1. Win a Broadcaster or one of 3 Teles! The annual Supporting Member Giveaway is on. To enter Click Here. To see all the prizes and full details Click Here. To view the thread about the giveaway Click Here.

All metal se amp, 2x6sk7,6v6

Discussion in 'Glowing Bottle Tube Amp Forum' started by Yoda2018, Mar 8, 2021.

  1. Yoda2018

    Yoda2018 TDPRI Member

    Age:
    47
    Posts:
    21
    Joined:
    Jan 11, 2020
    Location:
    Spain
    Hi everyone ,
    Found in my small stash of tubes a couple of 6sk7’s and 6v6 metal.
    So I thought about building a se amp with what I have lying around .
    I know most people say 6sk7’s are rf and useless, still, I’m just experimenting in order to learn.
    If anyone has an idea for a schematic I’d appreciate any input .
    I’ll go ahead and build something out and post schematics along the way, criticism is very welcome.
     
  2. danlad

    danlad Tele-Meister

    Posts:
    338
    Joined:
    Mar 24, 2015
    Location:
    Here and there
    Every journey I ever make on these sorts of things starts with this site...!

    http://www.valvewizard.co.uk/pentode.html

    The 6SK7 has a 250v example in the datasheet, so Merlin's worked example with the EF86 will give a good starting point for the anode and screen supply.

    I have a box of old TV and radio valves to mak some use of so I'm very interested in what you come up with.
     
  3. Yoda2018

    Yoda2018 TDPRI Member

    Age:
    47
    Posts:
    21
    Joined:
    Jan 11, 2020
    Location:
    Spain
    Actually i was using valvewizard´s page for the purpose...Fantastic site!
    One question I have and it might be a stupid one : should I set the two stages the same or how do you decide which settings for each stage?
    I´ve hardly seen amps with two consecutive pentodes ...I did build Bendyha´s Effing Else, which sounds fantastic.Other than that, not much is to be seen.
    What´s the rule to choose each stage´s settings?Why is it usually a pentode followed by a triode (or the other way around)?
    I konow I have 300v as b+, don´t necesarily need a tone stack (beacause i never use it), was thinking more of a gain+volume, or even gain+volume+master if that makes any sense.
    Again, not an expert at all, have built quite a few amps following opther schematics, and tweaked as far my knowledge allowed me to so no offence if someone needs to call me an ignorant!
     
  4. printer2

    printer2 Poster Extraordinaire

    Posts:
    7,190
    Joined:
    May 24, 2010
    Location:
    Canada
    Strap the 6SK7 as triodes. May still need some gain though if you want to distort it. A pedal up front would do it. Maybe make a booster.
     
  5. Yoda2018

    Yoda2018 TDPRI Member

    Age:
    47
    Posts:
    21
    Joined:
    Jan 11, 2020
    Location:
    Spain
    Don’t have a pedal...don’t need distortion either, just aiming for a little bit of breakup, if feasible with those tubes...
     
  6. danlad

    danlad Tele-Meister

    Posts:
    338
    Joined:
    Mar 24, 2015
    Location:
    Here and there
    Or even strap the first stage as a triode, then use the second as a pentode?
     
  7. Yoda2018

    Yoda2018 TDPRI Member

    Age:
    47
    Posts:
    21
    Joined:
    Jan 11, 2020
    Location:
    Spain
    Hi there,
    I gave it a go yesterday .
    I chose 33k for Ra and -6v curve on the graph
    Ra 33k
    Va 150v
    Rg2 150k
    Vg2 97v
    Rk 1k
    Vk 5.2v
    (Voltages indicated are measured ones)
    That’s about 5.85mA (Ia + Ig2).
    Specs indicate 9.2mA Ia and 2.6mA Ig2.
    Should I try to get closer to that? After plotting the load line I tried several resistor values and the ones I indicate are the closest measured/calculated.
    Other values just had nothing to do with the calculated ones.(voltage readings vs voltages calculated).
    Assuming these settings are ok I could move on to the next stage.
    Should I set the second 6sk7 equally? I ve wired a 1M pot between stages. Coupling capacitor is 0.047 (because I had one lying around).
    Any help would be welcome, am a bit in the dark here.
    B+ is 300v by the way.
    Thanks!
     
  8. Yoda2018

    Yoda2018 TDPRI Member

    Age:
    47
    Posts:
    21
    Joined:
    Jan 11, 2020
    Location:
    Spain
    Danlad and Printer2: triode strapping is an option as you say, but the idea is to try an all pentode amp.

    Depending on the results as pentode.
    Am not in a hurry...
     
