• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

SOTA Pyxi has arrived

wyn palmer said:
Yes. My attempt to end the tangential aspect of the ASR thread by posting the Korf measurements, interpretation etc. was a failure because, apparently, they did not fully either read or understand the point that Alex Korf was, in my opinion, so coherently and eloquently making. I'm (probably) not going back to that thread as it has become uninteresting.
Click to expand...

Ok this is slightly off topic but I can't resist. I have read that blog and it goes completely contrary to my findings, so either I'm not understanding (in which case I would love if someone would explain it all to me), or mr. Korf is not measuring what he thinks he is - which is my current conclusion. Let me also say that I've read most of the entries in the Korf blog and done so with enjoyment and I really appreciate what he's doing.

Now - I'm no scientist, but there seems to be something fundamentally wrong about that blog entry and his conclusions and it baffles me why he's seemingly unable to make the resonance frequency move around with compliance, since it's my experience it definitely does.

From the Korf blog:
Why can't we just measure the cartridge output through the preamp?

There are pass-through capacitors inside most preamps. These capacitors act as high-pass filters.
We'll see a false "resonant peak" at 5-7 Hz, but in reality it's just 1/f noise shaped by the presence of the capacitor.
Click to expand...

Maybe what he says here is true, but IME the cartridge/tonearm resonance peak will be clearly visible when measured from phono stage output, and will move lower/higher in frequency depending on cartridge compliance. Since the output is what happens in a real world scenario of playing a record, it would seem to me to be way more relevant than what ever you measure with an accelometer on a headshell. Now, my understanding on physics is limited, but it would also seem to me that for what he is trying to measure, he should have the accelometer fastened on inside the cartridge generator on the moving part (at the coils / magnets), since it's the cartridge suspension where the resonance happens. It's the movement that happens inside the cartridge that determines the output voltage.

So again, either I'm confused or he is not measuring what he thinks he is and the conclusions he reaches are not based on relevant data.

You'd also imagine if the whole thing was bogus as mr. Korf implies, someone like Ortofon would have figured it out in the 70+ years they have manufactured cartridges, or that a company like Brüel & Kjær would have found it out in their measurements, but apparently all it took was 15 minutes with an accelometer to prove it was all BS? I'm not buying that, especially since it doesn't align with my findings, which admittedly were not highly scientific and just a curious hobbyist fooling around, but still.

Now, to try to bring this back to topic at least in a loose definition of the word ( :) ) I will say subsonic / no-subsonic, there is no free lunch either way, I completely understand why you wouldn't want one in the chain, but I also understand why some people think it a necessity - and in some systems it really might be one. It depends on your system and use case whether you really "need" one or not. So in that sense I'd say it's one of those design compromises that happen unless you are building a "do it all for everyone" device. IME if you don't really need it, you're usually better off without one for subjective sound quality, but if you do need it, then it's quite invaluable.

Vinyl playback will always have noise in subsonic frequencies, it's of course best if you can deal with it at the source instead of having to rely on filtering. Also things like fluid damping can really help. But there is always going to be noise there, whether it matters for practical purposes depends on many things. Sometimes a sub-sonic filter might be a very good idea though, like if you're using modern stand mount speakers which will typically have very small drivers with a lot of excursion in a bass-reflex design tuned to about 40Hz to 60Hz - with such speakers the bass rolloff will be really steep below the tuning frequency, but the woofers will pump like crazy with any subsonic noise because the speakers are completely undamped at those frequencies. I'd say in such a case the benefits of limiting unnecessary woofer excursion (and resulting rise in distortion) out weigh what ever problems might come with a filter, like some phase shift in audible frequencies.

I recently made a move to such small stand mounts and watching the woofer excursion made me go crazy, I did my best to minimize it at the TT end but not to satisfying enough results, so I ended up building a crude subsonic filter into my phono stage, just a 2nd order passive filter at the output. Since I didn't do needle drops before / after, it's hard to compare, I'd like to think the sound got better but to be honest it was not a huge change either way. But it gave me better peace of mind, knowing my woofers are operating in their happy zone instead of wasting excursion on trying to playback subsonic noise.
 
muovimies said:
Ok this is slightly off topic but I can't resist. I have read that blog and it goes completely contrary to my findings, so either I'm not understanding (in which case I would love if someone would explain it all to me), or mr. Korf is not measuring what he thinks he is - which is my current conclusion. Let me also say that I've read most of the entries in the Korf blog and done so with enjoyment and I really appreciate what he's doing.

