Rocktron Silver Dragon- tried mods and Q.

[Guitarist]

I have a question to lower the noise floor. The heaters are running 12V AC which is creating hum and hiss. Would floating +9V DC into pin9 be a bad idea—seems so? Otherwise I might rectify ala boogie, but then should I ground (which/where?) those 33Ω resistors (post rectifier). This is a TOTALLY KILLER sounding pedal btw. Also the Intensity control oscillates past 4 for some reason.

Also C17 is 100pF and is always on (even with pedal "true" bypass) and bleeding these highs to ground, which maybe one can make smaller still. I have a buffered Maxon prior so not a big deal presently.

I just changed the tubes' plate resistors to metal-film 100K and it definitely sounds better, maybe quieter too. If you had a version with 47K that's perfect for a toneful 12at7 or 12ay7. This version had generic 150K carbon comps which made it lackluster.
I also paralleled the first one, 1R7, with a 220pF to hush the highs down a tad.

'also modded the input a bit (actually to the other side) to move the jack-in away from adjacent the rectifier diodes as they're seem to bleed hash at the input even though that jack is shielded, maybe overkill but it's done. Diodes create loads of noise and some was leaking through into that jack. Moved the original shielded cable (input end), lifted C13, cut trace to R16 but moving wire to that resistor, and added another cap right on the input jack (now next to the output jack). [If you were to try this and look, this might make sense;]

Much better now and has me forgetting the necessity of a noise gate.

[Guitarist]

The old Tung Sol 12ax7 just fizzled and browned out so I realized that was causing some oscillatation and noise—now fixed with a new tube. It's also quieter now. But what would be really nice is if one could triple or boost the plate voltage from that +9V. It's sounding a bit too soft especially compared to the English Muff'n next to it I'm out of those little EH toroids or I'd add one.
Perhaps if I run both effects from the same AC transformer I could run a wire to the plates from the Muff'n B+ (also adding a 10K and 10uF cap). Sorry just watching the Eddie VH interview at the Smithsonian here...

[Guitarist]

Bumper.

I'm close to trying tying +V after 1D4 to pin 9 of the tubes to float in some DC. Please no, am I going to blow up the Dragon?!

It's too noisy otherwise but may be quieter now that I'm wiring in a toroid-based and isolated PS from Digitech—ten outputs! Thanks...

[mozwell]

Both of these will generate a HV supply (full 250V) from a 12V or 9V input
You can also adjust the resistor dividers to get a lower output voltage (down to 48V if you want)
You may need to adjust some of the values of the resistors around the valve, and make sure the voltage ratings on any capacitors in the valve circuit are suitable for the voltage you select.

The last sch is a pedal i built (freestompboxes.org) the valve runs on approx +/-15V and the pedal sounds great.
It may be easy to modify your pedal something like it, to get more voltage on the valve
Currently the silver dragon has 16V or so on +V1, but the bottom of the valve goes to 0V through 1R22
If you change 1R22 (etc) to go to -V1, you will get 30VDC across the valve. You may need to add some level shifting capacitors on the input & output of the valve stage

PM me if you want vero layouts for the unit i built

[Guitarist]

Thank you Morwell. Those are nifty circuits but a bit beyond me as I'd have to debug my own work The pedal sounds better with the 12at7 and lower plate resistors now. The main issue is noise sense the heater-lift question.

I'm already seeing the lower plate's circuit referenced via -V1, is it not. Pin 8 cathode doesn't go to ground. So the voltage between plate and -V1 is double already isn't it (+V to -V). I haven't measured it, but your electro-logic is above mine

[J M Fahey]

Just as a side comment, going from tubes fed 9 or 15V to tubes fed 200V or thereabouts, "what tube designers intended" by the way, will absolutely change the pedal design, sound and features ... is that what you want?

Very low voltage fed stages can only offer little amplification, say 2x to 10x (with good luck) , some use them only as glorified clippers, while full voltage will turn each stage in one capable of, say, 25x gain, or 2500x if two cascaded .

Will the pedal stand such a huge increase?

It will also change clipping point big way, waveform, you name it.

The pedal was designed and tweaked with low effective gain in mind, who knows what will you come with if fed HV?

Just saying.

[mozwell]

I agree with Senor Fahey, it may make a lot of problems running the valves at high voltage, but you did ask if this could be done...

DC heaters may quieten the unit down some.
This should be easy to do as a "temporary" measure & if it does what you want, make it permanent.

Take the 12VAC, add a 1n4004 diode and a 4700uF 35V capacitor, this will give you approx. 16VDC
You will need to drop this to 12.6VDC for the heater, you can use a 27 ohm to do this. You already have 1R23 in the circuit, so you need to add a 18 ohm 1W
Don't forget to cut the connection from valve pin 4 to 12VAC

[Guitarist]

I was just preparing to add your parts for DC. I've not encountered this AC "ground" arrangement prior (with AC signal yes). I've check the board and that AC-in ground Tube Ground-2 does go to the filter cap -'ve IC12. Could it be that Rocktron had a two-prong, polarized AC adapter originally? I'm using a toroid. Also AC-2 connects with thick black wire to AC-5. Just checking.

