Phono Preamps and RIAA curve

Discussion in 'Audio Hardware' started by Magic, Feb 19, 2018.

Thread Status:
Not open for further replies.
  1. Magic

    Magic I'm just this guy, ya know? Thread Starter

    Location:
    Franklin TN
    I'm sure this is an over simplification but in thinking about this the other day I was wondering if the biggest reason phono preamps or the ones that "most" people like or that get the best reviews are due to their ability to compensate for the RIAA curve better than others. Not withstanding quality of components and not trying to start a debate of tubes vs solid state but just the ability to play back vinyl properly or more accurately. For instance why some sound more harsh, compressed, lacking mid range etc...
     
  2. chervokas

    chervokas Senior Member

    I'm not sure the differences in RIAA curve accuracy in modern phono pres are all that substantial. There are vintage tube circuits with some substantial departures from RIAA accuracy. But even the cheapest phono pres today are what +/- 0.5 to 0.3 dB accurate? I'm not sure these fractions of a dB RIAA eq deviations are big difference makers.
     
    2xUeL, SandAndGlass, amgradmd and 2 others like this.
  3. russk

    russk Forum Resident

    Location:
    Syracuse NY
    Phono Preamp can be measured for how accurately they follow the RIAA curve and are regularly in reviews. Most do pretty well, as mentioned above.

    Personally I think there's a lot more engineering/design expertise that goes into making something that equalizes and Amplifies such a small signal. Making them quiet and free from interference seems to give some designers apoplexy. So that and the voicing of the preamp of course are two of the biggest factors I think.

    Another reason I think you get some people loving a phono preamp and others hating it is that lots of people don't know how to appropriately match up a phono preamp to their cartridge and amp/preamp. Those that love a particular phono preamp either got lucky and found one that works and provides an appropriate amount of gain and correct loading that they need for their cartridge and amp/preamp or knew what they were doing and selected an appropriate unit. Those that dislike the unit didn't get an appropriate match and don't understand why.
     
  4. Magic

    Magic I'm just this guy, ya know? Thread Starter

    Location:
    Franklin TN
    I think I sort of overlooked that.

    Makes complete sense. Voicing is what I was trying to reason through.
     
  5. Wounded Land

    Wounded Land Forum Resident

    I think you are spot-on with this. This is something I didn't really understand myself until I got my Schiit Mani and started futzing around with different gain levels. As you may recall (because you helped me set the proper gain), I'm running an Ortofon 2M Blue through the Mani into an NAD integrated. I have it at the lowest gain setting (30 db) and it is spectacular. Running it at a higher gain completely changes the character of the sound...it's not just louder. It sounds like a different preamp.

    Of course, I have a pretty basic system. I can't even imagine the variables you guys with fancy cartridges have to deal with. :)
     
  6. Benzion

    Benzion "Cogito, ergo sum" Forum Resident

    Location:
    Brooklyn, NY
    I think it is because we all hear differently. I'd probably hate it at the lowest gain level - I nearly always like it at the highest gain level I can get away with, and not only because it gets louder, but because the sound is fuller and richer that way. The lower the gain level - the more lifeless it sounds to my ears.

    And it's not only us the consumers - the "professionals" also hear differently, and have their own preferences.

    KAB uses the figure of 325 mV as reference for desired output from the phono into the next stage.

    Quite a few others, including myself (although I'm not a "professional) think that a reference level should be 500 mV - i.e. regular combination of a 5.0 mV cart into 40 dB gain (or a LOMC after step-up).

    Tavish, IIRC, used the value of 700 mV as reference for Adagio, methinks - more than twice KAB's figure!

    There really isn't one objective truth to it - it's all based on preferences, I think.
     
    Last edited: Apr 28, 2018
  7. There are a bunch of different ways to configure filters to do RIAA. Each one of them sound a little different. Within the different filter configurations, different passive components tilt the sound different directions. One could easily spend a lifetime on just this without even getting into how the active amplification comes into play.
     
