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.

How would I replace el84 for ef80

Discussion in 'Shock Brother's DIY Amps' started by Len058, Feb 27, 2021.

  1. Len058

    Len058 TDPRI Member

    Age:
    47
    Posts:
    91
    Joined:
    Jan 31, 2021
    Location:
    Leeuwarden Netherlands
    My 15 watt amp kit is too loud for me and after building robrob's ef80 champ I realized it would fit my needs if I could replace the el84 output tubes with ef80 tubes.
    I know the sockets need rewiring but I have no idea how to anticipate the other differences between the tube types. Any help would be appreciated!

    The relevant part of the push pull circuit:
    pushpull.png
     
    Last edited: Feb 27, 2021
  2. Nickfl

    Nickfl Friend of Leo's

    Posts:
    2,530
    Joined:
    May 24, 2016
    Location:
    Florida
    Well, the good news is you power transformer is putting out about the right voltages, they may be a bit higher with the lower current draw of the EF 80s, but you're probably going to be in the ballpark. The bad news is you are almost certainly going to need a different output transformer. Can you give us the specs on the OT you have in there now?

    You'll also probably have to change the cathode resistor value and move the junction of the power tube grid stoppers and grid leak resistors so that the form a voltage divider to reduce the signal level on the power tube grids.
     
    Paul-T and dan40 like this.
  3. Len058

    Len058 TDPRI Member

    Age:
    47
    Posts:
    91
    Joined:
    Jan 31, 2021
    Location:
    Leeuwarden Netherlands
    Thanks for the help!

    The output transformer is an ATRA0427 pdf: https://www.die-wuestens.de/kd/ATRA0427d.pdf.

    8x intersected windings
    Raa = 9 Kiloohm
    N = 10-15W
    Connections for ultralinear operation
    Connections for 4,8,16 Ohm secondary

    Rob uses a Hammond 125b for the bassman micro and if I understand it correctly it has a higher resistance/inductance than my OT. I could just copy the part of the BassmanMicro power amp circuit but it uses a choke. The voltages seem in the ballpark but I could start with a higher resistance voltage drop to be safe. I have 100Ohm voltage drop resistor and the choke is 270Ohm.

    I think I'm starting to understand the Bassman micro and the Madamp circuits a bit better.
     
  4. dougsta

    dougsta Tele-Meister

    Posts:
    245
    Joined:
    May 25, 2015
    Location:
    Edinburgh, Scotland
    I've tested an OT designed for a 20W 2xEL84 PP amp (Jet City JCH22H designed by Mike Soldano, made in China) in my push pull EF80 amp, sounded great, lots of low end, the key was it's much bigger iron. Mike Soldano has a history of putting big iron in his amps.

    If you look at the primary impedance of the JCA22H OT, 1.8K - 0 - 1.8k, that's a lot lower than you "should" use. I think the key is that big iron means a lot more inductance.

    I'm confident your OT will sound good.

    Here's the OT beside my DIY "5 & dime" OT. (£5 for the cost of the donor transformer and the dime gives you an idea of the size)

    IMG_8017.jpg

    Doug.
     
    Len058 likes this.
  5. Nickfl

    Nickfl Friend of Leo's

    Posts:
    2,530
    Joined:
    May 24, 2016
    Location:
    Florida
    If you miss match your speaker impedance (put an 8r speaker on the 4r secondary or a 16 on the 8), you can double the primary impedance on that transformer to 18K. That's lower than what Rob's design specifies but it's in the ballpark and you could probably make it work as @dougsta said, you probably have a little more leeway because of the larger transformer.
     
  6. Len058

    Len058 TDPRI Member

    Age:
    47
    Posts:
    91
    Joined:
    Jan 31, 2021
    Location:
    Leeuwarden Netherlands
    Great! I have a 16Ohm speaker and the 8 and 16 outputs are there. I could also make a 4 Ohm output for my 8 Ohm speakers.

    I'm trying to compare Rob's EF80 designs to my EL84 circuit and added names to the parts to better understand my amp. Am I correct?

    If I look at the DeluxeMicro I would have to change the Coupling caps to 0.1 uF, the cathode bias to 300Ohm (or a bit more because my tranny is more powerfull) and lower the cathode bypass to 25 uF. I don't get how the junction would influence the grid load

    The screens are connected to B+1 in my amp but the Deluxe uses B+2 at the screens and B+1 goes to the OT so I think I have to change that. EDIT Ah I get it. Rob's amps have a minimal difference between B+1 and 2 but my amp has a huge difference so that won't work.

    Following the Deluxe circuit, I would have to remove the grid stoppers but that's a bit scary to me and I have a feeling I don't understand it properly. The DeluxeMicro also doesn't have resistors between the OT and the screens but I'm wondering if they are there to match the impedance between the tubes and the OT or to drop B+1 to B+2. Than my B+2 in the drawing would be B+3, which makes sense to me.