Thank you so much for the awesome schematic by the way.
I'll upload a sound sample once this is done.

[mozwell]

Had another quick look, attached is the slightly revised schematic as I realised I had a small error

AC in gnd connects via a wire from TUBE_GROUND_2 to TUBE_GROUND_1
This is just 0V for the board, it is not GROUND or EARTH that you would find on a mains supply plug.
The added diode connects to the 12VAC (or AC_5)
The output of the added DC supply goes to the end of 1R23

So remove the wire between AC_5 and AC_2, the AC5 end goes to the added diode anode, the AC_2 end goes to the added 18 ohm resistor.

Apologies for the error in my original schematic.

When you have made the changes to the board, unplug the valve & power the unit up.
Measure from pin 5 of the valve socket to pin 4, you should have 16VDC or thereabouts.
Measure between pin of the valve socket to TUBE_GROUND_2, you should have 0VDC

If all is good, power off, install the valve, power back on & recheck the voltage between pin 5 of the valve & pin 4, if you have approx. 12.6VDC you are good to go.
Interested to see if this fixes your noise problem.

[Guitarist]

Well, I completed the DC heater mod. I had to add a second 18Ω resistor to bring it down to 11.6V, better lower than higher. I just did a quick test into a mixer and the mod was a success! It's about the same noise floor as the newer Amptweaker which has a gate, so that's great. I'll note more, without 'phones test, when trying it through an amp...
thank you so much Mr. Mozwell, good work man! This was the easiest DC conversion I've tried....more to come.

[Guitarist]

Here's a quick, direct recording into Garage Band with a Fender Combo sim.

In alphabetical order: EH English Muffin' (Marshall Pre), Amptweaker TightMetal Pro, Rocktron Silver Dragon. The last seems to have a microphonic tube now but as they're so hard to swap out, it stays for now. Running the filaments on DC has certainly quieted it wonderfully. This is very unfair to the Silver Dragon but as it was $60 used and the Amptweaker $300, well it will quiet down with a new tube. Only the latest Amptweaker hasn't been modded.

I have a BBE Maxcom in the rack which has noise kill but it wasn't used here and is kinda set to work with the clean tones too. If there was a quick and easy diode trick for noise suppression, that would help. Still playing live you don't really notice.

[Guitarist]

Well, I just switched in a 12ay7 tube (pulling a 12at7) and there is still yet some low end rumble happening in the SS side. It's likely below 80Hz. Maybe 1C21, a 1uF or C31 47uF can be reduced?
Well, I just pulled C30 a 47uF unpolarized cap, hmm unpolarized . I have mainly smaller poly (.047 uF at 200V) and electrolytics. Can I use a 1uF polarized cap in here?

Edit: I just swapped in a .18 in C30 and it getting honky, like a wah part-way down. And the rumble is still there but I'm not sure where it's coming from. When I turn up the diode clipping dial, labeled Intensity, it comes in. Thanks.

[mozwell]

I would leave IC21 1uF as it is, its just a coupling cap
C30 & C31 could be reduced, they are coupling caps, try 1uF in both places

[Guitarist]

I just order some small 1uF poly caps but I don't mind using some polarized Elna Silmic II caps here. I just would like to know if I have the —'ve towards the following OPamp if that's OK? I guess there isn't +'ve on both sides is there. Sorry if that's a dumb question, I don't know SS well.

Other then that this pedal is amazing for what it does, molten tones aplomb.

[mozwell]

non polarised would be better, but if you use the polarised caps, the + end of C30 would go towards IC2-A pin 1 & the + end of C31 would go towards the 100 ohm resistor R37, but as the signal on the op amps is centred around 0V, it doesn't really matter which way around the polarised ones go.
Still best to use non polarised ones if you have them

Not sure if this will fix your "rumble"

While you are at it, add a 22uF 16V (or larger uF) in parallel with C5 & C10, the power supply caps on IC2
The + end of the cap in parallel with C5 goes to the +9V rail
The + end of the cap in parallel with C10 goes to the 0V rail
You could also add some small decoupling resistor on the +/-9V rails, between the output of the two 9V regulators & the 220u capacitors C11 & C14, 10 ohm 1/4W will do
The 78xx & 79xx series regulators can have noise on their outputs, the variable regs LM317 (etc) are much better in this regard.
The "rumble" you hear might just be some ripple on the power supply to IC2 getting into the output.

One other thing you could try, is between IC2B pin 5 & 0V, cut the track & insert a 470ohm 1/4W resistor

Quick question, is the valve side clean with no "rumble"