, its the same unstable pattern.

[stellavox]

John,

How about if you look at the 10 or 15Khz L & R sine waves one above the other on a dual trace scope - try to get them as close together as you can - then adjust the azimuth to get the peaks/valleys of each as close "in time" as you can.

Charles

[john]

Hi Charles,
                yes I did try that and its much better, i had one trace superimposed over the other and the apparent movement/misalignment is much less. On a professional machine like an ATR for example is it generally the same or more stable.

[john]

Or a Studer 810, Nagra or Tascam. Is it quality of transport related or electrical

[High and Outside]

You occasionally see the lissajous pattern completely stable, with no widening. That's a sign that everything mechanical is perfect: the piece of tape, the path over the heads, the other things the tape contacts as it moves, the surface of the head, the phase of the moon and the alignment of the stars. But more commonly there is a certain amount of bounce in this display. The transport is critical--experienced audio engineers will tell you that that the moving of tape over the heads is where the whole game is played. There are many aspects to that, and many people have contributed over the years to the knowledge base of how to do it well. Doing it right takes expense, in addition to good engineering practice, which is why an ATR-100 or a Studer A80 or 820 always cost more than their brethren.

[c1ferrari]

Thank you -- that was very helpful, Paul.

[microstrip]

I had this problem with a PR99 - although the frequency response was excellent, I could not get a nice and stable lissajous pattern at 10 kHz. The wow and flutter measured with MLR test tapes was normal, and in despair I changed the heads of the machine - it completely solved the problem. I later sent the heads for relaping to JFR and re-fitted them on the machine with success. Apparently the profile of the head was not allowing a stable contact with the head. BTW, the "flatness" of the heads was around 2mm before relaping.