[neobop]

Hmm something strange about where the cantilever meets the pivot point...






David

According to J. Carr, most all carts have the cantilever attached to the suspension with what is called a joint pipe, which is a slightly larger tube.  In this case I think the black part - the tube directly in front of the magnets, is the joint pipe.  The Jico tube is longer than OEM and seems wider with the grey stuff reinforcing the attachment. 
neo

[dlaloum]

Ahh that also explains the ease of replacing an entire cantilever - if you have the right solvent, you simply  dissolve the glue, slide out the old one and slide in the new....

Hence people like soundsmith having reasonably priced (relatively) "retipping" which replaces the cantilever rather than do the much more difficult task of actually removing the old needle and putting a new one on in correct alignment....

[neobop]

If we revisit the AT link provided by Scotty, we see that the black piece that holds both the magnets and the joint pipe is identified (correctly by David) as the pivot.
http://www.coolgales.com/brochures/AudioTechnicacartridgesbrochure.pdf

It also looks as if the filament (suspension wire) extends down the cantilever.  Is that possible with a solid exotic cantilever?  Behind the rubber damper the cantilever morphs into the suspension wire housing that is secured in the plug by the compliance screw.  I thought I read about someone replacing the pivot/magnets.  Maybe that's wrong, but I think it would only be possible if the cantilever doesn't end at the pivot.
neo

 

[neobop]

Maybe that drawing with the suspension wire going into the cantilever, was just showing dimensionality. 

Here's a link to further response by J. Carr about tip mass and high frequency resonance:
http://forum.audiogon.com/cgi-bin/fr.pl?eanlg&1200430667&openflup&11228&4#11228

"The joint pipe is usually made from aluminum or duraluminum, which although at 2.7g per cubic cm is heavier than beryllium (1.85g) or boron (2.46g), is certainly lighter than diamond (3.5g) or sapphire (4g). Keep in mind that a component that is closer in to the pivot (such as the armature and coils of an MC or the magnet of an MM) will affect the effective mass less than another component farther away from the pivot. The joint pipe must extend all the way back to the pivot, and in many cases it is what the magnet or armature are secured to.

Counter-intuitively, increasing the joint pipe length can raise the resonant frequency rather than lowering it, and this is because a shorter and lighter joint pipe can allow the cantilever rod to flop around at its root, while a longer and heavier joint pipe will result in less overall flexing.

Even among AT MMs, those that use a thin rod cantilever will most likely have the joint pipe, while those that use an aluminum pipe cantilever may be able to do without. This is because the suspension components (wire, dampers) come in specific sizes that need a stepped interface to mate with the significantly smaller diameter of a rod cantilever (a boron rod is 0.28mm in diameter). I don't think that the issue is company practice (AT vs. Jico).

Depending on how you calculate things, cantilever length has a more significant effect on HF resonance than tip mass. And while there are other equations that relate tip mass to HF resonance, these are only relevant if all other things are equal, which usually they are not.

For example, a designer can use a longer diamond tip (with higher self-mass) to extend the vertical reach of the cantilever and reduce the overall cantilever length, and this may increase the HF resonance. On the other hand, doing so will probably result in poorer crosstalk characteristics at high frequencies, as the longer stylus will start to allow torsional twisting in addition to horizontal and vertical movement. IOW there is no free lunch, but if a designer is simply interested in raising the HF resonance point, there are various tricks and tradeoffs available.

Also, HF resonance is affected by the overall cantilever stiffness (as can be seen in the longer joint pipe example above), and by damping. On the surface of things, using more damping to flatten things may appear better, but in practice doing so usually triggers more distortion. Experience shows that a rising HF response is less offensive to the ear (as it is a simple boosting of whatever is already present on the LP, which, given the 1/f spectral distribution of most music, may not be much) than distortion, which synthesizes new HF components that do not exist on the LP, and in the case of IMD will result in inharmonic distortion at lower frequencies that are particularly grating to the ear.

I find that cartridges sound better when the designer puts in a honest effort to increase cantilever stiffness while reducing moving masses, but does not try to steamroller the last few dBs into total submission."

Interesting stuff even if there's no formulas or revelations.  He's a hell of a nice guy, kind of reminds me of VDH and his Q&A forum, but that was more basic.  That's in a PDF now.
Speaking of VDH, you get interesting specs with his carts, including recommended arm mass and effective tip mass:
http://www.vandenhul.com/products/phono

neo

[dlaloum]

It all sort of makes sense, and is consistent with other bits of testing and analysis....

