Can someone show me how to run a Subcaster off a bipolar supply. Practical?

[Skruffyhound]

Yes, I know I'm all questions on the forum at the moment, but I'm a bit pressed trying to get a blues harp pedal sorted out for a mate and can't take my usual time (several years) to figure it out myself.

It looks like said pedal will have at least one module that requires a bipolar power supply and so I'm thinking about the advantage in clean headroom that would give to the Subcaster especially since high gain feedback seems to be a harp problem.

Unfortunately, I have no clue where to start.

I'm concerned that the heater draw if taken from the + side might skew the power supply. I'd thought to use an LT1054 or an ICL7660 (Max 1044-alike) to create the bipolar supply. That's just one of the issues though. I probably will ditch the LM317 and associated comps. to go with Rick/R.G.'s 18ohm 3 watt resistor to get 6.3V for the heater.

Anyway, if some kind soul could point the way I might have a crack at it, otherwise I'll make the 12V version, but it'll feel like an opportunity lost.

Other threads related to this build (which I'll collect together in a build thread if successful)
http://www.diystompboxes.com/smfforum/index.php?topic=100442.msg884050#msg884050
http://www.diystompboxes.com/smfforum/index.php?topic=100684.msg886953#msg886953

[frequencycentral]

- Run the heater from +12V using the 18R/2W resistor.

- Use ICL7660S or equivalent to create -12V rail.

- Connect the cathodes to -12V instead of ground. I would also use a cathode resistor and bypass cap, which would also mean recalculating the anode resistors so the anodes read half the supply rail, ie ~0V in this case. You'll find it sounds better with cathode resistors/caps and the anodes correctly biased. Adding voltage with the cathodes grounded is a waste.

[Skruffyhound]

19ohm 3 watt heater supply as shown to right below (sure I used an 18 ohm last time  ) The rest of the schem is obviously a vibracaster so take no notice of that



Edit: Rick posted while I was adding this.

[frequencycentral]

Yeah, but 18R/2W is easier to find (Banzai) and supplies a voltage within tolerance of the tube. 2W is more than enough.

[Skruffyhound]

Very helpful thanks mate.
Off to the breadboard I guess unless any other forumite has trodden this path and has the numbers stashed away for reference.

[Skruffyhound]

Yeah, but 18R/2W is easier to find (Banzai) and supplies a voltage within tolerance of the tube. 2W is more than enough.

I just found the ones I used last time (2009), 18 ohm 5 watt! I wasn't taking any chances was I

[Perrow]

This is almost exactly what I've been thinking about. I want to use a 6112 tube as a preamp for a TDA2030 amp and use the bipolar supply to help squeeze out all I can from that tube.

Will be watching this thread (for links to build thread ).

[Skruffyhound]

Ok, so after working my power difficulties out in the other thread (thanks to Paul (PRR)) I got back into this part of the project.

Not knowing anything about tubes per se, I trawled the internet for several days trying to find a similar project and see what kind of component values might be appropriate after reading Rick's comments above.
I didn't find much out there at all and by accident ended up back here looking at Aron's shakka tube project for the first time.


So making adjustments for the different pinout I breadboarded something that started as a Subcaster and morphed into a Shaka tube with 6111's (I tried 6112's too).
No dice, with a substantial boost in front it, it didn't sound awful, but alone it sucked. Literally, negative gain.
I need it to overdrive lightly and sound a bit tube-y but not anything high gain at all. It looks like this, literally a Shaka with the TL072 clipped off.

I couldn't figure out how to draw schematic pots properly in DIYLC, but you get the idea. Bias is also a pot with terminals 1 and 3 across the rails and wiper giving zero, or actually I found it sounded best when quite negative.
What idiot mistakes in layout or components did I make this time?

(oops, I can see R1 above should have been labeled 10k)
 

[frequencycentral]

Looks basically good. I'd dispense with the bias trimmer and tie the grid leak resistors to -ve. I'd also use a higher value for them, 1M maybe, 470K maybe.

What voltage are you measuring at the anodes?

[Skruffyhound]

I've had a really weird grounding issue that I still don't understand. Anyway, it's gone now. I moved the bias and -ve connection wire closer to the LT1054 on the bread board and ground started acting like it should and I got a little gain.

I re-breaded this to make sure I hadn't made any mistakes and then tried 1M on bias, much better thanks. I tried 2M too but it was better with the 1M's.

You were right about the bias trimmer too, now I've got it working it clearly wants to be all the way to minus anyway.

It's still a little harsh at full whack with a guitar, but it may be just fine, pulled back a little bit, for a blues harp (I'm making a harp stomp box).

Plates are at: Pin 1 = minus 3.1V , Pin 8 = 6.1V (up to 9V when strumming hard)

Did you ever make the "Le Craquement" with 6111's?

While I'm messing with this I'll try what I've got now with 6112's. Then with the Mosfet input from the Le Craquement.

Any other suggestions? I'm trying to limit it to 2 knobs for the dirt section (so four in total for the box).

[Skruffyhound]

Ok, I didn't get the Mosfet thing to work and it didn't sound as good with 6112's (where am I going to use those suckers) so I'm running with this:

[Skruffyhound]

Oops, bottom right cap = 100uF
Datasheet  recommends tantalum for the 10uF and the 100uF, but interestingly suggests that running a low value tantalum in parallel with electrolytic will give the desired ESR at
a reasonable price point. Absolute values of caps are not important, but beware making them too large.