Best phono stage?

[PeterA]

Right, and this low capacitance is also the reason I modified my XP-25 and removed the lowest 100pF loading capacitor option - and that entire knob/option and circuit is entirely useless for MC, though I suspect Pass included it in order to further affect the sound - tone controls.

Ack, do you notice much or any difference between the different load options on your Pass XP-25? I set mine to 1K ohm and have tried 47K, 500 and the lower settings. I really struggle to hear any difference. This is in sharp contrast to my old Pass Xono which had 200 or so different settings. With that phono, I could hear significant differences between settings.

Your comment that it could be included as a tone control implies to me that you hear differences between the settings. The Pass manual states that one will hear differences, so I understand where you get the tone control idea.

 

[Pb Blimp]

Ralph and Ack,

Thanks for your comments. I believe I understand the points regarding the need to find the right balance when cartridge loading to dampen RF without losing dynamics: too high impedance and you don't stop the rf causing phonostage distress while to low of impedance will attenuate voltage causing a reduction in dynamics etc. as stages are voltage amplifiers not current. (This is in fact what I was trying to say above.) Where you have enlightened me is that I always presumed a compromise in sonics was inherent in making a stage absolutely immune to rf since so few of even the best stages are truly immune. As I think about the ways absolute stability are derived I come back to the use of some sort of resistance in the equation which by definition results in voltage attenuation.

 

[ack]

Ack, do you notice much or any difference between the different load options on your Pass XP-25? I set mine to 1K ohm and have tried 47K, 500 and the lower settings. I really struggle to hear any difference. This is in sharp contrast to my old Pass Xono which had 200 or so different settings. With that phono, I could hear significant differences between settings.

Your comment that it could be included as a tone control implies to me that you hear differences between the settings. The Pass manual states that one will hear differences, so I understand where you get the tone control idea.

I hear HUGE differences with the various loading options, and quite easily demonstrable too, e.g. see http://www.whatsbestforum.com/showt...ing-A-Misnomer&p=278033&viewfull=1#post278033
The sound is really soft with very low resistor values, as expected and electrically explained (I played that LP at your place the other day)

 

[Atmasphere]

Ralph and Ack,

Thanks for your comments. I believe I understand the points regarding the need to find the right balance when cartridge loading to dampen RF without losing dynamics: too high impedance and you don't stop the rf causing phonostage distress while to low of impedance will attenuate voltage causing a reduction in dynamics etc. as stages are voltage amplifiers not current. (This is in fact what I was trying to say above.) Where you have enlightened me is that I always presumed a compromise in sonics was inherent in making a stage absolutely immune to rf since so few of even the best stages are truly immune. As I think about the ways absolute stability are derived I come back to the use of some sort of resistance in the equation which by definition results in voltage attenuation.

OK- I still feel like this needs correction! First comment in bold: Its not one of the best stages if it has RFI problems! IMO/IME that's simply poor design, and sorry if it cost a lot too but if it was RFI immune, it would sound better!

Second comment in bold: You really don't suffer any attenuation if stopping resistors are employed in the design. And you don't see any attenuation if loading resistors are employed- as long as they are at least 10x the source impedance of the cartridge, and so far, all I have seen over the years are.

However, Jonathan Carr and I had an interesting conversation at Munich a couple of years ago: in it, we discussed the mechanical effects of loading.

Since you are asking the cartridge to drive a low impedance load when using a loading resistor, the work being done to heat that resistor up has to come from somewhere and that somewhere is the stylus. That means you are making the stylus harder to move and that can't be a good thing. Dynamic impact, bandwidth and traceability are all affected (limited).

So there is a very real argument for eliminating the use of loading resistors, which are a band-aid to a poor design anyway; the argument being that if you want the most mechanical performance out of the cartridge, don't ask it to do work! Just let it do the job of tracking the groove.

This means that the preamp **must** be immune to RFI, no if ands or buts, else it cannot be considered in the list of 'the best'...

