vinyl coefficient of friction

[luckythedog]

Whether this would be sufficient to cause a pop or click though........

Here's my 2p worth on the physics of what has to be achieved to create a click or pop via friction, ie for the stylus to lose contact with at least one groove wall.

There are two reasons an upward stylus force is associated with friction:

1) Tracking offset angle (c 22 deg) means friction force is not applied in line with the groove, rather it is applied along the line of the tonearm which pulls into the inner groove wall (at c 22 deg), The groove wall is inclined at 45 deg, so this pull has an upward reaction equal to F*sin(22)*sin(45) where F is the friction force.

2) The cantilever has a VTA of c 20 deg, and the reaction force to friction therefore has an upward component equivalent to F*sin(20), again where F is friction force.

Because of 1) there is a tendancy for the stylus to ride up the inner wall first, which is why crackle-pop noise can often heard first hard panned to the channel associated with that motion.

Both factors combine though, and the threshold for losing contact with the outer wall, (the easiest to happen) is at a critical value for k (coefficient of friction) given by:

k_crit= 1 / [ ((Sin(tracking offset angle))*(Sin(45))) + Sin(VTA)]

For tracking offset angle = 22 deg, VTA = 20 deg

k_crit ~ 1.64

Note k_crit is independent of both VTF and the 'average' value of k, the coefficient of friction. It depends just on the angles involved.

k is typically 0.4 mean, so k_crit ~ ~1.64 represents a deviation of about +12dB from k=0.4 in that case. That is to say, for k = 0.4 if the random peak fluctuation in friction exceeds +12dB, the stylus will momentarily mistrack. But the lower the value of k, the higher the headroom before this happens, k_crit does not depend on k. So lower mean coefficient of friction k typically reduces crackle-pop noise.

To produce a click/pop in a typical case then the stylus has to momentarily encounter a high friction patch of vinyl, and stick such that the coefficient of friction increases by a factor of about 4 or so. For the inner wall, that is. The factor is higher for the other wall, because of the geometry involved.

And this tidies up more loose ends than one could ever hope for in terms of odd crackle-pop surface noise behaviour IMO !

 

[luckythedog]

PS : where k_crit > 1, stiction is implied. So in the case above where k_crit ~1.64, stick-slip friction would necessarily need to be involved to produce a click or pop. And if the friction coefficient when stuck exceeds 1.64, the stylus momentarily will lose contact with the outer groove wall and produce a click/pop.

 

[luckythedog]

@Luckythedog: Thank you for asking. Last antibiotics at 4AM tomorrow then I bring them both home in the morning. Looking forward to it as are brood 1. I do have some ebay deccas that I will dig out for a spin. On your analog rig do you have a collection of stylii for mono or have you picked one that seems to be sweet spot for most?

Glad to hear that, Bill. Sure that will be a big relief.

My mono setup has a 25g arm, and these days I mostly run a Serato S-120 (don't knock it), on a standard OM body series wired for mono (don't knock it). For special occasions I have a DL-102, which I conserve. The S-120 tracks brilliantly at 4g (don't knock it), it's just one of those well kept secrets IMO. For special occasions I have a DL-102, which I conserve.

 

[billshurv]

I don't knock anything I haven't tried! heck I only got a mono switch 3 months ago which was a revelation. I have a few A&R x77 bodies lying around that were going to be pressed into service. A true mono cart is out of budget for a while and choice is limited and nothing is cheap (but a lot is expensive, and of course you have the delicious irony of a mono SPU being available from ortofon!).

All good fun!

 

[Hiten]

Will longer light tonearm have less friction ? It will have less bias, so logically it would have less friction comparatively.
---
I had 'Ace of Diamond' vinyl of Sir Arthur Bliss, which had Elgar's 'Pomp and Circumstance' and 'Welcome the Queen' But at that time I didn't listened to it for quality recording (Late 80s) just enjoyed the composition. It did played loud if we talk about dynamic range. I had Pioneer three in one. Receiver, Cassette Payer and vinyl player (PL12). 12 watts RMS and it did shake the glass of my showcase. Nice memories
Regards.

 

[luckythedog]

Will longer light tonearm have less friction ? It will have less bias, so logically it would have less friction comparatively.

Longer tonearms have lower tracking offset angle, so the value for k_crit is higher with associated noise threshold benefits in principle. But the benefits are small when one looks at the numbers for practical length arms, and the VTA induced component always remains.

Moreover, because bias force derives from friction, random variation in friction means random variation in bias force. This is a significant stimulus to cart/arm resonant headshell wobble, see post #43 inc plot.

@billshurv, it was tongue in cheek when I posted 'don't knock it'. Being a DJ stylus, it has an extra barrier to cross to get a fair hearing, but I found it to be an affordable, robust mono spherical which punches above its weight in performance and has a bona fide presentation of mono sound IME. Series wire an OM cart body to produce true mono, and adjust loading correspondingly. Will track up to 10g VTF, if that isn't heresy..............just like they used to back in the day. I run it at 4g, FWIW.