    Done A Ton and Benzion like this.
  8. BayouTiger

    BayouTiger Forum Resident

    Was not the RIAA curve intended from day one to be a simple to implement EQ that could be accurately implemented by even a very simple preamp? Although I am certain that the implementation can definitely impact the SQ, I would not think it should any more than any other aspect of the design.

    I honestly think this is another are where folks today are way overthinking it when it comes to ny aspect of vinyl.
     
  9. TarnishedEars

    TarnishedEars Forum Resident

    Location:
    The Seattle area
    I think that RIAA accuracy is one of the lest significant factors in modern phono preamps since most deviate so little from the specified curve. Loading and gain certainly do play significant roles here as suggested above. However that alone does not explain why circuits with identical gain, identical loading, and essentially identical RIAA curves can sound quite different from one another.
     
    Benzion likes this.
  10. John Dyson

    John Dyson Forum Resident

    Location:
    Fishers, Indiana
    There are a couple of things about the RIAA curve and doing it correctly that a neophyte engineer might get wrong. Most important, beyond a dB or so error on the curve, is the cartridge match. Some cartridges have high inductance and/or odd characteristics that need a tight design on the frontend. Low capacitance on the frontend isn't always good depending on cartridge. A proper capacitance match is a good thing (I am assuming the traditional cartridge type -- high impedance.) Also, the front end should have low input current noise for traditional cartridge types, and very low input voltage noise for moving coils (since they have very low impedance, but very low signal voltage -- so a low resistance noise match is okay.) So, for the standard cartridges, it is a good idea to provide at least a couple of input capacitance variations. For the RIAA curve itself, there are some 'good' ways to do it, and some 'not so good.' The RIAA curve is often a heavy HF load for the node driving the feedback. It is important to have reasonbly good current drive into the TRADITIONAL RIAA feedback network. The simple two transistor circuit that had been traditionally used really had trouble doing the right thing. I tend to either like some of the better ICs (better does not necessarily mean expensive), or a well designed discrete frontend. Also, the traditional RIAA network shouldn't be taken for granted -- make sure that a good worst case analysis is done -- considering the drive impedance and the load impedance on the network. Make sure that you have 1% or better (metal film) resistors when needed, and good quality precision capacitors. The network doesn't have to be crazy accurate, just accurate.

    With the wrong circuit, there can be a lot of hidden distortion sources (the RIAA gain curve de-emphsizes HF -- partially hiding some kinds of distortion.) Make sure that the circuit design is never driving into nonlinearity. Nonlinearity can be made worse by the input stage impedance flopping around depending on the dynamically changing input stage operating current. Likewise, there can be distortion caused by the load of the RIAA network -- the driving impedance will most likely not truly be zero -- even for an opamp, and the drive current is always limited.

    There are so many caveats -- the design of any gain stage needs to be done from scratch, not assuming anything from a cookbook -- always study the freq resp, the distortion (both due to nonlinear gain and nonlinear impedances on input, network drive and output), and the noise (both open circuit and with an actual cartridge on the input.) I kind of forgot about the output impedance and sourcing ability -- but that kind of goes along with distortion also.

    John
     
  11. Just like the above post said: Good RIAA is not so simple. Sure you can build a circuit cookbook type active topology and get ok results but it won't knock anyone's socks off. If any frequencies deviate more than a half db I take a good hard look at the filter. Also most of the standard 1% resistors are plain sounding so more esoteric types have to be evaluated. Sometimes, at first seemingly unlikely components are the magic ones that bring a filter circuit alive. Be ready to hand match capacitors. Then there is the 4th pole, another wildcard that defies convention.

    Take my advice and stay away from 100% active RIAA.
     
    ctgregory likes this.
  12. SandAndGlass

    SandAndGlass Twilight Forum Resident

    In reading specs of phono preamps of modern tube design, most all of them indicate having a passive RIAA EQ.
     
  13. Agitater

    Agitater Forum Resident

    Location:
    Toronto
    I think that @chervokas said it well. RIAA compensation is essentially a known, fixed factor. Variation in passive RIAA EQ from one phono preamp to another is due as much to the variable tolerances in parts selected for any particular phono preamp as much as it is to the care and attention to quality that the particular phono preamp designer pays to his work. Slight variations from preamp to preamp are lost in the music.

