Physical vs Digital Azimuth cassette deck correction?

K

kylen

New member
OK - I guess I'll just blurt this out... :D

I just got a new Tascam 112mkII recently to do cassette-to-digital transfers and really don't want to fiddle around with the alignment screw for a number of reasons.

Can we discuss the virtues of physical (fix it at the source) vs digital (repair it after the fact) azimuth alignment?

I'm assuming an azimuth problem creates a situation between the left/right stereo tracks that results in a phase problem caused by one or the other tracks passing by the play head a little later in time than the other. If that's the case then, instead of poking a screwdriver into my new cassette deck, I can fix the time alignment using something like this, yes?
http://www.voxengo.com/product/audiodelay/

What are your experiences, criticisms and opinions in this area? Thanks!
 
kylen said:
OK - I guess I'll just blurt this out... :D
I'm assuming an azimuth problem creates a situation between the left/right stereo tracks that results in a phase problem caused by one or the other tracks passing by the play head a little later in time than the other. If that's the case then, instead of poking a screwdriver into my new cassette deck, I can fix the time alignment using something like this, yes?
Click to expand...
It still happens on half-track mono recordings, as anyone who used to own a ZX Spectrum can attest, so I don't see it being a stereo artifact. Whatever's physically happening, the practical offshoot is that you're losing the high frequencies unless the alignment is correct. I highly doubt you'll be able to regenerate them through software, not without increasing the hiss as well.

(On the Spectrum, where games and other software came on cassette tape, we used to align the cassette deck by ear, listening out for the highest frequencies. Even commercial tapes often had different alignment setups, which made life awkward, especially with high-speed loaders and copy-protection systems.)
 
jpmorris said:
On the Spectrum, where games and other software came on cassette tape, we used to align the cassette deck by ear, listening out for the highest frequencies.
Click to expand...
We may be talking about 2 different alignments - azimuth vs head height?

Here's a list of alignments I found on a web site and have performed some of these on different equipment (some using test equip, some using the ear):

Azimuth is the alignment of the heads perpendicular to the direction of tape travel. If the azimuth is misadjusted, any program recorded on both tracks will be played back slightly out of phase.

Zenith is the alignment of the heads in the same plane as the tape. For example, the bottom of the head can be kicked out too far, which may cause the top tracks not to fully contact the head, and would cause the tape to ride up on the head, snapping back down occasionally. The zenith can also be misaligned so that the top of the head is shoved too far out. The head should be aligned so that the tape rides against it with smooth contact to the entire surface.

Wrap is the alignment of the head so that the tape contacts the head at the center, aligned with the head gaps. The wrap adjustment causes the head to swing in an arc centered on the middle of the head, and moves the center of the head from the left to the right of center, depending on which way you adjust the screw.

Head height is adjusted so that the tape rides perfectly centered from top to bottom on the head. You can tell this immediately just by looking at the tape as it rides across the heads.

Tape-to-head contact is just that - the head should be positioned far enough into the tape path to ensure good contact with the tape. If it were too far back, insufficient pressure would cause playback to be inconsistent.
Click to expand...

Azimuth mis-alignment seems to mention the symptoms of a problem with the stereo tracks, maybe half-track mono has more to do with head height adjustment? I'm thinking on some cassette decks there were both azimuth and height - I suppose azimuth might be a way to affect height. Maybe I've overthinking this.
 
jpmorris said:
Whatever's physically happening, the practical offshoot is that you're losing the high frequencies unless the alignment is correct. I highly doubt you'll be able to regenerate them through software, not without increasing the hiss as well.
Click to expand...
OK - I think I've got it now, this little manual jogged my memory:
http://www.msxarchive.nl/pub/msx/mirrors/hanso/service_manuals/sonysdc500sm.pdf

The azimuth can affect the freq response of each of the 2 stereo tracks. Putting a stereo deck into mon to adjust azimuth just makes it easier to hear I suppose.

Now I'm understanding that without the best possible initial transfer and data/head alignment any corrections further downstream may probably involve some kind of trade off (this is my experience also). I've read this everywhere but was just wondering what you folks thought about after-the-fact repairs in a DAW kind of like Cedar Azimuth offers or folks like that.
 
One last thing (while I think about all of this).

Azimuth errors cause high frequency rolloff in mono. In stereo high frequency roll of as well as stereo timing errors (phase) can occur.

Azimuth adjustment affects perpendicular head-to-tape alignment along with (probably) some slight head height alignment.