Ortophase tests showed marked Golden ear preference for low or undamped cantilever - but totally undamped allowed the high end to rise too much - best compromise was light damping.

Cantilever length vs tip mass - the physics are so tightly interwound that I find it strange that he would seperate the two! But yes clearly the multiplying effect of the "lever" on effective mass means that length of cantilever can have a greater effect that the actual mass of the needle at the end (and the longer the cantilever the lower the needle masses proportion of the total)

The point about join pipes not being needed for aluminium cantilevers - and their contributions to rigidity is really interesting - and explains designs like the SAS with quite a long "pipe" telescoping out before the boron rod takes over.
It also implies that a longer attaching pipe could be a positive.... I must investigate my collection of needles and see whether I have a few that are very similar to each other some with longer or shorter or no join pipes and investigate where that then places the cantilever resonance.

And of course the final comment about a balanced design without too much excess...... wise words!

But I still love the Dynavector Karats - and they are definitely an example of excess in one particular direction

bye for now

David

[neobop]

The original definition (by Carr) of the joint pipe, included the end of the cantilever.  He was responding to the suggestion that AT MM don't seem to have a joint pipe, and a query about the relationship between tip mass and high frequency resonance. 

His pointing out the function of a longer joint pipe is something we should have thought of back on page 25 of this thread.  We know the benefits of both a shorter cantilever and increased rigidity, we just didn't connect the dots.  Unlike a boron or beryllium cantilevered stylus, in the case of an AT-95 replacement stylus, tip mass is of greater concern.  We're looking at straight aluminum cantilevers with bonded stylus.  How clunky can you get?  A longer joint pipe will effectively shorten the cantilever, but will also increase tip mass.  I don't see how it could not.  These replacement styli seem to track okay, so what's the problem, SQ ?

What is the acceptable range of effective tip mass?  Maybe best case 0.2mg to max around 0.7mg ? 
I noticed Ortofon stopped publishing eff tip mass.  I think that's in response to criticism of their HOMC tip mass. 
Later,
neo

[dlaloum]

The later version of the X5-MC is specced at 0.7mg - and I verified that with my measurements.

Seems clunky compared to the svelte 0.3mg of the OM30!

On the other hand it does make one wonder about the potential of a recantilevered Xx-MC fitted with a nice Ruby cantilever.... it might give some of the multimegabuck Ortofons a run for their money...

I concur - decent tip masses seem to max out at 0.7g

[neobop]

Remember the thread about Lyra Atlas?  To refresh our memories here's the German review:
http://www.fastaudio.com/workspace/uploads/downloads/stp_04_12_sd_atlas.pdf

When I saw that test report I was surprised by the 20KHz resonance, which is nearly identical to Kleos.  The close-up at the bottom of page 2 was even more surprising.  What's that big mounting platform under the tip?  It seems obvious that it's for keeping the stylus and cantilever acting as a single unit - structural stability.  A nude stylus is normally glued onto the end of the cantilever and this would help prevent rotation and separate movements.

Next question is, what's Atlas eff tip mass?  Atlas is widely acknowledged to be top drawer and that answer might shed some light (or not) on understanding tip mass range and numerical relevance.   

Another question is, why can't other modern magazines give us more than the subjective crap we're handed?  I'm supposed to take some silly reviewer's word that B is better than A ?? 
A has a blacker background, but B is more transparent.   
Stereoplay didn't have much but a FR graph.  Most of the clown for hire rags print nothing more than mfg. specs.  Sad but true.
Back in the day.....

Later,
neo

 

[dlaloum]

I'll grab my spreadsheet later today and calculate what a 20k resonance works out to in tip mass...   it won't be bad, but it won't be great either.

Looks to me like a voicing decision.... With mass control by that funky stylus plate.

[*Scotty*]

The Atlas construction looks decidedly less than impressive when compared to the nude square shank diamond stylus mounted in laser bored square hole in the cantilever as seen in the top of the line AT cartridges.

Scotty

[dlaloum]

20kHz resonance implies tip mass of 0.54mg...

