Tube boost + overdrive running off a 9 volt battery

[Jered]

  If you go high voltage, might as well build an amp.  http://www.harmonicappliances.com/powerman/design/pman_full_A.pdf The 6021's sound great at higher voltage. Haven't tried the 6111 or 6112 sub mini's at higher voltage.
  Jered

[brett]

Hi
with this kind of thought in mind -

in that case it would be better to use a 12V regulator for heaters and ditch the battery all together

- I built this today :



That's a faily low parts count - hence the pcb was only 2 1/2 x 1 1/2, including the B9a socket 
My wall-wart is the charger for my 14.4V rechargeable drill, which puts out about 19V.  The 7812 was getting very hot, so I used 3 diodes (denoted Dx*) ahead of it to drop the voltage to about 17V.  With 17V input and 12V output the 7812 wasn't burning me any more.

The results were a very quiet circuit (DC throughout and low gain probably help) and some promising initial results.

Here's some scope output (1V grids) for different tubes and input levels (sine waves):

12AU7, 200mV input


12AU7, 500mV input


12U7 (low voltage dual triode, built for car radios etc, pin for pin compatible with 12A*7s), 200mV input


12AX7, 200mV input


As you can see, the 12AX7 doesn't have much gain and is already clipping with 200mV of input. 
But the 12AU7 manages about 6V p-p of output (gain of 30) and seems to work very well.  Clipping is at just about the right amount of input (200mV p-p is about a fresh plucked string and 500mV a chord).  Both sides of the output started to clip at about 1V p-p input, and the output got fairly square at about 2V p-p.  So I think it will break up well with someboost in front.

I'll post some MP3s tomorrow, and try some more variations - maybe a small resistor on the cathode and a larger plate resistor on the second triode ?
cheers

[RLBJR65]

If you want longer battery life use 2 - 9 v in parallel.
More head room with 18v ? Try 2 - 9V in series using a center tap for the heater.

but make sure you run heaters with one battery only, 18V on heaters is no good.

Yes 18V on the heaters would be very bad! Guess I should have noted the 5 / 3 on the 18V pic is the heater connection as related to Dano schem.

It was just a suggestion to add a bit of room and still keep it a simple low voltage project as it was intended to be.

  If you go high voltage, might as well build an amp.  http://www.harmonicappliances.com/powerman/design/pman_full_A.pdf The 6021's sound great at higher voltage. Haven't tried the 6111 or 6112 sub mini's at higher voltage.
  Jered

Yep!

Nice work Brett!
I was going to recommend the 12U7. There are tons of used and NOS 12U7 around don't know of any manufacturers making new ones though.

[brett]

Hi
here is a clip.  Please excuse the playing (nerves).

First part is clean (some breakup due to overloading the sound card), then the same riff played through the Triode triode (with a little boost), then a few chords (Triode triode, no boost), then the riff again, same as before, just played "harder".  It's a bit hard to pick up in the clip, but there's a reasonable amount of difference between the second and fourth parts, and the circuit responds quite a lot to playing style and "attack".

http://www.aronnelson.com/gallery/v/BinOfBrett/Triode+triode+clean+-+boost1+soft+-+boost2_less+-+boost+2+hard+.mp3.html

With or without boost, I reckon this is a great little circuit.  Although most of my amps have tube preamps and are dirty, I would still use this as an extra valve stage.  It has a feel of its own.  It's a bit like a low gain pre-amp (e.g. the low gain channel in my Marshall JTM60), but a bit different, too.  Very crisp.

I think it would be an excellent front end for many solid state amps, offering pleasant compression and distortion.

Maybe to temper the highs, you could add some high frequency bypass caps in parallel with the plate resistors.  About 47pF would be right for the 470k resistor, and 100pF for the 220k resistor.  Also, the output is so high that a 5:1 voltage divider (e.g. 470k:100k) could be put ahead of the output pot.
cheers

[mojo_hand]

Possible improvement, depending on context:

I note that the original schematic and layout specify that the triode at pins 1-3 be used for the voltage gain stage, and 6-8 for the follower.

