tube OD pedal too loud, need help

[HTH]

hi all,

I've built a tube OD pedal with a single 12AX7 running plate voltage in the 130v region. Its really simple, just two basic gain stages cascaded much like a BF Fender normal channel...

V1a
33k grid stopper
1M grid leak
100k plate
1k5 cathode (no bypass)
22nF from plate into a 1M (log) gain pot

V1b
220k from gain pot into V1b's grid
100k plate
1k5 cathode with 22uF bypass cap
22nF from plate into a 100k (log) volume pot feeding the output jack

It sounds great, but the volume is much higher than the bypassed tone even with the volume set at 9 o'clock (i.e. barely on).

Would a larger pot (say 500k) give more range or do I need a voltage divider before the volume pot to dump some signal to ground?

[Steve Conner]

Originally Posted by HTH

22nF from plate into a 100k (log) volume pot feeding the output jack

First thing to try is to put a 1M resistor in series with the 22nF. This will form a 11:1 divider with the 100k resistance of the pot.

This will also increase the bass, as the time constant changes from 22nF/100k to 22nF/1100k. So you may need to change the capacitor to 2n2, or go the voltage divider route.

If it's still too loud, put 10k across the outer lugs of the pot, to up the ratio to 111:1.

A larger pot won't give noticeably more range, unless by chance it also has a better log taper.

[HTH]

Originally Posted by Steve Conner

First thing to try is to put a 1M resistor in series with the 22nF. This will form a 11:1 divider with the 100k resistance of the pot.

This will also increase the bass, as the time constant changes from 22nF/100k to 22nF/1100k. So you may need to change the capacitor to 2n2, or go the voltage divider route.

If it's still too loud, put 10k across the outer lugs of the pot, to up the ratio to 111:1.

A larger pot won't give noticeably more range, unless by chance it also has a better log taper.

cheers Steve,

I already tried lowering the pot's value to 25k by paralleling a fixed resistor across it - not really any better (maybe margainly so).

Will try your suggestions - cheers.

[Mark Hammer]

The time tested strategy is to arrange it such that the full rotation of the pot actually only covers a smaller range of the volume attenuation.

So, say you had a 50 volume pot. If you inserted a 47k/51k fixed resistor in series with the input of the pot, the resistor/pot combination would behave as if it it were a 100k pot that would never let you turn up past halfway. meanwhile, even though only a maximum of half the output is permitted to "escape" the circuit, yuo can govern that remaining 50% with the full rotation of the pot, allowing you to dial in nice small differences.

You can tailor the value of fixed resistor and pot-value to your tastes/needs. For example, you might find that a 75k fixed resistor and 25k pot does the trick for you, or an 18k resistor and 10k pot, and so on. Identify how much the maximum untainted output needs to be reduced to optimize usability, then identify what the optimum terminating resistance (pot or pot+resistor) should be for impedance-matching purposes, and pick your resistor+pot combo.

[HTH]

Originally Posted by Mark Hammer

The time tested strategy is to arrange it such that the full rotation of the pot actually only covers a smaller range of the volume attenuation.

So, say you had a 50 volume pot. If you inserted a 47k/51k fixed resistor in series with the input of the pot, the resistor/pot combination would behave as if it it were a 100k pot that would never let you turn up past halfway. meanwhile, even though only a maximum of half the output is permitted to "escape" the circuit, yuo can govern that remaining 50% with the full rotation of the pot, allowing you to dial in nice small differences.

You can tailor the value of fixed resistor and pot-value to your tastes/needs. For example, you might find that a 75k fixed resistor and 25k pot does the trick for you, or an 18k resistor and 10k pot, and so on. Identify how much the maximum untainted output needs to be reduced to optimize usability, then identify what the optimum terminating resistance (pot or pot+resistor) should be for impedance-matching purposes, and pick your resistor+pot combo.

Still haven't got round to doing any more with this, yet. However, I measured the 100k volume pot at the setting where it just gives a little volume boost and it measured around 10k, so a 10k pot and 91k resistor would be about right (I'm sure a 100k fixed resistor wouldn't really make too much difference though and it's a more common value).

cheers to all who've chipped in so far, I'll post an update (soon).

[Mark Hammer]

Consider 82k + a 25k pot. Pots are usually a bit lower than their stated nominal value, and you never know when you'll have a lower-level signal to feed the unit and may want a bit of additional boost. So, unless I've misunderstood you, the combo suggested here might give you what you want and might need some day. You'll probably never move the pot above the 1:00 or 2:00 position, but that's okay.

