V15V-SAS measured

[Ldg]

I was reading some vintage papers on stylus tracking and distortion - and I noted an interesting comment (perhaps obvious but still...) - when the needle resonance or cantilever resonance is reached - tracking ability is reduced by the resonance (yes I know - another "doh" - interesting to note the mention of the actual needle resonance... I think that was in one of the patent docs... shibata?, and the article might have been by Pramanik).

When vintage papers discuss needle resonance, it is almost certain to describe an assumed spring-mass system comprising vinyl compression and effective tip mass. This was widely accepted (still is, but not by me) and said to account for the resonant system you and I describe as 'cantilever resonance', DL, which has a completely different causal mechanism. One has to decide in which camp one stands as to causality here, because pragmatically there is only one resonant system.

Personally, I'm in the cantilever flex camp, though it is lonely over here. To me, this fits far better, and is not contradicted. Whereas vinyl spring/mass theory can readily be contradicted, and indentation can be shown to be very small imperically (indirectly through drag/friction).

Either way, there is significant non-linear generator motion at/near resonance. Also, significant force on the stylus which adversely affects trackability near resonance.

[dlaloum]

Here are some more measurements of the V15V-SAS - this time mounted in a standard V15V body...

I have set all parameters as close as possible to original V15V OEM specifications - the intention is to compare these to measurements taken from an original V15VMR...

All measurements have been done at 255pf and 46k
Test record for spot measurements is CBS STR130 (THD measurements were done the old fashioned manual way)

Test record for LF sweeps is Denon XG7002 - the sweeps shown here have been adjusted using an EQ curve set manually based on the table provided with the record... I have made a seperate posting regarding the best way to use this record...

I felt that the SAS damper brush was possibly less effective than the original and wanted to measure and compare the SAS damper brush to the Shure damper brush (soon as I get around to measuring the Shure one as well...)
The TT this is being measured on is a JVC with a Servo Damping arm - so the measurements were taken at three differing styli pressures (1g, 1.25g, 1.5g) in all the permutations of no damping, brush damping, and arm damping... Hence the confusion of measurements.

Straight spot frequency response measurements from CBS STR130

Not a lot of difference between the various VTF's and damping methods...

THD measured using the same CBS STR130 recordings

Interesting that increased VTF provides reduced THD - but only in the 5k to 8k range everywhere else it is much the same...
Perhaps keeping the cantilever under a bit of added tension reduces skew/twist related distortion?
As I get through some more measurements using the same settings, I will also look at an aluminium cantilevered Jico eliptical for comparison (different cartridge body unfortunately... but I have not seen this variation in THD in that frequency range noted in relation to VTF before...)

These last two plots are from the same data, the first is a close up of the second, showing the arm/cartridge resonance area, and the specific impact of the two damping methods and the no-damping configurations.
In all cases except one I have used the default Q-Damping settings recomended by JVC (which is for Q dial to be set equal to VTF) - the one exception was the 1g VTF reading where I found the Q=1 setting to be underdamped, so I did a secondary reading at Q=1.5 (as shown).





Interesting that the main resonance appears to be at around 5Hz (4 to 6Hz) - this appears to indicate that the SAS VN5MR stylus has a much higher compliance than quoted by Jico. (Jico quoted me 13cu @ 100Hz)

Remembering that this is being recorded on a JVC QL-Y5F, the V15V is mounted on a standard Technics headshell (7.3g) - which (using lds EMC spreadsheet) results in an effective arm mass of 14.5g and a total effective mass of around 20g - indicating a compliance of over 35cu (!).

Looking at the impact of VTF and damping on the resonance...

1) increased VTF appears to drop the resonance by around 1db
2) the SAS damper brush drops the resonance by 9db at 1g VTF, and around 8db at 1.5g - so it is actually quite effective!
3) Servo arm "Q Damping" is slightly more effective at controlling the main resonance (4.5Hz) than the brush, but noticeably more effective at controlling the Secondary resonance zone at 5Hz to 7Hz. Unsurprisingly the two together do the best job.
Would the brush + Q equate to a higher setting of Q alone - I don't know... But my gut feeling is that they behave slightly differently - and may be complementary.
Which also raises the question of whether the brush + Q damping is overdamping the system...
4) Damping did not have a measurable impact on THD. (I may have to try and measure IMD to see whether there is an impact there...)


Bye for now

David

[dlaloum]

From the same set of measurements....

I just added a column to read out separation then graphed it (isn't excel handy!)



So here is something interesting - the lighter the TOTAL VTF (the setting of the arm - as opposed to the effective VTF at the needle, which is where you deduct the contribution of the brush)...

So Seperation at 1g VTF is the same regardless of damping, seperation at 1.25g is a step down regardless of whether that is brush down (1g at the needle) or brush up. - I do wonder whether the brush is generating vibrations that are passing through to the needle via the stylus holder?. (an area of potential experimentation with immobilising the stylus carrier on the body using plasticine or BlueTack)

Highest separation is with the lowest VTF, lowest seperation is with the highest VTF

An interesting tradeoff, as highest harmonic distortion is with lowest VTF...

People noting changes in sound with brush down and/or with increased VTF may actually be noticing the change in the separation rather than the THD.

There is a sharp drop at around 260Hz which I assume is an arm resonance - by hearsay, I have read many reports of arm/headshell resonance being around the 200 to 300Hz range - which would correlate closely with this.

I have not worked out what is happening at 50Hz - but this might be the impact on measurements of normal power supply effects (Australia is a 50Hz country)

The other smaller drops may bear some examination...