 

[PeterA]

I hear HUGE differences with the various loading options, and quite easily demonstrable too, e.g. see http://www.whatsbestforum.com/showt...ing-A-Misnomer&p=278033&viewfull=1#post278033
The sound is really soft with very low resistor values, as expected and electrically explained (I played that LP at your place the other day)

Thanks Ack. So this must mean our Pass XP-25s are not stable if I understand Ralph's comments. I remember you used to used 500 ohm loading. Is that still the case? If so, why not 1K or 47K?

I don't follow your comment about the LP you played at my house? Which one and what does it explain electrically? We did not alter the loading from 1K ohms during our listening session.

 

[Pb Blimp]

OK- I still feel like this needs correction! First comment in bold: Its not one of the best stages if it has RFI problems! IMO/IME that's simply poor design, and sorry if it cost a lot too but if it was RFI immune, it would sound better!

Second comment in bold: You really don't suffer any attenuation if stopping resistors are employed in the design. And you don't see any attenuation if loading resistors are employed- as long as they are at least 10x the source impedance of the cartridge, and so far, all I have seen over the years are.

However, Jonathan Carr and I had an interesting conversation at Munich a couple of years ago: in it, we discussed the mechanical effects of loading.

Since you are asking the cartridge to drive a low impedance load when using a loading resistor, the work being done to heat that resistor up has to come from somewhere and that somewhere is the stylus. That means you are making the stylus harder to move and that can't be a good thing. Dynamic impact, bandwidth and traceability are all affected (limited).

So there is a very real argument for eliminating the use of loading resistors, which are a band-aid to a poor design anyway; the argument being that if you want the most mechanical performance out of the cartridge, don't ask it to do work! Just let it do the job of tracking the groove.

This means that the preamp **must** be immune to RFI, no if ands or buts, else it cannot be considered in the list of 'the best'...

Ralph, thanks again for your cogent explanation. I appreciate your patience. It still seams a bit odd to me that many of us have spent so much time over the years walking the tightrope with impedance to manage rfi if rfi immunity (phonostage stability) is easily achieved with no sonic compromise, but I hear you. Thanks!

 

[ack]

Thanks Ack. So this must mean our Pass XP-25s are not stable if I understand Ralph's comments. I remember you used to used 500 ohm loading. Is that still the case? If so, why not 1K or 47K?

I load at 1k as I said up above, and that's because my MIT MA-X phono cable has an impedance setting for that as well (the others are 47k and 100ohms). There is a sweet spot on the XP-25, and it appears to be between 500-1k ohms. I tried the 47k just recently, and the bass performance actually sagged. The XP-25 is also demonstrably unstable in the UHF if you were to compare my shielded unit with yours; this is why I copper-shielded the casing years ago, on top of the mumetal. I suggested years ago that you do the same, and I continue to think you should. See also what Ralph also said about RFI immunity. None of this is new, if you are a techie.

I don't follow your comment about the LP you played at my house? Which one and what does it explain electrically? We did not alter the loading from 1K ohms during our listening session.

The link I posted refers to an LP, which demonstrates the huge differences I mentioned in sheer dB of RMS volume; that LP I played at your place the other day, so basically you are familiar with it and you can try for yourself.

 

[ack]

OK- I still feel like this needs correction! First comment in bold: Its not one of the best stages if it has RFI problems! IMO/IME that's simply poor design, and sorry if it cost a lot too but if it was RFI immune, it would sound better!

Second comment in bold: You really don't suffer any attenuation if stopping resistors are employed in the design. And you don't see any attenuation if loading resistors are employed- as long as they are at least 10x the source impedance of the cartridge, and so far, all I have seen over the years are.

However, Jonathan Carr and I had an interesting conversation at Munich a couple of years ago: in it, we discussed the mechanical effects of loading.

Since you are asking the cartridge to drive a low impedance load when using a loading resistor, the work being done to heat that resistor up has to come from somewhere and that somewhere is the stylus. That means you are making the stylus harder to move and that can't be a good thing. Dynamic impact, bandwidth and traceability are all affected (limited).

So there is a very real argument for eliminating the use of loading resistors, which are a band-aid to a poor design anyway; the argument being that if you want the most mechanical performance out of the cartridge, don't ask it to do work! Just let it do the job of tracking the groove.

