cathode switching vs B+ (plate)switching

Discussion in 'Amps/Cabs Tech Corner: Amplifier, Cab & Speakers' started by kctx2, Oct 12, 2005.


  1. kctx2

    kctx2 Member

    Messages:
    114
    Joined:
    Aug 13, 2005
    To switch between an in-circuit 6V6 and an El84 (for sound testing differences), which is the preferred/best method. Use a SPDT switch to switch between cathodes (cathode resistor to ground) or between plate B+.
    Each tube has its own cathode resistor and grid resistor.

    Thankyou, for your time and your help.
    Keith
     
  2. VacuumVoodoo

    VacuumVoodoo Member

    Messages:
    1,566
    Joined:
    Jan 9, 2005
    Location:
    Gothenburg, Sweden
    Applying B+ to plate with no defined voltage on cathode may cause arcing between plate and grids and damage the tube. I would recommend switching B+ to the plates.
     
  3. John Phillips

    John Phillips Member

    Messages:
    13,080
    Joined:
    Mar 17, 2002
    Location:
    Scotland
    I would have thought so too... but most or all amp designs I know of which have the facility to turn off some of the tubes do it via the cathodes - including the Mesa Blue Angel, which does exactly this with 6V6s and EL84s. It may be that they don't want to switch the B+ since that would require two poles, as it's a push-pull amp - switching the cathodes only needs one. I assume you're talking about a single-ended though, so it should make no difference. The only other consideration I can think of is the voltage rating on the switch.
     
  4. VacuumVoodoo

    VacuumVoodoo Member

    Messages:
    1,566
    Joined:
    Jan 9, 2005
    Location:
    Gothenburg, Sweden
    One can arrange the switch so that the cathodes of the tube pair to be switched out of the curcuit get connected to a positive voltage (with negative bias still applied to the grids) high enough to put them well into the cut-off region. For the EL84/6V6 something like +20 to 30V should do it. You won't need a high voltage rated switch and your tubes will not arc. It can be done with a single pole switch.
     
  5. kctx2

    kctx2 Member

    Messages:
    114
    Joined:
    Aug 13, 2005
    It is actually a PP setup. An 18 watt proto
    so-to-speak that will switch between 6v6 pair and an EL84 pair. Once this is worked out the plan is to build a quad set (36 watt or so). Currently, the tubes are just swapped out. Works well, but a switch would be much better.
    Thankyou,
    Keith
     
  6. Wakarusa

    Wakarusa Member

    Messages:
    1,477
    Joined:
    May 7, 2004
    Location:
    Indianapolis, IN
    Wouldn't you have to switch the screen supply if you switched the plates?

    I seem to recall that hot screen + floating plate = dead tube.

    edit: I remembered why. The screen being positive with respect to the cathode and the plate being indeterminate makes the screen the best available anode -- so it melts down when it tries to be the tube's plate.
     
  7. kctx2

    kctx2 Member

    Messages:
    114
    Joined:
    Aug 13, 2005
    "Wouldn't you have to switch the screen supply if you switched the plates?"

    Your correct, I should have said B+ as in a standby switch configuration.

    Thankyou,
    Keith
     
  8. Rich M

    Rich M Member

    Messages:
    422
    Joined:
    Dec 3, 2004
    I've wondered about this after looking at the PA schematic of my MESA Mk IV. If you disconnect the cathodes, won't they float up to B+ thus GREATLY exceeding the max cathode-grid voltage rating of 200V?

    While I'm at it, any comments on whether using the triode switch will lower the heater voltgage enough to shorten tube life?
     
  9. John Phillips

    John Phillips Member

    Messages:
    13,080
    Joined:
    Mar 17, 2002
    Location:
    Scotland
    Ah ha - yes! I should know this, since I killed a tube a while ago exactly like that, by accidentally disconnecting the OT (with the screens still connected) while trying to troubleshoot something. And switching the screens off as well would involve even more switch complexity.

    And you can't just turn off the whole B+, since you need it to run the other tube...
    No, because the tube effectively becomes non-conducting (infinite resistance, since it is a vacuum) with the cathode disconnected, because there is no source of electrons within it - so actually the cathode voltage doesn't rise.
    Running a tube as a triode doesn't lower the heater voltage, but it does shorten tube life, at least in amp/tube configurations that are really designed to be run as pentodes.
     
  10. Rich M

    Rich M Member

    Messages:
    422
    Joined:
    Dec 3, 2004
    John,
    THanks for following up on my comments. I was tired when I wrote that. I meant the "tweed power" setting not the triode.

    Also, I guess I dont understand your comment about the vacuum. The electrons should have no problem coming off the cathode as long as there is a more positive eletrode around to attract them. The cathode is indirectly heated so it should still be able to supply electrons. You may be right, I'm just trying to understand. Thanks.
     
  11. John Phillips

    John Phillips Member

    Messages:
    13,080
    Joined:
    Mar 17, 2002
    Location:
    Scotland
    The electrons can't come from the cathode unless there is a current source supplying it - ie a circuit via ground back to the supply and the plate. If they did, it would rapidly becone extremely positively charged and attract them all back again (I'd guess this is actually probably what happens to some extent, with a very small 'cloud' of electrons near the cathode). But the rest of the tube is a very good vacuum, and has effectively infinite impedance - it's so high that there is no measurable voltage on the cathode when it's disconnected. I have to admit that I wondered about this at one time too, but it works perfectly well in practice - try measuring the voltage on a disconnected cathode.

    One thing that I have thought about though is that it would be better to turn off the filaments to the unused tubes too - this was raised in another thread. If the tubes are left running with no cathode current, the cathode coating will eventually become 'poisoned'. I don't know how serious this is in practice, but it bothers me because I mostly run my Blue Angel in 6V6-only mode, so the EL84s are potentially at risk from this. In fact, I'm not quite sure that the whole operation couldn't be accomplished simply by turning off the filaments and leaving the cathodes connected... except that there's then a risk of 'cold stripping' too. I'm not sure that this happens when the cathode is truly cold (as opposed to just a bit under-voltage) though.


    I'm not sure how far the 'Tweed Power' switch lowers the voltages - it seems odd that they do it by lowering the whole primary voltage rather than just the B+, but I think that is how it's done... don't have a schematic to hand.
     
  12. Rich M

    Rich M Member

    Messages:
    422
    Joined:
    Dec 3, 2004
    Well, I'll take your word since you measured it. Seems odd.

    The tweed switch is on the primary side so it would affect the filament voltage. So I don't use it. Never had much luck with a variac on my marshall, except when primary volts needed proper adjustment.
     

Share This Page