Now - I'm no scientist, but there seems to be something fundamentally wrong about that blog entry and his conclusions and it baffles me why he's seemingly unable to make the resonance frequency move around with compliance, since it's my experience it definitely does.

From the Korf blog:


Maybe what he says here is true, but IME the cartridge/tonearm resonance peak will be clearly visible when measured from phono stage output, and will move lower/higher in frequency depending on cartridge compliance. Since the output is what happens in a real world scenario of playing a record, it would seem to me to be way more relevant than what ever you measure with an accelometer on a headshell. Now, my understanding on physics is limited, but it would also seem to me that for what he is trying to measure, he should have the accelometer fastened on inside the cartridge generator on the moving part (at the coils / magnets), since it's the cartridge suspension where the resonance happens. It's the movement that happens inside the cartridge that determines the output voltage.

So again, either I'm confused or he is not measuring what he thinks he is and the conclusions he reaches are not based on relevant data.

You'd also imagine if the whole thing was bogus as mr. Korf implies, someone like Ortofon would have figured it out in the 70+ years they have manufactured cartridges, or that a company like Brüel & Kjær would have found it out in their measurements, but apparently all it took was 15 minutes with an accelometer to prove it was all BS? I'm not buying that, especially since it doesn't align with my findings, which admittedly were not highly scientific and just a curious hobbyist fooling around, but still.

Now, to try to bring this back to topic at least in a loose definition of the word ( :) ) I will say subsonic / no-subsonic, there is no free lunch either way, I completely understand why you wouldn't want one in the chain, but I also understand why some people think it a necessity - and in some systems it really might be one. It depends on your system and use case whether you really "need" one or not. So in that sense I'd say it's one of those design compromises that happen unless you are building a "do it all for everyone" device. IME if you don't really need it, you're usually better off without one for subjective sound quality, but if you do need it, then it's quite invaluable.

Vinyl playback will always have noise in subsonic frequencies, it's of course best if you can deal with it at the source instead of having to rely on filtering. Also things like fluid damping can really help. But there is always going to be noise there, whether it matters for practical purposes depends on many things. Sometimes a sub-sonic filter might be a very good idea though, like if you're using modern stand mount speakers which will typically have very small drivers with a lot of excursion in a bass-reflex design tuned to about 40Hz to 60Hz - with such speakers the bass rolloff will be really steep below the tuning frequency, but the woofers will pump like crazy with any subsonic noise because the speakers are completely undamped at those frequencies. I'd say in such a case the benefits of limiting unnecessary woofer excursion (and resulting rise in distortion) out weigh what ever problems might come with a filter, like some phase shift in audible frequencies.

I recently made a move to such small stand mounts and watching the woofer excursion made me go crazy, I did my best to minimize it at the TT end but not to satisfying enough results, so I ended up building a crude subsonic filter into my phono stage, just a 2nd order passive filter at the output. Since I didn't do needle drops before / after, it's hard to compare, I'd like to think the sound got better but to be honest it was not a huge change either way. But it gave me better peace of mind, knowing my woofers are operating in their happy zone instead of wasting excursion on trying to playback subsonic noise.
Click to expand...
Korf is NOT saying that the LF "peaking"due to the "resonance" does not exist. What he is saying is that for many cartridge systems that the system is sufficiently well damped that the "resonance" is actually just a high pass and is mostly benign. When this high pass is "excited" then you can see its effect. Thats why it appears in test records which generally supply a lateral pink noise excitation. I can state for a fact that this is the case with the cartridge etc. that I have as it exhibits almost no LF peaking- at most 0.5dB. This is also entirely consistent with seismic transducer theory and fact, so the transducer DOES behave linearly at, and even a bit below, the "resonance" frequency.
This is a measurable and mathematical reality and is entirely reasonable- i.e. the output is the multiplication of the high pass response of the seismic transducer times the electrical differentiating response of the cartridge times the excitation of the modulation.
All I ask is that one tries to actually understand a bit about seismic transducer theory (hence the ADI paper) which contradicts the idea that the system does NOT behave linearly in the region close to the "resonance" and hence is useless as a reproducing element there.
The B&K inference is meaningless in this case.
There are now c. 300 units out there that have no standard subsonic filter. Do you know how many complaints that I've received? The number is between -1 and +1.
Those units that include the "warp filter" successfully greatly reduced the ultra LF modulation, even though they do pretty well nothing to the lateral signal down to a fraction of a Hz.
 