    EL34-namen.jpg If
     
    Last edited: Feb 28, 2021
  7. Nickfl

    Nickfl Friend of Leo's

    Posts:
    2,530
    Joined:
    May 24, 2016
    Location:
    Florida
    It looks like you've labeled everything correctly in your attached file.

    Are you trying to rebuild your amp into a deluxe micro, or are you just trying to convert your existing amp to ef80 power tubes and using the deluxe micro as a model for the power section?

    If you're planning on keeping your preamp and phase inverter from your existing design it would help to know what the existing design is, looks like you have a long tailed pair pi in your existing circuit, which is already a significant departure from the deluxe.
     
  8. Len058

    Len058 TDPRI Member

    Age:
    47
    Posts:
    91
    Joined:
    Jan 31, 2021
    Location:
    Leeuwarden Netherlands
    I believe you're right it's LTP instead of cathodyne, thanks. I can't share the complete circuit because the designer wants to protect his thing and I respect that.

    I will have a look at the bassman micro instead. I want to keep the rest of the amp intact but if this fails I'll build the Bassman micro with the usable parts of this amp.
     
  9. Len058

    Len058 TDPRI Member

    Age:
    47
    Posts:
    91
    Joined:
    Jan 31, 2021
    Location:
    Leeuwarden Netherlands
    The phase inverter. I thought LTP's would be asymmetrical but this one isn't.
    EDIT Oh man I'm studying for too long. Cathodyne works without a tube.
     

    Attached Files:

  10. Nickfl

    Nickfl Friend of Leo's

    Posts:
    2,530
    Joined:
    May 24, 2016
    Location:
    Florida
    Is it a kit? Can you just say what kit and supplier? Just curious what the topology is too see what micro power amp would be the best match. Also, I'm skeptical of anyone who claims to have "original" designs they need to protect, considering were dealing with technology that was obsolete 50 years ago and almost all kits are clones of classic amps, which were themselves mostly lifted directly from tube manuals... but whatever :rolleyes:

    The bassman micro ef80 is probably your best bet. Notice that design moves the ef80 grid leak resistors position so they are between the grid stoppers and the grid, making a voltage divider, which cuts down the signal hitting the power tubes. I think you'll probably want that on this amp, but it's hard to say without knowing more about the circuit.
     
  11. Len058

    Len058 TDPRI Member

    Age:
    47
    Posts:
    91
    Joined:
    Jan 31, 2021
    Location:
    Leeuwarden Netherlands
    It's the Madamp M15mk1 kit. I see no point in keeping the circuit a secret because it's not really worthwhile to buy all the parts separately. But it's not really up to me.

    It's basically a 2 channel Marshall type amp. One half of a 12ax7 is used for the clean channel and 3 halves for the gain channel with a Marshall tmb circuit. No negative feedback.
    I don't mind changing the tone of the amp because I haven't really heard it at full throttle so if it sounds different but still good I'm happy.

    I noticed the move of the grid leak and also the funny master volume variable resister setup. I think that would be the solution.
    The phase inverter doesn't have matching plate load resistors, that puzzles me. I can always start at the end and work my way to the phase inverter than preamp if it doesn't work out.
     
  12. Len058

    Len058 TDPRI Member

    Age:
    47
    Posts:
    91
    Joined:
    Jan 31, 2021
    Location:
    Leeuwarden Netherlands
    After trying to read a lot of circuits I think I get it. The asymmetrical phase inverter is because of the feedback loop.

    If I compare the 6L6 Blackvibe and Soldano with the EF80 versions the tail of the PI goes from 10 to 18K in the Blackvibe and stays 10 in the soldano. I think I can leave the PI as is but order the different resisters and caps, just to be sure.

    I would remove C19 and replace R28/29 with 300 or 320 Ohm. Replace and move R26/27 to 470K before the grid leaks. Swap the grid leaks R24/25 with 82K ones. C15/16 to 0.1 uF because all EF80 amps change to that value.

    I think it's time to get me some parts.
     
  13. 2L man

    2L man Tele-Meister

    Age:
    61
    Posts:
    292
    Joined:
    Nov 23, 2020
    Location:
    Finland
    This is unusual. Often "bottom tube" which gets indirect drive its gain is increased using about 25% bigger anode resistor than a top tube anode resistor.

    On HiFi amp schematics I have seen a balance potentiometer between B2 and anode resistors which is adjusted using an oscilloscope when a new PI tube is installed and obviously adjusted regularly and possibly before every listening is began ;) For instrument amps a double triode which pairs are matched is sometimes recommended for LTP PI which is where I use few euros more expensive tubes.
     