    Last edited: Mar 10, 2021
  9. Yoda2018

    Yoda2018 TDPRI Member

    Age:
    47
    Posts:
    21
    Joined:
    Jan 11, 2020
    Location:
    Spain
    hi there,

    Schematics of where I am at ...
    I chose b+ of 300v and 9.2mA, bias at -4.5v with a 33k resistor
    Voltage readings are quite different from what the maths say (if they are correct that is).
    One doubt I have is whether I should sum Ik+Ia+Ig2 to get the full dissipation
    289-146.6=142,4/33000=4,31mA (plate) (according to graph should get about 100v)
    289+92=197/150000=1,81mA (grid2) (according to Merlin, I calculated the plate to grid factor and result is 3.5 so if my bias point is [email protected] i should have 1.82ma on the grid. 100v/0,00182=110k , closest resistor I had was 120k. But it just didn´t work so added a pot hence the strange 190k value.Tuned it so I could get the 4.5v bias voltage.
    Correct me if all of this is absolute crap.
    4,5/820=5,48ma (Cathode)
    That would be 10mA total if i add everything.

    Should I try getting closer to the knee of the curves ? -3v would be spot on then.With 40v on the plate at 8mA
    Don´t really understand how to calculate the gain and how much I should aim for.
    Second stage I tried to go for half after reading too much and understanding too little ...
    Grid leak : not clear either whether the 470k after the volume pot is correct. Supposedly you should´nt load the second stage too much.

    Anyways, will continue my obscure learning path .Perhaps choosing pentodes isn´t the easiest as first amp designed from the datasheets.
    But what the hell, it´s part of the fun


    I do feel like running around in circles ...

    Played around plugged into a Jensen c8r, output transformer is 7,2K , speaker 8ohms. Pretty nice sounding already...


    IMG_1804.JPG IMG_1802.jpg
     
    danlad and 2L man like this.
  10. 2L man

    2L man Tele-Meister

    Age:
    61
    Posts:
    306
    Joined:
    Nov 23, 2020
    Location:
    Finland
    Gain is how much output voltage change to input voltage change on loadline just dividing them. I would bias them bit colder closer the center of loadline. When bias is too much off from center other side of AC signal cuts first and now yours at least second stage run quite high current

    G1 voltages vs. Anode voltages seem bit strange? G1 -4,5V should make plate drop to 100V and G1 -6V should make about 150V to plate on that loadline if I understand it right? Perhaps Screen voltage make it like that? Or you crossed G1 voltages?
     
    Last edited: Mar 12, 2021
  11. printer2

    printer2 Poster Extraordinaire

    Posts:
    7,190
    Joined:
    May 24, 2010
    Location:
    Canada
    I would set the second one as a triode and play around with values to see where the first tube sounds its best. I did that before with a pentode, a pot for the cathode, plate and screen. I twiddled them around and found there was a combination which sounded better than the rest. I am guessing it would be more so given the variable gain of the 6SK7.
     
  12. Yoda2018

    Yoda2018 TDPRI Member

    Age:
    47
    Posts:
    21
    Joined:
    Jan 11, 2020
    Location:
    Spain
    As recommended by 2L man, I biased more centered
    Keeping the 33k plate resistor I have 153v, 98v on g2 (with 150k resistor) and changed rk to 1k, having 5.3v.
    I used Merlin’s factor calculation to set up g2
    Data sheets say 2.6mA @100v for g2 with 9.2mA for Ia
    9.2/2.6=3.54 factor
    My chosen bias point is [email protected]
    4.6/3.54=1.3mA for g2
    200v drop/0.0013=150k g2 resistor...
    1st stage set (for the moment...)

    Printer2 : sounds like a plan.and I like that pot arrangement, I’ll give it a go with 2nd stage as triode.
    I assume that would allow me to dial in first 6sk7 and then I can move along fiddling with the next one.Whether I keep it as triode will depend on results.
    Looking up how to wire pentodes as triodes I came upon this article which will require more reading in the future . For the moment and being the crappy student I am I just jumped to page 23
    http://www.audiodesignguide.com/New2A3/ETF06TS.pdf
     
    2L man and danlad like this.
  13. 2L man

    2L man Tele-Meister

    Age:
    61
    Posts:
    306
    Joined:
    Nov 23, 2020
    Location:
    Finland
    I follow your work because I just received few 6SJ7 metal pentodes!
     
  14. printer2

    printer2 Poster Extraordinaire

    Posts:
    7,190
    Joined:
    May 24, 2010
    Location:
    Canada
    Page 23 is what I had in mind. A lot easier chasing one squirrel at a time rather than two. You might even like a triode in front of a pentode, maybe even two triodes or two pentodes.
     
  15. 2L man

    2L man Tele-Meister

    Age:
    61
    Posts:
    306
    Joined:
    Nov 23, 2020
    Location:
    Finland
    Thank you for this find! Good study and done possibly as understandable as possible!
     
