|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
74.130.29.193
In Reply to: RE: Feickert Azimuth Adjust software - can I use a software Spectrum Analyzer instead? posted by tonyptony on February 06, 2016 at 07:56:25
Setting azimuth requires individual left and right channel signals. You need to adjust azimuth for minimum crosstalk, which indicates maximum channel separation. Yes, you can use a spectrum analyzer. I used the spectrum analyzer in Sound Forge software to measure crosstalk and optimize azimuth. In fact, I looked at the crosstalk response across the entire audio bandwidth rather than at just one frequency. The following graphs show the progression of setting azimuth for my DL-S1 cartridge in my Technics SL-1200 Mk2 turntable. The first graph is with the cartridge level. The second graph shows the cartridge tilted about one-half degree counter clockwise when viewed from the front. The third graph shows azimuth optimized with the cartridge tilted about one-degree counterclockwise.
.
.
..
.
Follow Ups:
That is a lot of difference in the amount of crosstalk. Do you keep it at the one degree counterclockwise orientation for normal use then?
> Do you keep it at the one degree counterclockwise orientation for normal use then?
Yes! However, when I tested a Fozgometer I found that different test records yielded different azimuth results. Nevertheless, I leave the DL-S1 at one-degree azimuth tilt after taking three days to zero-in on that adjustment. ;-)
I have a different DL-S1 on my Sota Millennia Vacuum and I keep it level with the record playing surface, e.g., zero-degree azimuth.
Best regards,
John Elison
Makes sense. Now I need to save for my own Fozgometer.
When you get your Fozgometer, you'll have to decide which test record you believe is correct. After all, the test record you choose will determine the azimuth setting for your cartridge. That's why I now believe it is just as good if not better to set azimuth to zero-degrees and be done with it. I have never heard any advantage to setting azimuth at the point of minimal crosstalk. I get just as nice a soundstage from my DL-S1 that is set to zero-degree azimuth as the one I adjusted for minimal crosstalk. However, to each his own!
Good luck,
John Elison
True, though not valid logic at all, one might gain confidence in some result if a large number of trusted test discs indicated a strong trend....
Did you happen note any other musical parameters that changed as a result of your dialing in lowest crosstalk? like THD for example
If azimuth is very important (and it is ) tell me about uni-pivots ...
The center of mass on unipivots is lower than most fixed bearing tonearms. Sometimes the counterweight is heavier below the arm stud than above so you can adjust its position to set azimuth. Other unipivots have weights that hang out the sides so they can be adjusted to change azimuth.
Personally, I don't like unipivots so I always buy fixed bearing tonearms. My favorite tonearm manufacturer is SME and my favorite tonearm is the SME V.
Best regards,
John Elison
I didn't make any distortion measurements, but it certainly wouldn't surprise me if setting azimuth reduced distortion on the test record used for setting azimuth. My rationale for not bothering further with azimuth was due to the fact that virtually every different test record yielded different azimuth results. This convinced me that ordinary music records probably had slightly different azimuth requirements, too. Therefore, I decided that zero-degree azimuth was just as likely to be correct for a given record as any other setting, and since I couldn't hear the difference anyway, it was probably a waste of time and money to worry further about azimuth. In my opinion, this is very valid logic.
Best regards,
John Elison
Sorry, I should have said "less likely" not unlikely in relation to the coil inducatance mismatch and zero error in channel separation using the vertical test tone channel mismatch.....
Regards Anthony
"Beauty is Truth, Truth Beauty.." Keats
" I have never heard any advantage to setting azimuth at the point of minimal crosstalk."
This is what I have found as well (in terms of using the test discs). The potential mismatch in coil inductances will tend to make it unlikely to ever be able to achieve zero difference between the channels.
I personally prefer to use the physical tip alignment as the reference and accept the difference in channel separation measurements for L and R relative to the specification. My observations with a tip azimuth error are a slightly smeared soundstage and subtle loss of detail. On silent grooves, one can hear that the rumble isn't "centred". Of course this assumes that the antiskate is also optimally set!
As a slight diversion, I discovered a more reliable way of verifying the optimum antiskate level relative to the calibrated value since different tip profiles would have subtly different drag coefficients depending on polish etc. I know that Line Contact designs would have a higher required value (which is also modulation amplitude dependent), but they are less than elliptical.
I believe that this can be determined more reliably using high level vertical modulation tones (rather than simply using the high level lateral tones which tend to give a higher than necessary value) combined with the usual analysis of the distortion components. A minimum level of 50um is very revealing. I used this to subtly adjust the antiskate for my OC9ML/II over the weekend by +10% relative to the calibrated value and the distortion components for the vertical modulation became the same for both channels as expected and the soundstage became rock much more focussed and smooth. In the R channel there was a collection of components around 5 kHz which completely disappeared after optimising the antiskate.
Prior to this I noted that the LF resonance values were not quite the same for each channel. After the adjustment, they came back in line. Since the LF resonance modulates higher frequency components, any mismatch would straight away give differences in the channel characteristics which would translate to a subtle difference in the perceived soundstage. In the same way VTA mismatch also contributes to additional distortion around 4 to 5kHz.
For me, the lateral "equal buzzing" is a little less precise when doing it by ear as I found when using fluid damping that any slight eccentricity of the disc would alter the buzzing amplitude in a particular channel. Sometimes on a marginal test, you would get a clean tone and then a slight buzz as the centre amplitude error increasd to maximum.
Anyway, just thought this might be of interest to you given you would have the suitable test tones at your disposal on your test discs.
Regards Anthony
"Beauty is Truth, Truth Beauty.." Keats
"The potential mismatch in coil inductances will tend to make it unlikely to ever be able to achieve zero difference between the channels."
