Bias Mod for Mesa Mark III

Discussion in 'Amps/Cabs Tech Corner: Amplifier, Cab & Speakers' started by aortizjr, Jan 20, 2005.

  1. aortizjr

    aortizjr Member

    Messages:
    144
    Joined:
    Nov 24, 2004
    Location:
    Santa Fe, NM
    Hi...

    So I just bought a Mesa Mark III Simul-Class head.

    Anyway I am looking at adding an adjustable bias.

    Here are the Schematics to the poweramp section:

    http://www.intensemetal.com/~aortizjr/projects/mesa-mkIII/mesaboogiemark3b.gif

    I see where the bias section is. But what mod actually needs to be done.

    I am guessing that I just replace the "factory selected" resistor to a pot and go from there. Except that I am having to bias both EL34's and 6L6's. So I would guess that I would have to mod it some way to have adjustable biases for both.

    Is this the case or should I just put the pot there and be careful with my tube selection?

    Or is there something completely different that I need to do with the mod?

    Any help at all would be great.
     
  2. WailinGuy

    WailinGuy Member

    Messages:
    1,266
    Joined:
    Aug 30, 2003
    Location:
    Silver Spring
    Unless you get into serious modification of the output circuit, It looks like you should just install a single pot that adjusts the bias supply voltage going to all four tubes. Here's why:

    Despite Mesa's marketing hype, the outside pair of power tubes does not run in class A. It runs in class AB, just like the inside pair. However, in an attempt to reduce crossover distortion, the outside pair runs at a higher idle current (they run hotter) and with a lower drive signal, compared with the inside pair. (If you look at the schematic, you can see that the outside pair has voltage divider networks just ahead of the control grids that reduce drive signal voltage at the same time as reducing negative bias voltage.) Setup with four matched 6L6's, the Simulclass power section will typically run the inside pair on the cold side (maybe around 25 ma at idle each, or even a bit less) and run the outside pair where I would normally want to run 6L6's in a 50 watt amp - around 35ma or so at idle.

    Now, I never quite understood the option of using EL34's instead of 6L6's for the outside pair. Generally, EL34's already require signficantly less negative bias voltage to operate normally, compared to 6L6's. So, using this combination results in a situation where typicall ALL FOUR power tubes will be biased on the cold side. I would think the tone would suffer when using EL34's, but I guess there are players that do prefer this setup. I suppose an adjustable bias would come in handy here, to allow all four tubes to run at a healthier idle current.

    (This is just my opinion, but I have to wonder if Mesa would have been smarter to forget about implementing the SimulClass idea and simply set the factory bias for higher idle current on their standard class AB output section. That's a much simpler way to reduce crossover distortion!)

    So my advice is to install the pot (but first measure the idle currents at the factory bias setting, so you can go back to the stock tone), and then adjust the bias so that the hotter running of the two pairs of output tubes is not running too hot (don't go much over 40ma per tube.)
     
  3. WailinGuy

    WailinGuy Member

    Messages:
    1,266
    Joined:
    Aug 30, 2003
    Location:
    Silver Spring
    One more thing...

    I've never triied this myself, but I've heard stories of techs converting Mesa SimulClass amps to run all four power tubes in standard class AB. In other words, the output section winds up looking exactly the same as that of a Fender Twin Reverb. This involves removing the resistors to ground before the control grids and adding 470 ohm/2 watt screen resistors on the outside pair of tube sockets and wiring them for pentode operation instead of pseudo triode (connect screens to screen supply instead of plates). Now, I don't know which pair of output transformer primary taps should be used for all four plates - the inside taps or the taps on the ends of the primary winding. Whichever is closer to approx. 2200 ohms primary impedance (you'd need a signal generator to measure this).

    Anyway, supposedly the amp owners were happier with their sound once the mod was done and the bias was adjusted somewhere in a healthy 30-40ma idle current range.

    Obviously, this mod gets more complicated if you still want to run EL34's and 6L6's at the same time. Then you DO need to have separate bias controls for each push-pull pair.
     