  14. Len058

    Len058 TDPRI Member

    Age:
    47
    Posts:
    91
    Joined:
    Jan 31, 2021
    Location:
    Leeuwarden Netherlands
    I think it's because there is no feedback loop. But I ordered the parts to make the PI like the Bassman Micro, including the NFB and presence control.
     
  15. Len058

    Len058 TDPRI Member

    Age:
    47
    Posts:
    91
    Joined:
    Jan 31, 2021
    Location:
    Leeuwarden Netherlands
    Yeah! It works!
    I changed the resistors to the values of @robrob 's designs and left the caps as they were except the cathode bypass cap, which I disconnected. I also kept the wiring to the OT as in the original design. So the grid is connected to the center tap of the OT. (correct me if I'm wrong)
    I rewired the tube sockets, altered the OT 16 to 8ohm and 8 to 4 ohm windings, and fired up the amp with a voltage meter at the plate of one of the ef80's. Plate voltage was way above the max 300V. So I connected the grid (el84 pin 9, EF80 pin 8) to B+2 (the same as the phase inverter). The plate voltage of the ef80 was around 250V, so in the very safe zone.
    I quickly measured the pre-amp B+* voltages and they were lower than the original design, but not so low that I would worry about it.

    I'll post the exact measurements later but I'm done for the night. It sounded great and before I realised it was already 01 in the night I thought breakup sound levels were very acceptable for daytime playing.
     
    Last edited: Mar 5, 2021
    robrob likes this.
  16. Len058

    Len058 TDPRI Member

    Age:
    47
    Posts:
    91
    Joined:
    Jan 31, 2021
    Location:
    Leeuwarden Netherlands
    I did some investigating and came to the conclusion that it might be because it doesn't matter that much if you don't use a matched AX7.
    I changed the resistors, added the fizz cap (stability in robs schematics) and checked the plates. At another forum I read that there should be a difference of approximately 7,2 volts.
    I tried all my 12ax7 tubes and they varied from 11V to 4V difference. I settled on the tube with 8 volt difference. So if you don't check the tube, there is a chance you're in the ballpark with two 100K resistors and miles off with a 100K/82K combination.

    I had a feeling the amp opened up a bit with the semi matched tube.
     
    2L man likes this.
  17. Tom Kamphuys

    Tom Kamphuys Tele-Holic

    Age:
    41
    Posts:
    715
    Joined:
    Sep 18, 2018
    Location:
    The Netherlands
    Blencowe's book gives an explanation for the 82k/100k. It is not because of a DC offset (which I assume you are referring to). That DC is gone after the coupling caps. It is because the gains of the inverting and non-inverting inputs are slightly different. If you do the math, the 82k/100k combination balances the output in the Bassman. This is however dependent on e.g. the tail resistor.

    This gives a hint of the wealth of information in the book: http://www.valvewizard.co.uk/acltp.html
     
  18. Len058

    Len058 TDPRI Member

    Age:
    47
    Posts:
    91
    Joined:
    Jan 31, 2021
    Location:
    Leeuwarden Netherlands
    That's interesting! But the plate resistors are in front of the caps. They divide the B+ over the plates, the signal is going the other way, through the caps where B+ can't go because it's DC instead of AC. More plate current on 1 of the two plates will compensate the gain loss at that plate.

     
  19. Tom Kamphuys

    Tom Kamphuys Tele-Holic

    Age:
    41
    Posts:
    715
    Joined:
    Sep 18, 2018
    Location:
    The Netherlands
    Oh boy, dare I say it? Yeah, I'm Dutch, no hierarchies. Here it is: Uncle Doug is wrong. Yes, the gains are different, yes R1<R2, but R1 is smaller to compensate for the *higher* gain of the inverting input. The signal that 'involves two tubes' is actually lower, because some of it is waisted in the tail resistor, which is compensated by the higher R2.

    http://www.valvewizard.co.uk/Common_Gain_Stage.pdf (see p.21 for a plate resistor vs. gain table).

    I agree fully with what you write @Len058, but somehow I feel we arrive at a different conclusion, although not explicitly stated. Maybe it is the last sentence. The amount of DC plate current is not that big of a deal. It is the AC current that determines the signal.
     
    Last edited: Mar 8, 2021
    dan40 likes this.
  20. Len058

    Len058 TDPRI Member

    Age:
    47
    Posts:
    91
    Joined:
    Jan 31, 2021
    Location:
    Leeuwarden Netherlands
    I'm glad you're educating me, @Tom Kamphuys . I'm just trying to understand what the * I'm doing and why.

    It will take some time to read everything 3 times and let it sink in. Uncle Doug made sound logic in my head, bummer he's wrong.

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