Troubleshoot for 12AU7 Valvecaster w/ SHO Boost

[tubetroubles]

I am building my first pedal and wanted to do an "always on" SHO boost followed by a switchable 12AU7 Valvecaster in a single enclosure. I built the pedal to my schematic and ran into nothing but silence when I tried to engage the valvecaster portion of the circuit. Without the switch engaged, the SHO boost on it's own sounds wonderful. Actually with the 25K trim pot/22K drain resistor combo the clean boost can get unbelievably loud. But, alas, when I switch on the valvecaster portion I get nothing. I ran some DC voltage checks and found that when the switch is off, the V+ voltage is right around 9V (which is what I would expect). However, with the switch engaged it drops off to 3.4V.

I know I'm probably missing something incredibly obvious, but I'm a total novice when it comes to circuit design and was wondering if you guys had any comments on the circuit or where I might have gone wrong. Appreciate the input!

[iccaros]

I can not see the picture added here at work, but if you can post your voltages, and what is your power source?

[artifus]

pin 3 & 8 ?

http://www.aronnelson.com/DIYFiles/up/12AU7-6111_Valve_Caster_Summary_Rev002.pdf

[puretube]

Besides the chance of partially cancelling signal through the combined Rk,
the big problem is the PS:
can it deliver the required ~150mA for the filaments?
That voltage-drop tells stories...

[tubetroubles]

Thanks for the feedback! Sadly I can't access aron nelson's website from work....have to check out that link when I get home.

...and whoops...Pin's 3 and 8 are not actually connected. Forgot to update the schematic to reflect that. I moved the gain pot over to the first stage cathode and wired the second stage (pin directly to ground as done on Matsumin's original design. This should resolve the partially canceled signal puretube mentioned, but still no sound.

For the incoming power supply, I'm using a 9V negative tip wallwart for the incoming power and it's a Godlyke rated for 2000mA, which should more than handle the 175 mA for the tube filaments.

[tubetroubles]

*(pin 8 ) not (pin sunglass smiley.... *

Also, one thing I should point out is that the filament heaters are always on regardless of what position the switch is in. In other words, 150-175 mA should always be going to the heaters. I've tried this pedal with the tube removed and installed (with the valvecaster portion off), and the result is the same. SHO boost sounds awesome, valvecaster is dead. The only things that change when hitting the switch are applying a voltage to the plates and switching the LED on (I think...).

When I get home, I'll do a more thorough voltage check and update the pic to show you what I'm getting in various locations on the pedal. Hopefully, that will help narrow down the problems.

[tubetroubles]

I was out of town all weekend and didn't get a chance to chase around the voltages, but I did figure out the source of my issues. The location of the LED was what was killing me. Before it was in parallel with the anode resistors. When the switch was engaged, it opened a low resistance path to ground via the LED. I have since moved it to the beginning of the power rail so it's in parallel with the 1K resistor and electrolytic cap.

Not sure exactly why this makes such a difference, but the pedal now works and the valvecaster acts as a massive treble boost (well...more of a bass cut). I'm gonna put some higher capacitance caps in the signal path to open it up a little, but other than that I think I'm good.

[CopyGuy]

Another thing with the treble boost may be the Gain pot bypass cap.  It can cause high pitched gain regardless of the pot, so maybe a small resistance in series with the cap could help.  I just used a low value bypass and clipped my treble in the tone circuit.

[tubetroubles]

Thanks for your reply. I actually removed the cathode following gain pot altogether as it had little impact on the gain with the SHO Boost in front of it. The SHO trim pot on the other hand has a huge impact on gain and can go from a high headroom clean boost to a great crunchy overdrive, depending on the setting. Once I get the sweet spot I want with overdrive I'm just going to set it and forget it, which is why I went with a trim pot inside the enclosure.

With the gain pot removed, I installed a passive master tone circuit with a 0.022uF cap. Sounds great and is really versatile. The range is pretty impressive and with the valvecaster portion of the circuit off, it can get unbelievably dark with just the SHO boost on...

Anyway, I've attached an updated schematic for those who are interested. The problem I was running into earlier was due to current inrush when I turned the LED on. By decoupling the anode resistors and LED from the SHO Boost drain resistor, I was able to fix it. I also added grid resistors at both the input jack and 12AU7 first stage grid which massively cut down on EMI/RFI noise.

[Jazznoise]

The 1n4148 shunt diodes at the front are from an old version of the Super Hard On that proved unreliable for preventing damage to the BS170. A 9v Zener from Gate to Drain does the job much better.

[amptramp]

Can you actually have a 9-volt battery supplying the heaters in this design?  The heaters are nominally 150 mA (although a bit less at 9 volts) and the battery is good for maybe 50 mA.

Zener diodes can be used for gate protection if the capacitance is low enough, but this is not always the case.  I have seen audio L-C filters where the C was a zener diode and the circuit was tuned by varying the reverse voltage (below the regulation voltage).  Some diodes have a capacitance of 0.015 µF near zero volts and much less at higher voltages, making them usable as audio varicaps.  There are high frequency zeners which have a standard diode in series and another diode antiparallel to handle reverse current.  You could do this with a pair of 1N4148's and get 8 pF.

[Jazznoise]

Mmm, I'm not entirely au fait with mostfet switching speeds but I am aware ZVex switched to the Zener diode after there was alot of complaints of the Super Hard On failing when a jack was plugged in while it was on. As far as I'm aware this seems to have remedied the issue - or at least made breakages a minority issue,