Tight Squeeze - Tiny Tube Amp

Discussion in 'Shock Brother's DIY Amps' started by SerpentRuss, Sep 25, 2021.

  1. SerpentRuss

    SerpentRuss Tele-Meister Ad Free Member

    Age:
    58
    Posts:
    223
    Joined:
    Feb 17, 2021
    Location:
    North Carolina
    First off, this project has been a total comedy of errors but it has taught me a lot of valuable lessons.

    It all started with my son's request for a small, tube headphone amplifier. The only tube amp that he owns does not have a headphone jack. I could add one, but what fun would that be?

    I had that request parked in the back of my mind for a while and then I had a power supply fail at work that had a neat, extruded aluminum case. I thought wow, that case would make a good enclosure if I could squeeze the amp into it.

    So I brought the box home and just started throwing stuff together without much of a plan other than it would just have a single tube pre-amp, a 6N2P because I have a box of them, and I would use a solid-state output amplifier to save space. That way I could avoid having an output transformer. The tube would lay down to make the project, kid's room friendly and it would include a headphone jack on the front and have a jack for an 8+ ohm load on the back.

    I happened to have a couple of TDA7266M chip amps. These are neat because they don't have to have an output capacitor and they also have a mute and standby function. At max supply voltage, they will put out about 7 watts. With my planned DC supply, the amp would be more like 2.5 watts.

    Previously I had used an ANTEK AN107 transformer for a small tube amp and I had an AN207 in my stash of stuff. The ANTEK AN207 is a 25 VA, 7-volt AC transformer with double primary and secondary windings. If you split the primary windings you can use the 2nd primary like a 120 volt AC isolated source. Put a voltage doubler on that, and you can power a pre-amp tube amp easily because the current draw is very low.

    The functional layout was supposed to be Primary Gain Stage -> Tilt Tone -> Gain Volume -> Secondary Gain Stage -> Master Volume -> Chip Amp. It required the doubler with filters for B+, a filtered DC supply from one secondary winding for the chip amp, running the heaters off of the other secondary winding. All squeezed into a 7x5x3 inch envelope. It was a nightmare.

    To make a long story short, I got everything in the case, but I made a wiring mistake that cost me the chip amp. Because of the way I put it together, I really was going to be hard-pressed to fix the mistake and so I pushed the mess aside and left it for another day.

    I have included a picture of the squirrel's nest in this post. The PT is under the cap board on the left. The vertical board in the center is the DC supply for the chip amp. The project board has the tilt-tone components, the load resistors for the tubes (to save space and keep most of the signal routing on the project board), and the chip amp. The front panel is made from 14 gauge perforated screen steel, the base and boards are made from a piece of phenolic floor covering taken from a computer floor tile. It's a regular junkyard dog. Too bad it didn't work. 20210920_225445.jpg
     
    Last edited: Sep 25, 2021
    Asmith, SRHmusic, gtrwrks and 4 others like this.
  2. SerpentRuss

    SerpentRuss Tele-Meister Ad Free Member

    Age:
    58
    Posts:
    223
    Joined:
    Feb 17, 2021
    Location:
    North Carolina
    So, in the end, after I slept on it for a few days, I decided I would try to put in a power tube amp section. The only small, decent OT I had was a Hammond 1750 AX. Its advertised pimary is 7500 ohm. The only tube I had on hand that was a good match was a 6CL6. Here are photos of the finished amplifier after I punted and went all tube. The headphone jack is top, the input jack is bottom, knobs left to right, Gain, Tone, Master. 20210924_210318_LR.JPG 20210924_210332_LR.JPG
     
  3. SerpentRuss

    SerpentRuss Tele-Meister Ad Free Member

    Age:
    58
    Posts:
    223
    Joined:
    Feb 17, 2021
    Location:
    North Carolina
    Here are some internal shots.

    20210924_204624_LR.JPG 20210924_204540_LR.JPG
     
    Asmith, gtrwrks, Deeve and 2 others like this.
  4. SerpentRuss

    SerpentRuss Tele-Meister Ad Free Member

    Age:
    58
    Posts:
    223
    Joined:
    Feb 17, 2021
    Location:
    North Carolina
    I paralleled the secondaries on the PT to supply the heaters since I didn't need one secondary to make DC for the chip amp. The B+ doubler and filters used cheap radial caps, I think they were less than 2$ each. They are all 47 uF. I will upload an as-built schematic when I complete it, but it's very much like any two-tube, single-ended amplifier.

