Coupling capacitor values vs. low frequency response question ?

Discussion in 'Tube Audio' started by oldschoolpunk, Apr 21, 2005.

  1. oldschoolpunk

    oldschoolpunk dog boy

    Messages:
    13
    Location:
    Texas
    As my first venture in a long time back into tubes I am recapping a small 6bq5 se. The coupling caps between the driver (12ay7) and the 6bq5 are only .02uf@1000vdc.
    In searching the web I found a lot of old amps that use a smilar small value coupling cap for both hifi and guitar ( my amp is hifi). However I found this comment
    at

    http://www.harpamps.com/weber/Guitar51.htm

    Am I right that in the lower the frequency the greator the voltage drop across the coupling capacitor ( in effect increase the lowend rolloff with the cap) :thumbsdn: ? If I raise the value of the coupling cap could I increase low end response and conversly at the same time could I use it to trim any 60 cycle hum rather then a choke :thmbsp: ?

    Man, and I thought I had it tough when I was building speakers :scratch2:
    Tony
     
  2. Tom Bavis

    Tom Bavis Audiophool Subscriber

    Messages:
    9,466
    Location:
    Macedon NY
    Remember that flat frequency reponse without distortion is NOT a requirement for an instrument amp... increasing the coupling cap may allow driving the output transformer into saturation, adding a desired "flavor" to the tone...

    Often small amps had small coupling caps for just that reason - a large value would give more bass, and distortion in the bass. gets eve more complicated if the amp uses feedback...
     
  3. Fisherdude

    Fisherdude Regular Dude - Super Mod Super Mod Subscriber

    Messages:
    27,738
    Location:
    Out West.
    It's best to assume that the designer of this amp was not ignorant of the cause/effect relatonship of changing capacitor values.

    While all components are built to a price point, that didn't usually extend to reducing the value of a coupling cap to the point where bass response was degraded. If you want to experiment with different types of audiophile caps, or slightly different values, give it a go. However, if you make radical changes, i.e. extending to decimal point/factor of ten changes, you could be asking for problems, like oscillation.
     
  4. oldschoolpunk

    oldschoolpunk dog boy

    Messages:
    13
    Location:
    Texas
    Thanks guys!

    I had not taken in to account that even though the output tubes could handle it that it could drive the transformers into over saturation. Thankfully caution won when I place the call for parts today and I went with orginal values. I did order two sets of values for the cathode bypass caps. Currently they are 50uf and I am going to pop them up to 470uf (across a 150 ohm resist)
     
  5. NOSValves

    NOSValves AK Subscriber Subscriber

    Messages:
    3,415
    Yes raising the value will increase low end response to a point. But there are limiting factors in the tube itself in SE operation and then the transformers saturation point. Most likely you wouldn't want to raise it much. If you go to much the amp will start motorboating. This is something that you can just simply play with to your liking trust your ears.

    Craig
     
  6. GordonW

    GordonW Speakerfixer Subscriber

    Messages:
    19,310
    Location:
    Marietta/Moultrie GA USA
    Also, note that many tube amps have feedback circuits that ROLL OUT the feedback at the lowest frequencies (that's what the RC circuit in the feedback circuit does) to a lower level than the "passband". This will, to an extent, "cancel out" some of the low-end-truncation effect of a small coupling cap, by somewhat INCREASING gain (feedback reduces gain, less feedback == higher gain) at the bottom end.

    So, in some cases, increasing the coupling cap won't only extend the bottom end... it can sometimes cause a "hump" or peak in the low-end frequency response down low... a sort of "one note boom" at a very low freuqency. Don't let this stop you from experimenting... but do be aware, if this type of behaviour takes place, this is probably WHY it's happening...

    Regards,
    Gordon.
     
  7. Thatch_Ear

    Thatch_Ear Addicted Member

    Messages:
    7,786
    Seems like a good idea to me. Way out of my league but always interesting to get some simple reasons why things work.

    One guy told me about getting better bass response by using a higher value in uFs lytic on the cathode to ground in SE amps and maybe a couple of years later another guy who also designs and builds gear told me that it was acting like a low pass. OK, I still don't know why it would act like a low pass but it is another step past,"If you do this you can get better bass."
     
  8. Mike Stehr

    Mike Stehr Poverty Audio

    Messages:
    1,480
    Location:
    East Wa.
    2 x 3.1416 x R x C = F? Where R is the grid resistor, C is the chosen cap value, and F is the break point frequency.

    I can't recall if it's 30% below the fundamental frequency......
    Say you what the roll-off at 30 Hz. You would have the break point frequency at around 3 Hz.

    I'm not sure if this is right, and I was dealing with a PP 6BQ5 amp at the time when I used this equation.

    Thatch,

    On my SE Magnavox 6BQ5 amp, I increased the shared cathode cap on the output tubes to a 100 uF or something. It does work like a bass boost, and it helped the little amp out on the LF response.
     
    Last edited: Apr 22, 2005
  9. wallacefl

    wallacefl Active Member

    Messages:
    276
    Just a follow up question from a tube noobie...I am thinking of building a Grounded Grid preamp and note the standard coupling capacitor is 1uf...would replacing these with a 4.7 be too much in your opinion?
     
  10. Kegger

    Kegger Anything can be S "MODed" Super Mod Subscriber

    Messages:
    12,226
    Location:
    Michigan,warren
    We would need to see the whole design.
    (you realize you posted to a 7 year old thread) :scratch2:

    I would suggest start a new thread on the question.
    But if the cap in question is the output coupling cap you should be fine.
     

Share This Page