    Voicing a phono preamp is another matter. I think that most phono preamp designers strive to extract from an LP groove exactly what was pressed into it. The only limitation for any particular PP designer are the restrictions imposed by his R&D/production budget and the target retail price. What that affects most of all is the noise floor of the PP and the power supply for the PP. The most expensive PPs are almost always the absolute quietest. The greater thing is that a remarkable number of entry-level and mid-priced PPs achieve remarkably quiet backgrounds too, most of which can be traced back to a designer who made the decision to pay attention to the choice of power supply.

    The difference between a moderately quiet, basically listenable PP and one of the better quality mid-to-high priced PPs cannot be emphasized too much and can often be astonishing merely because of the quieter noise floor. After that has been achieved, factor in the attention paid by the designer to timbral accuracy (through circuit design, choice of electronic parts) and related issues and we end up with different sound at different price points.
     
    H8SLKC, punkmusick and SandAndGlass like this.
  14. The Tavish Adagio and the Grado PH1 are both full active. Those are two that come to mind off the top of my head.

    Parasound JC3 is one pole of active (for the bass basically) and one pole of passive for the highs.
     
    H8SLKC likes this.
  15. allied333

    allied333 Audiophile

    Location:
    nowhere
    I believe the better sounding preamps uses minimal amplification stages with negative feedback. The use of cathode followers, etc degrades the sound. I prefer a simple two tube preamp with 12AX7s or 6922 tubes. The 12AX7 has most varieties of tube manufactures for tuning the amplifier sonics.
     
    H8SLKC and SandAndGlass like this.
  16. Davey

    Davey NP: Portishead ~ Portishead (1997)

    Location:
    SF Bay Area, USA
    Not a big deal, but I believe the Adagio has one passive stage and one active, which according to the designer, is for better overload margins.
     
    Last edited: Apr 30, 2018
  17. You are correct! I skimmed the Stereophile article before but went back for closer read. Actually Tavish uses the same topology as the JC3, passive high cut filter feeding active bass filter.

    Also noticed the price of the Adagio took quite a jump since that 2016 article :eek:
     
    SandAndGlass likes this.
  18. Benzion

    Benzion "Cogito, ergo sum" Forum Resident

    Location:
    Brooklyn, NY
    What hasn't?
     
    morinix likes this.
  19. Davey

    Davey NP: Portishead ~ Portishead (1997)

    Location:
    SF Bay Area, USA
    Yea, it did, though it was priced kind of low for introduction in that 2016 time frame at $1490, so they had to bump it up the next year to $1690, and then this January to $1790. You've been pretty fortunate (or maybe forethoughful) to keep your preamp in steady production for a few years without any price increases, isn't that right?
     
  20. We had to raise shipping by $2. :hide:
     
    TheVinylAddict likes this.
  21. TheVinylAddict

    TheVinylAddict Look what I found

    Location:
    AZ
    $2?

    Highway robbery. How do you sleep at night? :)
     
    morinix likes this.
  22. TheVinylAddict

    TheVinylAddict Look what I found

    Location:
    AZ
    Therefore, according to this - stay away from these two? (although I think later in the thread you discovered Tavish is not Full Active?)
     
  23. The postmaster made me do it! It’s all his fault :D
     
  24. Half active topology is not bad. With the high frequency filter being passive the worst effect is avoided. Tavish Adagio is passive for the high frequency. Davey corrected me on that.

    All this active/passive stuff is my, and others’, opinion. Plenty of EE designer types will disagree with me. I’ve had some heavy exchanges on Diyaudio about this.
     
    TheVinylAddict likes this.
  25. TheVinylAddict

    TheVinylAddict Look what I found

    Location:
    AZ
    OK... thanks. Not trying to put you on the spot... well, maybe I am... :)

    Understood it is not something that is etched in stone, but happy to hear opinions. Thanks....
     
    morinix likes this.
Thread Status:
Not open for further replies.

Share This Page

molar-endocrine