This means that the preamp **must** be immune to RFI, no if ands or buts, else it cannot be considered in the list of 'the best'...

On the bolded, this is one of the aspects in the upcoming Spectral Phono that has my juices flowing for quite some time - no loading.

 

[Atmasphere]

Ralph, thanks again for your cogent explanation. I appreciate your patience. It still seams a bit odd to me that many of us have spent so much time over the years walking the tightrope with impedance to manage rfi if rfi immunity (phonostage stability) is easily achieved with no sonic compromise, but I hear you. Thanks!

I think the problem is poorly understood in the industry! I know this is true as both Jonathan Carr and I spend a lot of time explaining how it works; Jim Hagerman had to write this excellent article http://www.hagtech.com/loading.html .... -And no-one reads it...

BTW it is a lot harder to make solid state stuff behave because the junctions of the input device can act like a rectifier. RFI can be particularly problematic in such cases.

On the bolded, this is one of the aspects in the upcoming Spectral Phono that has my juices flowing for quite some time - no loading.

I think you will find that it is loaded at 47K. That's the standard in the industry. That's how we load the cartridge (but we could easily load higher), although we do it in the balanced domain. You do get into a problem with higher impedance inputs- noise pickup and the like (and resonance issues if you are working with MM; loading MM cartridges is a whole 'nuther topic), so 47K is a pretty good compromise.

 

[ack]

I think you will find that it is loaded at 47K. That's the standard in the industry. That's how we load the cartridge (but we could easily load higher), although we do it in the balanced domain. You do get into a problem with higher impedance inputs- noise pickup and the like (and resonance issues if you are working with MM; loading MM cartridges is a whole 'nuther topic), so 47K is a pretty good compromise.

I'll measure it when I get it. Right now I can tell you it's got no loading options.

 

[awsmone]

I have to say I find this discussion refreshing
I have spent years playing with loading of mc and never satisfied
Then went to step ups and mm inputs bypassing the mc sections and now have much better sound no tick and pops little noise and huge dynamics
Moving to low cap ic was the next step
Each of these steps was an improvement
I didn't know why not using the loading options was better just I could hear it
Plus step ups give you far more tailoring of the sound than just loading resistors

I know there is a reflected impedance but I assume that since these are not active devices this might be different issue

 

[PeterA]

OK- you have this wrong. Reducing the circuit's susceptibility to RF also reduces the issues of loading at the same time.

The reason is that loading is done to affect the phono preamp and not the cartridge, unless you are using a high output MM cartridge. What the loading does is to reduce the Q of the tank circuit caused by the inductance of the cartridge and the capacitance of the cable. This circuit resonates when the cartridge makes energy. RFI is thus injected directly into the preamp! The loading resistor detunes the tank circuit and thus the RFI. But if the phono preamp is immune to RFI, the loading will make no difference.

'Stopping' used in the context I mentioned earlier will not affect bandwidth or dynamics; in fact if anything its the other way 'round, since what stopping does is it gets rid of oscillation. When that is gone, the circuit is lower noise which is good for dynamic contrast.

So if you see a preamp that requires loading, it means that its unstable, and can't possibly be the 'best' (if that exists in the first place...). Even when the loading is in place you'll still hear more ticks and pops than you should. That's not accurate reproduction!

Ralph, why do cartridge manufacturers give a recommended loading specification for their cartridges if loading is done to affect the phono preamp and not the cartridge? They don't know the capacitance of the cable that will be used nor the quality of the phono preamp. Are they assuming the phono preamp is unstable?

 

[gian60]

Dyssanayake is doing,after 50 years of experience, a tube phono preamp with external power supply,
where the signal,mm or mc, go directly to the tube,and he says is incredible sound also because the best transformer reduce the quality of signal,after tube,i cannot explain well,he put one resistance
to load the mc.
Here a photo with him that is building the power supply of the phono

 

[gian60]

And he use,he has a big stock,only component of years 50/60,he has a lot of trans Western Electric and Yamamoto and bought a lot of things that use in aviation and on the rocket,
bought in Aviano,a military center

 

[ack]

Ralph, why do cartridge manufacturers give a recommended loading specification for their cartridges if loading is done to affect the phono preamp and not the cartridge? They don't know the capacitance of the cable that will be used nor the quality of the phono preamp. Are they assuming the phono preamp is unstable?

