Cartridge azimuth adjustment using a voltmeter and a test lp

"The generally accepted method is to use the amplifier output terminal to measure the voltage output of the test lp- but would it be a problem to measure the voltage at the speaker terminals?" 

I'm wondering where that idea came from.  It certainly has the potential of giving you some sort of answer, but I have never heard of that method (measuring at the amp output).  Also, keep in mind that you are trying to minimize channel to channel crosstalk, not channel balance.  If you measure at any point beyond the cartridge itself, you run the potential of being deceived by any channel imbalance downstream from the cartridge, as you seem to realize.  Best to do it at the phono cable.  Problem with that is the extremely low voltage of the signal at that point.  This is why we have the Fozgometer and similar devices to add amplification as close as possible to the source. Even really good voltmeters do not have the sensitivity to set azimuth direct from the cartridge, unfortunately. That's to set azimuth electrically. The opposite philosophy is to square up the cartridge with respect to the groove, which will give you a mechanical azimuth adjust.

This is probably one of those things you just have to wear yourself out trying, but on the off chance its not, I went through all this and decided its a colossal waste of time. Tried a couple different methods. Because people do make it seem like there's some wonderful sonic benefit to getting it right. If you find it, congratulations, and let me know.

lewm does a really good job explaining how hard it is to try and measure. I had the same problems. But here's the thing. The whole time I was trying to measure I was also listening. Because almost always measurements can only get you so close, and you have to do the rest by ear. Especially with turntables. Well with azimuth I never did hear anything, even when deliberately moving it what I thought would be quite a ways off kilter. 

Finally, you might want to click on over to Soundsmith and listen to Peter Ledermann talk about setting azimuth. Once I understood everything he talks about my experience made a lot more sense. Who knows, between the three of us might even save you a lot of time and trouble?

The issue is that the coils have the proper geometric relationship to the record. Cartridges now a days are well constructed the stylus being in the proper relationship to the coils. The parameter most affected is separation. Thus you can assume that if the stylus is perfectly perpendicular to the groove azimuth and separation are at their best.
The best and easiest way to do this is to align the cartridge, set VTF, place a pocket mirror on the platter and the stylus on it. View the stylus directly from the front with good lighting and adjust the azimuth until you see a perfectly symmetrical "hourglass" (the stylus and it's reflection in the mirror) The mirror greatly increases your accuracy in this adjustment. This is also the approach that will minimize record wear even if the coils do not have the exact correct relationship to the stylus. IMHO you will not do significantly better with meters or oscilloscopes. If your cartridge is poorly made you will align the coils, miss align the stylus and potentially wear out your records. All you need is a pocket mirror and a sharp eye.
It is also possible to do this by viewing the cantilever and it's reflection in the mirror. They have to be right in line with each other. 
Geoffkait, as luck would have it you don't have to turn up your hearing aids before you try this and please don't try to plug the mirror in.

Several months ago I found this method of adjusting azimuth by ear.
It works brilliantly.
http://durand-tonearms.com/Support/Support/azimuth.html

The best way to set the azimuth is by ear (see the note at bottom of this page). Note that if you hear a defective azimuth adjustment as an obvious problem of balance--if the sound is predominantly to the left or the right--, that could be because the azimuth is off (er, check the balance on your preamplifier first, of course!). But there could be many other causes, such as problem with room acoustics, deficient electronic component in one channel, etc. It's best to first ensure that there's nothing unbalanced in the rest of the system before attributing the problem to azimuth. It is however true that a center image can sound slightly off center if the azimuth if not perfectly adjusted.

Here are some tips to help you with this process:

Use a mono recording with relatively sparse material (for example, of a female vocalist). As you adjust the azimuth, you will perceive that the voice has the best “presence” when the adjustment is right on; that indicates that the stylus is perfectly vertical and centered in the groove; if it is even slightly off, the voice will sound somewhat recessed and unfocused. There is a very fine line between being right on, or not quite there yet; usually, a fraction of a degree is all it takes. So you really have to be very subtle when adjusting this parameter. Start with the cartridge vertical. This is just a starting point; remember that with most cartridges, the stylus is not exactly vertical, so just looking at the cartridge won't help much. A perfectly vertical stylus is what we are trying to achieve, and that's not something you can see with the naked eye.