This isn't usually the best way to do it with a 12A*7.  With battery power or regulated DC going through the heaters, it doesn't matter which triode you use in which position, but if you're using AC or less-than-pristine DC, the triode at pins 6-8 will be quieter, due to the location of the heater pins on the tube.  (You can find mention of this in the data sheets on many 12A*7 series tubes.)

Just thought I'd throw that out there, since this design seems to be giving rise to a lot of variations, and some of them might benefit from swapping the triodes.

[Krinor]

So, put simple what you're saying is that it might be better to use pins 1-3 for the follower and pins 6-8 for the gain stage ?
Will there be a noticable aural difference in such a low voltage circuit ?

[brett]

It definately matters for circuits with AC heating and high gain, such as the Real McTube.

Any of the all-DC designs are ok regardless of order.

By the way, the Triode triode has plenty of gain in both triodes, so it's two gain stages, not a gain stage and a follower.
cheers

[mojo_hand]

So, put simple what you're saying is that it might be better to use pins 1-3 for the follower and pins 6-8 for the gain stage ?

Exactly.

Will there be a noticable aural difference in such a low voltage circuit ?

If you're using AC or rectified AC for the filaments (or a hummy DC wall wart), hum will be reduced by using 6-8.  How audible it will be depends on the circumstances.  If you used shielded and/or twisted pair cabling for both signal and heaters, and approximated star grounding, it should be quiet enough that you'll notice the difference.  If you did a bad job wiring it, and you've got a noisy fuzz feeding it signal, I'm sure you'd never know one way or the other.  It's not a big deal, and it's a total non-issue in any circuit using clean DC for the filaments, I just mentioned it because few circuit tricks require zero materials and have zero downside, but I like using those few that do exist.

[mojo_hand]

By the way, the Triode triode has plenty of gain in both triodes, so it's two gain stages, not a gain stage and a follower.

D'oh!  I mis-remembered the schematic.  I hereby renounce my use of the word "follower" in my earlier post.  Everything else still applies, however.

[Krinor]

So, put simple what you're saying is that it might be better to use pins 1-3 for the follower and pins 6-8 for the gain stage ?

Exactly.

Will there be a noticable aural difference in such a low voltage circuit ?

If you're using AC or rectified AC for the filaments (or a hummy DC wall wart), hum will be reduced by using 6-8.  How audible it will be depends on the circumstances.  If you used shielded and/or twisted pair cabling for both signal and heaters, and approximated star grounding, it should be quiet enough that you'll notice the difference.  If you did a bad job wiring it, and you've got a noisy fuzz feeding it signal, I'm sure you'd never know one way or the other.  It's not a big deal, and it's a total non-issue in any circuit using clean DC for the filaments, I just mentioned it because few circuit tricks require zero materials and have zero downside, but I like using those few that do exist.

Thanks for sharing. My build seems quiet enough, but I'll try these tricks in the next one I make and see if it gets even better.

[Caferacernoc]

Great job Brett! The pictures you made with your scope show the 12au7 and 12u7 looking very similar. Any sound differences?

[bancika]

There are tons of used and NOS 12U7 around don't know of any manufacturers making new ones though.

where?

[brett]

Hi

The pictures you made with your scope show the 12au7 and 12u7 looking very similar. Any sound differences?

The 12U7 is definately a softer, cleaner tone.  Even when boosted fairly hard, the 12U7 was mild.  The 12AU7 broke up much earlier and harsher. 

It's probably a bad analogy, but I was thinking that the 12U7 had a lighter, overdrivish, Fendery feel, while the 12AU7 is a dirtier, more Marshally tone. (Old Marshall, like an 18 watter or JTM, not a JCM or high-gain Marshall)

cheers

[RLBJR65]

where?

I have a surpluss store near me that has a box of them. The problem is that I don't think they would go and look if you called etc.
If you want some let me know I can go in a couple of weeks. The last one I bought was $4.00.
I just Googled "12U7 tube" and found 2 places on the first page

[Caferacernoc]

"The 12U7 is definately a softer, cleaner tone.  Even when boosted fairly hard, the 12U7 was mild.  The 12AU7 broke up much earlier and harsher. "

Awesome. Thanks for the info. I have some 12au7 tubes already and 6011's on the way to play with on these low voltage type designs. I will get some 12u7's also.