For comparison -

SAS (res = 28kHz) has tip mass of 0.28mg

Pickering XLZ7500, res=26kHz, tip mass = 0.32mg

Signet TK9e res= 25kHz, tip mass = 0.35mg

Shure V15VMR res=31kHz, tip mass = 0.23mg

Dynavector Karat 23rs res= 50kHz, tip mass = 0.09mg
Digitrac 300SE res=20.5kHz, tip mass = 0.52mg

[neobop]

Scotty,

Nice photo. 

Is that an ML on an aluminum cantilever, like a 440?

[neobop]

Seems like Ortofon left one on their spec sheet, and it's one of those I thought they were trying not to display.

Ortofon MC-3 Turbo = 0.75mg
Tracking ability >70um

http://ortofon.com/hifi/products/cartridges/moving-coil/high-output-mc/mc-3-turbo

[*Scotty*]

Yes it is.  I also found a picture of the ML 150 stylus with the beryllium cantilever which looks the same except for the gold color.
Amazing technology, I am still planning on transplanting a MLX 150 stylus on to my AT 440 ML body as finances permit. I guess I will have to be satisfied with a Boron cantilever.
Scotty

[neobop]

Wow. 
I just revisited Miyajima site to check on Kansui specs.  That's the one Throwback uses, or I should say, one of the ones.   Kansui is 16 ohms and 2.3mV - shibata/aluminum.  You wouldn't know it from the specs, but Miyajima carts have nonmagnetic cores!  Most MCs with nonmagnetic cores have very low output and high resistance like the DL-S1.  They also usually have high inductance for the amount of output.  Because you need tons of gain for a cart like DL-S1 (0.15mV, 33 ohm) the relatively high inductance combines with cable capacitance and might cause high frequency overload in extended bandwidth preamps.  That causes a particularly irritating intermodulation distortion throughout the mid/treble region.  The solution in that case is either load it down like crazy (I think Elison in the Asylum uses 22 ohms) and need even more gain, modify your preamp, or try another cart or preamp. 

Today there are carts like the AT ART 7 that have nonmagnetic cores without the unusually high resistance.  At Miyajima site there's a 10 minute video illustration of their cross ring design.  Highly recommended viewing.
http://www.miyajima-lab.com/e-stereo.html

Later,
neo

Note:  The AT Art 7 has an output of 0.12mV, inductance is 25mH.  That's a lot more inductance than I thought.  MC inductance is usually measured in micro henries. 
http://eu.audio-technica.com/en/products/cartridges/product.asp?catID=8&subID=57&prodID=4495

[*Scotty*]

Looking at how the specifications have been typed, I would bet that the m in this case is intended to mean micro. Using lower case u as a prefix in front of the upper case H for Henries is counter intuitive. The u in this case comes from the Greek for μικρός (transliterated as mikrós) which means small. There is also no font available for the Ohms symbol which is Omega ( Ω) in the Greek alphabet. I noticed that W was used for the Ω in the published specs.
That's my theory and I'm sticking to it.
If we look at the specs for the AT-F7 we find 12ohms and 25uH which is a little more believable.
http://eu.audio-technica.com/en/products/cartridges/product.asp?catID=8&subID=57&prodID=4028
Scotty

[neobop]

I think you're right.  It's got to be some kind of mix-up or typo.  I don't think it's possible for a MC of that output to have 25mH.  A HO Grado has 45mH.
The ART-7 was patterned after the 50ANV which has the same output with a nonmagnetic core.  It has 7uH. I have a question in for a guy who has an ART-7, asking if he has a spec sheet.

If you can deal with that output, the 7 is around $1K on Amazon - shipped from Japan.  LpGear still has the 50ANV - price reduced to < $1.5K.

Those Miyajima carts are intriguing.  Two have compliance similar to a 103 and the other two are about 15 or 16 cu. 

neo

[neobop]

I'll grab my spreadsheet later today and calculate what a 20k resonance works out to in tip mass...   it won't be bad, but it won't be great either.

Looks to me like a voicing decision.... With mass control by that funky stylus plate.

There is/was a poster on VE called Desktop who is a retired speaker designer.  He worked for a big company and also designed his own.  I think it was about 6 or 7 years ago, he spoke of a movie probably made by Dynevector, of a cantilever in action - while tracking.  This video was made using laser microscopy interferometry.  I think that's what he said.  Magnification was probably minimal  <100X ?   Interferometers are used for scanning a surface.  They're widely used in precision manufacturing, optics, electronics (phase testing), aerospace etc. 