  16. Yoda2018

    Yoda2018 TDPRI Member

    Age:
    47
    Posts:
    21
    Joined:
    Jan 11, 2020
    Location:
    Spain
    curious to see what you come up with!
     
  17. Yoda2018

    Yoda2018 TDPRI Member

    Age:
    47
    Posts:
    21
    Joined:
    Jan 11, 2020
    Location:
    Spain
    How many times can you go around the circle ?
    I´ve tried all sorts of combinations...triode to pentode, pentode to triode, pentode pentode...With and without cathode bypass caps...
    All sorts of values and all sorts of settings.

    And keep coming back to the same point.
    1st stage stays as pentode , values as seen on schematics.
    2nd stage : tried the recommended or typical values on the datasheet.
    But am not satisfied .
    Sound is good, nicely balanced, plenty of volume for a se home amp.
    But it´s a tad harsh, even playing around with the tone pot on the guitar it´s a bit too percussive. I am not getting the breakup i am aiming for, even at full volume.

    There might be different causes for this :

    -speaker used is a Jensen c8r, in some of my different settings the cone was flapping. i do not know what causes this, is it too much gain, is it the ot, is it the
    impedance between stages?

    -I dont understand in valvewizard´s pentode page why at the bottom he chooses 1.6mA to calculate the gain. Is it because of the chosen bias point?
    If that´s the case then I should have 4.6mA x 33k = 155. not that it means much to e but compared to most triodes it´s a hell of a lot of gain.

    This morning i decided to set both pentodes the same and am going to try with a tone stack (just bass/treble + volume) between the two 6sk7s.

    And am going to give it a go with a 12" speaker I have lying around.

    This is a much harder quest than i thought it would be, but i came across loads if interesting reads.
    The last one being https://livinginthepast-audioweb.co.uk/index.php?p=valvegaincalcs
    It´s a tad more complicated as an explanation than valvewizard´s and haven´t really had time to understand it.
    Bit by bit...

    Maybe i am just on wrong quest and should go back to copying other people´s schematics and using usual tubes but that ain´t no fun!

    Will update with results with the new schematic and 12" cab...

    By the way guitar is an epi 1966 reissue or whatever it´s called with a Fralin P90 (just another variable to what´s coming out of the amp).


    IMG_1843.jpg
     
    jimytheassassin likes this.
  18. Commodore 64

    Commodore 64 Friend of Leo's

    Age:
    45
    Posts:
    4,258
    Joined:
    Mar 1, 2010
    Location:
    Kent, OH
    Try running the signal into a pentode and triode in parallel. I've got an amp that has a pentode in V1 and you can use a switch to change it to parallel or cascade into a V2 triode.

    Some people prefer triode into pentode rather than the reverse (musically speaking).
     
  19. Yoda2018

    Yoda2018 TDPRI Member

    Age:
    47
    Posts:
    21
    Joined:
    Jan 11, 2020
    Location:
    Spain
    Yep why not, do you have or know of a schematic that shows how this could be done ?
     
  20. CoolBlueGlow

    CoolBlueGlow Tele-Afflicted

    Posts:
    1,543
    Joined:
    Nov 20, 2011
    Location:
    Arkansas
    Just keep in mind that on those metal can 6V6, the metal can is connected to pin 1.

    Some people forget and by habit when building an amp use the pin 1 lug on the tube socket as a convenient tie-off point for all kinds of things, including HV plate supplies. They do this because they think pin 1 is “not connected to anything on the tube”, or because they are building monkey-see wire by wire knockoffs of their favorite designs, and THAT builder used pin 1 as a tie off for HV, etc.

    Some folks even look at the tube datasheet and fool themselves, by “confirming” pin 1 is blank - because it is blank... for 6V6G (G=glass outer shell) designs. On the typical manufacturer datasheet, the connection of pin 1 to the metal shell is not well noted. Some sheets don’t even mention it.

    You need to know that Pin 1 of a 6V6 (aka “metal can”) is connected to that metal shell. If for build convenience reasons you tie some HV (or whatever) to that #1 socket lug, when you grab that metal can tube to pull it out, (or touch it) you will get a nasty and possibly fatal surprise.

    A quick continuity check with any common meter will confirm this. Ditto for the 6SK7. Flying without my master tube data book handy here, so do that continuity check from metal can to all pins on any metal can tube. Otherwise you may let out your personal magic smoke. That would be bad.

    Just a word to the wise.

    CBG
     
    jimytheassassin likes this.
IMPORTANT: Treat everyone here with respect, no matter how difficult!
No sex, drug, political, religion or hate discussion permitted here.