A loading resistor will shunt part of the tiny cartridge current to ground - hence the higher dynamics with higher resistor values, as I show; and vice versa (see also comments below from Genesis) - because of how the resistor is connected (in parallel to the coil), so they probably advise from the cartridge perspective and with respect to that aspect of the picture; that resistor may also affect the frequency response; and I think that's why they usually recommend a value "greater than something" as opposed to a fixed value.

To do it right, a phono stage should really be designed with a phono cable together, and Genesis for example supply one http://www.genesisloudspeakers.com/gphono.html "The Genesis Phono-Preamp is designed from the base of the tone arm onwards, and a Genesis designed cable is supplied with a mini-DIN connector". The Spectral, I am sure, will be talking about MIT cables again.

Here's more from Gary Koh from the same link (and I also highlight what I said about dynamics):

There isn't a plethora of loadition options. The inductance of the coil of the cartridge interacts with the capacitance of the tonearm cable to create a reactive peak. This peak can be as high as 32dB, and is the usual reason given for all the loading options for a phono stage. However, when the graphs of this reactive peak are shown in the marketing literature of many phono stages, the frequency scale is often omitted. In computer simulations, we can easily see that this reactive peak is in the megahertz range – far outside our hearing. However, if this peak is present, it can easily cause problems for a phono-preamp. High loading (if used to tame this peak) also reduces dynamics and transparency.

 

[Pb Blimp]

A loading resistor will shunt part of the tiny cartridge current to ground - hence the higher dynamics with higher resistor values, as I show; and vice versa (see also comments below from Genesis) - because of how the resistor is connected (in parallel to the coil), so they probably advise from the cartridge perspective and with respect to that aspect of the picture; that resistor may also affect the frequency response; and I think that's why they usually recommend a value "greater than something" as opposed to a fixed value.

To do it right, a phono stage should really be designed with a phono cable together, and Genesis for example supply one http://www.genesisloudspeakers.com/gphono.html "The Genesis Phono-Preamp is designed from the base of the tone arm onwards, and a Genesis designed cable is supplied with a mini-DIN connector". The Spectral, I am sure, will be talking about MIT cables again.

Here's more from Gary Koh from the same link (and I also highlight what I said about dynamics):

And cartridge too since inductance (in addition to cable capacitance) is a critical part of the ultra high frequency propagation problem as well.

Ack, have you seen the engineering behind how Spectral is achieving RF immunity without some type of resistive damping which historically attenuates voltage (and therby dynamics). It looks like a free lunch.

 

[Jeffy]

I load my Aesthetix Io at 47K using the Miyahima Madake Cartridge. When I had the ZYX Universe II I had to load it to 480 ohms as it would be too thin.

 

[Atmasphere]

I'll measure it when I get it. Right now I can tell you it's got no loading options.

So you can't use it with a MM cartridge, as unlike LOMC cartridges, MM's need some loading to prevent distortion at audio frequencies, as well as to prevent resonant peak interactions with the cable. We provide a simple loading terminal for this.

I have to say I find this discussion refreshing
I have spent years playing with loading of mc and never satisfied
Then went to step ups and mm inputs bypassing the mc sections and now have much better sound no tick and pops little noise and huge dynamics
Moving to low cap ic was the next step
Each of these steps was an improvement
I didn't know why not using the loading options was better just I could hear it
Plus step ups give you far more tailoring of the sound than just loading resistors

I know there is a reflected impedance but I assume that since these are not active devices this might be different issue

The thing is, if you use a transformer, you have to load it properly to prevent it from ringing (which is basically harmonic distortion). Usually the correct loading is a combination of a resistance and a capacitance. The correct loading of the transformer will vary from cartridge to cartridge. The manufacturer should have information on how to do this.

Ralph, why do cartridge manufacturers give a recommended loading specification for their cartridges if loading is done to affect the phono preamp and not the cartridge? They don't know the capacitance of the cable that will be used nor the quality of the phono preamp. Are they assuming the phono preamp is unstable?