Before you start changing the azimuth, listen to a short excerpt (2-3 minutes) several times. Identify as many elements as you can: the different vowels the singer is singing, the consonants, the mouth noises; then listen to the instruments: try to pay attention to each one individually. Is there perhaps one that seems to be more separated from the others? Don't listen to instruments or voices in the low register, they won't tell you much in this process. Percussion instruments in the middle or high register can be very useful: listen to the attack, then the decay.

Once you think that you've identified one instrument/voice that you "know" well on that recording, modify the azimuth, in one direction or the other. Let's say, you start by going toward the left (armwand rotated counterclockwise); rotate the arm by a very small amount--when you get close to the right spot, the rotation can be a fraction of degree. Listen to what happened. Then move again, in the same direction. Listen again. Is the sound getting more present, or not? Is the instrument getting more 3-D like, perhaps moving to the front a little, perhaps gaining more separation from the others? Or perhaps nothing changed, or it got worse (more recessed, less focused). Modify again, in the same direction. If nothing is improving, you might be going in the wrong direction. Come back to vertical, (on the Talea™ you can follow your changes on the little scale in the azimuth window) and do the same procedure in the other direction (clockwise). At some point, the sound will seem to change for the better. If it isn't, remember that some cartridges are less sensitive than others to azimuth changes. Or perhaps, you're exhausted by now and can't focus anymore. Don't worry, leave it aside for a while, and just enjoy listening to music.

Note that it's not impossible that the best sound is with the cartridge perfectly vertical! Sometimes the stylus is perfectly aligned.
Some possible clues that the azimuth setting is good:
- Source is larger
- In some systems source is closer (with more depth of stage)
- The sound you are focusing on is more separated from the others than before (more 3-D like)
- Sound source seems louder (consequence of previous points)
- More difficult: trueness of timbre. Listen to instrumental timbres (attacks, decay), as well as particularity of vocal timbre (sibillance and other subtle noises at beginning or ending of consonants, etc)

As with any fine adjustment, it takes time and patience to get it right. It's the same thing with tracking force and VTA/SRA adjustments: you can do it by eye, or follow the manufacturer's recommended setting, and hope for the best. Or you can experiment, try other settings just to see what happens, and suddenly discover uncharted territories. Through practice, your ears will get better at hearing the fine differences, and this in turn will take you to new heights of musical enjoyment.
Nobody said playing vinyl was instant gratification...

Just as a performer needs to understand how to take advantage of his fine instrument, anyone who has engaged for a significant period of time with the setup of a sophisticated tonearm knows that it can take a very long time to understand how it reacts to minute variations and how to anticipate its reactions and play with them. The more you play with it, the more you realize what it can do and how to make it sound its best, and the more it gives you back. And when you get it right, you can sit back and enjoy the new level you've just reached in your analog experience!
______________

Note: There have been other methods suggested in the past. A particularly popular one was offered by Victor Khomenko in the late 90s and seems to have been widely accepted since. In our experiments, we've found that, while this method works very well at a given frequency (most people seem to use a 1 KHz tone for this purpose), it is unfortunately not consistent throughout the frequency range found in music. We discovered that after adjusting the azimuth for perfect balance at 1 KHz with this method, voltage readings with a 100 Hz and a 8 KHz tone give widely different results; one channel would be greatly emphasized at the lower frequency, and the other one at the higher frequency. So while the method is theoretically sound, it fails to address the reality of the musical signal, which is far more complex than a single sine tone. So, until a better and more reliable method is established, our ears will do nicely... and they're free...
Additional note: to be fair, it has been suggested that crosstalk is not constant on all cartridges. Some cartridges (a few?) demonstrate excellent consistency across their frequency range, while many don't. For what it's worth...