[HTH]

I tried a 100k fixed resistor into a 10k log pot and it was much better.

Not quite there yet though, gonna try a 220k fixed resistor into the 10k pot.

How about running a 100k to ground off the volume pot's wiper to dump some signal to ground?

[dai h.]

Ian, you can "turn down" the stage on the other side of the cap as well changing the ratio of the plate resistor. Think of the plate resistor as a pot maxed out with the ground going to AC ground (via the local filter, or "decoupling" cap) instead of DC ground (as is the "regular" pot on the other side of the output coupling cap). So if the plate R was a pot, the wiper would be connected to the output coupling cap, and as that wiper moved closer to (AC) ground (which is B+) the signal is reduced. Don't know if that will give what you want out of it but will reduce volume.

[Mark Hammer]

Originally Posted by HTH

I tried a 100k fixed resistor into a 10k log pot and it was much better.

Not quite there yet though, gonna try a 220k fixed resistor into the 10k pot.

How about running a 100k to ground off the volume pot's wiper to dump some signal to ground?

Sure. That'll help out. The suggested added resistor is going to be in parallel with the ground leg of your pot. This will change the taper so that the ground leg goes low "faster". I.E., where you might have had to turn down to 11:00 to get what you want, you'll be able to do that now at 1:00.

[HTH]

Originally Posted by dai h.

Ian, you can "turn down" the stage on the other side of the cap as well changing the ratio of the plate resistor. Think of the plate resistor as a pot maxed out with the ground going to AC ground (via the local filter, or "decoupling" cap) instead of DC ground (as is the "regular" pot on the other side of the output coupling cap). So if the plate R was a pot, the wiper would be connected to the output coupling cap, and as that wiper moved closer to (AC) ground (which is B+) the signal is reduced. Don't know if that will give what you want out of it but will reduce volume.

are you talking about using a split load plate resistor Dai?

[HTH]

Originally Posted by Mark Hammer

Sure. That'll help out. The suggested added resistor is going to be in parallel with the ground leg of your pot. This will change the taper so that the ground leg goes low "faster". I.E., where you might have had to turn down to 11:00 to get what you want, you'll be able to do that now at 1:00.

thanks Mark,

I'm gonna give this more tweaking over the weekend when I can have the pedal running quite loud into an amp (its currently 9:30pm here and I don't think the neighbours would be very happy with my 4x12 chucking out uber-bass at this time of night, ha ha).

[HTH]

also, thinking I might attenuate the signal a wee bit by throwing an 18w Marshall-style tone control in there. I'll give a bit extra control over the tone and attenuate the level a little bit.

[dai h.]

Originally Posted by HTH

are you talking about using a split load plate resistor Dai?

yeah, that would be the fancy term for it lol. Same stuff Mark is talking about, with the 10k, 91k etc. applies. Picture the triode with the plate resistor facing down as analogous to the pot after the coupling cap except the plate resistor's ground goes to *AC* ground instead of *DC* ground (as the pot's ground side is). And feedback is another technique that can reduce gain but I've not done much experimenting with that.

also since gain is 1st gain X 2nd stage gain, if you don't need so much, then you could try putting the triodes in parallel maybe have two bypass caps to switch when using a single coil or hum. sort of like norm./treble ch. on a marshall. Oh yeah another one to reduce gain is series R with the bypass cap since you have one on the second stage (22uF).

[HTH]

Originally Posted by dai h.

yeah, that would be the fancy term for it lol. Same stuff Mark is talking about, with the 10k, 91k etc. applies. Picture the triode with the plate resistor facing down as analogous to the pot after the coupling cap except the plate resistor's ground goes to *AC* ground instead of *DC* ground (as the pot's ground side is). And feedback is another technique that can reduce gain but I've not done much experimenting with that.

also since gain is 1st gain X 2nd stage gain, if you don't need so much, then you could try putting the triodes in parallel maybe have two bypass caps to switch when using a single coil or hum. sort of like norm./treble ch. on a marshall. Oh yeah another one to reduce gain is series R with the bypass cap since you have one on the second stage (22uF).

its funny you should mention feedback - thats on the list of things to try, however not to reduce the volume (as I'm wanting at the minute) - just to tighten up the bass end and make it punchier.

with the two triodes in series, there isn't as much gain as you'd think (unless the volume pot is on full), so I'd rather keep it cascaded for the minute. the gain and volume pots are very interactive, just like many old amps - it's really down to balancing them out to get the right tone, but its still much louder than the bypassed signal.