Connect with us

Very low noise SINGLE audio op-amp in DIL-8 ?

Discussion in 'Electronic Components' started by Eeyore, May 15, 2007.

Scroll to continue with content
  1. Eeyore

    Eeyore Guest

    There are lots of really good op-amps suitable for or specifically designed for
    low noise audio use but the most popular configuration is the dual package (for
    many good reasons).

    Having established that my client for the refurbished 'boutique' mic pre-amp
    doesn't mind if it's non-original provided there's an improvement, does anyone
    have any favourite candidates to replace the venerable NE5534 ? Bipolar or fet
    input, I don't mind, the noise source impedance at the op-amp non-inverting
    input is ~ 1200 ohms.

  2. Eeyore

    Eeyore Guest

    It's retrofit into an existing socket.

    The 5534's en is ~ 3.5 nV/sqrt Hz btw.

  3. Why not just use half of a dual you know and love?

    Best regards,
    Spehro Pefhany
  4. John Larkin

    John Larkin Guest


  5. Fred Bartoli

    Fred Bartoli Guest

    Eeyore a écrit :
    Why not just keeping it?
    At 0.4pA/rtHz, the 5534's current noise will be unnoticable.
    And since 1200R is 4.5nV/rtHz, total noise will be 5.7 (for a 2dB NF).
    Getting down to a total 5nV/rtHz will require a less than 2nV/rtHz
    opamp, which you'll probably don't like (cost).
  6. Eeyore

    Eeyore Guest

    A blast from the past ! I must look at it again and there's the op-37 too IIRC.

    The best I've found so far is National's LME49710 (I didn't realise they did a single
    part) which boasts typical mid-band en of 2.5 nV/sqrt Hz and stupendous linearity and
    wonderful DC figures too. I expect In will be better as well.

  7. Eeyore

    Eeyore Guest

    Ok, I have some samples on order, Farnell doesn't stock them yet. This part actually
    smacks of a deliberate intention to make a form-fit-function equivalent upgrade for 5534s
    ( and 5532s with the LME49720).

    AD have better of course but it's a dual in SOIC. It's tempting to make a converter board
    and 'throw away' the unused half !,2877,AD8599,00.html

    1.15 nv/sqrt Hz op-amps ! -105dB THD @ 20kHz ! Yummmmm. And they're not actually insanely

  8. qrk

    qrk Guest

    LT1028 is around 1nV/rtHz above 100Hz.
  9. Gareth

    Gareth Guest

    What about AD797 or AD829?

    Or LT1028,C1,C1154,C1009,C1026,P1234,D3480

  10. Isn't the AD797 supposed to be the "ultimate" audio opamp from AD?
  11. John Larkin

    John Larkin Guest

    LT1028 is a fabulous amp *if* your signal is very low impedance. We
    use them as current-shunt amps with ppm precision. The feedback
    resistor current is so high that we put another, cheap opamp in the
    loop just to drive the feedback, so we don't heat up the 1028 and
    spoil its beautiful offsets.

    Current noise is high.

    Maxim did a MAX1028 for a while, but I think LTC stopped them somehow.

  12. Phil Hobbs

    Phil Hobbs Guest

    Probably by ordering some, which always causes Maxim to discontinue parts...


    Phil Hobbs
  13. ISTR reading that it tends to go quite unstable,( layout, gravity,
    leylines, Hubble const. 440Hz, etc,) but I havent checked it out on

  14. Guest

    I've never used that AD797, but I suggested it once as a possible
    alternative to the LT1028, and the guys that tried it were very happy.
    The output stage of the LT1028 can be a bit cranky with some loads,
    and the AD797 is apparently rather nicer.

    It is a relatively fast op amp, good for 20V/usec output slew rate and
    the data sheet does emphasise the need for good decoupling and careful
    layout, so there are people around who could find it unstable.,2877,AD797,00.html
  15. Carl Ijames

    Carl Ijames Guest

    I've used the Analog Devices AD829 as the second stage in a preamp and
    it worked well. Specced at 1.7 nV/rtHz. They also have the AD797 at
    0.9 nV/rtHz if it's fast enough for you. You can sort their op amp list
    by noise if you haven't already found that magic button.
  16. Boris Mohar

    Boris Mohar Guest

    If you can live with +/-5 V try MAX412
  17. MooseFET

    MooseFET Guest

    LT1027 is very like an op-27
  18. MooseFET

    MooseFET Guest

    On May 15, 4:10 pm, wrote:
    The LT1028 can be tricky near unity gain. They really hate being
    smacked into the rails and don't like hitting slew rate limit. Other
    than these things they are actually not too bad to work with.

    Since this is high end audio, you may want to look at parts like the
    LT6230-10. These give better highs.
  19. MooseFET

    MooseFET Guest

    And below about 100KHz. The noise rises in the 100K-1M band.
  20. ["Followup-To:" header set to]
    Thanks, Phil, you made my day.

Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day