    20210924_204548_LR.JPG 20210924_204605_LR.JPG
     
    Asmith, SRHmusic, gtrwrks and 2 others like this.
  5. SerpentRuss

    SerpentRuss Tele-Meister Ad Free Member

    Age:
    58
    Posts:
    223
    Joined:
    Feb 17, 2021
    Location:
    North Carolina
    The B+ voltage settled out at 261 volts. The pre-amp tubes and the OT are all getting exactly the same voltage. There is a 110-ohm screen-saver resistor on the 6CL6 socket and the voltage there is about 252.

    The cathode resistor for the 6CL6 is a 270, 3-watt with a 47 µf bypass cap. The voltage drop across this resistor is 7.04 volts for an idle current of 26 mA, about 90% dissipation. It's warm but the enclosure is well-ventilated and the fins and aluminum structure help keep the total temperature under control.

    Pre-amp tubes have 1.8 K cathode resistors, 100k, load resistors and 33 µF bypass caps.

    A "feature" of my stellar, seat-o-the-pants wiring job is I got the tilt-tone reversed, so treble is all the way left and bass is all the way right. The tone control is really the star of this little amp, it works remarkably well though it is very interactive with the gain as well.
     
    Last edited: Sep 25, 2021
    RodeoTex, Asmith, stormsedge and 3 others like this.
  6. NSB_Chris

    NSB_Chris Tele-Holic

    Age:
    56
    Posts:
    622
    Joined:
    Sep 2, 2017
    Location:
    New Smyrna Beach Florida
    Thanks for posting the evolution. Can you post your schematic for posterity.
    How are you silencing the speaker output when you go headphones only? Do you have a dummy load on the output if no speaker is attached?
     
    Telekarster likes this.
  7. Telekarster

    Telekarster Friend of Leo's Gold Supporter

    Posts:
    2,931
    Joined:
    Aug 14, 2019
    Location:
    Earth
    That's a cool little amp man! Good job!
     
  8. SerpentRuss

    SerpentRuss Tele-Meister Ad Free Member

    Age:
    58
    Posts:
    223
    Joined:
    Feb 17, 2021
    Location:
    North Carolina
    Yes, I am using a switching jack on the rear speaker output. You can see a 10-Ohm, 5-watt power resistor (White) under the jack in the photo.

    The headphone jack is always fed. There is a 110-Ohm dropping resistor off of the OT output on the positive side feeding both the tip and the sleave of the stereo headphone jack (top jack on front panel).

    My son has been using the amp constantly. He has a stereo insert cable plugged into the headphone jack going into his Focusrite audio interface. He's been experimenting with mic and cab simulations in his DAW (Ableton? Reaper? it seems to change quite a bit). He has also tested the amp with the 10-inch speaker in his Pro Jr., which is again dead in the water, and my next project. This Pro Jr lived on a back porch, broken and unplayed for over 10 years and the traces are lifting off the circuit board and breaking/burning. The last failure resulted in the loss of the bias voltage which left the tubes all nice and toasty. I am going to probably put in a hand-wired deluxe or perhaps a matchless spitfire. Right now, I leaning toward the spitfire circuit because I like self-biasing amp designs. I tend to throw things together, but I want to do a righteous job on the Jr. rebuild.


    JackResistor.JPG .
     
    NSB_Chris likes this.
  9. SerpentRuss

    SerpentRuss Tele-Meister Ad Free Member

    Age:
    58
    Posts:
    223
    Joined:
    Feb 17, 2021
    Location:
    North Carolina
    One of the many things I've learned from this project is the absolute necessity for breadboarding any SS components you might use in an amp design before you solder them on. I was foiled by the TDA7266M however because it uses an odd-ball offset pinout. So the spacing is like a TO package, but it won't plug into a breadboard because the second row is not in line with the first.

    TDA Pin-Out.JPG

    When I put the chip onto the project board, I bent the even pins up, cut off the ones that weren't used, and ran pig-tails to unused rows on the project board from the remaining pins. So pins 2, 4, 6, and 8 were hand-wired, the odd-numbered pins were soldered directly.

    Where I screwed up was accidentally bypassing my input resistor when I put the board in the amp and tied in the volume pot. The input resistance of the chip amp was 25k. It needed additional protection from the swing of the preamp. I was testing the first iteration of the amp with headphones, and at the time I had a 1K resistor feeding the headphone jack instead of the 110-ohm resistor I have now, so the headphone output was very attenuated. When I turned the amp up after letting the pre-amp warm-up, I fried the input of the chip amp. After all that work, it lasted a glorious 15 seconds. It did sound pretty good, though.

    Pulling the board out and replacing the chip seemed like a lost cause, I would have to unwire a bunch of stuff just to swap it. That's why I decided to go with a tube output stage.
     