  4. RussB

    RussB low rent hobbyist Silver Supporting Member

    Messages:
    11,071
    Joined:
    Jul 3, 2004
    Location:
    CT
  5. RussB

    RussB low rent hobbyist Silver Supporting Member

    Messages:
    11,071
    Joined:
    Jul 3, 2004
    Location:
    CT
    Stock bias supply, with "snipped" R117

    [​IMG]

    Bias supply, as modded by me (new caps there, too)

    [​IMG]

    [​IMG]

    Class A sockets, w/2.2k 2W resistors added, as per Mike B @ Mesa...the bias supply (negative voltage) resistor going to the outer pair of tubes can be switched to give a nice balance between the 2 pairs of tubes. I have a 150k there now. Factory is a 220k. Right now the 6L6's run @ 34mA, and the EL34's run @ 38mA (41.5 in Class A) with 450 plate volts. I'm using SED tubes.

    [​IMG]

    [​IMG]


    EDIT: OK, why don't the pics show up! All I see are the links... :mad:
     
  6. aortizjr

    aortizjr Member

    Messages:
    144
    Joined:
    Nov 24, 2004
    Location:
    Santa Fe, NM
    Wow...

    That is excellent info. Exactly what I was looking for. Thank you so much!

    I have been looking around and such. My amp actually doesn't come in till Friday of this week.

    But I will be sure and post what I do.

    Anyway I am trying to pre-order parts as best I can.

    Where did you get your trim pot from?

    I have a bunch of 50K 1/2 watt ones running around, but looking at other amp schematics, they recommend at least 1 Watt. Of course all the resistor values on the schematic are unspecified.

    I would assume that 1/2 watt should be fine.

    I think I am going to try JJ, SED, and EH tubes on this go-round.

    Also, what modifications need to be made to run all 6L6 and the "Simul-Class" become basically 60/100?

    Thanks!
     
  7. RussB

    RussB low rent hobbyist Silver Supporting Member

    Messages:
    11,071
    Joined:
    Jul 3, 2004
    Location:
    CT
    HoffmanAmps.com

    I don't know the wattage of the pot, but he recommended it for a bias control. Cermet pots are a good choice as well...I ended up buying a few, but the 50k carbon pot is working just fine

    I don't know, and I frankly don't understand why you'd want to eliminate the "Simul-Class" aspect of the amp...just guessing, but I wonder if the PT's are the same for the 60/100's and the Simul's
     
  8. WailinGuy

    WailinGuy Member

    Messages:
    1,266
    Joined:
    Aug 30, 2003
    Location:
    Silver Spring
    My guess is that the power transformers are the same, but, as I implied in my previous post, the output transformers are different. I don't know if you'd get the proper primary impedance by connecting all four power tubes to the outside taps or to the inside taps of the Simul OT. Perhaps modding a Simul amp to operate 60/100 is not such a good idea, especially if you use the stock OT.

    I have to wonder if a big part of the Simul sound is the fact that the outside tubes run in triode mode (screens connected to plates). (BTW, the Mark IV allows you to switch between triode and pentode operation for the outside pair. When I owned a Mark IV, I preferred the pentode setting.) Triode mode definitely "softens" the tone and induces some sag in the attack response. This is just my opinion, but I think it's silly to run the inside pair of tubes cold and then run the outside tubes hotter in order to "fill in the spaces" where crossover distortion would occur if the inside pair were to operate alone. Furthermore, reducing the drive signal to the outside pair seems like a rather contrived way to minimize crossover distortion (at the expense of output power). I think it makes more sense just to run all four tubes at a healthy bias point for good tone and maximum power/headroom. Then, you can still provide the option to run either one or both of the pairs in triode mode for a softer tone and reduced power if and when you need it.
     
  9. aortizjr

    aortizjr Member

    Messages:
    144
    Joined:
    Nov 24, 2004
    Location:
    Santa Fe, NM
    Ok...

    got the amp and it came with original Mesa tubes which is nice. At least I can get the current settings from the Mesa "special" tubes.

    Now with the EL-34 resistors... do you think it would be worth it to get a couple of like 250K or 300K pots and use those instead of the resistors?

    Or maybe a 200K pot and 100K resistor. I will crack the case tonight and make sure that there is room. But I would guess that it would make the bias adjustments easier for those EL34's.
     