[Lonzo]

What a great looking circuit; I've been trying to get into a tube build for some time. Thanks for sharing circuit.

Dose anyone have any thoughts on the affect of eliminating the tone control?  Would there be some additional gain obtained as a result? 

Thanks for your thoughts,
Lonzo

[Krinor]

I don't think it makes such a big difference. At full tilt there is a tiny bit more gain than when it is turned all the way down. This tone control works fine, but it's not that effective. Personally I leave mine at about 1 o'clock. If your point is that you want to save some space and leave out a pot, you might still want to shape your tone by making a preset treble bleed. For instance you could replace the pot with a 50k resistor (to put it in the middle position). Or put sockets in and experiment with different values until you find the tone you like.

Good luck. This circuit is very easy to make and a lot of fun too. Use a 12 volt wall wart!

[Faber]

I was thinking about this circuit and I came up with something for someone who built this for a second time, or someone who is WAY too ambitious (like me).

Okay, so I looked at the argument about 12V vs 9V and I was thinking about using a 12.6V transformer and a full wave rectifier TUBE.  Now, this might be too bulky for such a small pedal, but it would kinda be more like a real tube amp...  Maybe I'm just crazy... errrr... maybe I'm crazier than I thought I was!

Didn't I also see something about the tube heaters needing (or wanting) 12.6V???

Not to mention, 2 TUBES LIGHTING UP!!!

[Zben3129]

As per tube recto, there would be no need, except for look, as a tube recto and diode recto do the exact same thing in this particular situation.

For the 9v and 2 tube argument, I agree. With both heaters + circuit drawing power, this thing is a 9v battery eater. In my pedal, I power by a 12vac wallwart, then rectify internally with diodes.

Zach

[mojo_hand]

I was thinking about this circuit and I came up with something for someone who built this for a second time, or someone who is WAY too ambitious (like me).

Okay, so I looked at the argument about 12V vs 9V and I was thinking about using a 12.6V transformer and a full wave rectifier TUBE.  Now, this might be too bulky for such a small pedal, but it would kinda be more like a real tube amp...  Maybe I'm just crazy... errrr... maybe I'm crazier than I thought I was!

Didn't I also see something about the tube heaters needing (or wanting) 12.6V???

Not to mention, 2 TUBES LIGHTING UP!!!

A 12.6v transformer will not give you the voltage you want for the filaments, unless you skip rectification and just use the AC.  If you hook the heaters up in series, anything from 9v to 13v will work, although 9 is pretty low to do the job, and anything over about twelve and a half will reduce tube life.  Something like 11-12 is probably ideal.  If you're hooking up the filaments in parallel, halve those voltages.  A 12.6v transformer produces 12.6 VAC, after that goes through a bridge rectifier and hits filter caps, what you end up with is 12.6 * 1.414 volts, less the drop across the rectifier diodes.  In practice, this works out to maybe 16.5-17v, which is high enough to fry the heaters very quickly.  If you use a different rectifier topology, you end up with 12.6 * 0.707 volts (minus losses), which might get you 8 volts, with luck.  Tube diodes have higher internal impedance than silicon, so the voltage drop would be greater, but I really don't think by enough to drop your 16.5-17v to 12.6 or less.

If you want to do use tube-rectified DC for the heaters, you need a transformer in the 7-9v range (or half that if hooking them up in parallel).  Or higher voltage and a regulated power supply.  Do remember to get a bigger transformer, since you'll need to provide the rectifier with heater current, too.

Bear in mind that tube rectification's main mojo (power supply sag) does not apply to Class A circuits, which this one is, so the only benefit you're likely to get for the added expense and bother, will be cosmetic.  If you want to make it more like a tube amp, there are probably better ways of doing that.  You could, for example, use something like a 20k:2k output transformer.  This would cut the output at clipping from close to 9v to a more manageable 0.9v, lower the output impedance, and probably make it sound a bit more like the real deal.  Since you're very ambitious, I'll leave figuring out how to do that as an academic exercise.  It's not hard, and learning is good.