• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

changing (de)coupling cap values

Jump to Latest
K

Klimon

Member
2005-04-01 1:06 pm
Boutersem
  • #1
    Hello,

    I've just changed my amp's (cheap and crappy) input decoupling caps and coupling caps between El34's (in pp) and driver (6U8A) with paper in oil: in short: dramatical results, much clearer, lifelike sound, for the first time I can FEEL bassguitar: much better bottomend.

    I found some similarly entitled threads, but there never seems to be any conclusive, general answer...so I'll put my question: is there any use in augmenting the values of these caps? From previous threads (about input caps) I've learned that some say the higher you go, the better the bottomend, but your midrange sounds less good, others recommend a higher value without reserve. If you'ld recommend higher values, how much higher? (now the input caps are 0.2mf, the coupling caps are 0.47 each (=2/channel, PP).

    Another tweak I remember having read about, but can't find back: is it a good idea to remove the input decoupling cap + resistor, knowing that in my sources (cd, minidisk) there are output decoupling caps? Could I measure how much DC is present and then decide whether to remove them or not?

    Thanks in advance for your much appreciated reactions!!!:)
     
    analog_sa

    analog_sa

    Member
    Paid Member
    2002-08-14 1:47 pm
    Cascais
  • #2
    You seem to liberally intermix the terms coupling and decoupling. From what i understand you have only modified coupling caps so far.

    Depending upon the topology input coupling caps are generally not needed in tube designs not counting the odd fixed bias or grid leak.

    AS far as values go i always choose the smallest i can get away with. Good caps are generally expensive and i see no point paying a premium for bigger values. Yes, bigger caps inherently seem to have a more 'bassy' sound but it seems like a very expensive way to tone-correct.
     
    K

    Klimon

    Member
    2005-04-01 1:06 pm
    Boutersem
  • #3
    Thanks for the reply, analog. I think I'll remove the output caps from my sources then. Do you know if it's possible to 'overdo' it with the amp coupling cap values? I have plenty of Philips old stock pio caps, so money is not a criterium here. I know I can experiment, but a little advice from more experienced tweakers is always a blessing + I'm already spending my days soldering - listening - soldering - listening - ad infinitum!!!

    Greets!
     
    P

    percy

    Member
    2004-04-25 5:47 pm
    CA
  • #4
    Klimon said:
    Hello,

    I've just changed my amp's (cheap and crappy) input decoupling caps and coupling caps between El34's (in pp) and driver (6U8A) with paper in oil: in short: dramatical results, much clearer, lifelike sound, for the first time I can FEEL bassguitar: much better bottomend.

    I found some similarly entitled threads, but there never seems to be any conclusive, general answer...so I'll put my question: is there any use in augmenting the values of these caps? From previous threads (about input caps) I've learned that some say the higher you go, the better the bottomend, but your midrange sounds less good, others recommend a higher value without reserve. If you'ld recommend higher values, how much higher? (now the input caps are 0.2mf, the coupling caps are 0.47 each (=2/channel, PP).

    Another tweak I remember having read about, but can't find back: is it a good idea to remove the input decoupling cap + resistor, knowing that in my sources (cd, minidisk) there are output decoupling caps? Could I measure how much DC is present and then decide whether to remove them or not?

    Thanks in advance for your much appreciated reactions!!!:)
    Click to expand...


    f = 1 / (2*pi*R*C)

    Generally this is good estimate for the coupling caps, where R is the grid resistor of the next stage, f is lowest frequency you'd like to pass.

    Beyond this point (cap value) I'd say the Quality of the capacitor matters more than the value.
     
    K

    Klimon

    Member
    2005-04-01 1:06 pm
    Boutersem
  • #6
    I calculated the theoretical lowest frequency for input stage and output stage, with the standard C (and R) values... This results in 1.7hz and 3hz... I'm quite sure I didn't miscalculate, so those caps won't be blocking any low frequencies I guess... :bigeyes: Off to the next tweak! (diy stepped attenuator and better Q+higher value power supply caps+ bypass with pio) - thanks for sharing!!! :angel:
     
    K

    Klimon

    Member
    2005-04-01 1:06 pm
    Boutersem
  • #8
    I understand your first thread; I used the word blocking in another meaning (as: not letting through), just to avoid confusion...;) Going further on the blocking (your blocking:) , it would maybe be a good idea to go the other way around and to lower the C's values; hereby not influencing the lowend (as 3hz seems ridiculously low - maybe 20 hz is enough in the real world, margin included) but reducing the negative effect caused by blocking. All nice in theory, but will I really hear a difference? (...say yes and I might try it)
     
    K

    Klimon

    Member
    2005-04-01 1:06 pm
    Boutersem
  • #10
    Almost there...

    When removing the input coupling caps, should:
    a. the resistor to ground be also removed
    b. the resistor value be recalculated with the formula (supra)
    c. the resistor left unchanged?

    My electronic knowledge is only very basic, so I try to look at schematics from people whose ability I do not doubt, as diyparadiso.com here in Belgium (in general, he does not use input caps or resistor to ground, but there's always a resistor in series between input and cathode). Andrea Ciuffoli (audiodesignguide.com) doesn't use input caps either, nor a resistor in series before cathode, but always a resistor to ground. Different people, different styles?
     
    Yvesm

    Yvesm

    Member
    2004-07-09 6:23 pm
    Ardeche
    www.dissident-audio.com
  • #12
    Klimon said:
    Sorry, in my previous post I used cathode instead of grid. The technical terms for these resistors seem to be: grid stopper resistor and grid leakresistor (which I don't know what to do with)...
    Click to expand...

    Grid leak resistors are between grid and ground (or bias source).

    They are intented to fix the DC grid voltage and MUST exist.
    Sometimes, a volume control do the job.
    In low level stage, they are something between 100k to 1M.

    Grid stopper are in serie with grid.
    They prevents for some kinds of parasitics oscillation and are only added when needed.
    Theyre values are in the 500 to 10K range.

    Much more left to say about !!

    Yves.
     
    You must log in or register to reply here.
    Top Bottom