His measurements still don't make sense to me, he seems to say the resonance frequency doesn't change with the compliance of the cartridge and that matching cartridge compliance / tonearm effective mass is not in anyway necessary. Maybe what he is trying to say just flies over my head, that is entirely possible. However from practical stand point I have enough experience with different cartridges and tonearms and how they perform together that I can say in general matching cartridge compliance with tonearm mass matters. I don't think it's always a hard and fast rule, but it's not irrelevant either.

There are plenty of well regarded phono stages out there with no subsonic filter and I've happily used many for years. After a move I had to make changes in my system including new speakers and I found myself in a situation where subsonic filtering suddenly started to seem like a very good idea :) Which was the entirety of my point, they have their uses. I'm in no way implying the Pyxi is "defective" for lacking one, and I have no interest in getting into any sort of debate about it. The "warp filter" is actually a pretty clever way of doing it, I would imagine the lateral stuff is rarely the problem, so leaving it alone is probably a good thing in most cases.
 
muovimies said:
His measurements still don't make sense to me, he seems to say the resonance frequency doesn't change with the compliance of the cartridge and that matching cartridge compliance / tonearm effective mass is not in anyway necessary. Maybe what he is trying to say just flies over my head, that is entirely possible. However from practical stand point I have enough experience with different cartridges and tonearms and how they perform together that I can say in general matching cartridge compliance with tonearm mass matters. I don't think it's always a hard and fast rule, but it's not irrelevant either.

There are plenty of well regarded phono stages out there with no subsonic filter and I've happily used many for years. After a move I had to make changes in my system including new speakers and I found myself in a situation where subsonic filtering suddenly started to seem like a very good idea :) Which was the entirety of my point, they have their uses. I'm in no way implying the Pyxi is "defective" for lacking one, and I have no interest in getting into any sort of debate about it. The "warp filter" is actually a pretty clever way of doing it, I would imagine the lateral stuff is rarely the problem, so leaving it alone is probably a good thing in most cases.
Click to expand...
He explains it poorly. I'll see if I can do better.

The reason that the "peak" does not exist in a predictable way- for his system- is that in essence he is relying on an external stimulus (a warp or less usefully, an off axis central hole) to generate the output due to the multiplication of the three terms that I described above. Depending on the parameters of the system the response may not show any peaking at all, or a peaking that changes with changes in the elements in an unpredictable way.
In any case the response consists of a series of harmonics of the excitation "filtered" by the linear cascade of the response of the seismic transducer and the differentiating action of the cartridge. For a suitably damped system this just looks like a linear high pass with minimal peaking and in any case the output is by definition a series of harmonics of the excitation frequency modified by the frequency response of the system.
The magnitude and number of these harmonics depends on the width of the "warp", for example, together with the peak amplitude, which causes the eventual output signal to be in some sense non-intuitive.
Shorter duration warps, even when the amplitude is less, can cause more peak output because the magnitude of some of the higher frequency harmonics i.e. the ones that are in the high pass transition region, are larger- i.e. the vertical velocity can be locally higher, but the region of increase is of shorter duration. The harmonics are also not of constant amplitude versus frequency, so the relative frequency distribution of the "resonance" and the excitation can result in an unexpected output frequency characteristic.

Look, the bottom line in this is that the concept that the "seismic transducer" - that is the cartridge system- does not produce a usable output until greater than 2x the "resonant" frequency, and this is worse with increasing damping, is just nonsense, and is in fact completely backwards to what measurements and math indicate.

In essence two ideas have become conflated in the original argument- the "undesirable" presence of ULF content that causes unwanted excursions of the woofers and the creation of chaotic signals due to the action of the "seismic transducer" in the presence of an external stimulus.
The latter is still subject to the linear action of the mechanical system, the former is purely of interest due to the behavior of the speaker system, and in my systems case, for example, the bass units cannot be induced to leave their linear response region without the main power amps "clipping" and so, other than a visible excursion if the covers are removed, has no consequence.
Add a high pass if you wish. Most preamps and power amps have built in high pass restrictions, and many have optional high pass filters.
In my experience, a single order high pass has minimal effect on the excursions unless it negatively affects the LF response of the phono stage, so a higher order (greater than 2nd) high pass or a damping ratio-controlled 2nd order warp- style filter which leaves the lateral signal effectively unchanged is essential to do the job right!
 