 

[billshurv]

series wire? I thought you paralleled for mono? Ah google is indeed my friend.

There are more OM carts made than probably anything else, and ortofon know DJs are a demanding lot, so no reason why it shouldn't be a lot better than its price suggests!

 

[Calvin]

Hi,

wiring the cartridge coils in series?
Wow, never thought of that.
Why should that be so, as Bandwidth drops seriously due to the doubled inductance.
The electrical amplitude response will probabely fall below even 10kHz.

I also think that for example Ortofon wires their Mono-cartridges in parallel as the 600Ohm/300mH specified for their 2M Mono SE suggests (each coil then 1k2/600mH).
Simulation of series connection (2k4/1.200m) would result in -3dB frequencies for 47k/100p, 300p and 500p load of 7.9kHz (dampend response), 9kHz (linear response) and 8.2kHz (peaking +0.5dB@4.4kHz).

The same simulation with paralleled coils (600Ohm/300mH) results in -3dB frequencies of 34kHz (peaking +0.35dB@16.45kHz), 24kHz (peaking +3,7dB@14.75kHz) and 19.2kHz (peaking +5.5dB@12.1kHz).
The peaking response suggests to tune the loading R value to lower values.
Simulating now with a fixed an practical Cload of 300pf and Rloads of 47k 33k 22k and 18k results in -3dB frequencies of 21.8kHz (peaking +1.28dB@12.3kHz), 19.6kHz (peaking +0.24dB@9.4kHz), 16.5kHz (linear response) and 12.6kHz (dampened response).

conclusion: Maybe less than 10kHz bandwidth is sufficient for Mono records reproduction, but for higher bandwidth parallel the coils and tune the Input resistor of the amp to lower values by paralleling a second resistor (120k||47K=33k8, 39k||47K=21.3k).

jauu
Calvin

 

[billshurv]

If you look at some calcs sreten made about 5 years ago he doesn't agree with your numbers Wiring Stereo cartridges for Mono - Page 2 - Vinyl Engine. Yes loading is critical in series, but if you are building a dedicated mono rig you would do that.

It's certainly a valid option.

 

[billshurv]

D'oh. This is what happens when everyone is posting as a guest on a thread. You read when low on sleep and misquote. So Apologies to Sreten, it was his thread but not his Sims.

 

[Hiten]

Allow me to go little off topic. Can a cartridge be made band limited to play 78 Shellacs which already have limited bandwidth ? We can then remove surface noise from getting amplified.
Regards.

 

[billshurv]

It's a resonant system so in theory yes. In practice may be as easy to add a roll off electronically. If you digitise the signal then all sorts of options appear.

 

[luckythedog]

If you look at some calcs sreten made about 5 years ago he doesn't agree with your numbers Wiring Stereo cartridges for Mono - Page 2 - Vinyl Engine. Yes loading is critical in series, but if you are building a dedicated mono rig you would do that.

It's certainly a valid option.

Yes, I prefer series wiring. You get twice the signal amplitude, and more predicatable behaviour when impure vertical stylus movement (ie not straight up vertical), inc noise. There is a significant noise advantage to using a true mono configuration of any nature, of course.

For series wiring, it's necessary to adjust MM loading, because inductance doubles. 15kHz bandwidth is necessary for mono, and if anyone other the dog can actually hear that content, it's nicely there.

 

[scott wurcer]

I was away for the hoilday but I would like to add my vote for ways to measure the stylus temperature. I've used Liquid-crystal thermography on chips and the resolution in both temperature and space is ideal here.

Thermal Imaging: Liquid-crystal thermography characterizes heat issues - Laser Focus World

 

[Calvin]

Hi,

@billshurv
The VinylEngine thread and my findings in #128 don´t contradict, the opposite is true.
The VE Sims use different coil impedance values, which are non practical.
A 300mH coil is rather low in value and wouzld feature lower Resistance than 1k (compare Ortofon 2M Mono: 350mH, 700Ohm).
Also the chosen 200pF of load capacitance are rather on the low side.
300pF-500pF are more practical.

In my Sims I choise the coil parameters of the Ortofon 2M Mono SE with 1.2kOhm and 600mH.

Attached are the schematics and two amplitde response sims.
Pic2 compares the amplitde response of the 2M Mono SE in parallel (green) and in series connection (red) feeding into a fixed Rload of 47kOhm and variable Cloads.
It shows that the series connection profits from +6dB of voltage level at the expense of reduced Bandwidth (-3dB@8.9kHz, -8.5dB@20kHz).
It also shows that 47kOhm as too high as load value for the parallel connection as HF-peaking occurs from very low pF values on.

