• Welcome to ASR. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Fun with vinyl measurements

Somehow, I keep missing TRS-1007's :facepalm:

One just went on Discogs last week for $3.43...sigh
? How do these appear? Having in on the want list does not seem to work.
 
Anyone want to help vet this? https://docs.myhi.fi/tools/tonearm-lf-mechanics.html

It's built off the physics of LD's "loafer" spreadsheet. I added the contact force/wear multiplier plot.

Technical (AI) explanation:

The system is modeled as a single degree of freedom damped harmonic oscillator with base excitation. The mass is the combined tonearm effective mass and cartridge/fixings mass. The spring is the cartridge cantilever compliance. Damping comes from two sources: the cartridge suspension (derived from the ratio of static to dynamic compliance at a known frequency) and the tonearm (bearings, fluid dampers, or other mechanical damping).

The undamped natural frequency follows from Hooke's law: fₙ = √(k/m) / 2π, where k = 1000/compliance and m is total effective mass. The damped resonant peak frequency depends on the system's overall damping and may differ slightly from the undamped natural frequency.

The frequency response uses the cartridge output transfer function, which represents the relative motion between stylus and arm body — this is what generates the cartridge's electrical signal. It has a high-pass characteristic: zero output at DC (the arm tracks the groove perfectly), a resonant peak where relative motion is amplified, and unity gain at high frequencies where the arm can no longer follow the groove modulation.

The contact force plot uses the base-excited displacement transmissibility: T(f) = √((1 + (2ζr)²) / ((1-r²)² + (2ζr)²)), where r is the frequency ratio and ζ is the overall damping ratio. This approximates how the mechanical resonance modulates the contact force between the stylus and the groove wall. At resonance, the cantilever deflection is amplified, swinging the contact force above and below VTF. Above resonance, the arm's inertia prevents it from responding, and the contact force returns to VTF. The contact force envelope shows VTF ± (k × excitation × T) at each frequency. This is an approximation — the transmissibility captures the qualitative behavior correctly but does not precisely model the cantilever force at all frequencies.

The wear multiplier applies Archard's wear law with a Hertzian contact exponent of 1.5. Because wear rate is proportional to force^1.5, the cycle-averaged wear from an oscillating contact force is always greater than the wear from steady force at the same mean value. The multiplier is computed by numerically integrating force^1.5 over one oscillation cycle and normalizing to the steady-state wear at VTF.

Cartridge damping ratio (Tc) is derived from the difference between static and dynamic compliance: Tc = (k × √((Z/k)² - 1)) / (2π × f_dynamic) / c_critical, where Z = 1000/dynamic_compliance and c_critical = 2mωₙ. This captures the viscoelastic loss in the cantilever suspension. Tonearm damping (Ta) is either entered directly or computed from the logarithmic decrement method using two successive peak amplitudes of a decaying oscillation.

The trackability analysis runs a time-domain simulation of forced vibration at a specified frequency and amplitude. The simulation uses only the cartridge damping in the equation of motion, consistent with the original Luckydog spreadsheet model. This is valid for the typical trackability test frequencies (100 Hz–1 kHz) where the modulation frequency is far above the arm-cartridge resonance and arm damping has negligible effect. The minimum VTF required is the peak dynamic force perturbation during the simulation, representing the tracking force needed to prevent stylus liftoff at that modulation.
 
I will have difficulties confirming with my setup. Compliance at 10 Hz is around 22 for the JICO SAS, and the Moerch DP-8 arm is not easy to predict with its asymmetrical mass and lateral damping around the pivot. Have only experimental data where vertical resonance is just below 9 Hz.

 
When you're not skiing ( as you are supposed to do in Norway at Easter), you can play with the PhyPhox (Iphone) app's gyroscope and turntable.
Let the phone spin on the turntable, the phone takes and data as csv then exported to Excel for presentation

phyphox – Physical Phone Experiments

The data from Phyphox (Z-axis) comes in radians per second. That value is converted to rpm as Z *60/2pi. Then you can calculate it% deviation from average or deviation from 33.33

Below you can see the startup from my belt-driven player, as you can see it takes that many seconds before it swings into the average value in rpm.

From the oscillations it should maybe be possible to calculate platter intertia(effective mass)/belt compliance? Or maybe it the the suspension with 3hz resonance that comes into play..?