Last edited:
I did my tests a bit differently, measuring the cartridge output (left channel from HFN PN track B4). As I've an EPA-500 setup and several copies of each armwand, along with enough VM95 bodies for them all, I threw together the E, H, M, and L wands which are advertised to have EMs of 6g, 8g, 10g, and 18g. I went with the conical stylus, and set VTF to 2g and swapped the same stylus between arms/bodies. AS, VTA, etc. stayed the same. Conical because this was a quick and dirty experiment and I didn't want to spend much time on alignment.

The punchline is that the plots follow what traditional calcs predict for the most part. The only outlier is the 18g wand where the resonance is where a 13g wand should be. I've not verified the wand EMs and am going by Technics literature.

EDIT: forgot to mention that I defeated the dynamic damping for the armwands, of course.


EMTests-02.png


IMG_2380.jpeg
 
Last edited:
JP said:
I did my tests a bit differently, measuring the cartridge output (left channel from HFN PN track B4). As I've an EPA-500 setup and several copies of each armwand, along with enough VM95 bodies for them all, I threw together the E, H, M, and L wands which are advertised to have EMs of 6g, 8g, 10g, and 18g. I went with the conical stylus, and set VTF to 2g and swapped the same stylus between arms/bodies. AS, VTA, etc. stayed the same. Conical because this was a quick and dirty experiment and I didn't want to spend much time on alignment.

The punchline is that the plots follow what traditional calcs predict for the most part. The only outlier is the 18g wand where the resonance is where a 13g wand should be. I've not verified the wand EMs and am going by Technics literature.

EDIT: forgot to mention that I defeated the dynamic damping for the armwands, of course.


View attachment 2947190


View attachment 2947191
Click to expand...
How do they look with the dynamic damping active?
 
Anybody here try one of these with a Pickering 7500 cart? I have an Emotiva phono pre and the Emotiva XSP gen 2 preamp with MC inputs. However, it still seems very low with this cart. So I am looking into phono pres that would be a good match for this cart and then other carts like my Graces.
 
Hitachiht550 said:
Anybody here try one of these with a Pickering 7500 cart? I have an Emotiva phono pre and the Emotiva XSP gen 2 preamp with MC inputs. However, it still seems very low with this cart. So I am looking into phono pres that would be a good match for this cart and then other carts like my Graces.
Click to expand...
I can't find any info on the cart. What's the rated sensitivity?
 
This is all I could find and someone said they are using this: Equinox MC head amp
Pickering XLZ7500S
Specifications are:
# Diamond Stereohedron II (line contact tip) 2 x 8x71
# Output 0.33mv (MC level)
# Frequency range 10 to 50k
# Tracking force 0.5 gram to 1.5 grams
# Tracking ability 100 microns
# Compliance 30 cu
# Channel separation 35db
# Load 50 - 250 ohms, pf no limit
# Effective mass 5g
 
Hitachiht550 said:
This is all I could find and someone said they are using this: Equinox MC head amp
Pickering XLZ7500S
Specifications are:
# Diamond Stereohedron II (line contact tip) 2 x 8x71
# Output 0.33mv (MC level)
# Frequency range 10 to 50k
# Tracking force 0.5 gram to 1.5 grams
# Tracking ability 100 microns
# Compliance 30 cu
# Channel separation 35db
# Load 50 - 250 ohms, pf no limit
# Effective mass 5g
Click to expand...
The Pyxi should have plenty of output, a bit more than 2V rms for the loudest sections.
 
Hitachiht550 said:
This is all I could find and someone said they are using this: Equinox MC head amp
Pickering XLZ7500S
Specifications are:
# Diamond Stereohedron II (line contact tip) 2 x 8x71
# Output 0.33mv (MC level)
# Frequency range 10 to 50k
# Tracking force 0.5 gram to 1.5 grams
# Tracking ability 100 microns
# Compliance 30 cu
# Channel separation 35db
# Load 50 - 250 ohms, pf no limit
# Effective mass 5g
Click to expand...
By the way, as far as I can tell the Equinox does not have any actual specifications- unlike the Pyxi which is fully characterized with the results available on the data sheet, so it's hard for me to make a comparison. It does cost 1/20th of the price of the Equinox and does not come with "balanced" inputs (completely unnecessary in my view), and the specs of the Pyxi lack the usual audiophile verbiage, so it has to be vastly inferior...
 
You must log in or register to reply here.
Back
Top Bottom