True I'm assuming there is not much that can be done by the end user about things like channel to channel inductance tolerance, or incorrect orientation of the coils with respect to each other without rebuilding the cartridge. Though I am potentially putting this into another class, ie problems which can be addressed by setup changes intended to be adjusted by the end user. The difference John Ellison saw seems really significant to me. I don't know that I'd worry about it for normal listening but I might for archival work. What are the theoretical downsides to this approach assuming one can trust their test disc signal that is, ie minimizing crosstalk via azimuth adjustment?
> What are the theoretical downsides to this approach assuming one can trust their test disc signal that is, ie minimizing crosstalk via azimuth adjustment?
The downside is the very high probability that your test record has absolutely no relationship to any of your other records with respect to azimuth. I concluded this from the fact that of eleven different test records I own, only two exhibited the same azimuth requirements.
Best regards,
John Elison
2 records out of 16 is definitely not a strong trend. Thanks for your comments.
Realistically 2 to 3 degrees is quite possible with a given cartridge. If we take MC cartridges, then the Feikert method will help optimise electrical azimuth i.e coil relationship to groove wall. However, depending on quality control the stylus tip may not be mounted perpendicular to the cantilever. So if the cantilever is correctly mounted with respect to the coil orientation, but the tip is not, you will end up with what I personally find to be a slightly smeared presentation with a definite loss of focus and susceptibility to sibilance.
The more extreme the stylus cut (i.e large bearing radius), the more critical tip azimuth needs to be since the stylus physically won't sit in the groove correctly past a certain angle of rotation. Elliptical and spherical are more tolerant in the sense you can rotate them further without issues with touching the correct portions of the wall. However, the pear shape contact area basically means you have different effective tracing radii on each wall and my experience is that you get really bad sibilance.
Consequently, I prefer to sacrifice the electrical azimuth specification and ensure the tip is perpendicular to the groove. In extreme situations, the channel imbalance can be very large, but you have less tracing distortion (as with elliptical) or you contact the grooves in the ideal segment thus avoiding the portions close to the shoulder which may be damaged.
For MM, I have some *superb* examples of Stanton/Pickering styli where I had cantilever rotations of 8 degrees and in one case 10 degrees. That's what happens when you buy NOS I guess....Anyway, the spec of the cartridge is typically 35dB for the 681, 881 etc. Well when the tip is correctly sitting in the groove, a 10 degree rotation gives you a very out of spec channel imbalance (can't remember specifics but something like 4 dB) and channel separation of 20dB or less. I can accept this better than sibilance! FWIW, those styli just sit unused.....they were Stereohedron styli too!
Regards Anthony
"Beauty is Truth, Truth Beauty.." Keats
Going to need to digest this for a bit I think. Thank you for your words. Very much appreciated.
This image may help in the understanding. The image is of a Soundsmith OLC seen from the front. It is not quite as sharp as I would like, but there appears the camera focus plane on the Zeiss microscope I was using is not the same as the optical focus plane as I had it pin sharp in the eyepiece!It appears to be similar to if not actually a Fritz Gyger tip. You will notice how large the internal angle is between the scanning surfaces - it is very wide in order to extend the length of the scanning surface. An elliptical or spherical is only about 55 degrees IIRC.
If you have a tip azimuth error past a certain level, the stylus simply won't fit in the groove properly.
Regards Anthony
"Beauty is Truth, Truth Beauty.." Keats
Ah. Rereading when I had more time also made a difference. Thanks for explaining. It does make more sense after I've thought about it now. Best regards.
Did you use R and L channel pink noise tracks for this or did you use some sort of sweep? Can you tell me what analyzer settings you used?
I was going to use the 1KHz right and left tracks on the UA disc with a fairly narrow RBW to keep the extraneous junk from noisying up the signal, but I like the full frequency approach you took. Any further details on how you did this would be appreciated.
BTW, I corrected my first post where I erroneously said I was going to use mono tracks for this, which I knew was incorrect but didn't catch.
I used the Denon test record stipulated in the graph's title, which contained individual sinusoidal spot frequencies. I don't remember what settings I used on the spectrum analyzer. The spot frequency measurements were manually loaded into a Microsoft Excel spreadsheet for graphing.The two graphs below are identical. The second graph simply connects the dots with smooth lines.
.
.
Edits: 02/06/16
John, Are you using Sound Forge for the Mac? If so, what version gives you the audio spectrum analyzer function? Thanks.
It was version 7 for PC. I don't use a Mac.
They make several products for the Mac, but on their website it's difficult for me (as one who is not familiar with the jargon) to determine which ones will do what I want to do, which involves having audio spectrum analysis function on a Mac laptop. Perhaps I will contact them and ask which Mac-specific problems mimic version 7 for PC.
Looked around some and can't seem to locate one.
eBay has a feature you can enable which notifies you with an email when rare items come up for sale. I was eventually able to piece together complete collections of both the Denon and CBS test disc suites. Though I'm sure I paid a bit more than retail on some.
I could kick myself for not buy Denon's Wow & Flutter test record, XG-7004, when I had the opportunity. I bought only three, XG-7002, XG-7003, and XG-7005.
Is there a way to set azimuth with volt meter and test record?
If you intend to use a broadband AC voltmeter to measure signal level and crosstalk, you will also need a narrow bandpass filter centered on the test frequency in order to isolate the crosstalk signal from vinyl surface noise. Otherwise, you will not get an accurate measurement of the crosstalk signal because it will be obscured by the noise floor.Best regards,
John Elison
Edits: 03/04/16
Nothing is easy.Thanks John!
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: