Adjusting gain , Plate resistors vs. cathode resistors ?

Discussion in 'Amps/Cabs Tech Corner: Amplifier, Cab & Speakers' started by ChrisGS, Mar 9, 2006.


  1. ChrisGS

    ChrisGS Member

    Messages:
    154
    Joined:
    Jul 20, 2005
    Is there a difference in tone from adjusting the plate resistors vs. adjusting the cathode resistors ?

    Thanks ,
    Chris .
     
  2. ChrisGS

    ChrisGS Member

    Messages:
    154
    Joined:
    Jul 20, 2005
    Does increasing the value of the B+ dropping resistors have pretty much the same effect as increasing the value of the plate resistors at the preamp plates ?

    Thanks ,
    Chris.
     
  3. scottl

    scottl Member

    Messages:
    17,042
    Joined:
    Apr 11, 2003
    Location:
    Cherry Hill, NJ
    No....

    The ratio of plate to cathode sets the bias of the tube. A larger plate resistor (or smaller cathode resistor) will cause the tube to draw more current and increase gain. The voltage measured at the plate will be lower. Reversing this will cause less gain and higher plate voltage. Sometimes you increase both resistors. This slightly increases gain and harmonic content. Bias is still "tube handbook" values but the whole thing is scaled up.

    Just lowering the B+ will change tone a little but the tube bias is still the same.
     
  4. billdurham

    billdurham Member

    Messages:
    293
    Joined:
    Dec 29, 2005
    What he said! I think one thing that needs to be qualified is if you are refering to "Gain" as distortion. I have mistakenly associated the two as the same. At least with regard to tube amps. Part of my mistake is comparing tube circuits to opamp circuits..where gain is a function of feedback and where distortion is not reached until the gain causes the output signal to exceed the voltage swing of the DC rails. My assumption with tubes was that gain/distortion was strictly a voltage thing..higher plate voltage, more distortion......wrong! There is more gain with higher plate voltage...but there is less distortion. The way to get more distortion is to lower the plate voltage on the first tube...but you get lower gain.
    Scott...correct me if I am wrong.

    BD
     
  5. WailinGuy

    WailinGuy Member

    Messages:
    1,268
    Joined:
    Aug 30, 2003
    Location:
    Silver Spring
    Actually, a larger plate resistor will cause the tube to draw LESS current. But, because there is a larger voltage drop across a larger value resistor, you will obtain a larger signal (more voltage swing) off the plate.

    In general, you will want to increase both resistors (plate and cathode) if you want more gain or a fatter or darker tone from the gain stage. Typical values you find in guitar amps are:

    100K, 820 (Marshall - normal channel input stage)
    100K, 1.5K (Fender - all preamp stages)
    100K, 2.7K (Marshall - bright channel input stage)
    220K, 2.2K (Vox - input stage)

    Don't forget that the cathode bypass cap has a lot to do with the gain and tone of the tube stage.

    In general, you want to bias the gain stage so that its not too unbalanced. In other words, a sine wave input signal shouldn't cause the tube to reach saturation well before it reaches cutoff, or vice versa. It should be idling somewhere near the middle of the two extremes in order that the signal gets clipped on both the bottom and the top of the waveform. On the other hand, a certain amount of asymmetry (but perhaps not too much) is often desirable in guitar amps, since this helps generate even-ordered (musical sounding) harmonics when the tube starts to clip. That is why the pair of resistor values chosen don't necessarily conform to textbook or hi-fi specifications.
     
  6. scottosan

    scottosan Supporting Member

    Messages:
    363
    Joined:
    Nov 29, 2003
    Location:
    Arlington, Texas
    To my ears, increasing the plate resistors early in the preamp, seems to give the amp a more compressed and softer feel. Does this equate to gain, yes, but to me its more of a feel, while juicing gain at the cathodes tends to be more sensative to the amount of gain per small change.
     
  7. tybone

    tybone Member

    Messages:
    1,079
    Joined:
    Oct 3, 2004
    Location:
    Ontario
    We have a right proper geek fest here lads. I guess adding a bypass capacitor across the cathode resistor will simply change the cathode resistance (and gain) in a frequency dependant way?
     
