Self designed tube pedal/preamp problem

[omt]

I have built and designed this thing (please note that i have no clue about what i am doing ) and all i get is a loud noise and a very weak distorted guitar signal, so i wondered if any of you guys know what i have done wrong (or if i have done anything right at all).
i wold be glad for any help.  

(heaters are conected to 12v Dc and the filaments is glowing, v+ is 12v dc +)

[Ge_Whiz]

At least tell us what the tube is! 12AX7?

[omt]

(shematic uploaded)

[B Tremblay]

What are the plate voltages?  Also is pin 4 connected to +12V with pin 5 grounded and pin 9 unconnected?

If the heater pins are correctly wired, try replacing the 200k resistors with 500k trimpots and adjust until you get sound.

[omt]

pin 4 is connected to v+ and pin 5 to ground pin 9 unconected. How do you mesure plate voltage?? (i m a noob)

[zeta55]

If V+ i s 12VDC, the plate voltage should be a voltage drop bellov. In this case I guess we're talking a starwed plate V+ design. Measure the plate voltage after the 200K resistors and ground.

/Krister

[omt]

I mesured 10,6V (after the 200k resistors and to ground)

[H.Manback]

Ummm, where are the cathode bias resistors The triodes have to be biased in a way that the grid voltage is negative to the cathode in idle mode. You do this by adding a resistor (and optionally a capacitor) between the cathode and ground.

Now that said, I have no idea if the same applies to starved plate designs.

What I would like to know is, how did you come up with this design if you have like you said no clue about what you are doing?

[omt]

okay i have some clue, i took bits and pices from the chandler tube driver, the real mctube and the hotbox and substituted what i had lying around but i was werry unsure, this design worked with only one tube(two stages) and had a great sound.

[omt]

But with two tubes it dosent work , now i only get a LOT of noise and a werry faint guitar signal

[omt]

About the catode bias resistor, i read that it can be dropped for a diferent sound(more gain)

[George Giblet]

When you say "noise' what do you actually mean: hum, hiss buzz?    If you are getting a hum then it could be that your 12V rail has got too much  ripple (because the tube heaters pull a reasonable amount of current).  Try adding a large cap (say 1000uF or more) across the supply.  You should also add an RC filter on the plate supply and not the heaters.  Look at how the mctube does it - it's like that for a reason.

Once you solve that problem, if you are still getting a weak signal I suggest playing around with the bias on the grids and the plate resistor values.  You have some grids unconnected and others to ground.  Look at Aron's shaka tube for an adjustable bias scheme.

[omt]

Thnx for the tip, il try that tomrrow now i am going to bed (its 2.30 here in norway)

[B Tremblay]

Add 1M resistors between grid and ground of the last two stages.  Add the trimmers as I suggested earlier, as well as 1k5 resistors between each cathode and ground.  Then, adjust the trims so each plate is close to 9v.

[R.G.]

Good suggestions.

It's oscillating, as George points out, because you have four stages connected to the same power supply connection. Break the power supply between the first and second tube. Put in a 10K resistor between them. Bypass the first tube and second tube independently with 100uF caps. That will get rid of the oscillation, if you have done grounding well.
The second tube grids do need pulled down, as noted.

Put 1.5K resistors in each cathode, bypass them with 22uF caps.

[slajeune]

Hi All,

omt, as b. tremblay suggested, start by adding the missing 1M resistor to ground after the .01uF (stage 3 and 4).  I wouldn't add cathode resistors yet.  Start by having sound going thru it.  Furthermore, in preamp stages, it is not required to have a cathode resistor / bypass cap.  There are two ways to bias a tube:

- cathode biased (i.e. with a cathode resistor and an optional bypass cap)
- Grid leak biased.  This is done by changing the value of the 1M resistor to ground.  This can be increased to bias the tube.

As you can see in the following circuit:

http://www.sopht.ca/schematics/sopht_12al8-v2.jpg

There are no cathode resistors on any stages.  The tube bias is developed by resistors R1, R5, R6 and R8.

So, I would say start by adding the grid to ground resistors.  Once this is up and running and you have some sound comming thru it, you can tweak the circuit to your needs by biasing each stage indivitually.

Hope this helps.

Thanks,
Stephane.

[omt]

Okay update: i tryed adding the 1M resistors, after that i herd sound, but the noise was still louder than the guitar so i tried addsing a large cap on the powersuply (4700uF) and splitting the powersuply with 10 k resistors. after i added the cap i messured 20V on the powersuply and it came smoke out of the 100k pot when i conected the power.

[B Tremblay]

Put 1.5K resistors in each cathode, bypass them with 22uF caps.

Would there be any chance for a clean sound with four stages set up like that?  I once tried a circuit with only two stages like that and found it provided an overabundance of distortion.

[omt]

Why do i suddenly get 20V when the capacitor is conected ?? I know i shold get a higher voltage but from 12-20 is pretty much

[H.Manback]

@omt: Probably you get a higher voltage now because you were measuring it with a big ripple before you added the capacitor. The meter evened it out and didn't measure peak value. Now that you have cleaner DC it measures it correctly.

@B Tremlay: Well if you have four stages, obviously you need less gain per stage to prevent the tubes from distorting the signal. You can do that by attenuating the signal being passed through to the next stage (with a voltage divider).