I will also do a HF frequency response plot using the same Denon record shortly (1kHz to 48kHz) - and I will plot seperation for that sweep as well - I expect that plot will clearly show the cantilever resonance(s) as drop(s) in seperation.

Bye for now

David

[Ldg]

Noble effort, DL !

A few guesses at why harmonic distortion and channel seperation might vary with VTF :

1. The obvious one : VTA varies as suspension sits at different heights. Not only mechanical tracing distortion, but generator neutral position might shift. Need to reset VTA after varying VTF, to remove this unwanted variable.

2. Stylus sits in the groove at a slightly different 'rotation', if the stylus is not perfectly vertically mounted and/or geometrically perfect. Perhaps !

3. Pre-stress, or pre-flex, in the cantilever varies. Perhaps !

People noting changes in sound with brush down and/or with increased VTF may actually be noticing the change in the separation rather than the THD.

Or, more likely, something completely different. Such as lf stability induced effects such as modulation distortion, etc etc etc.

[dlaloum]

I just reviewed my measurements of frequency response for the V15V-SAS, at factory loading...
There is a high frequency resonance at 23.3k and another at 47.4k - LD, does the second one make sense as an eigen frequency?

I have noticed that my VN5VSAS is not tracking as well as my VN5xSAS - but I have had the latter over a year and it has been run a lot - so the suspension seems to have loosened up a bit, where the VN5VSAS is probably still getting there.

bye for now

David

[dlaloum]

Typical impules of V15VSAS - used it to measure the values required for Loafer.XLS


Other impulses look well nigh identical - including manual needle drop...

[Ldg]

There is a high frequency resonance at 23.3k and another at 47.4k - LD, does the second one make sense as an eigen frequency?

Well, it's not near the ratio between primary and 2nd. But it is close to the ratio of 2nd and 3rd eigenfrequencies, for one end retained and one end free to rotate, it's about -1.5% off. But, then again, it's also close to 2....about +2.5% off. What tolerance would you estimate on the measurements, DL ? Any chance of posting the plot before any manipulation, ie just the raw response, I couldn't see it on this thread ?

Each eigenfrequency corresponds to a mode of vibration. To make it fit, one would have to say the lowest mode doesn't happen, or doesn't show up. It would be around 7kHz if the 2nd and 3rd have those frequencies.

Before getting too excited, my guess is the tolerance isn't going to allow us to seperate the ratio accurately enough to tell it apart from exactly 2. Curses, Red Baron. Exactly 2 would suggest a harmonic ratio, could be many sources in the path, and one would have to examine the whole measurement path before declaring it real. If it dropped on a known eigenfrequency ratio, that would be more interesting !

I have noticed that my VN5VSAS is not tracking as well as my VN5xSAS - but I have had the latter over a year and it has been run a lot - so the suspension seems to have loosened up a bit,

You might measure impedance@100Hz with your impedance test disc, compare and see ?

[Ldg]

Typical impules of V15VSAS - used it to measure the values required for Loafer.XLS


Other impulses look well nigh identical - including manual needle drop...

OK, its been puzzling me all day, but I think I've cracked what this is. It is a high frequency transient, the extremely fast leading edge of a transient displacement. Each dot is a sample at 96kHz, and the timecode axis is min:sec, and each major division on the x axis is 50uS. Is that right, DL ?

Although this is completely useless for loafer.xls, which looks at low frequency mechanics, I think you've perhaps accidentally stumbled on something quite fascinating, DL. From what you say, it's repeatable for needle touchdowns too....I wonder if it is actually the lesser spotted top mechanical resonance, be it due to cantilever flex or vinyl indentation (as one prefers)??? The frequency works out to about 14.3kHz, which is not implausible........and if not this, what on earth is it ?

How interesting !

[dlaloum]

Yes it is consistent - and yes the timecode is min:sec and each dot is a 96kHz sample.

Ahhh I was not aware that the loafer requirement was for a low F pulse... So I simply went to a recorded track, searched for a click and zoomed in on it.
Sounds like I should search for a "pop" rather than a click....

And also this was consistent with a manual touchdown - the surface touched down on was my VTF scale (stainless steel) - perhaps it would change is I touched down on a softer surface? (plastic?)

[Dismord]

A fascinating discussion but it again reminds me of the frustration I feel at manufactures of phono preamps not building in sufficient flexibility on their moving magnet inputs. For decades there's been an industry snobbery that appears to be telling us only moving coils deserve to see a wide range of input options. GRRRR!

[dlaloum]

No it is much more snobby that that - the attitude has been mostly that high inductance/high output cartridges are irrelevant... and that no one but an entry level stereo buyer (and certainly NOT an audiophile) would use an MM/MI

[Dismord]

No it is much more snobby that that - the attitude has been mostly that high inductance/high output cartridges are irrelevant... and that no one but an entry level stereo buyer (and certainly NOT an audiophile) would use an MM/MI

Happily living with a Garrott 'Optim' FGS with ruby cantilever I have some argument with their groundless prejudice.

[Thomas_A]

Regarding fundamental resonance of the Shure V15 with JICO SAS I measured compliance it to be around 35 horizontal and 21 vertical.

viewtopic.php?p=176777

Thomas

[dlaloum]

Hi Thomas, I remember that - which is why I was so surprised when mine measured out to 54cu.... so I remeasured it and got the same result. (This being a 6months old VN5VSAS stylus)

I do need to go back and revisit the N97xSAS stylus and see whether the results are the same - I have a feeling that Jico have been playing with the suspension and damping on the SAS, and that either the models differ, or they have varied it over time... or final option - there is substantial production variation between individual styli. (that would be the not so good option the last one!)