So then what is the meaning of dual azimuth system ? The head NEVER MOVES, just the motor and the electronics of the head that only switches. You are definitely mistaking for a walkman that has a rotating head or a head that shifts in height (just like an 8-track recorder). The rotating heads are the ones that are likely to have azimuth errors whereas the shift heads are likely to have crosstalk (hearing either the forward or reverse side information faintly going backwards).

Nobody said anything about rotating heads in a Walkmans. there is no room for Rotating heads in a Walkman. Rotating heads are found in Mid to Hi-End Cassette Decks. 

"The rotating heads are the ones that are likely to have azimuth errors whereas the shift heads are likely to have crosstalk"

On contrary, Rotating heads are very much precise than any other azimuth system because once they are set they will never have any kind of errors unless they are faulty.

Crosstalk could be found on any cassette player if they are not shielded properly therefor the signal is leaking to the other Chanel. another cause for crosstalk is when the head main screw is loose or there is a defect to the head base and it is not flat.  

Are you serious ?! Have you forgotten one thing bad about rotating heads ? That's right, they loose their azimuth over time the more times you play both forward and reverse ! Also some walkmans use a rotating head (rotates in about 180) or a shift type head (that shifts in height) in some Walkmans as their auto reverse method.

Yes I am serious! as I said, once they are properly set they are more precise then stationary heads. Stationary heads have limitations, they will never have the same precision for both forward and reverse.

Interesting, Can you post a picture of a Walkman with a rotating head? I never seen any  before I will be happy to see such a beast

Rotating heads are more prone to trouble and they were used in some walkmans, Panasonic use them in their newer models, they use a spring and gear to rotate the flip the head around, which would crack overtime and would sometime go out of aligment due to spring tension, metals and so forth, will deform, causing 'play" and thus deviations from the original (correct/new) locked position and it will slowly knock itself out of alignment. That why Nakamichi, when with the flip the tape design and using a stationery 4 channel head instead. I works on ton of these autoreverse deck and nothing but problem when it comes to head azimuth

There was some better decks that used rotating three head design held the azimuth and alignment much better.

Not only Permalloy heads wear out quicker but they have inferior sound quality compared to Amorphous heads. It's a strange thing why the Sony TC/WM-D6C would want to do this to newer versions.

Today I "discovered" that many SONY walkmans have also an automatic bi-azimuth system, and I didn't realized about that until today.

The system is found in many models based on the same basis. Digging a bit in my collection and in my database and image resources I found that the first model to have it was the WM-R202, which is the basis that many other models uses. I can confirm that my R202 have this bi-azimuth system. surprisingly, SONY didn't say anything about that ¿¡?!

This is the head on my WM-504, which is identical to the one found in the R202:

As you can  suppose, that little thing below the head is the key. but as you see, it doesn't work by pushing it from below (as AIWA mechanism). It has to be slided to the left:

So if SONY release the R202 back in '86, and AIWA it's PX303 in '89, then SONY was the first one to release such system; am I right?

I have three WM-EX909 with dual head system. Here is my black 909:

And here it's dual head:

Hope it helps!

It not automatic bi-azimuth, theirs adjustment screws on the otherside that hit the pole as you turn the screws and move it for alignment.

Checking my WM-100 service manual, it does seem to have Bi-Azimuth. Two azimuth screws on the board side. Not automatic like a Dragon though.

Even if a Walkman had Bi-Azimuth, marketing wise it would just be another bullet point to pad the list- I don't think many consumers knew what azimuth is, let alone the benefits of bi-azimuth. And adjusting it was not something many owners would do.

It is  Bi-Azimuth, just not auto.

wonderful thread. being utterly unqualified from the technical point of view, below i just give my musical tribute to the topic of the (bi-) azimuth

aiwa? sony? you decide which is which. (auto)reverse mode activated 

the R202 it's not automatic, just bi-azimuth, of course. My mistake.

but I don't understand what you say. Could you say it in other way?

Nakamichi Dragon's NAAC is completely different from Dual Azimuth. The first adjusts for every tape while the second one is only for one tape which the tape is recorded at a different azimuth.

OK, what is the weakness of stationary heads? If you are talking about the way how manufacturers set the azimuth, then in some ways, "YES" if the alignment is not at 90 degrees. If it is the adjustable version, then both sides are aligned immediately if they are at 90. 4-track heads are EXACT precision. Only not good with 4-track heads are that you get a little bit of small crosstalk of the recorded music on the right channel due to parallel magnetic pole. Multitrack recorder heads have this effect more pronounced because the tracks are wider and equal guardbands whereas the autoreverse and music recorders used thinner tracks and more guardbands to separate side A and B more. 

If it is a stationary 4-track head, it will be perfectly aligned (90°. That means both sides are at 90°. If it is a fixed azimuth, it can vary quite a lot, some can be 85° to 92°. That means that it is rare to find one at 90°. If it is bi-azimuth, be sure to adjust them both to 90°.

I feel obliged to resolve the misunderstanding about the alleged concept "bi-azimuth". There is no such thing. Let me explain; stationary heads almost always have two similar adjustment screws. These are used to calibrate height AND azimuth. By adjusting only one of the screws, you cannot adjust either height or azimuth alone, the other parameter will also be affected. To do an azimuth adjustment without affecting height, you have to turn both screws the same amount, but in opposite directions. To do a height adjustment without affecting azimuth, both screws must be turned the same amount, in the same direction.

Fixed 4-channel heads with tape guides on both sides are not immune to azimuth problems. If such a head is not exactly aligned, the tape edge may curl up against one side of the tape guide - the result being both azimuth error and damage to the tape.

Say you have a hypothetical cassette deck, with perfect dimensions and geometry of the cassette well and the cassette guides. The capstans are perfectly straight, perfectly perpendicular to the tape path, and perfectly parallel to the tape plane. The entire mechanism is perfectly rigid, and completely unaffected by the forces exerted on it by the motors, the cassette, and the touchpoints to the tape.

Into this cassette deck you insert a perfect cassette - the geometry is perfect, and the rigidity and vibration damping is perfect. The tape path inside the cassette shell, including the guide rollers, are perfect.

The tape of this cassette will travel in a perfectly straight path. It will do so in both directions, if we assume our perfect cassette deck has auto-reverse. The perfect azimuth angle for this situation will be exactly 90.0000000 degrees. In both directions.

For any lesser, real-life cassette player, the optimal azimuth angle will be 90 degrees. In both directions. The imperfections of cassettes, the tape deck-cassette interface, and the cassette mechanisms, cannot be known in advance. But the statistical mean will be 0. It would be ludicrous to adjust, say, the azimuth angle to 89.9 degrees for the reverse direction, because you believe this will match more tapes.

The only situation in which this will be sensible - applying an azimuth offset between the two tape directions - is when you can predict flex/deformation/imperfection in the mechanism, so you know that the forces exerted on it will affect azimuth.

Which means the Aiwa "bi-azimuth" feature is meant to compensate for inferior rigidity in the cassette mechanism itself, and not to find a better azimuth match to cassettes than you could otherwise attain.

And there is of course nothing automatic about it. Two set screws adjust the "bend" that will be applied to the head in each direction. Introducing a flexible head platform into an already flexy mechanism does not sound like such a great idea to me. Design choices like these are among many reasons that authorities on the field - most notably Doctor Walkman - does not recommend any non-Sony walkmans for serious, critical listening.

The only known automatic azimuth correction systems are NAAC (Nakamichi) and MAAC (Marantz). Automatic azimuth adjustment requires two distinct features;

1. A head with 4 (for uni-directional) or 8 (for auto-reverse) channels. Each stereo channel is split into two, and the two signals are compared for phase difference. If there is a phase difference, the playback azimuth is wrong, by an amount/angle that can be deduced from the phase difference.
2. An actuator or motor to adjust the azimuth. This has to be separate from the tape transport motors, for obvious reasons.

I can actually think of a more sensible justification for this arrangement, namely to compensate for imperfections in the head.

A two-track head can always be oriented so that the gaps are aligned to a line that is exactly 90º to the theoretical, perfect tape path.

A 4-track auto-reverse head is trickier. It is difficult to produce four 1μm head gaps exactly in line, but you there is always an orientation of the head where 2 of the gaps, either gaps 1-2 or 3-4, respectively, are aligned to a line that is 90º to the tape path. So the two set screws can be used to factory-set the azimuth in both directions, allowing a cheaper and less accurate production process for the heads.

It still stands, though, that the flexible head bridge will in itself cause problems. I am pretty sure that by applying pressure - even the slightest pressure - to the walkman lid over the head bridge will result in audible azimuth changes.

Summing up; the "bi-azimuth" arrangement is probably supposed to (I don't know anything about it's effectiveness) fix problems caused by imperfect heads. It would probably be better to put that investment into manufacturing better heads in the first place, and mount them on as rigid a platform as possible.

What are you now talking about, a 4-track head is basically the same way as a 2-track head, they did not reduce the size of the tracks, except for multitrack recorders, where the tracks are thicker and guardbands are thinner. It is the same as producing a 2-track head, only you use 4.

Ehm, how do you exactly adjust the azymuth of a wm-504 if the slide doesn't hold?

The servicemanual has no mention of it.

After reading this thread it's helped me a lot. Had no idea a small adjustment makes a huge difference. In my case I have a WM 103 and a J501 that always had one side play worse than the other. Always thought it was simply wow and flutter and that the pinch roller was the cause. After some small adjustments both play perfectly now.

Damn!!! So much knowledge . Thanks guys.