    SRHmusic and cometazzi like this.
  10. SerpentRuss

    SerpentRuss Tele-Meister Ad Free Member

    Age:
    58
    Posts:
    223
    Joined:
    Feb 17, 2021
    Location:
    North Carolina
    Thanks, I still have to put some kind of pilot in it, if I can ever get it away from my son to take it back apart.
     
    Telekarster likes this.
  11. SerpentRuss

    SerpentRuss Tele-Meister Ad Free Member

    Age:
    58
    Posts:
    223
    Joined:
    Feb 17, 2021
    Location:
    North Carolina
    OK, Here is a preliminary, as built, schematic. I have to warn everyone that I tend to work off of napkins and pieces of cardboard, and this is my first attempt to use DIYLC. It's a pretty impressive program. Bottom Line, if something looks wrong, it probably is. For the most part, this is a simple amp and there are hundreds of amps like it. The major difference is the 6N2P preamp. It's quite like a 12AX7 except the heaters are 6.3 volts, so no connection to pin 9 at all, or you can ground it. Mine is floating. The 1.8K cathode resistors would have been 1.5K, but I didn't have any, not even 1/4 watt. Since it works so well, I probably will not change them. The headphone dropping resistor was chosen on the third attempt. I have big AKGs and my son has the same set. I am also pretty deaf from doing industrial work for over 30 years. If you build something like this, a 1K volume pot as a headphone trim might be a good idea, that way you can drive the amp to where you want it, and then adjust the volume. Just me thinking about next time.


    I'm also including the screenshot of the sweeps from the tone stack calculator (Great Program). I used the Big Muff, template as a starting point since it is a tilt tone control.

    This is about a $90 dollar build, minus cabinet. The PT, delivered is less than $20, the OT can be had for about the same, The tubes can be found for way less than $5 dollars each. that gives you $40 bucks for the tube sockets, jacks, and other little bits. It's a cheap amp to build.

    TightSqueeze.JPG



    TightSqueeze Tone.JPG
     

    Attached Files:

    Last edited: Sep 26, 2021
    Yer Crazy Uncle and Telekarster like this.
  12. teleman1

    teleman1 Friend of Leo's

    Age:
    66
    Posts:
    2,743
    Joined:
    May 16, 2003
    Location:
    Arizona
    May I call you, DAD?
     
  13. SerpentRuss

    SerpentRuss Tele-Meister Ad Free Member

    Age:
    58
    Posts:
    223
    Joined:
    Feb 17, 2021
    Location:
    North Carolina
    LOL,

    Since you're a couple of years older than me, it might be a bit awkward.

    My guitar-playing son is a senior in high school, the youngest of four, the only one at home now, and the only one of my children that has a passion for music. So, we're as tight as two people that are 40 years apart can be. While I wish he was into the amp thing, he's a player. His biggest compliment is, I like it. He likes this.
     
    SRHmusic, teleman1 and Mad Kiwi like this.
  14. SerpentRuss

    SerpentRuss Tele-Meister Ad Free Member

    Age:
    58
    Posts:
    223
    Joined:
    Feb 17, 2021
    Location:
    North Carolina
    HELP NEEDED!

    So tonight I got the amp back and did a bit of work and ran upon a mystery. First, I neatened up some of the solder joints inside and I clipped one of the 7 VAC secondaries, and changed the value of the artificial center tap resistors to 68K. This got the heater voltage down to 6.5 from 6.8, which I'm hoping will help with valve longevity. I also added a pilot lamp, inside the case. It shines out via the holes in the front panel.

    OK, now for the mystery. I have a low-frequency oscillation. Evidently, I did not turn both the gain and the master volume up all the way, so I never noticed it. Of course, my son had no trouble finding the issue. It's a pretty bad oscillation around 27 HZ. I know this because I have a graphical voltmeter that has a frequency counter. At its worst, it's putting out about 1.7 VAC to the test speaker. Sounds bad and steals power. My first thought was it was the Cathode bypass capacitor causing some kind of resonance. The easiest test was to increase the value by running another cap in parallel. I thought it might make it worse and it did make it a tad louder and lowered the oscillation to 24 HZ. So I lifted both caps and the issue went away. Of course, the amp doesn't sound as good either.

    So, I have a 270-ohm cathode resistor. Initially, 47 uF gave me a 27 HZ oscillation. I think that should have given me about a 12 HZ roll-off knee. When I paralleled it with 33 uF that gave me 80 uF, which lowered the roll-off to 7 HZ.

    Two questions: What value should I try, 16 uF, 22 uF? Also, is there anywhere else in the circuit I can fight this oscillation? Do I need to decrease the value of my coupling caps?

    Thanks for any insight you can provide.
     
