screen grid resistors?

Discussion in 'Amps/Cabs Tech Corner: Amplifier, Cab & Speakers' started by jzucker, Feb 12, 2006.


  1. jzucker

    jzucker Supporting Member

    Messages:
    20,267
    Joined:
    May 20, 2003
    Location:
    Home of the ex-world champion Cavs
    Do these make a difference in tone?

    I had a 6L6 amp with 1k screen grids and replaced them with 470s and I could swear the amp "opened" up after that. Was it my imagination?
     
  2. electroid

    electroid Member

    Messages:
    202
    Joined:
    Feb 12, 2006
    Location:
    Atlanta, GA
    The amp will produce slightly more power and may have a little more output gain with lower value screen resistors. However, the tubes will draw more current through the screen grid. If you install a tube with an improperly aligned sor under-rated screen grid, it could draw too much current and short out. You could in theory, operate the amp with even lower values than 470 ohms but it would be risky.
     
  3. jzucker

    jzucker Supporting Member

    Messages:
    20,267
    Joined:
    May 20, 2003
    Location:
    Home of the ex-world champion Cavs
    This amp is running plate voltages of 485 and I've currently got the bias set at about 25ma / tube. You think I'm pushing it with the 470ohm resistors?
     
  4. scottl

    scottl Member

    Messages:
    17,042
    Joined:
    Apr 11, 2003
    Location:
    Cherry Hill, NJ
    It is fine Jack. I have experimented extensively with 1k versus 470 ohm in 6L6 amps. First off all, 470ohm is the standard. You are not risking anything. Only if you use EL34 tubes is it an issue.

    My observation is that the 1k screen resistors have a definate squish to them that absolutely may be interpreted as less open or more compressed.

    I am currently running 470ohms in my protomules......

    You may want to bump that bias and see how she sounds. 25ma may be real low. Especially if measured with the 1ohm to ground method.

    Scott
     
  5. jzucker

    jzucker Supporting Member

    Messages:
    20,267
    Joined:
    May 20, 2003
    Location:
    Home of the ex-world champion Cavs
    Thanks Scott. I was measuring the bias current using the transformer shunt method. One tube is at 24.5, the other's at 28.5. I'm being a bit conservative because the plate voltage is fairly high compared to most of the fenders I've worked on. I'd like to knock it down to 460 or so. You ever use a zener diode to do that? I was thinking of possibly hard-wiring a weber copper cap in there...

    They're JJ tubes BTW. I've heard they can take higher plate voltages.
     
  6. scottl

    scottl Member

    Messages:
    17,042
    Joined:
    Apr 11, 2003
    Location:
    Cherry Hill, NJ
    Good idea on the dpdt switch....

    So, do you agree with my description of the tonal effect?? Some people think you can't hear it.

    Scott
     
  7. bob-i

    bob-i Member

    Messages:
    7,194
    Joined:
    Oct 16, 2005
    Location:
    Central NJ
    I prefer the 470ohm screens myself. Some people like the more compressed feel of 1K.

    The bias of only 25ma is very low like Scott said. I'd kick that up to about 70% dissipation. That may lower the B+ just a tad.
     
  8. jzucker

    jzucker Supporting Member

    Messages:
    20,267
    Joined:
    May 20, 2003
    Location:
    Home of the ex-world champion Cavs
    If I kick the low one up to 28-30, that'll put the high one at 33-35. Will that be ok with 490V on the plates?
     
  9. Matt H

    Matt H Guest

    as you increase the current draw, the plate voltage will drop some.

    just be sure to keep a calculator nearby for rough and ready dissipation checks. :)
     
  10. John Phillips

    John Phillips Member

    Messages:
    13,080
    Joined:
    Mar 17, 2002
    Location:
    Scotland
    I think you can hear the difference too - it's not huge, but probably real. Some older amps have no resistors at all, and they can sound good like that, but they tend to be quite hard on tubes so I always fit the resistors, and I use the lowest value (within reason) that works reliably - which is usually 470 ohms for 6V6/6L6 amps and 1K for EL34s.

    I don't think the type of resistor affects the tone at all - I always use wirewound resistors since they're the most resistant to heat, which matters since they're usually on the tube sockets, and the most robust during short-term overloads - which happens if a tube blows. I prefer the hard-glazed type, they're the toughest of the lot. 2.5W or 3W for 470s, 5W for 1Ks.

    I do NOT believe in the idea of using under-rated screen resistors as 'fuses' as is sometimes suggested to 'protect' other parts - the burning resistor can cause very serious damage directly or indirectly, and I honestly think it's one of the worst justifications for doing something in an amp I know of. Resistors are not fuses and should not be used as such because they are not designed to fail safely. If you want extra protection against tube shorts and don't think the main fuse is enough, fit the amp with a HT fuse.
     
  11. Tapp

    Tapp Member

    Messages:
    198
    Joined:
    Aug 12, 2005
    John, the big question is: what is your desired value for EL84's? I've read a few opinions and most seem to like the lower values for 84's ala 100R's like the Vox amps. I've got a homebrew 2-EL84 amp based on a Vox with a Fender tone stack placement, no neg feedback, Vox PI values, and I originally had high screen R's (like 2k2's). I've since come down to 1K's but I'm thinking of going lower.

    Tapp
     
  12. John Phillips

    John Phillips Member

    Messages:
    13,080
    Joined:
    Mar 17, 2002
    Location:
    Scotland
    I've never had any trouble with 100-ohm 1W. I've never experimented with higher values, although I came across some evidence that one of the new-production types needs much higher (I think it was from something written by Paul Cornford), like 680 ohms.

    I did have a funny one a month or so ago though... someone brought me an AC30 that he'd bought for £10 in blown-up condition, and had another tech fix it (one who's excellent at modern solid-state and digital gear, but has very little experience with tubes). It seemed to be working, but was very quiet. When I checked, I found that the other tech had fitted (to a beautiful standard of workmanship, with heat-resistant sleeving and very neatly) four nice shiny new screen resistors... 100K ones :). He must have misread the schematic, and not had enough 'tube amp instinct' to know it was the wrong value.

    So I can definitely say that 100K is too high :p.
     
  13. Tapp

    Tapp Member

    Messages:
    198
    Joined:
    Aug 12, 2005
    LOL John! Yeah that K will get you if you're not careful. Not the first or last time a K is mixed up with an R (or vis versa).

    Have you read any of Kevin O'Connor's books? He's a big proponent of higher screen grid R's. At the time of his first books the current production tubes probably needed the added screen grid protection, but I think the modern tubes have gotten much better in this area. JJ EL84's seem to be really decent and fairly robust for modern tubes.

    Tapp
     
  14. VintageJon

    VintageJon Member

    Messages:
    155
    Joined:
    Nov 3, 2005
    Location:
    Austin TX
    For EL84's 100 Ohms seems to be the standard and I've not seen any problems with it being used.

    JZUCKER- 35mA @ 490V is 17.15W of IPD. Max would be 17.5W, but we haven't figured screen current into the equation. It's likely to be 2mA or less with a 6L6. so you are probably OK. I couldn't recommend going any higher in cathode current though.

    -Jon
     
  15. jzucker

    jzucker Supporting Member

    Messages:
    20,267
    Joined:
    May 20, 2003
    Location:
    Home of the ex-world champion Cavs
    Thanks Jon. Is there a FAQ for the calculations ? I found a few charts but they didn't seem to take all the variables into consideration.
     

Share This Page