  8. billdurham

    billdurham Member

    Messages:
    293
    Joined:
    Dec 29, 2005
    If you look at the Marshall 1987 and Fender AB763 preamps..V1 in particular...Fender has a 25uf cathode bypass on both sides of the tube, and from what I have heard..this provides enough freq response for most every note that a guitar can produce. Marshall on the other hand uses a 330uf cap on the normal channel...which extends the low end down to damn near DC and a .68uf cap on the bright channel..which rolls off the low end pretty drastically. I'm not sure how the cathode bypass cap results in gain...maybe its the resistor, by way of bias that is determining the gain??

    BD
     
  9. VaughnC

    VaughnC Supporting Member

    Messages:
    14,952
    Joined:
    Jan 4, 2002
    Location:
    Pennsylvania
    Being that the signal being amplified ends up going through the cathode resistor, the cathode resistor lowers the AC gain as well as setting the bias. So, while bypassing the cathode resistor with a capacitor restores the gain, a capacitor's reactance is inversly proportional to frequency: Xc=1/6.28*freq*capacitance. So, the cathode caps & interstage coupling caps are areas where you can tune an amp to your personal taste.
     
    jaysamson and Balok like this.
  10. VacuumVoodoo

    VacuumVoodoo Member

    Messages:
    1,566
    Joined:
    Jan 9, 2005
    Location:
    Gothenburg, Sweden
    Lower supply voltage & lower plate resistor value--> lower gain, less headroom, higher distortion

    Higher supply voltage & higher plate resistor value-->higher gain, more headroom, lower distortion.

    Tables on page 3 in the datasheet should be rather selfexplanatory

    http://www.mif.pg.gda.pl/homepages/frank/sheets/010/e/ECC83.pdf

    Alex
     
  11. scottl

    scottl Member

    Messages:
    17,042
    Joined:
    Apr 11, 2003
    Location:
    Cherry Hill, NJ
    http://www.tone-lizard.com/Mods_and_Odds.htm

    Check the link out. There is an informative chart regarding the relationships.

    I misspoke a little.... I meant current starving when referring to increasing the plate....
     
  12. ChrisGS

    ChrisGS Member

    Messages:
    154
    Joined:
    Jul 20, 2005
    Hey guys ,
    Thanks a lot for all the replies . To be honest with you , I'm a little too drunk to read this stuff tonight , but tomorow after some coffee , I can try to get some of this stuff in my head .

    Thanks again,
    Chris .
     
  13. billdurham

    billdurham Member

    Messages:
    293
    Joined:
    Dec 29, 2005
    Alex,

    thanks for the link...very good info. It would be really cool to setup a jig such that you could substitute caps and resistors and hear what diff they make.

    BD
     
  14. Wakarusa

    Wakarusa Member

    Messages:
    1,477
    Joined:
    May 7, 2004
    Location:
    Newnan, GA
    Yeah. You have to evaluate the circuit DC characteristics to determine the bias point of the tube (and I disagree with an earlier post... changing B+ while holding plate and cathode resistors constant will change the bias point -- i.e. the amount of current the tube draws with no input signal). So, from a DC perspective, the cathode bypass cap looks like an open circuit.

    When we flip around and do AC analysis on the circuit, the cap value does matter and determines the impedance (and therefore current flow) seen at different frequencies. So a "fully bypassed" cathode looks like a short circuit to AC frequencies in the range of interest (typically around 50Hz to around 6KHz). An "unbypassed" cathode has no cap, so the AC frequencies see the full cathode resistor. "Partially bypassed" means that the bypass cap is tuned to act essentially as a high-pass filter.

    Related tidbit -- a common complaint when folks add capacitance in an oscillator circuit to slow down a fast tremolo is that the tremolo sounds weak when turned slow. This is usually because the oscillator driver will have a fully bypassed cathode at audio frequncies (say 25uF), but only partially bypassed when the oscillator frequency (5-7Hz maybe?) is taken into account. The solution is to increase the size of the bypass cap :)
     
  15. donnyjaguar

    donnyjaguar Member

    Messages:
    4,234
    Joined:
    Nov 4, 2005
    Location:
    Canada
    Here are some of the rules-of-thumb that I live by. These can be applied to whatever amplifying device you're using:

    More current = more linearity = lower output impedance = less gain = more heat = shorter lifespan

    Less current = less linearity = higher output impedance = more gain = less heat = longer lifespan

    As for the bypass capacitor value, this isn't a "try it and see" value, it can be calculated: 1/2PIxFx(R/10) where F is the lowest frequency you want to amplify and R is the cathode resistor. This will set the -3dB point at the frequency of choice. Things get interesting when you are running push-pull circuits with a common cathode resistor. Because the frequency across the resistor is twice the output, you only need to bypass it to twice your lowest desired frequency. This equals a lower value (and cost).