  15. NSB_Chris

    NSB_Chris Tele-Holic

    Age:
    56
    Posts:
    622
    Joined:
    Sep 2, 2017
    Location:
    New Smyrna Beach Florida
    This doesn't make sense to me. The artificial center tap resistors are just creating a voltage reference for the heater lines and represent a low load on the heater circuit compared to the valve heaters. To lower the voltage you would have to ether put resistors between the secondary supply and the heaters or put so much load on the unregulated supply that it would sag a bit. Since increasing the artificial center tap resistors would effectively lower the current load on the heater supply a bit, if anything, it would increase the supply voltage. I would think that adding the pilot light to the heater supply would be more likely to bring the heater supply voltage down. Maybe I am missing something.

    Could it be that the oscillation is coming from the fact that the preamp tube plate supply is common with the output tube plate supply? I have seen others state that not putting filter nodes between input gain sections and phase inverter sections etc. can lead to oscillations as the valves interact with each other. Maybe that is what is going on here.
     
    andrewRneumann and King Fan like this.
  16. King Fan

    King Fan Poster Extraordinaire Ad Free Member

    Posts:
    6,995
    Joined:
    Jan 1, 2013
    Location:
    Salt Lake City
    The oscillation is way beyond me, so lemme think about this bit. I agree that changing the artificial CT resistors isn’t a fix? But maybe… you were using two heater secondaries and now you’re using one, increasing the loading? Beyond that, as I suspect you know, Merlin has ways to reduce heater voltage. In a perfect world of course we target 6.3 +0/-5%.

    I'll stay tuned to hear smart folks tackle the oscillation.
     
  17. SerpentRuss

    SerpentRuss Tele-Meister Ad Free Member

    Age:
    58
    Posts:
    223
    Joined:
    Feb 17, 2021
    Location:
    North Carolina

    Thanks for your response. On the heater, I took away one set of secondaries and that got the most bang for the buck. Reducing the artificial CT resistors increased the current in the heater circuit. It got me about another .06 volt drop. If you think of the heater transformer as an unlimited resource, making this change wouldn't help, but every increase in load causes a bit more voltage sag. If I had 3 tubes, the 7 VAC supply would be perfect with one set of secondaries. As it was, I didn't have room for a couple more power resistors and I wouldn't want the extra heat anyway.

    I actually looked at adding another filter drop for the pre-amp for exactly the reason you state, and I just don't have the space to do it now. If I had planned with it from the start, it would be a different story. Right now, with a much-reduced bypass cap (10 µF) on the 6CL6, the amp sounds fine and the oscillation is gone. I am probably going to make a refined version of this as a Christmas gift for a guitar-playing friend of mine, and it will be slightly larger and better planned.
     
  18. NSB_Chris

    NSB_Chris Tele-Holic

    Age:
    56
    Posts:
    622
    Joined:
    Sep 2, 2017
    Location:
    New Smyrna Beach Florida
    Good discussion! Glad you got the oscillations down! I would not have known that reducing the bypass cap would do the trick. I would not have even thought to do that. Have to remember that for the future

    Your original schematic said that you had 470 ohm false center tap resistors and then stated you changed them to 68K resistors. That would have gone in the direction of decreasing leakage current and if anything would have increased the heater voltage level. Maybe you meant 68 ohm resistors?

    Ahh, I see you had paralleled the two heater secondaries and just took one out. That makes sense! King Fan figured it out right away but I didn't realize what you were saying.

    When you paralleled the two heater secondaries, how did you confirm which ones to join. Did you connect two leads and then measure the total voltage between the remaining two to confirm the correct connections?
    I am asking because I may need to do this in the future.
     
  19. Lone_Poor_Boy

    Lone_Poor_Boy Tele-Holic

    Posts:
    642
    Joined:
    Sep 14, 2021
    Location:
    Colorado
    You build amps like I cook. Well done!
     
  20. SerpentRuss

    SerpentRuss Tele-Meister Ad Free Member

    Age:
    58
    Posts:
    223
    Joined:
    Feb 17, 2021
    Location:
    North Carolina
    I think I'm right about lowering the center-tap resistors as well, because I tested before and after. Electrically, think about them now being another heater element in parallel with the other two, since the two 68K resistors connect to each other via ground. The ground reference just "un-floats" the AC in reference to ground.

    I trusted the start-stop color code from Antek. They suggest in their literature putting the secondaries in parallel for maximum amperage. I'm pretty sure something strange would have happened if I had gotten it reversed. But yes, if all the wire had been the same color I would have checked them with a volt meter first.
     
IMPORTANT: Treat everyone here with respect, no matter how difficult!
No sex, drug, political, religion or hate discussion permitted here.