  10. RussB

    RussB low rent hobbyist Silver Supporting Member

    Messages:
    11,071
    Joined:
    Jul 3, 2004
    Location:
    CT
    I tried that...I used 250k pots were the 220k resistors to pin 5 of the outer (EL34) sockets. The thing is, you'll need a 4 head bias checker, because when you twiddle ONE pot, it changes the bias on all the other tubes as well!
     
  11. aortizjr

    aortizjr Member

    Messages:
    144
    Joined:
    Nov 24, 2004
    Location:
    Santa Fe, NM
    Ahh.. that sounds like a pain....thanks for the info.

    The problem with the resistors is that I don't have a wide selection. But they are pretty cheap so I will just get a range of them and experiment.

    So I cracked the case and R117 (the factory select one) is clipped. Interesting...

    Looks like a lot of the resistors are just 1/2 watt. So I will go ahead and use my 50K 1/2 watt trim pot.

    Mine needs a cap job as well (which I knew from the seller). However all of them check out except for one of the big 30uF ones near the input.

    In fact it is the same one that is marked with an X in your "Before" pictures. Mine is marked with a V. But only registers like .2uF. That one looks like it is leaking.

    So I figure might as well replace them all while I am at it. I don't think I will do some of smaller (in size) ones. Only one seems to have really drifted, so maybe just that one.

    OK.. so off to ordering parts....

    Thanks again for all of your help and info. I will post more as things progress.
     
  12. RussB

    RussB low rent hobbyist Silver Supporting Member

    Messages:
    11,071
    Joined:
    Jul 3, 2004
    Location:
    CT
    The 220uf/300v are no longer available from Sprague. I used "Powerlytic" 53D 220uf/350v. I got replacements from Triode Electronics (but they're currewntly out-of-stock!) I think AES might have them...

    From what I gather, Mesa clipped all the R117's for the US market. One note, with the bias pot, I have to be careful with 6L6's in the outer sockets...I tried the old SED 6L6's that came with my amp in the outer sockets (after the bias-adjust mod) and they were drawing over 100mA! They red-plated in no time! :eek:
     
  13. aortizjr

    aortizjr Member

    Messages:
    144
    Joined:
    Nov 24, 2004
    Location:
    Santa Fe, NM
    Yep I am having the same problem with finding the 220uF.... I found some at Antique Electronics, but they are out of stock as well.

    When I checked the capacitance with my meter I was getting like 540uF. So I might just get 250uF or to match closely, like a 120uf and a 300uF...

    Anyway.... we'll see... thanks for warning with the 6L6's.
     
  14. aortizjr

    aortizjr Member

    Messages:
    144
    Joined:
    Nov 24, 2004
    Location:
    Santa Fe, NM
    Alright got all the caps in. The question I have now is how you went about removing stuff off the PCB and installing the new components.

    I am trying to put new caps and the bias Pot. However, my PCB is soldered from the bottom.

    Did you remove the board from the chassis? Did you clip the old components and solder the new ones to the old wires?

    As of now I can't even free the old ones from the joints. Admittedly I am not the most experienced at soldering, but this shouldn't be that bad.
     
  15. RussB

    RussB low rent hobbyist Silver Supporting Member

    Messages:
    11,071
    Joined:
    Jul 3, 2004
    Location:
    CT
    I use a solder-wick.

    Sometimes you need to ADD a bit of solder to the connection. Then apply the solder-wick and heat until the solder flows into the wick. Use a small braid solder-wick. Don't hold the heat to the joint very long...PCB can be a real bitch if you lift a trace (don't ask me how I know!)


    Grab the component with the needle nose pliers, you may have to apply just a tad more heat, and yank the component up...repeat for the other end.

    Sometimes it's easier if you cut one end of the component off, so you can remove each half seperately...

    I have a Weller WES51, and set the heat to ~750. I use a chisel tip, or a small pointed one.

    When you solder the components back on, a little solder is enough! Avoid globs...they will just hang down below the PCB. It would be a major PITA to remove the board!
     
  16. RussB

    RussB low rent hobbyist Silver Supporting Member

    Messages:
    11,071
    Joined:
    Jul 3, 2004
    Location:
    CT
    clipping the old components and soldering the new stuff to the leads is a good idea...the chance of burning a trace is GREATLY reduced.
     

Share This Page