    You can also use this calculation (if you're clever) to determine minimum capacitor values in your power supply. You can also use it to set up a pentode voltage amplifier for the right gain characteristics. What's more it can be used for bypassing the resistor in your NFB loop so that the amplifier won't start screaming at 100kHz when the phase through the transformer starts to change.

    Now before you go and change the 25uF in your Fender (43Hz w/1k5) to 100uF (10Hz w/1k5) to improve the bass, keep in mind one thing. Tube output transformers don't do that great a job of converting bass frequencies. The general rule of thumb is to keep the very low bass out of them altogether. If you don't, it'll sound like crapola and possibly cause the dreaded motorboating or machine gunning that we've all heard at one time or another. I always use 66Hz as the lowest frequency I want to amplify in a guitar amp. Hell, a bass guitar's open E is 66Hz so why go lower than that?

    Group delay is another thing to keep in mind going through an amplifier. I won't get into that unless someone really cares.

    Now you have some of my tricks.

    DJ:RoCkIn
     
    finom1 likes this.
  16. Shea

    Shea Member

    Messages:
    540
    Joined:
    Sep 7, 2004
    All true, but it should be mentioned that the reason why the bass gets flubby in a Fender when you crank it is because the amp circuit as a whole passes too much bass, not too little. So if you tried to fix it by increasing the size of a cathode bypass cap, you'd be going in the wrong direction. You'd want to reduce it. Some people tame the bass in Fenders by replacing one or two of those cathode bypass caps with 10, 5, or even 1 uf bypass caps. It works.

    But IME, once you've cut enough bass out of a Fender to keep it from flubbing at high volume, then you'll have lost that big & deep clean tone. So, it's a tradeoff. If you keep pushing the gain and cutting the bass in a Fender, you'll end up with something that sounds too much like a Boogie.

    Shea
     
    finom1 likes this.
  17. Wakarusa

    Wakarusa Member

    Messages:
    1,477
    Joined:
    May 7, 2004
    Location:
    Newnan, GA
    Well, that and a really soft power supply ;)
     
  18. RL in Fla

    RL in Fla Member

    Messages:
    3,198
    Joined:
    Feb 2, 2003
    Location:
    Future Cleanup Site
    B+ caps or tranny sag Todd ?
     
  19. Wakarusa

    Wakarusa Member

    Messages:
    1,477
    Joined:
    May 7, 2004
    Location:
    Newnan, GA
    Yes :)

    If I had to pick one over the other I'd go with poor regulation in the power transformer (tranny sag) as the greatest contributor.
    A while back I was showing around a prototype (that managed to get named "Bob". Don't ask.) that folks:
    - swore was 100W
    - thought would make a nice bass amp
    - kept looking in the back to count the tubes

    In fact, it runs a pair of 6L6GC in AB1 fixed-bias with around 450VDC on the plates. On a calibrated wattmeter it maxes out at around 42W. So where does the huge bottom end come from?
    - An oversize PT that, in this circuit, has about 3% regulation
    - Relatively heavy filtering (though not enough to make it sterile or too stiff)
    - Relatively high voltage in the preamp
    - A power supply topology that decouples each and every gain stage from the others.
    - Some modest regulation of the screen supply.
    - Good speakers
     
    finom1 likes this.
  20. RL in Fla

    RL in Fla Member

    Messages:
    3,198
    Joined:
    Feb 2, 2003
    Location:
    Future Cleanup Site
    Thanks as always (to all involved , BTW :AOK ) .

    My "sort of" suspicion was tranny also , because the 5F Allen I built had bass to burn (for a single 6V6) and his b+ caps were smaller values than the AA764 Champs/Vibros , but his trannies are mo'fo's ;) .

    Got (what turned out to be) a hulk Champ (worked but had rusted octal sockets & some other black tape/"wirenut wizardry" inside :rolleyes: ) that's stripped down to nuttin' but 'al-you-minimum' & I'm gonna try a 5F/764 amalgamation on it .
    I'll use the Allens & save the 764 PT & OT for a rainy day thang . :p :D
     

Share This Page