1775219565950.png


DD vs beltdrive guess whow is who... 5 units moving average to filter out some noise, sampling rate 100Hz, so filter is 50ms
1775224067966.png
 
Last edited:
When you're not skiing ( as you are supposed to td in Norway at Easter), you can play with the PhyPhox (Iphone) app's gyroscope and turntable.
Let the phone spin on the turntable, the phone takes and data as msn then exported to Excel for presentation

phyphox – Physical Phone Experiments

The data from Phyphox (Z-axis) comes in radians per second. That value is converted to rpm as Z *60/2pi. Then you can calculate it% deviation from average or deviation from 33.33

Below you can see the startup from my belt-driven player, as you can see it takes that many seconds before it swings into the average value in rpm.

From the oscillations it should maybe be possible to calculate platter intertia(effective mass)/belt compliance? Or maybe it the the suspension with 3hz resonance that comes inro play..?

View attachment 522073
Just downloaded it and made one spin.
IMG_1138.png
 
use Multi: and Z..or Absolute),,, .X and Y are sideways movement, your 33 rpm rotation is about 3.49 rad/s
1775224497771.png




Belt and DD
1775224755179.png


Complete data set
1775225047605.png
 
Last edited:
I will have difficulties confirming with my setup. Compliance at 10 Hz is around 22 for the JICO SAS, and the Moerch DP-8 arm is not easy to predict with its asymmetrical mass and lateral damping around the pivot. Have only experimental data where vertical resonance is just below 9 Hz.

Doesn't need to be with real values for a sense check. The default values are my AT150MLX on EPA-501H with a guess at the arm damping figure - the CW has both magnetic and oil damping. Later I'll play the with log decrement analysis to come up with the damping figure.
 
Got this with Z
IMG_1143.png
 
Yes, and then you can export data to excel to calculate rpm, % deviation , and filter ( moving average is the simplest) etc

Maybe @JP, could )make the WF app accept Phyphox Z ( or absolute) data) from the gyroscope capture . It is CSV format in rad/sec. Then much more people can make W&F . .. a W&F library..

PS I am still figuring out the difference between Z and Absolute…googled it but the concept is slow to sink in..
EDIT…
Since file format version 1.4 (phyphox 1.0.6) there is another output abs which gives the absolute (sqrt(x*x+y*y*z*z)) for 3D sensor data.

Still not sure if z or absolute should be used..
IMG_0339.png


In phyphox, the
Gyroscope (rotation rate) experiment provides both the individual z-axis (along with x and y) and the absolute(magnitude) gyroscope data, depending on which view you are looking at.
  • Gyroscope x, y, z:Provides raw data for angular velocity (in rad/s) around the three axes.
    • z-axis: In phyphox, the z-axis is perpendicular to the screen, pointing out of it. A "turn" around the z-axis happens when you rotate the phone like a steering wheel or a merry-go-round while keeping it flat.
  • Absolute Gyroscope (Gyro abs): Gives the magnitude of the rotation rate vector (
    1775280706068.gif

    1775280706074.gif

    ), representing the total speed of rotation regardless of which axis it turns around.
    Phyphox
    Phyphox +4
Key Considerations:
  • Raw Data: The gyroscope sensor directly measures how fast the phone is rotating.
  • Turns: A "Turns" tool in phyphox allows for measuring the number of revolutions, with specific options for calculating the total number of rotations based on the z-axis or the absolute rotation.
  • Orientation: The z-axis is perpendicular to the phone screen.
    Phyphox
    Phyphox +2




phyphox
Did you notice the change in X and Y when rotation starts? It is a different pull from the belt when it runs ,, that shifts the platter position slightly ( at least in my suspended belt drive ). A visual circular bubble level confirms it..
 
Last edited:
Pasted in 8 sec of phypox data( converted rad/sec to rpm too) in a Shaknspin -txt file and ran the W&F analyser from JP... got this, with Denon 51F
1775290304570.png



A longer file works fine too.
1775291262130.png


for some reason I only get the histogram, not the polar plot..
1775291583563.png
 
Last edited:
What did you do different in the 2 plots, something made a difference..
One was x the other z.

( This is with the new Linn belt, having a bit more of the 1 Hz pollution. I am running this one in a bit to see what happens. Think this is what is seen in the wiggling.)
 
Both show z, one with start up , the other without 2693,2694
 
Both show z, one with start up , the other without 2693,2694
Post #2689 was my initial fast test, shows only "Gyroscope x". Second one I chose Z. Shows the same variation as below (new Linn Belt, 1 kHz pollution.

index.php
 
Back
Top Bottom