Pic3 compares the amplitude responses of the series connection with chosen fixed RL and CL (red) against the parallel connection with fixed CL and varying RL (green).
It prooves that the same level of linearity can be achieved for the parallel connection if the load resistance is reduced, but positively preserving the higher bandwidth limit.
The RL can easily be tuned by connecting a resistor in parallel to the amp´s input resistor.

If the amp can take the higher voltage level without issues and the restricted bandwidth sounds ok, then nothing speaks against the series connection.
If not the parallel connection would be the right choice.

jauu
Calvin

 

[billshurv]

OK think I am getting you. You are writing as if you are talking with a whiteboard which makes you hard to follow! You quote a low inductance MM then sim with a high inductance, which hardly seems to be fair!

luckytd is using an ortofon concorde which is 450mH, 750Ohm. Up to my ears in nappies and posset so don't even have time to SIM that but certainly will be more likely to give enough HF response.

There are of course at least 2 other options, a bit of eq (its a dedicated mono rig) or the cordell vinyltrac.

What I don't have a full handle on it how the 2 arrangements differ with vertical information. I note in the 2010 discussion no conclusion was reached. Maybe electrical summing is the only way.

Still worth discussing the off the wall stuff even if at the end of the day its not the best way.

 

[luckythedog]

If the amp can take the higher voltage level without issues and the restricted bandwidth sounds ok, then nothing speaks against the series connection.
If not the parallel connection would be the right choice.

Series connection requires retuning of MM loading really, ie changing CR loading. Parallel wiring has its own, entirely different, set of issues.

Personally, I would use the options in order of preference ( best first):

True mono cart
Series wiring
Electronic summing in preamp (done well)
Parallel wiring

 

[Calvin]

Hi,

I disagree in part.
The C-part of an amp´s input impedance seems generally be called ´loading´, but actually the R-part is rather the correct loading mechanism.
The C-part functions rather as frequency decisive component, together with the pickup coils it builds a resonating circuit.
The R-part of the coils and the R-part of the amp-input summed up function to dampen the Q of the resonator to sufficiently low values.
Lower R-values increase the loading, increase damping, reduce the peaking.
Wether the coils are wired in series or wired in parallel doesn´t matter for the functioning as such.
It only affects the set of parameters of the resonating circuits.
As typical the lower inductance pickups react more sensitive, with higher Q-values (hence more peaking) on the C-part, but also with a higher resonating frequency.
If the peaking is too much with Rcoil+47kOhm (Q too high), we need more damping to flatten the amplitude response.
As the sims show Rcoil+25kOhm would give a very flat electrical amplitude response for the 2M Mono SE pickup.

The decision for series or parallel connection is just trading voltage level against bandwidth.
Depending on that decision either the C_part of the amp-input or the R-part may be dominant to taylor the amplitude response.
I don´t see ´entirely different set of issues´ for the parallel connection.

jauu
Calvin

 

[billshurv]

Z=load. Surely we can agree on that. Lets face it our model misses out the entire mechanical part of the system

But I do have a confusion thinking about this last night and being stupid. assuming

H=horizontal (or mono) signal
LV=left vertical
RV=right vertical

With a pristine mono record in series connection you get 2H and parallel you get H. With damage/dirt on the (say) left wall you get 2H+LV (so noise is supressed). with ivparallel coils one is trying to put out H and the other H+LV. Given the complex electromechnical system there, what actually comes out?

purely vertical movement in either case clearly still comes through and the DL-102 wins!

 

[luckythedog]

I don´t see ´entirely different set of issues´ for the parallel connection.

Consider what happens when noise is only in one channel, say due to stylus motion of momentarily riding up one groove wall as cropped up an example of friction induced noise on this thread.

Then, in parallel connection, the generator channel is loaded/shunted by a frequency dependant impedance (the other cartridge channel), which for the most part of the audio band is abnormally low, at lf mostly c 500R resistive from the coil resistance of the quiet channel.

Firstly, this is an abnormal load - c +40dB current at lf.

Secondly, the shunt load is frequency dependant, with steep variation right across the audioband. So generator output current for any given level is significantly frequency dependant in the audioband. This is shown in the attached plot, which shows the ratio of shunt current when one channel is driven for parallel wiring versus normal load current, for a frequency sweep across the audioband. 0dB is normal load current, into 47k/200pf.

Now extend this principle to, say, playing back stereo content. But also for 'imperfect' mono, where stylus motion is not perfectly lateral, and in interactions between asymmetric noise and mono signals. How a cartridge handles such overload is cartridge specific, depending on the electrical, magnetic and mechanical systems within the generator. Results vary IME.

It only crops up for parallel wiring, not for series, of course. Thus series and parallel wiring typically sound fundamentally different IME, even after proper RC tuning.