They have no idea what cable or phono section, and frankly I think some of them use unstable phono preamps as well! But since they can't be sure the phono section the customer uses is competent, they have to recommend a loading value that is generally going to be in the ballpark, assuming a one meter cable and low capacitance. But you will also note that nearly all of them also at the same time recommend 47K as a load as well!

A loading resistor will shunt part of the tiny cartridge current to ground - hence the higher dynamics with higher resistor values, as I show; and vice versa (see also comments below from Genesis) - because of how the resistor is connected (in parallel to the coil), so they probably advise from the cartridge perspective and with respect to that aspect of the picture; that resistor may also affect the frequency response; and I think that's why they usually recommend a value "greater than something" as opposed to a fixed value.

To do it right, a phono stage should really be designed with a phono cable together, and Genesis for example supply one http://www.genesisloudspeakers.com/gphono.html "The Genesis Phono-Preamp is designed from the base of the tone arm onwards, and a Genesis designed cable is supplied with a mini-DIN connector". The Spectral, I am sure, will be talking about MIT cables again.

Here's more from Gary Koh from the same link (and I also highlight what I said about dynamics):

The first statement in bold: The loading does not affect the cartridge at audio frequencies; bandwidth is unaffected.

The second statement in bold: This statement appears false. You don't have to worry about the cable as a preamp designer if the phono circuit is stable. If it is not, it might be important to include the cable so its effects are predictable, but in thinking about that I can't come up with a reason why since the cartridge inductance is still a variable. So my surmise is this is a way to sell a cable (there is an advantage of not having the mechanical connector, but its probably more convenient to have the cable integrated into the tonearm rather than the preamp as seen on the Triplanar; you get the same advantage). FWIW, if the cable is balanced and the phono section is balanced, the sonic artifact of the interconnect cable will be minimized to the point of inaudible, on account of the extremely low impedance balanced system that results from this setup.

 

[ack]

So you can't use it with a MM cartridge, as unlike LOMC cartridges, MM's need some loading to prevent distortion at audio frequencies, as well as to prevent resonant peak interactions with the cable. We provide a simple loading terminal for this.

Yes, you are right about MM, but we leave them out of these discussions because no one seems to be using moving magnet cartridges around here, nor do I know anyone who does. Capacitance is critical with MM.

The first statement in bold: The loading does not affect the cartridge at audio frequencies; bandwidth is unaffected.

I can buy that, though you may see folks on vinylengine and elsewhere depict alleged frequency response variations in the treble as a result of loading.... but I think they may be measuring the phono, rather than the cartridge.

The second statement in bold: This statement appears false. You don't have to worry about the cable as a preamp designer if the phono circuit is stable. If it is not, it might be important to include the cable so its effects are predictable, but in thinking about that I can't come up with a reason why since the cartridge inductance is still a variable. So my surmise is this is a way to sell a cable (there is an advantage of not having the mechanical connector, but its probably more convenient to have the cable integrated into the tonearm rather than the preamp as seen on the Triplanar; you get the same advantage). FWIW, if the cable is balanced and the phono section is balanced, the sonic artifact of the interconnect cable will be minimized to the point of inaudible, on account of the extremely low impedance balanced system that results from this setup.

Yes, if the preamp is stable, then what I said does not apply - the subject here is instability, though. And I don't think Genesis want to sell cable - Gary has written about what he does, on this forum, some time ago - anymore than Spectral selling MIT, which they don't.

 

[Atmasphere]

I can buy that, though you may see folks on vinylengine and elsewhere depict alleged frequency response variations in the treble as a result of loading.... but I think they may be measuring the phono, rather than the cartridge.

Back in the old days before I sorted this stuff out, I used to change the loading on my preamp until I got the highs right- IOW it did work as a tone control to a certain degree. But when I measured the response by ringing the cartridge, I found that bandwidth was unaffected until you got well above 1MHz. I consider that negligible in this case

But how the preamp behaves when powerful RFI is being injected into its input is a whole 'nuther thing! That is where the idea of changes in bandwidth seems to come from.