Anyway, Desktop's description of the cantilever was nothing short of amazing.  It whipped around and wiggled in ways that looked impossible.  At one point it looked as if it would fold up on itself.  I never saw this video, but I also have no reason to doubt this story. 

That clunky looking plate on the Atlas cantilever might be very effective "ground control".  I'm sure it's there for a reason.  If you look at the photo again, there's a hole that goes clear through the boron cantilever and the structure is secured on top.  I don't know what the plate is made of, maybe titanium?  but it must cost a bundle.  Lyra (Scan-Tech) gets their diamonds and probably cantilevers from Ogura.  (Anybody notice stylus price increases lately?)  Ogura and Namiki  are the high price spread and I would guess that cantilever/diamond costs them more than a 440 costs us, retail.  What price glory?
neo

 

[neobop]

Some interesting developments on Agon thread.  Turns out ART7 has 8uH inductance.  That's good news if I want to try one.  I had to use another phono preamp for the DL-S1.

There's also more elucidation about cantilevers, specifically AT.  Most of the stuff we already know.  J. Carr raised the possibility of misalignment with the pole pieces if compliance (damper compression) is changed.  I think AT stylus fitment is self aligning to some degree, but that seems like a real possibility given the size of the gaps.  J. Carr has been tremendously helpful.  We're lucky to have his input.   

David,
What are your thoughts about this?
neo

P.S.  I forgot to mention, the stylus mounting plate on Atlas is there to keep the tip in place - from breaking/falling off etc.  It's kind of disappointing and points out the advantage of a replaceable stylus.  I don't know the warranty or retipping policy, but on a $9.5K cart.....

[dlaloum]

Interesting comment about the AT setup and pole piece positioning - makes sense....

If I was AT, I would have a standard compression - ie a standard production line that pulls that wire/string the same amount for all the styli, but I choose the materials for the doughnut such that it compresses variably - and the doughnut thickness is then based on knowing how much it will compress - which defines where the magnets end up relative to the pole pieces.

The alternative would require variable tension system - so they can define what tension is needed for each stylus - and then the variable of doughnut thickness, compressibility etc... would need to be worked into the design - one more variable to control.

Dunno.... would love a tour of the AT stylus production line, guided by one of their design engineers!

The stylus mounting plate seems a bit of a kludgy setup - perhaps Lyra don't have access to the type of high precision laser tools used by Dynavector? (laser cut hole in the cantilever, press fit square cut diamond into laser cut hole, possibly add a dab of glue at the back...)

The approach with the mounting plate seems to imply a cruder lower tech craft approach rather than precision through higher tech....

The Technics Boron tube cantilevers with laser cut holes and pressure fit needles seem to remain the acme of stylus technology - never to be repeated..... sad really.

Going back to the AT MM designs - I don't think the pole piece positions are quite that critical - given that the stylus mounting is probably thermoplastic made by pressure injection, I would assume that there would be a level of uncertainty in the holder itself.
The more precise metal versions used in the AT25 series have long been discontinued (and there is another possible advantage to that vintage design - the metal mounting might allow for more precise and closely positioned pole pieces - with magnetic advantages in addition to the fact that these are true torroidal designs...)

Looking at how the styli snap into the bodies, and we can actually see the tip of the pole pieces in the opening - I think there is quite wide latitude there.... MC's are probably a lot more sensitive to this type of positional variation than MM's are.

On the other hand the magnetic "system" of an air coil MC is more theoretically "perfect" than the standard MM - probably less eddy current, hysteresis, etc.... That would make for a better midrange from around 2kHz up... (effect should be of less than 2db in magnitude but that is still quite substantial!)

The magnetics are starting to make me come around to thinking that MC might in fact be the superior system ultimately - or at least that low output is the superior system...

When I measured the XLZ7500 and the XSP3003 - both with the same needle - there was a substantially greater trough in the 2k to 8k frequency range on the high output body than there was on the low output body - this puzzled me at the time.... but with many months of further marinating in the back of my mind, I am theorising that the much lower signal levels involved reduce the undesirable effects substantially.

It also makes me wonder whether the only models that had laminations in the Stantering families were the low output ones? - Laminations never seem to have been mentioned in ANY Stantering blurbs - and the midrange troughs on the Stanterings seems greater than on equivalen Shures and AT's.... (Laminations, or perhaps HotPressedFerrite like Technics used?)

bye for now

David