Any comments on these opinions anyone? Thanks jpmorris...all for now! :)
 
Digital post facto treatments are limited, and anything involving physical alignment is particularly problematic. Personally I wouldn’t stress about it too much with a brand new 112MKII. See how it sounds. Azimuth is only one aspect of alignment, though some use the term for alignment in general. :)
 
Kylen,
You seem to understand it pretty well. Yes ajusting azimuth does affect head height secondarily but it depends on how far you adjust azimuth off standard and also how the head is set up. Ideally you would adjust just the head height as a separate exercise without affecting the other parameters but how many machines will do that? I have one old Akai which does it and it's a rare feature.
The Cedar or similar auto azimuth corrections do, as you say, only correct for stereo phase or timing errors. They do nothing about errors in azimuth within one track.
Whereas if you mechanically align it in monoed stereo, you get both in one hit.
One of the big problems with cassette recordings is azimuth DRIFT. It will always drift and wander to some degree. If the azimuth is significantly off to one side of optimum, it doesnt matter which, there will be more drift and attenuation than if you had peaked the azimuth manually and it only drifted around the optimised peak. Stephan Kudelski has a great little section on this in his Nagra manuals.
I have come to hate the sound of azimuth errors and so it's a bit of a passion with me to always try and get it right. it takes so little time! Running through a stereo monitoring amp with a mono switch helps aligning.
Tim
 
OK - thanks Beck I'll sleep on all of this and see how the transfers sound. I already know I have to re-EQ and a couple of other things in many cases anyway.
 
Tim Gillett said:
One of the big problems with cassette recordings is azimuth DRIFT. It will always drift and wander to some degree. If the azimuth is significantly off to one side of optimum, it doesnt matter which, there will be more drift and attenuation than if you had peaked the azimuth manually and it only drifted around the optimised peak. Stephan Kudelski has a great little section on this in his Nagra manuals.
Click to expand...
Thanks for that info Tim. Yes I am getting drift on certain cassettes - it improved significantly when I went to the Tascam but I can hear the tracking of the worn tape wandering from time to time - azimuth drift. Yes I think that makes sense - I'm getting a "sound" picture in my head anyway...you're a screwdriver azimuth guy (I read in other posts) so I appreciate the insight...I'll see if I can find the Nagra excerpt by Stephan - Thanks!
 
Beck said:
Digital post facto treatments are limited, and anything involving physical alignment is particularly problematic. Personally I wouldn’t stress about it too much with a brand new 112MKII. See how it sounds. Azimuth is only one aspect of alignment, though some use the term for alignment in general. :)
Click to expand...

I knew what I meant when I said this, but looking at again I could have said it better. Ok... Take 2... :D

Digital post facto treatments are limited, and using digital correction to remedy a physical alignment issue is especially problematic.

There, I feel better now. :)

Anyone who isn't me (which is most people) may not have read the first one like I was thinking it.
 
Last edited:
jpmorris said:
(On the Spectrum, where games and other software came on cassette tape, we used to align the cassette deck by ear, listening out for the highest frequencies. Even commercial tapes often had different alignment setups, which made life awkward, especially with high-speed loaders and copy-protection systems.)
Click to expand...
The ZX81 was fussier. I can remember drilling a hole in the tape cover of a cassette deck at the time to make the azimuth easier to adjust.
 
OK - thanks for shaking the gray cells. Skew is what I was trying to think of from my Customer Engineer days and computer 9 track tape drives.

Skew and azimuth variation can create timing errors between different tracks OR can create frequency rolloff of any single track. Does that fly with you guys?

Here's some into I found at http://www.tpub.com/neets/book23/101b.htm
SKEW

This magnetic tape recording specification only applies to multi-tracked magnetic tape recorders.

SKEW DEFINITION

Skew is the inter-track fixed and dynamic displacement, or change in azimuth, encountered by different tracks across the width of the magnetic tape as it passes the magnetic heads. In other words, it's the time difference between the tracks on a multi-tracked magnetic head.

A typical skew specification might read "+/- 0.15 microseconds between adjacent tracks on the same head stack at 120 inches per second."

This means that one of the tracks on a magnetic head could lead, or lag, the track next to it by as much as 0.15 microseconds at 120 ips. This specification applies to both fixed and dynamic skew.

Fixed skew can be caused by

magnetic tape recorder electronics,
gap scatter in the magnetic head stack,
azimuth alignment of the magnetic head stack, or
fixed difference in tension along the tape path
You can minimize most fixed skew by adjusting the magnetic recorder's electronics or by realigning the magnetic heads.

Fixed skew errors usually do not show up when magnetic tapes are recorded and reproduced on the same tape recorder. Since fixed skew errors are additive, they'll usually show up when you record on one magnetic tape recorder and then reproduce on another.

Dynamic skew errors are caused by either the magnetic tape transport or the magnetic tape itself. If the tape transport guides are worn or sticking, the magnetic tape won't properly pass over the magnetic heads. It'll drift and pass the magnetic head at an angle (like a car skidding on an icy road). If the magnetic tape itself is warped or isn't uniform across its width it, too, will cause dynamic skew.
Click to expand...
Thanks for the opinions or trouble (could go either way :D ) depending on your nature...it's my nature to be curious till I get a fairly comfortable grip on things. :)
 
You must log in or register to reply here.
Back
Top